92230 v2 ANNEXES Kazakhstan Agricultural Insurance Feasibility Study October 2011 The World Bank THE WORLD BANK Table of Contents Executive Summary ......................................................................................................... iii Context and Scope of the Study ..................................................................................... iii Crop Risk Assessment .................................................................................................... v Review of Compulsory Crop Insurance Program .......................................................... vi Strategy and Options for Strengthening the Current Crop Insurance Scheme ............... x Opportunities for New Crop Insurance Products ....................................................... xxiii Tailoring Crop Insurance to the Needs of Small Farmers in South Kazakhstan ....... xxix Chapter 1: Introduction and Objectives of the Study .......................................... 32 Importance of Agriculture in Kazakhstan ..................................................................... 32 Agricultural Crop Production in Kazakhstan ................................................................ 33 Government Policy for Agriculture .............................................................................. 35 Exposure of Agriculture to Natural and Climatic Disasters ......................................... 35 Government objectives for crop insurance ................................................................... 36 Objectives and scope of the study................................................................................. 37 Report Outline............................................................................................................... 39 Chapter 2: Crop and Weather Risk Assessment ......................................................... 40 Objectives and Scope of Agricultural Crop and Weather Risk Assessment ................ 40 Data Availability for Crop and Weather Risk Assessment ........................................... 40 Climate and Agro-ecological Regions .......................................................................... 43 Overview of Spring Wheat Crop Production in Kazakhstan ........................................ 45 Key Climatic Perils and Impact on Crop Production and Yields ................................. 51 Assessment of crop production risk exposures ............................................................. 55 Chapter 3: Review of Kazakhstan Crop Insurance Program .................................... 60 Policy and Regulatory Framework for Crop Insurance ................................................ 60 Compulsory Crop Insurance Policy Terms and Conditions ......................................... 61 Government Financial Support to Crop Insurance in Kazakhstan................................ 68 Performance Assessment: Technical Results, Liabilities, Reinsurance........................ 69 Assessment of the Technical, Operational and Institutional Features of the Compulsory Crop Insurance Program ............................................................................................... 82 Evaluation of Crop Insurance Effectiveness for Key Stakeholders .............................. 87 Chapter 4: Strategy and Options for Strengthening the Current Crop Insurance Program ........................................................................................................................... 90 Phase 1: Returning the Obligatory Crop Insurance Scheme to Profitability and Financial Stability ......................................................................................................... 91 Phase 2: Transition towards a Market-based Crop Insurance System ........................ 108 Phase 3- Transformation into a fully Commercial Crop Insurance Scheme for Kazakhstan .................................................................................................................. 118 Chapter 5: Opportunities for New Crop Insurance Products .................................. 134 Named Peril Crop Insurance ....................................................................................... 134 Area-Yield Index Crop Insurance ............................................................................... 140 Crop Weather Index Insurance ................................................................................... 154 Chapter 6: Tailoring Crop Insurance to the Needs of Lower Income Smaller Farmers .......................................................................................................................... 174 Identification of Appropriate Crop Insurance Products .............................................. 174 Tailoring Crop Insurance for Different Client Levels ................................................ 177 Organisational and Operational Systems for Small Farmer Crop Insurance .............. 183 Identification of Operational Linkages to Bundle Programs ...................................... 190 Annexes .......................................................................................................................... 194 Annex 1: Spring Wheat Production Systems in Kazakhstan and Crop Risk Assessment ................................................................................................................. 195 Annex 2: Mandatory Crop Insurance Program Statutory Features ................... 233 Annex 3: Kazakhstan: Mandatory Crop Insurance Results ................................ 255 Annex 4: Individual Grower MPCI Opportunities for Kazakhstan .................... 275 Annex 5: International Experience with Crop Insurance Pools ........................... 289 Annex 6: Named Peril Crop Insurance Opportunities in Kazakhstan ................ 297 Annex 7: Area-Yield Index Insurance Opportunities in Kazakhstan.................. 302 Annex 8: Weather Index Insurance Opportunities in Kazakhstan..................... 325 Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the crop production and yield data and crop insurance financial data included in this work and on which basis some analyses have been made and from which some conclusions have been drawn and recommendations made. The World Bank cannot be held responsible for any insurance result or other financial outcome which might arise from any decisions or actions taken by Insurers or any other party as a consequence of the contents of this report. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. ACKNOWLEDGEMENTS The report was prepared by the World Bank in partnership with the Second Agriculture Post Privatization Assistance Project (APPAPII) and was authored by a team led by Sandra Broka (Senior Rural Finance Specialist, ECSS1, World Bank) and Meiram Akchukakov (Program Coordinator APPAP II). The team was comprised of Ramiro Iturrioz (Senior Agricultural Insurance Specialist, Insurance for the Poor Program, GCMNB, World Bank – Technical Leader); Talimjan Urasov (Operations Office, ECSS1, World Bank); Charles Stutley (Consultant, ARMT, ARD, World Bank); and Andrea Stoppa (Consultant, ARMT, ARD, World Bank); Bakhyt Sattybaeva (Consultant, APPAPII); Lunara Umralinova (Consultant, APPAP II); Marina Gabdulinova (Consultant, APPAPII); Aigerim Malik (Consultant, APPAPII), and ARKA Consulting the local consultant firm appointed by APPAPII for this project. The team acknowledges the contributions of all stakeholders, including the Ministry of Agriculture (MoA), and in particular, the Department of Investment Policy and External Relations and the Department of Strategic Planning in Agribusiness and Innovation Policy; JSC Fund for Financial Support for Agriculture, Hydro Meteorological Service of Kazakhstan (Kazhydromet); National Statistics Agency, Agency for Financial Market and Financial Institutions Regulation and Control RK, National Space Agency of the Republic of Kazakhstan, JSC, Union of Farmers of Kazakhstan, Scientific Research Institute of Economy of Agro- Industrial Complex and Development of Rural Territories, A.Barayev Kazakh Scientific & Research Institute of Grain Farming; Akimat of Enbekshilder Rayon; Akimat of Bulandinskky Rayon; Akimat of Altynsarin Rayon, KazakhInstrakh Halyk Group Insurance Company; JSC Insurance Company ―Pana Insurance‖; JSC Grain Insurance Company; ―Agro-Insurance‖ Mutual Insurance Society; ―SFK-Insurance‖ Mutual Insurance Society ―Dostyk 05‖ Co Ltd. The authors are grateful to the peer reviewers John Nash (Lead Economist, LCSSD, World Bank), Olivier Mahul (Program Coordinator; FCMNB, World Bank); Gary Reusche (Operations Officer, IFC); and Vikas Chaudhary (Agriculture Specialist, ARD, World Bank). The team gratefully acknowledges funding support from the Commodity Risk Management Multi-donor Trust Fund. -i- ABBREVIATIONS APPAP II Second Agricultural Post Privatization Assistance Project ARKS National Agency of Statistics Republic of Kazakhstan AYII Area-yield index insurance CRAM Crop Risk Assessment Model Centner Grain production unit equivalent to 100 Kilograms CF Commercial Farmer (usually with between 100 Ha and 1000 Ha). Also termed Individual Farm or Peasant Farm. CTL Constructive Total Loss clause (used to define a total crop loss) CWII Crop Weather Index Insurance CoV Coefficient of Variation EKO East Kazakhstan Oblast ENSO El Niño Southern Oscillation FAO Food and Agriculture Organization of the United Nations FFSA Fund for Financial Support for Agriculture FMCIA Farmer‘s Mutual Crop Insurance Association GDP Gross Domestic Product GRK Government of the Republic of Kazakhstan KHM Hydro Meteorological Service of Kazakhstan (alternatively known as Kazhydromet) KZT Kazakhstan Tenge (monetary currency of Kazakhstan) IU Insured Unit LIC Loss of Investment Costs (crop insurance policy) MFI Microfinance Institutions MoA Ministry of Agriculture MoF Ministry of Finance MPCI Multiple Peril Crop Insurance MT Metric Tonne equivalent to 1,000 Kilograms NSA National Space Agency NGO Non Governmental Organization NKO North Kazakhstan Oblast Oblast Administrative region of Kazakhstan PE Production Enterprise (Large grain farm usually with greater than 1,000 Ha). Also termed Agricultural Enterprise. PML Probable Maximum Loss PPP Private-Public Partnership Rayon Administrative area within an Oblast RK Republic of Kazakhstan SKO South Kazakhstan Oblast TSI Total Sum Insured TSU Technical Support Unit VAR Value at Risk WKO West Kazakhstan Oblast WII Weather Index Insurance - ii - Executive Summary Context and Scope of the Study 1. Agriculture is a very important socio-economic sector in Kazakhstan. Agriculture occupies a key position in the economy of Kazakhstan: approximately 7.3 million people (47.2 percent of the total population) currently live in rural areas and agriculture employs more than 22% of the labour force in the country. Agriculture contributes 5.92 percent of Kazakhstan‘s GDP. The country is one of the major global producers and exporters of grains (mainly wheat). Other principal agricultural products include meat, wool, cotton and milk. Farming areas occupy more than 220 million hectares (about 74% of the country's total area), of which cereal growing areas occupy about 13-14 million hectares. The major grain crop is spring wheat which is predominantly grown under extensive low-cost production systems in northern and central Kazakhstan. 2. In northern and central Kazakhstan spring wheat production and yields are highly influenced by climatic and biological factors. On account of the very uncertain climatic conditions, Kazakhstan has the highest year on year variation in national average wheat production and yields of any major wheat producing and exporting countries. Drought is the most pervasive peril affecting rain-fed crop production in northern Kazakhstan and severe droughts are experienced every two to five years. Spring wheat crops can also be damaged by hailstorms, early autumn frost and pests and diseases of wheat. This study estimates that on average about 14.71% of the total value of the national spring wheat crop is lost due to drought and other perils every year, valued at KZT 66.5 billion (US$ 443 million)1 and in extreme drought years such as 1998 physical losses were in the order of 7 million metric tons losses equivalent to as high as 42% of the total expected value of wheat production. Very severe drought losses have been experienced in spring wheat most recently in 2008 and again in 2010. 3. The Government of the Republic of Kazakhstan (GRK) introduced a national compulsory crop insurance scheme in 2005 in order to provide grain producers and other farmers with a minimum level of protection against catastrophe climatic events. The scheme was enacted through the Compulsory Crop Insurance Law dated March 10, 2004 and became operational in 2005. The law established the terms and conditions for the implementation of a compulsory Loss of Investment Cost (LIC) insurance policy providing comprehensive protection against the loss of production costs invested in growing a range of strategic grain, oilseed and other field crops. The Kazakhstan crop insurance scheme is based on a public private partnership (PPP) implemented by the private commercial and mutual insurance companies and supported by government financial subsidies on claims. The implementing agencies are the Ministry of Agriculture (MoA), through the Direction of Strategic Planning, the Fund for Financial Support for Agriculture (FFSA), the private commercial insurance companies and the Farmers‘ Mutual Crop Insurance Associations (FMCIA) and finally the local authorities in each Oblast and Rayon. Under the PPP, GRK provides financial contributions to the crop insurance scheme through an Indemnity Fund, termed the ―Fund for Financial Support for Agriculture‖ (FFSA) and indemnifies 50% of the costs of all crop insurance claims. 1 In this report, a current 2011 exchange rate of Kazakhstan Tenge, KZT 150 = US$ 1.00 has been used, unless otherwise stated. - iii - 4. The GRK’s decision to introduce a national compulsory crop insurance scheme for the most important crops grown in the country should be viewed in the context of government’s policy toward agriculture. The GRK recognizes the importance of the agricultural sector in diversifying economic growth, reducing rural poverty and in contributing towards improved food security with a strong emphasis on maintaining rural welfare for the country‘s predominantly small farmers, as set out in the strategic three year plan for agriculture 2009-11. Government‘s introduction of compulsory insurance was mainly designed as a measure to ensure that small farmers and wage-paid agricultural labourers would be guaranteed a minimum level of financial protection in the event of catastrophe crop production losses. Since the late 1990s the government has significantly increased its financial support to the crop sector with direct input subsidies, subsidised credit and output-price support measures. Under its goal to meet World Trade Organisation, WTO, accession terms, Kazakhstan will need to reduce its direct subsidies to agriculture; however, WTO legislation does not prohibit government from increasing it financial subsidies to agricultural crop and livestock insurance. 5. Overall, the crop insurance scheme has not performed well. Over the past 6 years the obligatory crop insurance scheme has achieved very high levels of uptake, but the scheme has also encountered major operational and financial problems. Crop insurance penetration is very high in Kazakhstan averaging 74 percent of the eligible cropped area for the period 2005-2010 – the high uptake is a function of the compulsory nature of the scheme. Notwithstanding the high level of penetration, insurance results have been very poor over the past three years with the result that many commercial insurers have ceased to support the scheme. From an operational viewpoint the scheme is very costly to administer and it is under-rated in several regions: as the terms and conditions of cover are determined by government and fixed by law the insurance companies have little say in risk acceptance and underwriting decisions and today only three companies continue to support the scheme. Finally private commercial insurers and farmer‘s mutual insurers are very exposed to catastrophe losses as the scheme is not currently reinsured against excess losses. 6. During 2011 a comprehensive study aimed at the review, refinement and improvement of the current compulsory crop insurance scheme in Kazakhstan has been performed by the World Bank, under the risk management component of the Agricultural Post-privatization Assistance project (APPAP II). The study has reviewed the Kazakhstan compulsory crop insurance scheme and has made recommendations for its strengthening and transition over time to a more market-based agricultural insurance system. The study has also assessed the potential to introduce new crop insurance products and programs to complement the existing Loss of Investment Cost (LIC) policy. The study has included the following specific components: a. Review of the Compulsory Crop Insurance Scheme in Kazakhstan. A detailed diagnostic review has been carried out of the technical basis of the LIC policy, the institutional and organizational features of the public private partnership scheme and its operating systems and procedures and finally the financial performance of the scheme. This review identified a series of key issues and challenges for scheme management to address. b. Phased strategy to transform this scheme into a financially sustainable market oriented system. Drawing on local expertise and international experience and best practice a phased strategy has been identified over the next three to five years to strengthen the existing compulsory crop insurance scheme, to return it to profitability and to transform this into a sustainable market-based system which is supported both by the public sector and international reinsurers. - iv - c. Agricultural risk assessment. A formal crop risk assessment has been performed for spring wheat, the most important export crop. This risk assessment was intended to assist policy makers, planners and crop insurers in the planning, design and rating of new crop insurance products. Owing to the size of the country, it was agreed that the scope of the risk assessment would be limited to spring wheat grown in the 8 most important Oblasts of Kazakhstan namely Kostanay, Akmola, North Kazakhstan (NKO), East Kazakhstan (EKO), West Kazakhstan (WKO), Pavlodar, Karaganda, and Aktobe. d. Crop insurance product development. An analysis has been conducted for spring wheat of four new crop insurance product types in Kazakhstan which in future may either complement or replace the current LIC only product including, two traditional indemnity- based covers, i) individual grower multiple-peril crop insurance (MPCI), and ii) crop hail cover and two new index-based products, iii) area yield index insurance (AYII) and iv) crop weather index insurance WII. For the purpose of the new product feasibility studies, the research was focused on spring wheat grown in Altynsarinski and Auliyekolski Rayons in Kostanay Oblast; Aktogaiski and Zhelezinski Rayons in Pavlodar Oblast; and Bulandinski and Enbekshilderski Rayons in Akmola Oblast. e. Challenges for developing small farmer crop insurance in Kazakhstan. The final objective of this study has been to identify ways to tailor the provision of crop insurance to the needs of resource poor farmers located in southern Kazakhstan. Tole-bi Rayon in SKO Oblast was selected for more in-depth study. Crop Risk Assessment 7. A detailed risk assessment has been conducted of weather risks and their impact on spring wheat crop production and yields in north Kazakhstan Region. The risk assessment comprised the following components: i) a review of the availability and quality of time-series crop production and weather data availability in Kazakhstan for spring wheat risk assessment and insurance design and rating purposes; ii) a review of climate and the agro-ecological regions and spring wheat crop production systems in the selected Oblasts of Kazakhstan, iii) a detailed statistical analysis of spring wheat production and yields and the climatic constraints to production including an analysis of rainfall data and the relationship to national and rayon-level spring wheat crop production and yields and finally iv) application of a Crop Risk Assessment Model, CRAM, which uses time-series rayon-level production and yield data to estimate values of risk, expected losses and expected claims costs for spring wheat in the eight selected Oblasts in Kazakhstan. This latter analysis is very relevant to crop insurers‘ understanding of risk accumulation and maximum expected losses in spring wheat. (Full results of the risk assessment are presented in Section 2). 8. The results of the crop risk assessment of rayon level crop production and yields for spring wheat in the northern and central regions of Kazakhstan shows that this crop is heavily exposed to losses caused by droughts. This is evidenced by the average loss cost estimated by the CRAM for the 17-year period, 1994 up to 2010 of KZT 66.5 billion (14.71 percent of the total value at risk of spring wheat production) and a calculated 1 in 100 year PML of KZT 246.8 billion (54 percent of the national spring wheat gross value of production). The highest average annual expected losses in spring wheat apply to Aktobe Oblast (average expected loss of 22% of the spring wheat crop value) and WKO (average losses of 40% of the spring wheat crop value) located in western Kazakhstan. Except in a few rayons situated in the north of Aktobe and in the south west of WKO, in most of the rayons located in these two Oblasts, spring wheat average yields are both low and highly variable on account of the much lower and erratic rainfall -v- experienced in western regions of Kazakhstan. Conversely, spring wheat production is much less risky in the northern Oblasts of NKO, Kostanay and Akmola and which receive higher and more stable precipitation. 9. The need for accurate and independent measurement and recording of crop production and yield data and corresponding weather data at local up to national levels is critical to the design and rating and implementation of any crop insurance scheme. On the basis of this study it is apparent that the GRK has very efficient Meteorological, Agricultural and Statistics Services. The production and weather data availability for the design and rating of crop insurance products and programs is, in general, very good. Kazakhstan has a modern and efficient national meteorological service, Hydro Meteorological Service of Kazakhstan (alternatively known as Kazhydromet - KHM), which has provided time series rainfall data for a sample of weather stations in the selected rayons in order to permit a detailed assessment of rainfall and yield relationships to be made and to provide the basis for the design and rating of prototype WII products. There is, however, a key constraint at present namely that the density of weather stations in such a vast territory such as Kazakhstan is inadequate to implement a national commercial micro-level CWII scheme in the near future. Kazakhstan also has good spring wheat crop area, production and yield statistical records at Rayon, Oblast and National levels and which are collected by the National Agency of Statistics Republic of Kazakhstan (ARKS). This data has enabled prototype individual grower MPCI and Area Yield Index crop insurance products to be designed and rated for Kazakhstan. (Detailed in Sections 4 and 5). Review of Compulsory Crop Insurance Program 10. Section 3 of this report provides a detailed review of the technical, operational, institutional and financial features of and challenges faced by the compulsory crop insurance scheme. Technical Challenges 11. The Kazakhstan compulsory crop insurance policy is a Loss of Yield Policy which indemnifies the Insured when the value of the harvested production falls short of the costs invested in growing the crop due to the action of insured perils: it is also termed a “Loss of Investment Costs” (LIC) Policy. There are several advantages of the LIC Policy including (a) that it provides comprehensive multiple-peril crop insurance, MPCI, protection to the farmer again the loss of his production costs invested in growing the crop and (b) that it can be used in situations where there is inadequate or no information on individual farmer historical crop yields. There are, however several potential drawbacks of the Loss of Investment Cost policy including the need for in-field yield-based loss assessment where partial losses are involved and often the difficulty of establishing objectively the salvageable amount of the crop and its sale value and whether this salvage value exceeds the insured investment costs leading to a claim. 12. The coverage levels provided by the LIC crop insurance scheme in Kazakhstan are extremely low and do not provide adequate levels of financial protection to the farmer in the event of loss. Although the compulsory scheme offers a series of production costs based sums insured varying from low costs to high costs, in practice it appears that practically all farmers in Kazakhstan elect the cheapest or lowest sum insured coverage option of about KZT 3,500/Ha (slightly less than US$ 25/Ha nationally for spring wheat) because they will pay the least amount of premium for this option. However, this means that on average farmers are only insuring between 20% to 30% of their total production costs. The very low crop insurance cover levels are therefore, often inadequate to put farmers back into production in the event of major crop losses. - vi - 13. The LIC crop insurance premium rating methodology in Kazakhstan should be revised. Currently, crop insurance premium rates are calculated for each crop at an Oblast level when they should to be calculated at Rayon level or even higher levels of disaggregation in order to take into account differences in risk either at the local Rayon level or at the individual farmer level. Crop insurance premium rates are fixed by Law: these rates were last adjusted in 2008 and now need to be adjusted on an actuarial basis. Crop insurance premium rates are set for each crop type and group of Oblasts according to minimum and maximum rates and the Insurers are not permitted to charge higher rates even if these are actuarially required. 14. The basis of indemnity on the LIC Policy requires strengthening. Crop output valuation prices are determined at the time of harvest as opposed to being pre-agreed and specified in the policy wording. This means that neither the insurers nor the government can calculate their liability in the event of claims: when output prices for what are low the policy is much more exposed to losses than when wheat prices are high. This issue should be addressed by introducing fixed agreed harvest valuation prices at the time of signing the policy agreement. Operational Challenges 15. The obligatory nature of the Scheme prevents underwriters from exercising proper risk selection and control over the underwriting of their crop insurance portfolios ..Insurance companies are obliged to accept all crop insurance risks (individual farmers) even where these are poor risks (i.e. they do not observe the correct crop production and husbandry practices for growing the insured crop) and / or where they are located in such high risk areas that they would normally be considered uninsurable by commercial crop insurers. 16. Policy sales are currently permitted to continue up to the time of sowing when farmers are in a good position to predict whether the growing season will be poor and this exposes the program to moral hazard. Where pre-existing drought conditions are developing farmers may modify their behavior by (a) they elect to buy the maximum ―Normative Costs‖ Sum Insured level in the expectation of receiving a claims payment and (b) they may incur less than their normal levels of expenditure on crop husbandry and inputs because they know they are likely to lose their crops and in which case they can expect to claim on their insurance policies (this is termed moral hazard). 17. Private Insurance Companies do not have their own networks of locally based qualified agronomists to conduct pre-inspections on the insured farms at the time of sowing in order to confirm whether the farmer has complied with the correct sowing practices, seed rates etc. As such cover is open to moral hazard. The costs of establishing such a network and in inspecting each and every farm would be prohibitively expensive to the insurers, and under the current rating system they are not able to increase rates to cover their A&O expenses. 18. Loss adjustment requires the participation of several parties, it is expensive, and sometimes lacks transparency. In Kazakhstan up to 5 persons are involved in adjusting the crop losses at the local level and this is a very time consuming and costly exercise. There is a need to rationalize the loss assessment procedure and to reduce the costs of this exercise. Institutional Challenges 19. The Kazakhstan compulsory Crop Insurance scheme represents a public-private partnership (PPP) which is highly regulated by government and underwritten by the private commercial (and mutual) insurance sectors, with financial claims subsidy support provided by government through the FFSA. As such the insurance companies are unable to accept or reject - vii - individual crop risks (farmers), which is a fundamental principle of crop underwriting. In most countries where governments support agricultural insurance PPP‘s their main function is to provide legal and regulatory support and financial subsidies. 20. Crop insurance in Kazakhstan is written by Commercial Insurers and Farmers’ mutual insurance associations. The level of participation of the private insurance companies owing to the poor underwriting results of the scheme is very low (only three commercial insurers continue to operate in the scheme) and there is a danger that unless the scheme can be returned to profitability the private insurance sector companies may cease to provide their support in future. Conversely, the participation of Farmer‘s Mutual Crop Insurance Associations is gaining importance in the scheme. As of 2011, there are more than 38 farmers‘ mutual insurance associations offering crop insurance in Kazakhstan. 21. Private insurance companies and farmer’ mutual associations providing crop insurance are not equally regulated. While private insurance companies are subject to the regulation of the Agency for Financial Market and Financial Regulation and Control, the farmers‘ mutual crop insurance associations are regulated separately by the law of the Republic of Kazakhstan on Mutual Insurance. Non-life Private insurance companies are required to have a minimum capital of KZT 1.2 billion (US$ 8.0 million) in order to operate. On top of the minimum capital requirements, private insurance companies are frequently monitored on their solvency, net retentions, and the implementation of risk management procedures. Conversely, the farmers‘ mutual associations offering crop insurance are not regulated and supervised by the Agency for Financial Market and Financial Regulation and Control. Therefore, farmers‘ mutu al associations are not subject to any minimum capital requirements, or controls over net retentions and solvency requirements. In 2010 it is understood that several mutuals had not collected enough premiums to pay the full amount of claims incurred during the crop season 2010. There is a need to ensure that the private companies and mutuals are regulated equally under the crop insurance scheme. Government Support 22. GRK provides major financial support to the Compulsory Crop Insurance Scheme. This support is provided in two ways by i) compensating the Insurers for 50% of all the claims incurred and ii) funding of the administration and operating expenses of the Fund for Financial Support for Agriculture, FFSA. The 50% claims compensation fund is administered by the FFSA which is responsible for monitoring and managing the financial transactions on this insurance scheme on behalf of government and for approving the claims reimbursements to individual insurance companies. Over the past 6 years, GRK has provided KZT 4.7 billion to the FFSA of which 93% has been allocated to settling the 50% of claims and 9% to the A&O expenses of the FFSA. During this period the FFA has reimbursed the Insurance Companies a total of KZT 3.84 billion equivalent to 46.7% of total paid claims or an average of KZT 641 million per year (US$4.3 million per year). Financial Challenges 23. The compulsory crop insurance program in Kazakhstan has experienced poor overall underwriting results over the period 2005 to 2010. The long-term average loss ratio for the 6- year period 2005-2010 is 140% and in four of the six years the scheme has operated at a financial loss in terms of the gross claims to premium or loss ratio exceeding 100%. The average net loss ratio to the insurance companies after the reimbursement of 50 percent insurance losses (claims) from the government is 75%. Over the past three years (2008 to 2010), the results have - viii - deteriorated badly with an average loss ratio of 182% (99% after operation of the government claims subsidies) and with a worst loss year in 2010 with 261% loss ratio. 24. Scheme performance varies widely across different regions and the very poor underwriting results in Aktobe and WKO are making the scheme financially unviable. The pattern of claims varies widely by geographic region. The best performing Oblast is NKO which over the past 6 years has contributed 22% of total scheme liability, but only 3% of claims and has a long-term loss ratio of only 24%. At the other extreme, Aktobe and WKO in western Kazakhstan which have collectively accounted for only 4.8% of total scheme liability over the past 6 years have incurred 41% of all claims and respectively have 6-year long-term loss ratios of 381% and 507%. These two Oblasts are severely prejudicing the financial viability of the whole of the national crop insurance program and measures of controlling the claims costs in these two Oblasts must be introduced. 25. The private insurance companies and mutuals do not have any access to reinsurance on their 50% retained claims and they are extremely exposed to catastrophe drought losses. This is a major issue which will need to be addressed under any future reforms and strengthening of the Kazakhstan crop insurance scheme. Evaluation of Crop Insurance Effectiveness for Key Stakeholders 26. The Kazakhstan compulsory crop insurance scheme was launched with very well intentioned social and economic objectives, but it appears that it is failing to meet the requirements of its key stakeholders including farmers, crop insurers and government. The program was originally conceived as a mechanism to ensure that all farm workers and small peasant farmers would receive a minimum indemnity in the event of crop failure due to drought or other natural and climatic perils. The obligatory nature of the scheme was intended to be a short-term measure which would enable the insurance sector to develop a sound and stable crop insurance market based on a PPP between private and public sectors, while at the same time providing time to educate farmers in the benefits of crop insurance so that they would continue to purchase cover once the scheme was made voluntary. After 6 years the scheme has failed to develop a strong crop insurance market and to educate farmers. The fact that the program is very unpopular with many farmers also suggests it is failing to meet its social objectives. Finally government faces major uncertainties over its financial exposure to claims (Figure 1). - ix - Figure 1. Key Issues facing Main Stakeholders involved in the Compulsory Crop Insurance Scheme Source: Authors Strategy and Options for Strengthening the Current Crop Insurance Scheme 27. A phased approach is recommended for strengthening and improving the obligatory crop insurance scheme in Kazakhstan and for gradually converting this into a fully market- based commercial crop insurance system involving 3 distinct Phases: (i) strengthening the current obligatory scheme and achievement of financial stability (short-term, 1-3 years); (ii) transition towards a market-based crop insurance system (short-term, 1-3 years); and (iii) the transformation into a fully market-based Pool crop insurance system backed-up by international reinsurance under a suitable PPP for Kazakhstan ( medium-term, 3-5 years). Phase 1: Short-term Measures to Strengthen the Existing Compulsory Scheme Legal & Regulatory 28. In the short-term it is recommended that the Obligatory Crop Insurance Law should be amended to permit insurance companies greater flexibility in determining the premium rates and other terms and conditions of cover they provide under the LIC Policy . While it is assumed that crop insurance will continue to be obligatory by law for farmers in the short term, the law should be amended to permit insurers to set their own terms and conditions of cover and especially premium rates (Section 4.6 to 4.11). Technical Strengthening of the Design and Rating of the Loss of Investment Costs Policy 29. A series of practical measures have been identified to strengthen the design of the LIC Policy and which are listed in Box 1. Full details are provided in Sections 4.12 to 4.25, and key recommendations for strengthening are highlighted below. -x- 30. There is a need to consider increasing the sums insured. Currently the LIC policy does not provide adequate levels of protection to the majority of farmers. If the sum insured coverage levels were to be increased under this program, this would provide farmers with higher levels of protection, but would have to be accompanied by actuarial increases in the premium rates charged and the amount of premium farmers have to pay. With increased sums insured, the Insurance Companies would face a much higher financial liability in the event of severe drought losses and it would be essential for the insurers to have a comprehensive reinsurance protection program in place. For government the higher coverage levels would mean a correspondingly higher budgetary allocation would need to be made to cover the 50% claims reimbursements. There is a need to set the sums insured in each Rayon according to: i) the actual production costs of different types of farmer in each Rayon and ii) the risk exposure in each rayon. 31. There is a need to revise the crop premium rating methodology in order to i) introduce a system of Rayon-level premium rates and ii) to update the premium rates on an actuarial basis. The 6-year long-term gross loss ratio at end 2010 was 140% (equal to a 70% loss ratio net of the government 50% claims reimbursement), suggesting that on average the scheme has operated on a break-even basis. The premium rates were last revised by government in 2008 and since then, however, the scheme results have deteriorated very badly over the past three years during which time the average loss ratio has been 182% (loss ratio of 91% net of the government 50% claims subsidy). In some Oblasts including NKO and Kostanay the scheme has performed very well, but in others including Aktobe, WKO, and Zhambyl the claims performance has been very poor. There are also major differences in the pattern of losses among rayons in individual Oblasts and there is a need to revise the Oblast rating methodology by introducing a system of Rayon-level actuarially determined premium rates for each crop. Finally it is recommended to introduce a bonus-malus2 system into the crop rating methodology (See section 4.21) Box 1. Measures to Strengthen the Design of the Loss of Investment Costs Policy Item Detail Criteria for Acceptance of Even if government decide in the short-term to maintain Compulsory crop Risk / Compulsion of Cover insurance for all producers special consideration will need to be given to farmers located in Aktobe and WKO. Insured Perils Maintain current coverage to include loss or damage to crop production due to: ―Adverse weather events‖, as defined. Sales Cut-off date Introduce a policy sales cut-off date 1st April Cover Period From the time of crop emergence and full stand establishment (e.g. wheat 10 cm stage in crop) through to completion of the crop harvest. Insured Unit (IU) Strengthen the definition of the IU which is defined as the ―individual field‖. For farms of less than 250 hectares, consider re-defining the IU as ―the total area of all fields of the same crop grown in the same location or farm Sum Insured Government should amend the Obligatory Law to permit Insurance Companies to have the option to establish an agreed sum insured with each farmer according to their own circumstances and insurance requirements. Premium Rates  Government should amend the Obligatory Law to permit insurance companies to set their own premium rates for each crop and zone.  Actuarial rating should be introduced to reflect differences in risk exposures between Rayons in each Oblast and possibly differences in technology levels and risk exposures between farmers. 2 Under a bonus malus system, a farmer who does not submit a claim is rewarded at the renewal date of his policy by receiving a reduction in his premium rate (termed the ―bonus‖), while a farmer who submits high claims will have his rate accordingly increased at the renewal (termed the ―malus‖). - xi - Bonus Malus System It is recommended that underwriters introduce a Bonus-Malus system on the compulsory crop insurance scheme. The objectives of the bonus-malus are:  to introduce individual farmer experience rating  to reduce the submission of speculative claims notices by farmers. Basis of Indemnity and The Crop Sale Price which is used to value actual production in the event of Claims Settlement a Partial Loss should be pre-agreed based on an average historical farm-gate sales price for each crop in each Rayon and stated in the Policy Wording. Source: Authors Measures to Improve Scheme Profitability 32. The financial viability of the crop insurance scheme is being adversely impacted by the inclusion of Aktobe and WKO Oblasts. Over the past 6 years Aktobe and WKO have collectively accounted for only 4.8% of total scheme liability, but have contributed 41% of total claims and respectively have actual loss ratios of 381% and 507%. Over this period, if Aktobe and WKO had not been included in the scheme, this would have led to a reduction in total premium earnings of 13% only, but would have led to a huge saving in claims costs for insurers of 40% of actual claims (or a reduction in claims of KZT 3.3 billion – US$ 22.0 million) and a reduction in the 6-year long-term gross loss ratio from 140% to 96%. Following application of the government 50% reimbursement of claims the ―As If‖ loss ratio for the scheme would have been reduced to about 48%. (Table 1). These two Oblasts (in common with much of the northern and central regions of Kazakhstan) were traditionally livestock grazing regions which were converted to cereal (mainly spring wheat) production, during Soviet times. The underlying problem, however, in most of Aktobe and WKO Oblasts is that soils are mostly poor, average annual rainfall is very low and in a normal year is barely adequate for growing spring wheat in most of the rayons located in these two Oblasts. As such these are very marginal Oblasts for spring wheat crop production and no matter how a crop insurance policy is structured in these two Oblasts, it will always be very heavily exposed to loss and the technical and commercial premium rates that would have to be charged would be so high (an average commercial rate prior to application of government claims subsidies of between 37% Aktobe and 66% WKO) as to make the scheme commercially unviable in these two Oblasts. There are, however, a few areas in these Oblasts where spring wheat production is more stable and where crop insurance might possibly be considered in future, albeit at very high rates, including Taskalisky and Kaztalisky Rayons located in the south west of WKO and Karagaly and Martuk Rayons in northern Aktobe Oblast. Table 1. Comparison of Crop Insurance Actual Results with and without Aktobe and WKO Item Actual Portfolio Modified Portfolio Percent (With Aktobe/WKO) (Without Aktobe/WKO) Reduction No. of Policies 140,961 134,193 -5% Total Insured Area (Ha) 73,770,915 69,453,803 -6% Sum Insured ('000 TH) 242,631,438 231,048,598 -5% Premiums ('000 TH) 5,861,958 5,102,670 -13% Average Premium Rate % 2% 2% -9% Claim payments ('000 TH) 8,222,776 4,910,314 -40% Loss Ratio % 140% 96% -31% Loss cost % 3.39% 2.13% -37% Source: Authors analysis of FFSA crop insurance data 2005-2010 - xii - Options for WKO and Aktobe 33. The first option which is being examined by the insurance companies is to continue operating the LIC scheme in Aktobe and WKO on the understanding that government will agree to reimburse these insurers for a higher percentage of claims in these two Oblasts. Currently government reimburses insurers for 50% of the value of all paid claims. Under this option, GRK would agree to indemnify a higher percentage of claims incurred in Aktobe and WKO only: according to the insurance industry they have already requested government to consider raising its share to 70% or 75% of total claims in these 2 Oblasts in 2012. Over the past 6 years, government‘s 50% contribution to claims costs have amounted to about KZT 4.1 billion or an annual average of KZT 0.685 billion per year. If Government agreed to increase its share of claims in Aktobe and WKO, the ―As If‖ annual total cost to government would increase to: i) Option A 75% claims share in Aktobe and WKO: KZT 0.823 billion per year or an increase of 20% and ii) Option B 100% claims share in Aktobe and WKO: KZT 0.961 billion per year or an increase of 40% (this latter 100% claims subsidy option shows the maximum expected costs to government if it wishes to continue including Aktobe and WKO in the LIC scheme and assuming the insurance companies act purely as administrators in these 2 Oblasts, but do not accept any claims liability: the ―As if‖ 100% claims cost to government in these 2 Oblasts would be an average of about KZT 550 million or US$ 3.7 million per year). While this option may provide government a short-term solution to retaining the support and underwriting capacity of the insurance industry and in increasing the profitability of the program after the government‘s increased claims reimbursements, in the medium to long-term this does not tackle the fundamental issue that commercial spring wheat crop insurance is not financially viable in most of the Rayons located in Aktobe and WKO Oblasts. 34. The second option for government to consider would be to take Aktobe and WKO out of the crop insurance scheme altogether and to establish a separate and formal disaster relief scheme for producers of annual crops in these 2 Oblasts. The argument for taking Aktobe and WKO altogether out of the scheme centre on: i) these are very marginal rainfall areas for growing annual grain crops and ii) from a crop insurance viewpoint these Oblasts (with the possible exception of a few Rayons listed above) are effectively uninsurable with 6 year long-term loss costs of 22% and 40% respectively and catastrophe losses every other year, with loss ratios (claims to premium ratios) as high as 1000% in major drought years. It is therefore necessary for government to consider whether to continue to promote annual crops in these Oblasts, but to use a separate disaster relief fund to compensate farmers in severe drought years. The cost of this disaster relief compensation program for farmers in Aktobe and WKO, assuming the same 100% original claims costs and compensation levels as under the current LIC insurance scheme, would be exactly the same as the 100% claims compensation option outlined previously with an average cost to government for these 2 Oblasts of about KZT 550 million (US$ 3.7 million) per year and with a peak of about KZT 1.6 billion (US$ 10.7 million) in a very severe drought year such as 2010. Options for some form of macro-level catastrophe drought Weather Index Insurance cover are also examined in this report. Strengthening the Scheme Operating Systems and Procedures 35. In conjunction with the design changes identified to strengthen the LIC policy, there are a series of potential measures scheme administration should consider in order to improve underwriting and claims operating systems and procedures and to reduce the costs of these operations. Box 2 summarizes some of the key identified operational areas which require strengthening. In the specific case of loss assessment the current procedures whereby a committee of up to 5 individuals from five different organizations are involved in the in-field loss assessment must be rationalized and made more cost-effective and the insurance companies given - xiii - a more central role in loss assessment, whilst at the same time ensuring that impartiality and fairness are maintained for both the Insured and the Insurer. In the short-term Kazakhstan does not have independent firms of certified and approved loss adjusters which specialize in crop loss assessment, but in the medium terms options should be explored for creating such specialist entities. Box 2. Options for streamlining and reducing the costs of policy marketing, pre-inspections and loss assessment systems and procedures Requirement Detail Pre-Inspections need to be  A system of sample pre-inspections should be considered for the introduced large farms in northern Kazakhstan. Pre-inspections are needed in order to minimize moral hazard. Loss Notification and Loss  The Law should be modified to simplify the loss notification Assessment Procedures need obligations of the insured to be streamlined and made The Law should be modified to enable loss assessment functions to more cost-effective be managed by the insurance companies and procedures streamlined to reduce the costs of loss assessment.  A Bonus-Malus system should be introduced to dissuade farmers from submitting claims save where a major loss has occurred and which is likely to lead to an indemnity. Use of Remote Sensing to The national space agency (NSA) is already involved in applications of support in-field loss remote sensing to estimate crop sown area and production and yield assessment estimates for the MOA and to monitor crop status during the growing season. It is recommended that the crop insurance scheme managers should review potential supporting roles by the NSA for loss assessment. Review Policy Marketing and Currently most policy sales are made through local Agents located in each Distribution Channels with a Oblast and sub-region. The Agents are currently paid 10% brokerage by law. view to reducing costs Alternative crop insurance marketing and distribution channels should be promoted in order to reduce costs including sales though cooperatives and farmer associations, input dealers, rural banks, grain merchants, etc. Farmer Awareness Programs In conjunction with the proposed improvements in operating systems and on Crop Insurance procedures, greater emphasis needs to be paid on farmer awareness programs about the basis of insurance and indemnity on the current LIC program Source: Authors Institutional Strengthening 36. In the short term there is a major challenge to find ways of encouraging more private commercial insurance companies to support the crop insurance scheme. Several measures have been identified under this study which will hopefully be attractive to local insurers including i) the introduction of actuarially determined premium rates , ii) the measures to reduce insurers‘ liability in Aktobe and WKO and iii) recommendations for strengthening loss assessment procedures and for giving insurers more direct control over this important function. However, in the medium term it is also probable that Insurers will insist of being given greater control over risk selection and underwriting function if they are to join the scheme. 37. There is also a need to create a level playing field for commercial insurers and the farmer mutual insurance associations. Mutual insurance associations should in future be regulated under the Agency for Financial Market and Financial Institutions Regulation and Control and be required to follow the same guidelines in terms of the capital requirements and in terms of constitution of insurance reserves as per the commercial insurance sector. Failing to - xiv - doing so creates biased competition and the possibilities for the mutual associations to fail to meet the financial obligations. The Need to Cap Catastrophe Losses (Excess of Loss Protection) 38. Both the Crop Insurance companies and the Farmer Mutual Crop Insurance Associations in Kazakhstan are very exposed to Catastrophe losses which exceed their reserves and options for enhanced reinsurance protection need to be considered. GRK currently provides free of cost proportional reinsurance protection equal to 50% of the claims to the private insurance companies and mutuals in Kazakhstan. Currently, however, neither the private insurers nor the mutuals have any reinsurance protection on their 50% retentions and they are therefore very exposed to major systemic drought losses. In the case of the private insurance companies their ability to absorb catastrophe losses is much higher than the mutuals because of their much larger size, their capital and claims reserves and their diversified non-life insurance portfolios under which crop insurance only represents a fraction of their overall premium earnings and overall liability. In 2010 there is evidence that some mutuals could not meet their claims obligations in full because the claims exceeded their premium earnings: they therefore had to prorate down claims. There is an urgent need to design catastrophe excess of loss reinsurance protection for both the Private Insurance Companies and the Mutuals companies on their 50% retentions through some form of non-proportional or Stop Loss Reinsurance protection. In the short-term it is unlikely that the Kazakhstan obligatory crop insurance scheme will be able to meet the standards required by international reinsurers to attract their capacity and therefore any non-proportional reinsurance solutions will probably have to be provided by GRK. Some indicative rating analyses have been carried out for a GRK aggregate (i.e. over the whole crop insurance scheme) Stop Loss Reinsurance3 protection for the spring wheat crop and under the assumptions of the existing LIC scheme sum insured and actuarially determined rates, for priority levels of 70% of gross net premium income (GNPI) and 100% GNPI. This analysis has been conducted separately assuming Aktobe and WKO continue to be included in the LIC compulsory crop insurance scheme and then without Aktobe and WKO. While caution must be exercised as this is a preliminary rating analysis, it is considered robust and suggests that the indicative costs of providing full value (i.e. up to 100% of TSI) aggregate stop loss protection excess of 100% and 70% priorities would be in the order of KZT 0.27 billion (US$ 1.8 million) to KZT 0.31 billion (US$ 2.1) per year. (Section 4.2). Premium Subsidy Considerations 39. If government were to switch its support from a share of 50% the claims to some form of non proportional excess of loss protection, the insurers would need to increase their premium rates and in this case GRK may also need to consider whether farmers can afford these rate increases or if it will be necessary to introduce premium subsidies . Over the past 6 years, government‘s 50% of claims liability has been an average of KZT 685 million per year (US$ 4.6 million), but in severe drought years such as 2010, government‘s 50% claims share was much higher at KZT 1.4 billion (US$ 9.3 million) and the total cost has been KZT 4.1 billion (US$ 27.3 million) in claims compensation. If instead, over the past 6 years, government had provided 50% premium subsidies, the total cost would have been somewhat higher at KZT 5.86 billion (US$ 39.1 million) total or an average of KZT 0.98 billion (US$ 6.5 million) per year. 3 Stop Loss Reinsurance is a type of non-proportional reinsurance protection which is designed to cap an Insurers claims liability at a pre-agreed amount (value) which is referred to as the priority. Any claims above the priority are then transferred (ceded) to the reinsurer to settle, up to the limit of the reinsurer‘s liability which is defined in the reinsurance agreement. The priority is often expressed as a percentage of the premium income that is underwritten by the insurance scheme, net of any policy cancellations and or returns of premium, hence the term Gross Net Premium Income (GNPI) - xv - 40. GRK should study very carefully the issues surrounding premium subsidies before deciding whether to switch from the current system of claims subsidies to premium subsidies . The current system whereby government compensates 50% of the claims costs and then caps premium rates at approximately 50% of the technically required rates: in some regions of the country current premium rates are above the actuarially required rates and in other parts of the country actual rates are far too low. On the one hand this results in distorted crop insurance price signals in the market and on the other hand the 50% claims compensation does not provide the local insurers with the catastrophe protection on their retained claims that they require. Finally international reinsurers are not willing to support an under-priced scheme. While the authors are very cautious about recommending premium subsidies, it would be preferable in Kazakhstan to have an actuarially rated and commercially priced program and for government to then decide whether to provide financial support in the form of premium subsidies and also in the short term in the form of non-proportional stop-loss reinsurance protection and then in the medium term to promote the participation by international reinsurers. Phase 2: Transition towards a Market Based Crop Insurance System 41. The proposed transition over the next few years to a market-based crop insurance system in Kazakhstan is centered on 1) the introduction of individual grower multiple-peril crop insurance MPCI either as a complement to, or as a substitute for the current Loss of Investment Cost Policy and 2) the introduction of formal excess of loss reinsurance protection for the crop insurance industry. Under this transition it is assumed that crop insurance would probably continue to be obligatory for farmers during this interim phase and that Aktobe and WKO would no longer be included in the crop insurance scheme because this crop cannot be commercially insured in these two. It is intended that farmers in these two Oblasts are protected by a separate disaster-relief mechanism. Individual Grower MPCI Cover for Spring Wheat 42. The transition from the existing LIC cover to a more standardized individual grower Multiple-Peril Crop insurance (MPCI) Policy would be relatively simple. The crop insurers of the LIC policy have gained considerable experience in underwriting loss of yield-based multiple crop insurance and in conducting in-field loss assessment to establish actual yields and the amount of loss. This experience would enable them relatively easily to design, rate and implement individual grower MPCI. 43. MPCI and LIC insurance products are slightly different. The main differences of an individual grower MPCI product to the LIC Policy include: i) the establishment of a pre-agreed Insured Yield at the time of policy subscription (the Insured Yield is usually calculated as a percentage of the individual farmer‘s historical average or normal crop yield or t he local area average yield), ii) a pre-agreed unit Valuation Price which is applied to the Insured Yield to calculate the sum insured and iii) loss assessment involves the measurement of the actual yield which is compared to the Insured Yield and the amount of shortfall is indemnified accordingly at the pre-agreed valuation price. This basis of insurance and indemnity which is based on loss of yield is potentially much more transparent and understandable for farmers than under the existing LIC policy and loss assessment is also much more objective as yield loss is measured rather than - xvi - expected shortfall in production costs compared to the estimated value of the remaining crop (salvage revenue). 44. The international experience of MPCI is that the product is very popular with farmers, but on account of the high premiums associated with this product, most schemes are dependent on government support in the form of premium subsidies. The international literature on MPCI often highlights the drawbacks encountered under voluntary schemes of anti-selection and moral hazard, or the difficulties of establishing average farmer yields and corresponding premium rates, through to high premium costs requiring government support in the form of premium subsidies and the often very high costs of individual grower in-field loss inspection and loss assessment. While these arguments are indeed very valid, they apply as equally to the existing Loss of Investment Cost policy in Kazakhstan. Currently issues of anti-selection are less of a problem because the scheme is obligatory for all farmers, but because it is a loss of yield multiple peril scheme it shares the drawbacks of other MPCI schemes. 45. In Kazakhstan the main challenge for introducing individual grower MPCI centres on the procedures for establishing an individual grower “normal average yield” and then in establishing Premium Rates for different MPCI coverage levels in each Rayon. On the basis of international experience it is believed that the quality of the rayon-level crop production and yield data in Kazakhstan is sufficiently good to use the 17-years of available spring wheat yield data to design and rate an individual grower MPCI program. As the historical Rayon-level spring wheat crop production and yield data is available for both Production Enterprises and Commercial Farmers, separate coverage levels and premium rates can be offered to each type of farmer if required. 46. A detailed rating analysis has been conducted for MPCI cover for spring wheat grown in the 6 main Oblasts (excluding WKO and Aktobe) located in northern Kazakhstan The statistical rating methodology used in this study to establish individual grower MPCI rates conforms to the MPCI rating procedures which are adopted by the insurance industry. The spring wheat MPCI rates presented in this report are indicative commercial premium rates for a 60% target loss ratio, but it is stressed that final decisions over rates will be taken by insurers and their reinsurers. For Insured yield coverage levels of 20% to 30% of yield the MPCI rates would be roughly comparable to the full (unsubsidized) LIC Premium rates, but for higher levels of coverage rates would be correspondingly higher. (Section 4.62 to 4.65). 47. It is likely that if individual grower MPCI is introduced into Kazakhstan that GRK will need to consider supporting the higher premiums with premium subsidies. Some preliminary estimates are made in Section 4 of this report of the costs of government under different uptake scenarios. Risk Financing and Reinsurance 48. If individual grower MPCI is introduced for spring wheat and other crops, and higher coverage levels and sums insured are offered to farmers, this will have important implications for insurers and governments liability in the event of severe drought losses. For these reasons a detailed analysis has been conducted in Section 4 of the probable maximum losses associated with increased coverage levels for the spring wheat program. The analysis suggests that for a maximum 50% coverage level the expected losses that might occur every 10 years could be in the order of KZT 36.3 billion (US$ 242 million or 17.13% of the value of the TSI), and for the 1 in 100 year PML a loss of KZT 99.7 billion (US$ 665 million or 47% of TSI) - xvii - 49. Given the catastrophe risk exposures of spring wheat production to drought in Kazakhstan, it is extremely unlikely that the insurance sector would be willing to assume the increased liabilities implied under the proposed MPCI program unless government is willing to provide reinsurance support for this initiative. Currently government is providing 50:50 quota share reinsurance protection to the private commercial crop insurers and mutual crop insurers but as previously noted this protection does not cap their exposure to catastrophe losses. Therefore for the purposes of this study some preliminary analyses have been conducted for non- proportional stop loss reinsurance protection for the spring wheat MPCI program. 50. Some preliminary modeling has been conducted to establish the indicative reinsurance pricing for an Aggregate Stop Loss Reinsurance Protection for the spring wheat MPCI program. The modeling has been conducted assuming full value protection and priority levels of 70%, 100% and 150% of GNPI for the four MPCI insured yield coverage levels. The analysis shows that for a full value Aggregate stop loss reinsurance protection for losses excess of 100% GNPI, the stop loss reinsurance pricing would be in the order of KZT 991 million (US$ 6.6 million) for 20% coverage level, rising to KZT 5.8 billion (US$ 38.7 million) for 50% coverage. If the lower 70% of GNPI priority was to be adopted the Aggregate Stop loss pricing would be correspondingly higher. 51. In Phase 2 it is possible that some of the larger private commercial crop insurers may be able to arrange their own reinsurance programs with international reinsurers. If these insurance companies can demonstrate that even under an obligatory crop insurance scheme that they are underwriting a balances portfolio by risk region and that they are adopting technically based MPCI rating and that they have strengthened their MPCI loss assessment systems and procedures, then they should be able to arrange both proportional and non-proportional reinsurance through leading international reinsurers of this class of business. Phase 3 Transformation into a Commercial Pool Crop Insurance Scheme supported by International Reinsurance 52. In the final phase it is assumed the scheme would be transformed into a fully market- based public-private partnership agricultural insurance scheme for Kazakhstan, central features of which would include i) voluntary crop insurance, ii) formation of an Agricultural Pool coinsurance system to crowd in private commercial insurers (and possibly the mutual crop insurers if they can conform to insurance market regulations), iii) introduction of a formal program of international reinsurance protection for the pool and iv) backed up by suitable support from government and which might include financial support for premium subsidies and or catastrophe risk financing. Transition from Obligatory to Voluntary Crop Insurance 53. As part of the transition to a market based crop insurance system, policy makers in Kazakhstan will need to consider making crop insurance voluntary. Kazakhstan is one of a small minority of countries to adopt obligatory crop insurance and almost unique in trying to implement obligatory crop insurance through the private commercial insurance sector. International experience tends to suggest that agricultural insurance should be a voluntary class of insurance. In many countries (including India, Philippines, Mexico), there are very close linkages between public sector crop-credit provision and crop insurance and banks make their lending conditional on the farmer having crop insurance in place at the time of receiving his loan. In - xviii - other words, the banks make their lending conditional on the loanee taking out parallel crop insurance (mandatory basis) while crop insurance is voluntary for non-loanees. 54. If crop insurance is made voluntary in Kazakhstan it is likely that there will be a major reduction in the demand for crop insurance in the short-term while the farming sectors adjust to the realities of a demand-driven voluntary crop insurance system. At this stage it is not possible to predict how great the contraction in demand by farmers for voluntary crop insurance may be, but this is likely to be significant. Under a voluntary system, crop insurers would be free to select which types of farmer, which crops and which regions they are willing to underwrite under the crop insurance program. Rationale and Features of an Agricultural Insurance Pool in Kazakhstan 55. Coinsurance Pools for agricultural insurance have proved to be very popular with private and mutual insurers in many countries including most notably, the Agroseguro4 Pool in Spain, the Tarsim Pool in Turkey, the Philippines livestock insurance pool, the Austrian Hail Insurance pool and various other pool arrangements in China, Argentina, Malawi, Mongolia and Ukraine. 56. The rationale for recommending the formation of a coinsurance pool in Kazakhstan centres on a number of key factors which include: i) The very small number of private commercial companies which are currently supporting this scheme and the need to crowd in commercial insurers if crop insurance is to remain a viable proposition in Kazakhstan. LIC/MPCI crop insurance is a catastrophe class of business and many insurance companies are reluctant to risk their capital on it; however, under a Pool agreement individual companies can participate with very small shares of the overall risk if they wish. ii) The prohibitively high start-up investment costs for individual insurance companies in creating their own internal crop underwriting and claims departments and then in developing regional networks of marketing and sales agents and trained crop inspectors and loss assessors to administer the scheme implementation. Pools offer the opportunity to create a single centralised insurance underwriting and claims management and loss assessment capability (often termed a Managing Underwriter, MU) and for individual Pool members to contribute towards the running costs of the MU whilst benefiting from the advantages of economies of scale.. iii) The lack of common standards at regional level in the underwriting of crop risks and especially in the in-field loss assessment capabilities of individual insurance companies and the farmer mutual insurance associations. Under a Pool Agreement the MU would be responsible for coordinating all underwriting and loss adjustment activities and in ensuring that common standards are adopted throughout the country. iv) A lack of consistency in crop rating and competition which is driving down the crop insurance premium rates to unsustainable levels. Under a Pool Agreement 4 AGROSEGURO stands for the Agrupación Española de Entidades Aseguradoras de los Seguros Agrarios Combinados (Spanish Group of Insurance Entities of the Combined Agrarian Insurance) - xix - all insurers would issue standard crop insurance policies and they would all adopt the same premium rates for each crop in each zone and region; v) The difficulties of arranging commercial international reinsurance protection for individual Kazakhstani insurance companies with very different underwriting standards and portfolios. Under a Pool agreement, a single reinsurance program would be purchased by the MU and it would be much cheaper to transact a single reinsurance contract for the Poolc 57. It must also be recognized that there are potential drawbacks of introducing Pools to underwrite agricultural insurance. Classical economic theory would argue that by forming a pool (with monopolistic or oligopolistic tendencies) this will reduce competition particularly over pricing of insurance products. These arguments do not apply directly to Kazakhstan where there is currently no competitive market for agricultural insurance, where the single product available and the premium rates are fixed by law. While it is fully accepted that the introduction of a pool would lead to standard crop insurance policies and uniform technical premium rates being offered by each pool member, the aim will be over a number of years is to expand the range of crop insurance products available in the Kazakhstan market, to achieve economies of scale in key areas such as loss assessment and which will hopefully translate into lower overall commercial premium rates and to improve the insurance services provided to farmers. 58. The proposed Kazakhstan Agricultural Insurance Pool would involve the active participation of the public and private sectors. The central feature of the new system would be the creation of an Agricultural Coinsurance Pool which would be designed to underwrite all classes of agricultural insurance business. It is recommended that the Pool coinsurers also create a separate Managing Underwriting Company (MUC) which would be responsible for underwriting the scheme on behalf of the coinsurers and which would handle premiums and loss assessment and claims settlement on behalf of the pool coinsurers and which would also negotiate reinsurance on behalf of the Pool Coinsurers. The main function of the Pool coinsurers would be to market crop insurance through their sales distribution networks. The Public-Sector including the Insurance Regulator, MOA and FFSA would play very important roles under the proposed Pool system. The Fund for Financial Support to Agricultural Insurance, FFSA would continue to act as the main public sector implementing agency on this pool agricultural insurance scheme. Under the proposed Pool system, the FFSA‘s roles would be amended to no longer include participation in field-level loss assessment and would be expanded to include: i) Coordination with the Crop Insurance Pool‘s Managing Underwriting Company (MUC) in the development of the technical studies required for the design of new crop, livestock, forestry and aquaculture insurance policies and programs into the system ii) management of the government‘s financial fund for the support of agricultural insurance and disbursement of funds (including as appropriate, premium subsidies and catastrophe reinsured claims payments) to the MUC on behalf of the Pool Coinsurers, iii) maintenance of crop insurance underwriting and claims data- bases and iv) provision of information and advice to farmers. An outline institutional framework for the Agricultural Insurance Pool is shown in Figure 2 which draws on the experiences of the organizational structures of the Spanish and Turkish agricultural insurance pools. - xx - Figure 2. Organizational Framework for Kazakhstan Agricultural Insurance Pool Scheme Fund for Financial Agency for Financial Market and Ministry of Support to Agriculture Financial Institutions Regulation Agriculture and Control (FFSA) Financial support: Policy, planning, premium subsidies, research and Insurance legal & catastrophe development regulatory reinsurance Pool Management Board Kazakhstan Non-Life International AGRICULTURAL Insurance Companies Reinsurers INSURANCE POOL Farmer Mutual Crop Insurance Associations Managing Underwriting Company Farmer Associations, Cooperatives, Large Farmer Production Rural Banks and other Aggregators Enterprises Small and Medium Farmers Source: Authors 59. Details of the proposed operating systems and procedures for the Kazakhstan Agricultural Insurance Pool are set out in Section 4. The MUC would be responsible for the functions of product design and rating, underwriting and risk acceptance, claims administration and loss assessment and for negotiating common account reinsurance protection on behalf of and reporting to the pool coinsurers. The potential for the MUC to organise and train an independent crop loss assessment capability in standardized and timely and accurate field-based loss assessment procedures is highlighted as one of the major potential advantages of a pool system. The MUC, as a single entity acting on behalf of all pool coinsurers should be able to achieve major economies of scale and A&O cost savings. Government Support to the Kazakhstan Agricultural Insurance Pool 60. As part of the transition to a market-based Pool Agricultural Insurance System in Kazakhstan, the role of Public-sector support to this scheme should be reviewed. To date GRK‘s main support to the obligatory crop insurance scheme has been in the form of free Quota Share reinsurance of 50% of the incurred claims, but this is likely to change under the transition to a market-based agricultural insurance scheme. 61. The GRK has several potential roles to play in supporting the Pool, including: i) to provide legal and regulatory support including a review and reform of the Obligatory Crop Insurance Law No 553-II of 2004, ii) enhancing crop insurance data and information systems and infrastructure support, for example investment in weather stations to support crop insurance, iii) to support the insurance companies through implementation of crop insurance awareness programs for farmers, iv) catastrophe risk financing and v) premium subsidy support. (Full details are reviewed in Sections 4.102 to 4.109). Voluntary Crop Insurance Portfolio Financial Estimates 62. Some provisional financial estimates of Total Sum Insured, premium and costs of premium subsidies to government have been calculated for the spring wheat MPCI program over the next 5 years. The estimates are based on a spring wheat MPCI scheme with an average - xxi - 40% coverage level and demand projection uptake rates over the next 5 years under a voluntary program starting in year 1 with 10% uptake and rising to 50% uptake after 5 years. On this basis, total scheme liability in year 1 might be in the order of KZT 17 billion (US$ 113 million) rising by year 5 to KZT 85 billion (US$ 567 million) with corresponding year 1 commercial premium of nearly KZT 2.1 billion (US$ 14.0 million) rising to KZT 10.7 billion (US$ 71.3 million) by year 5. The provisional estimates of the costs to government of premium subsidies assuming 50% premium subsidy level would be nearly KZT 1.07 billion (US$ 7.1 million) in year 1 rising to KZT 5.35 billion (US$ 35.7 million) by year 5. Estimates are also provided for premium subsidy levels of 25% and 65% of premium (Table 3). Table 3. 5-Year Uptake Estimates for Voluntary Spring Wheat MPCI Total Sum Insured and Premium Income and Costs of Premium Subsidies (40% Converge level) MPCI 40% MPCI Crop Insurance Uptake Scenarios next 5-years (KZT Million) Item Coverage Year 1 Year 2 Year 3 Year 4 Year 5 100% Basis 10% 20% 30% 40% 50% Total Sum Insured 169,697 16,970 33,939 50,909 67,879 84,849 Total Commercial Premium 21,410 2,141 4,282 6,423 8,564 10,705 Cost of GRK Premium Subsidies: 25% of premium 5,353 535 1,071 1,606 2,141 2,676 50% of premium 10,705 1,071 2,141 3,212 4,282 5,353 65% of premium 13,917 1,392 2,783 4,175 5,567 6,958 Probable Maximum Loss: PML 1 in 100 years 66,470 6,647 13,294 19,941 26,588 33,235 PML 1 in 250 years 79,688 7,969 15,938 23,906 31,875 39,844 Source: Authors Risk Financing and Reinsurance 63. It is recommended that in future the Agricultural Insurance Pool would purchase common account reinsurance protection from international reinsurers in order to protect the program against catastrophe losses. The support from international reinsurers may include both proportional and non-proportional reinsurance. It is likely in the initial stages that for any stop loss cover, international reinsurers will only be willing to provide layered stop loss reinsurance protection in order to limit their liability to catastrophe claims and that the Government of Kazakhstan may therefore need to also participate in the structured risk financing program by providing catastrophe reinsurance for low frequency but high severity losses. An example of layered insurance and reinsurance is presented in Figure 3 below. For the proposed spring wheat MPCI program some illustrative commercial insurance and stop loss reinsurance layering and pricing involving both international reinsurers and government of Kazakhstan and full results of this analysis are presented in Section 4. 64. The access to the international agricultural reinsurance markets will benefit the local industry by having access to the expertise and services of specialized reinsurers . The service and expertise that the international agricultural reinsurers can provide is critical for the development of agricultural insurance schemes, particularly during the first years of operations. International agricultural reinsurers can provide their expertise and services to the local industry in the fields of agricultural insurance product research and development, pricing and underwriting and claims management - xxii - Figure 3. Example of Agricultural Risk Layering 50 Size of the Loss Government 40 Risk Reinsurers Transfer 30 Insurance Companies 20 Cooperatives Risk & Mutuals Pooling 10 Agricultural Risk Producers Retention 0 Minor Small Medium Large Catastrophic Type of Event: Mahul & Stutley 2010 Opportunities for New Crop Insurance Products 65. As part of this World Bank study an assessment has been conducted of the potential to design and implement new crop insurance products including named peril frost and hail cover, area-yield index insurance (AYII) and crop weather index insurance (CWII) for specific types of Kazakh farmer and for different regions according to the key risk exposures. Section 5 of the report presents the findings of this assessment. Named-Peril Crop Hail for Spring Wheat Farmers 66. Crop hail has been the second most important cause of insured claims on the Obligatory crop insurance scheme over the past 6 years and the preliminary findings of this study are that farmers in specific regions of Kazakhstan may be interested in a hail only crop insurance policy. Hail is a moderate to severe problem in many parts of Kazakhstan with peak months of hail exposure occurring between May and July (Section 2). Over the past 6 years of operation of the Obligatory LIC scheme, hail has been the second most important cause of loss after drought, accounting for about 2.5% of the total area lost due to insured perils. On the basis of the field visits it appears that there may be demand under a voluntary scheme by farmers in some regions for a hail-only insurance policy in spring wheat, including pockets of medium to high hail risk in parts of NKO, Akmola, Kostanay (e.g. in Altynsarin and Kostanay Rayons) and Pavlodar oblasts. . 67. Crop Hail insurance is a simple and well understood class of crop insurance and it is widely applied throughout the world to the insurance of wheat and a wide range of cereals, horticultural and tree fruit crops. Single peril hail insurance is the simplest and best known type of indemnity-based crop insurance which has operated for more than 100 years in Europe, North America, Argentina, Australia and New Zealand. Today there is a large body of accumulated experience with crop-hail damage-based insurance & indemnity products and wordings are readily accessible through international associations of hail insurers, premium rates can initially be set based on transferred international experience and so long as suitably high each and every loss deductibles (or franchises) are maintained, the rates are generally not high, and finally - xxiii - standardised damage-based loss assessment procedures can be accessed from the international hail associations and training provided to local staff. This experience could therefore be transferred very easily to Kazakhstan and then tailored by the crop insurers to meet local requirements. Outline policy design features for a spring wheat hail insurance policy for Kazakhstan are outlined in Sections 5.9 to 5.17. 68. It is concluded that crop hail insurance should be relatively easy to design and implement in Kazakhstan as a commercial crop insurance product. A preliminary hail rating exercise has been conducted for spring wheat in selected rayons in Kazakhstan which suggests that it should be possible to design hail cover at affordable rates to producer: in low hail prone regions for a standard 6% damage excess (franchise) average hail rates of between 2.5% to 3.5% should be feasible, rising to 5% to 6% in medium hail risk regions. Since this is a non- catastrophe crop insurance product, it should be relatively easier for the crop insurance companies and possibly the Farmer‘s Mutual Insurance Associations to underwrite this product with limited access to reinsurance protection. It is likely that the demand for single-peril crop hail insurance for spring wheat will be quite low in the initial stages of implementation of this product as hail exposure is not as widespread as the drought risk exposure. There would be an important start-up cost namely to design the suitable crop hail loss assessment procedures for Kazakhstan and to then identify a core of loss assessors who would receive specialist training in hail loss assessment procedures in wheat. Finally, it is likely that there will be demand for crop hail in other crops, for example cotton and horticultural crops grown in southern Kazakhstan and over time there should be potential to develop and expand a crop hail portfolio in Kazakhstan. Area-Yield Index Insurance (AYII) for Spring Wheat 69. On the basis of this feasibility study it appears that there may be considerable potential in Kazakhstan to design and implement Crop Area-Yield Index Insurance (AYII) as an alternative to or as a complement to the existing individual grower LIC and new proposed individual grower MPCI crop insurance programs. Outline proposals are presented in Section 5 of this study for a prototype Area-Yield Index product and program for spring wheat grown in Kazakhstan. The AYII proposals relate only to the 6 main wheat growing Oblasts of Akmola, EKO, Karaganda, Kostanay, NKO and Pavlodar: Aktobe and WKO Oblasts and their Rayons on account of the commercially uninsurable risk exposures in these two Oblasts. Features, Advantages and Disadvantages of AYII 70. AYII represents an alternative approach to MPCI insurance which aims to overcome many of the drawbacks of traditional individual grower MPCI insurance. The key feature of this product is that it does not indemnify crop yield losses at the individual field or grower level; rather, an Area-Yield-Index product makes indemnity payments to growers according to yield loss or shortfall against an average area yield (the index) in a defined geographical area (e.g., the total sown area of spring wheat grown in a single Rayon). In the context of Kazakhstan it is proposed that the AYII product would operate at the Rayon level (termed the Insured Unit – IU). 71. In the context of Kazakhstan a key potential advantage of AYII over individual grower MPCI is the ability to offer higher levels of insured yield coverage at lower rates because losses are adjusted against an area yield index and not at the individual farmer level . Other advantages of the AYII approach are that moral hazard and anti-selection are minimized, and as the costs of administering such a policy are much reduced, this offers the potential to market this product at lower premium costs to farmers. There are also major cost savings in AYII loss assessment as this is not conducted on an individual farmer and field by field basis, but rather according to a pre-agreed random sampling of crop yields on plots within the IU. However, the - xxiv - AYII has one main important drawback. The main drawback of an AYII policy is ―Basis Risk‖ or the potential difference between the insured area-yield outcome and the actual yields achieved by individual insured farmers within the insured area. Basis risk arises where an individual grower may incur severe crop yield losses due to a localized peril e.g. hail, but because these localized losses do not impact on the district or rayon average yield, the farmer who has incurred crop damage does not receive an indemnity. In addition, basis risk may arise where individual farmer crop production and yields in the same rayon are highly heterogeneous and some farmers whose average yields are above the area-average may in fact receive indemnities even though they have not incurred any significant yield reduction or loss on their own farms. 72. AYII is potentially a flexible crop insurance product that can be implemented at the micro-level for individual farmers, or alternatively as a meso-level product that is designed to protect the credit portfolio of a regional financial institution. In Kazakhstan there may be scope to design AYII both as a micro-level individual grower product for medium to large wheat (or other cereal) producers and to then design it as a meso-level product to protect the cooperative or MFI loan portfolios to large numbers of small rural households in individual Rayons in Southern Kazakhstan (discussed further in Section 6). 73. AYII has been adopted in several countries including India, the USA and more recently into Brazil, the Ukraine and is now being tested in Peru, the Philippines and in Vietnam. India has more than 30 years of experience with implementing AYII for food crops and oilseeds under the National Agricultural Insurance Scheme (NAIS) which is a public sector program for small and marginal farmers which is linked on an automatic or compulsory basis to crop credit provision and which is heavily subsidised by government in the form of capped premium rates and government compensation of excess claims. As such there are many similarities between the NAIS and current obligatory LIC scheme in Kazakhstan. The NAIS is the world‘s largest crop insurance program which is currently insuring about 25 million farmers each year. Since 2010 the NAIS has been undergoing major transformation both to strengthen the AYII policy and to move the scheme towards a market-based system and many of the issues and lessons from the NAIS are of potential interest to crop insurance planners in Kazakhstan. In the USA AYII has been offered for a wide range of cereal and oilseed crops for more than a decade, but is a relatively small program in contrast to the popular individual farmer MPCI covers that are available. In the Ukraine, which has very similar spring wheat production systems to Kazakhstan attempts have been made since 2002 to introduce both AYII for cereals and also weather index insurance, but with limited success: in the case of AYII there were difficulties in accessing quality Oblast and Rayon time series crop production and yield data because of the disruptions surrounding independence and the product was not widely accepted by farmers who preferred individual grower MPCI cover. The Ukraine experience can again provide useful lessons for any future planning and design of AYII for Kazakhstan. (See Section 5 for further details of the Indian and Ukraine AYII programs). AYII Design Features, Coverage Levels, Sums Insured and Premium Rating 74. In Kazakhstan the proposed AYII insurance policy for spring wheat would operate at the Rayon level, which is the lowest level of disaggregated time series crop area, production and yield data available through the National Agency of Statistics Republic of Kazakhstan (ARKS). A preliminary premium rating exercise has been conducted for the spring wheat AYII program for all eligible Rayons in the 6 Oblasts. The rating exercise was based on spring wheat 17-years of historical time series spring wheat crop production and yield data. This exercise adopted internationally recognised rating procedures and full details of the calculated technical and indicative commercial premium rates are presented in Section 5. - xxv - 75. Under a spring wheat AYII program for Kazakhstan the coverage level in each Rayon should be set in accordance with i) the underlying risk exposure and frequency, and ii) the commercial premium rate that can be afforded by the targeted farmers. In order for a crop insurance scheme to be both affordable to farmers the Insured Yield Coverage levels in each Rayon should be set at levels to enable commercial premiums of no more than about 10% to be charged and on this basis coverage levels of up to 50% could be offered in most Oblasts / Rayons and as high as 70% coverage in NKO (Section 5). Operational Considerations 76. The procedures followed by the National Statistical Agency of Kazakhstan for estimating actual average yield in each Rayon are technically sound for the implementation of AYII. However, it is recommended that, in case AYII is implemented, that the insurance companies enter into a formal agreement with the National Statistical Agency of Kazakhstan to provide the results of their crop-cutting yield estimates for each Rayon. Under an AYII scheme it is also likely that insurers and their reinsurers will wish to implement some form of independent monitoring of the area-yield estimation procedures at Rayon level to verify that standards of accuracy in the measurement of yields are maintained. Insurance companies may also wish to establish formal agreements with the NSA in order to access its remote sensing services to estimate crop sown area, crop production and yields and to monitor crop status during the growing season. Portfolio Estimates Financial and Reinsurance 77. Some provisional portfolio financial estimates have been calculated for AYII cover for spring wheat assuming a voluntary program and 5% incremental uptake rates per year over the next 5 years. Under the assumptions of a 50% Insured Yield coverage level and 5% uptake rate of AYII insurance per year over the next 5 years, the total sum insured might rise from KZT 10.6 billion in year 1 with corresponding premium income of KZT 692 million rising after 5 years to KZT 53 billion with premium income of KZT 3.5 billion. The costs to government of different levels of premium subsidy levels are also estimated. It is recognized that these uptake estimates are extremely ambitious for a voluntary insurance scheme and would need refinement following a more detailed AYII demand study. Conclusions on Area Yield Index Insurance 78. AYII for spring wheat is technically and operationally feasible in Kazakhstan, but until further research has been conducted into the potential demand for this cover it is very difficult to predict likely uptake rates under a voluntary crop insurance program. 79. Farmers’ demand for and willingness to pay for AYII crop insurance will also ha ve to be studied further before any decisions are made to proceed with the design of an AYII program. This feasibility study has identified a very low level of interest in the obligatory LIC crop insurance scheme by farmers and it is probable that this would apply equally to voluntary crop insurance in future. Similarly the experience from Ukraine has been of low voluntary demand for this product. It is therefore recommended that a formal AYII crop-insurance demand assessment study should be implemented by the key stakeholders in Kazakhstan. 80. AYII for spring what could possibly also be underwritten either as a Meso level product designed to protect the season loan portfolio of agencies which are lending to cereal producers (banks, input suppliers or MFIs) or as a Macro-level AYII cover for government to operate for small family farms in southern Kazakhstan. - xxvi - Weather Index Insurance (WWI) Opportunities for Kazakhstan 81. The analysis carried out in this feasibility study shows that developing Weather Index Insurance (WII) contracts for hedging the drought exposure of spring wheat in the North of Kazakhstan is technically feasible. However, challenges in the possible scale of implementation, in the commercial viability and in farmers’ interest for WII may limit the scope of application of this class of insurance products. The details of the analysis are presented in Chapter 5, where the prototypes developed for this study are illustrated and the operational and commercial challenges of WII discussed. Features, Advantages and Disadvantages of WII 82. The essential feature of WII is that the insurance contract responds to an objective parameter (e.g., measurement of rainfall or temperature) at a defined weather station during an agreed-upon time period. The parameters of the contract are set so as to correlate, as accurately as possible, with the loss of a specific crop type suffered by the policyholder. All policyholders within a defined area receive payouts based on the same contract and measurement at the same station, eliminating the need for field loss assessment. 83. The suitability of WII to transfer weather risks depend on how strong is the correlation between the weather parameter and the crop yield, and how spatially correlated is the risks. WII is best suited to transfer weather risks, where these risks are well-correlated over a widespread area and where there is a strong correlation between weather and crop yield. The strongest relationships typically involve a single crop, a marked rainy season, and no irrigation. WII is less suited to transfer weather risks where more complex conditions exist. Localized risks, such as hail, or where microclimates exist (for example, in mountainous areas) are not suitable for WII. Similarly, the scope for WII is limited where crop production is impacted by many or complex causes of loss or where pest and disease are major influences on yields. 84. The features of spring wheat crop production in north of Kazakhstan, indicate that WII is a potentially suitable risk transfer product for this crop. Spring wheat is almost a monoculture in North Kazakhstan, the region enjoys a marked rainy season, and crop production is fully rain fed. Furthermore, the analysis presented in this report of cumulative season rainfall deficit and spring wheat yields for a sample selection of stations and rayons, in general, exhibits very high correlations and this suggests there is considerable scope for GRK to continue research into the development of rainfall deficit WII cover for spring wheat in the northern region. 85. As for AYII, basis risk is the key constraint of WII. ―Basis‖ can be defined as the difference between the loss experienced by the farmer and the payout triggered by the weather index. It could result in a farmer experiencing a yield loss, but not receiving a payout or also in a payout being triggered without any loss being experienced. WII works best where losses are homogenous in the defined area, and highly correlated with the weather peril. 86. WII can be retailed at different business levels. At the micro level, the policyholders (the insurer‘s customers) are individual farmers, households, or small business owners who purchase insurance to protect themselves from potential losses caused by adverse weather events. At the meso level, the insurance policy is issued to an organization that has economic interests that are contingent to the results of agricultural activities including, for example, a financial institution or a cooperative that lends to the rural sector that wants to protect itself for eventual default of the loans given to farmers due to unfavorable weather conditions during the crop season. or an input supplier. At the macro level, the insurance policy is settled to a government or a national organization. The insured interest in the case of macro level coverages is usually - xxvii - related to government disaster relief for small farmers (see Mexican Macro-level index programs are presented in Box 6.1) or with food security issues. Weather data and contract design 87. Under this study, a prototype WII product design prefeasibility analysis for drought peril in spring wheat production in 6 selected rayons in north Kazakhstan region. The prototype WII product design prefeasibility analysis has been performed based on rainfall information provided by Kazhydromet (KHM) for 9 weather stations situated in the selected rayons and/or in neighboring rayons, which fully complied with the best practice for the design and operation of WII. The prototype WII product design prefeasibility analysis followed a widely used methodology developed by the Agriculture Risk Management Team (ARMT) of the World Bank and specific additional indexing procedures that were developed for the spring wheat environment of North of Kazakhstan. 88. WII may be technically feasible to transfer drought risk in spring wheat crop production in north Kazakhstan region. In six out of the nine rayon yield-weather station combinations analysed, it was possible to develop meaningful rainfall deficit WII structures. However, while this applies for the areas surrounding the specific weather stations analyzed, an actual full scale implementation of a micro level (farm level) WII program may be hindered by the relatively low density of the weather network. Taking as a reference the nine cases examined, in Kazakhstan, distances between contiguous weather stations start from a minimum of 70 km, (against conventional wisdom that for rainfall WII, stations should be no further than 20km to 25 km apart), which is probably too large for assuming the possibility of granting a full micro WII coverage of the entire territory. 89. The potential implementation of macro-level or meso-level WII contracts, which are less reliant on the weather station network than the micro-level WII, could have good chances of being relatively rapidly implemented. While in the medium term it may be possible to overcome the structural constraints related to the insufficient density of weather stations for the development of micro-level WII, in the short term a widespread full scale implementation of WII at farm level does not seem to be realistic. In this regard, the implementation of WII contracts at meso or macro levels, that may be less influenced by the density of the weather network, could have greater chances of being rapidly implemented than the micro-level approach. Basis Risk 90. Preliminary analyses conducted under this study suggest the existence of a relevant basis risk dimension for individual farmer WII cover. Basis risk is defined as the potential mismatch between the actual financial losses due to an insured event suffered by the farmer on his farm, and the payouts received from the insurance. During the study a basis risk assessment of WII was performed for a limited sample of farmers situated within 25 kilometers of the selected weather stations for the 2010 spring wheat crop year. The results of this analysis suggest the existence of a basis risk dimension and it highlights the importance of carefully evaluating the basis risk component embedded in farm level WII products. Cost of WII 91. In case of being implemented to cover drought on spring wheat crop production in North Kazakhstan, WII would be a relatively expensive product. The analysis carried out in the feasibility study shows that Indicative Commercial Premium Rates of the technically acceptable contract structures range from 13.7% in relatively good crop areas to 24.2% in marginal areas. - xxviii - This is a clear indication of the high potential cost of using WII contract for insuring against drought. Meso level WII and reinsurance 92. A specific form of meso level application of WII could be the use of index contracts as a reinsurance coverage for insurance companies involved in agricultural insurance programs. The analysis carried out in the study shows that it is actually possible to structure drought insurance products for spring wheat production by calibrating weather indexes on Rayon level yield records. These structures could technically form the basis of a reinsurance transaction. Conclusions on WII 93. The analysis carried out in this feasibility study indicates that WII for spring wheat in the North of Kazakhstan is technically feasible. 94. Despite the positive technical findings, the actual density of the weather measurement network does not seem to make a full scale implementation of farm level WII a realistic option in the short term. While potential actions to address this constraint may be undertaken including GRK‘s current investments in new weather stations, for the time being individual farmer WII cannot be considered a readily implementable alternative to the current LIC insurance scheme. 95. In addition, the high potential cost of WII products developed in the analysis, together with a preliminary analysis of basis risk patterns, suggest the need to carry out further research to assess the potential interest of famers for such an alternative insurance approach. Tailoring Crop Insurance to the Needs of Small Farmers in South Kazakhstan 96. The final section of this report presents some of the international lessons and experience on strategies and programs to address the agricultural risk transfer and insurance needs of small farmers and which may be applicable to the small household mixed crop and livestock farming sectors which are mainly located in southern Kazakhstan. Identification of Appropriate Crop Insurance Products 97. No “one-size” fits all. To date, Crop Insurers in Kazakhstan have offered a single Loss of Investment Cost (LIC) crop insurance product to mainly medium and large cereal producers located in northern Kazakhstan and there has been very little debate about the appropriateness of this product to small and marginal farmer in southern Kazakhstan. 98. There is a wide range of crop insurance product types available internationally and this report has recommended that in future Kazakhstan’s crop insurers should aim to develop and introduce several of these alternative crop insurance products. The products that have been recommended for individual farmers include named peril (hail) crop insurance, MPCI loss of yield insurance and new non- traditional index products, area yield index insurance (AYII) and crop weather index insurance (WII). The suitability of each crop insurance product type should be studied carefully in the context of the types of farmer (commercial, semi-commercial, subsistence) and farming systems (irrigated versus non-irrigated) in southern Kazakhstan. - xxix - 99. Traditional individual farmer MPCI is a risk management tool that is often appropriate for commercial and semi-commercial farmers: however, it cannot provide solutions for subsistence farmers. There is much evidence today that traditional individual farmer multiple peril crop insurance (MPCI) does not work for small and marginal farmers and usually ends up being heavily subsidized by governments. For most small subsistence farmers producing food crops for on-farm family consumption, crop insurance is a luxury few of them can afford, hence governments‘ intervention to make crop insurance more affordable through premium subsidies. In Kazakhstan crop insurance is unlikely to be a useful intervention for the very small rural households were these are mainly subsistence producers. 100. For subsistence farmers it may be much more cost-effective for governments to examine alternative social safety nets or, where they elect to use insurance, to consider some form of macro-level weather index programme to permit early payments to be made in the event of a major natural disaster. To date, several countries including Ethiopia, Malawi and Mexico have designed macro-level rainfall deficit index covers that have been designed to provide national and or regional governments with immediate cash liquidity following a natural disaster and to enable the government to provide an early response. Tailoring Crop Insurance to Different Client Levels 101. Crop Index Insurance (including both AYII and WII) is potentially a very flexible instrument which can be designed to provide risk transfer solutions at different levels of Aggregation. Index products can be offered at three levels: i) at the individual farmer-level (termed Micro-level insurance) , ii) at an intermediate level of aggregation as a financial business interruption protection for banks and other lending organizations such as cooperatives and MFI‘s (Meso-level insurance) and finally iii) this is an instrument that regional and or national governments can use to insure against major systemic perils such as drought (Macro-level insurance). 102. In Kazakhstan there may be opportunities for government to use macro-level index insurance as a catastrophe drought insurance mechanism. Mexico has for nearly a decade operated a system of AYII and WII catastrophe climatic insurance programs which provide insurance protection to the state-level governments against crop failure among their small scale subsistence farmers. This macro-level index approach may offer solutions to severe drought losses in WKO and Aktobe which are not be insurable under a commercial individual farmer insurance program. The applications of such a macro-level cover for small subsistence farmers in South Kazakhstan Oblast (SKO) appears more limited because of the highly heterogeneous farming systems which do not lend themselves to macro-index solutions. Insurance Delivery Systems for Small Farmers 103. Insurance companies throughout the World face major challenges in trying to identify cost-effective ways of delivering and administering agricultural crop and livestock insurance programs for small famers. In Kazakhstan this problem is accentuated because of the very low sums insured adopted under the Obligatory Crop Insurance scheme with an average sum insured over the past five years of only KZT 3,287/Ha (US$ 22/Ha) and an average premium rate of only 2.42% generating an average premium of about KZT 80 per insured hectare (US$ 0.53/Ha). In other words, a farmer with only 100 Ha would on average generate a total premium of about US$ 53 and this sum is far too low to enable Kazakhstan‘s commercial crop insurers to cover their A&O costs on such small farm units. - xxx - 104. In southern Kazakhstan there may be considerable potential for Commercial Insurers to enter into a “Partner-Agent” relationship with rural organizations (e.g. the Cooperatives or MFIs) which have an existing rural distribution network and a large farmer membership. Under a Partner-Agent Model, the Insurance company enters into a formal contractual agreement with the Agent under which the Agent assumes responsibility for marketing and promoting the Insurer‘s policies to its membership, for collecting premiums from Insured‘s and paying these over to the Insurer, for notifying claims to the Insurer, and finally in some cases for distributing claims settlement payments to the Insured‘s. Usually the Insurer will agree to pay the Agent a commission for its services. This model would potentially enable the private commercial insurers in Kazakhstan to deliver crop insurance more cost-effectively to large numbers of small and medium farmers. Such a model could also be used for delivering livestock insurance to the small mixed cropping and livestock families predominantly located in southern Kazakhstan. Mutual Insurance as a Solution for Small and Marginal Farmers in Kazakhstan 105. The Farmer Mutual Insurance Associations have been heavily promoted by GRK since 2008 and if their financial status could be strengthened the Mutuals might be the ideal institutional vehicle to underwrite Kazakhstan’s small and marginal crop and livestock producers. Currently about 38 Farmer Mutual Crop Insurance Associations (FMCIAs) are underwriting the Obligatory LIC crop insurance scheme in Kazakhstan. These Mutuals have very limited financial reserves and none are formally protected by any form of insurance or reinsurance protection. The individual Mutuals are therefore very exposed to catastrophe losses which exceed their reserves. In the event that claims exceed their reserves such as happened in 2010, Section 4 reported that the Mutuals had to pro rata down each claim settlement made to their members who incurred losses. International experience shows that when catastrophe claims occur which cannot be paid by the Mutual this often leads to the collapse of the Mutual. 106. If the Mutuals are to remain solvent and to underwrite crop and or livestock insurance for small and marginal farmers in Kazakhstan, ways of providing some form of catastrophe reinsurance protection must be developed. In the short-term it is unlikely that the private commercial insurance sector in Kazakhstan or international reinsurers would be willing to provide excess of loss reinsurance protection to the Mutuals. It is likely therefore that such a program would have to be offered through the Public Sector and given the FFSA‘s experience with administering the financial claims subsidies on the Obligatory LIC program, the FFSA would be best placed to administer some form of excess of loss program for the Mutual Crop Insurance Associations in Kazakhstan. In this context, policy makers in Kazakhstan may wish to study the Mexican ―Fondos de Aseguramiento‖ (Self Insurance Funds, SIFs) program which is an example of a very successful small farmer mutual crop and livestock insurance scheme which is formally reinsured by Agroasemex, the national Agricultural Reinsurance Company. See Section 6 for further details of the Fondos program. - xxxi - Chapter 1: Introduction and Objectives of the Study 1.1. The Republic of Kazakhstan is located in Central Asia. It is the ninth largest country in the world with an area of 2,724,900 sq km (1,049,150 sq miles) and is landlocked. Kazakhstan shares common borders with China, Kyrgyzstan, Turkmenistan, Uzbekistan and the Russian Federation: the total length of its borders amount to 12,187 km. The territory of the Republic of Kazakhstan (RK) stretches some 3,000 Km from the low reaches of the Volga in the West to the foothills of the Altai Mountains in the East, and about 2,000 Km from the West Siberian lowland in the North to the desert of Kyzylkum and the mountain range of Tien Shan in the South. 1.2. Kazakhstan, with a GDP US$ 6.3 billon US$ is the largest economy in Central Asia. The country was formerly part of the Soviet Union and gained independence in 1991. Administratively Kazakhstan is comprised of 14 regions, or Oblasts. Map 1.1 shows the administrative divisions of Kazakhstan. Map 1.1. Republic of Kazakhstan. Administrative Division Source: Wikipedia (2010) Importance of Agriculture in Kazakhstan 1.3. Agriculture is a very important economic sector in Kazakhstan. Agriculture occupies a key position in the Kazakh economy. Agriculture contributes 5.92%5 of Kazakhstan‘s GDP. Although agricultural output contracted sharply during the transition phase following independence in 1991, output has steadily recovered since 1998. During 1998–2010, total agricultural output increased at an average real annual rate of 5.21%. Most of this growth was a consequence of increases in crop production (+84 percent) while livestock grew in real terms by a modest 7%. Despite the recovery of agricultural production, its share in the economy shrank because of strong growth in other sectors, particularly in the extractive oil industry. 5 The Economist Intelligence Unit (2011). GDP values calculated as average for the period 2006-2010. (The Economist Intelligence Unit Limited, 2011) - 32 - 1.4. The agricultural sector is also very important from a social perspective in Kazakhstan. Approximately 7.3 million people (47.2% of the total population) currently live in rural areas in the country6. Agriculture employs more than 22% of the labor force in the country. Although the country is classified as a middle-income country, rural poverty is widespread. The transition from a centrally-planned to a market economy brought about a decline in many state-provided agricultural services, and enduring problems of soil degradation and a lack of infrastructure have resulted in abandonment and underutilization of agricultural land. 1.5. The natural and climatic conditions in Kazakhstan are favourable for agriculture. The country is one of the major global producers and exporters of grains (mainly wheat). Other principal agricultural products include meat, wool, cotton and milk. Farming areas occupy more than 220 million hectares (about 74% of the country's total area), of which cereal growing areas occupy about 13-14 million hectares. The area under pastures totals 185.7 million hectares or 68% of the total farming area. Individual farmers, collective farms and organizations utilize 81% of all farming areas and 98 % of all pastures. 1.6. Livestock is also a very important agricultural activity in the country. For a country with a long nomadic history, it is not surprising that stockbreeding is the traditional and dominant agricultural sector. No less than three quarters of all agricultural land is used for grazing. Sheep breeding is predominant, while cattle breeding and the rearing of pigs, horses and camels are also well developed. Animal husbandry typically accounts for about 45% of the production value in agriculture in Kazakhstan. Primary meat products include beef, veal, chicken, horse, lamb, pork and rabbit. Agricultural Crop Production in Kazakhstan 1.7. Kazakhstan is an important crop producer and exporter in Central Asia. Kazakhstan is an important producer and exporter of high-quality wheat. In 2010 Kazakhstan was ranked as the 6th largest wheat exporter, by volume, in the world. Average annual production of wheat is about 13 million tons, but output is highly dependent on weather and in recent years has fluctuated between 10 and 17 million tons. Between 2 and 8 million tons of wheat is exported annually, mainly to destinations in Europe (including Russia and Ukraine), northern Africa, and Central Asia. Kazakhstan also produces around 2 million tons of barley, and a small amount of oats, corn, and rice, but wheat is by far the country‘s most important commodity. The production of oilseeds (sunflower seed and rapeseed) is increasing but total oilseed output remains well below 1.0 million tons. The country also grows a small amount of cotton in southern Kazakhstan, with annual lint output of about 100,000 tons. 1.8. Crop production is mostly concentrated on grain crops in northern Kazakhstan. Kazakhstan consists of 14 administrative regions, or Oblasts. About 75 percent of the country‘s wheat is produced in three oblasts in north-central Kazakhstan: Kostanay, Akmola, and NKO (See Map 1.2 showing main crop producing areas). Kostanay alone plants about 4 million hectares of wheat per year. Spring wheat occupies 95 percent of the total wheat area in Kazakhstan and virtually all of the wheat in the three north-central oblasts. Minor grains include spring barley and oats (which are grown in the same region as spring wheat), winter wheat (southern Kazakhstan.), and rice (southern Kazakhstan, mostly in Kzyl-Orda Oblast). Oilseed area has nearly doubled in the past five years but still accounts for only about 6 percent of the country‘s total crop area. Sunflower, which is Kazakhstan‘s main oilseed crop, is grown mostly 6 Embassy of the Republic of Kazakhstan in United States. Web page: http://www.kazakhembus.com/index.php?page=national-goals-and-initiatives - 33 - in eastern Kazakhstan. Rape seed is grown in north-central Kazakhstan. Cotton is grown only in southern Kazakhstan. Map 1.2: Main Wheat Crop Production areas in Kazakhstan and Russia Source: Authors based on ARKS 1.9. The crop farming sector is highly heterogeneous in terms of farm structure and productivity. The Northern region is dominated by large Agricultural Enterprises (also termed Production Enterprises) specialized in crop production, whereas smaller mixed Peasant Farms (also termed Commercial Farms) and Household plots predominate in the South. The total agricultural land area increased by about 6% between 2003 and 2007 including, growth of 4% in the area of arable land and 6% in the areas of hayfields and pastures. The share of agricultural land farmed by Production Enterprises has declined from 59% to 50% during 2003-2007, while land farmed by individual Commercial Farms has increased from 41% to 49%. Commercial farmers have slightly reduced their arable land (by 0.14 million ha) while increasing their hayfields and pastures (by 8.47 million ha) between 2003 and 2007. In contrast the Production Enterprises have increased their arable area (by 0.92 million ha) and decreased their hayfields and pastures (by 5.14 million ha). 1.10. The Agro-Industrial sector of Kazakhstan still suffers from a number of problems which result in low productivity and profitability. Kazakhstan‘s labour efficiency in agriculture is five times lower than in the Eastern Europe, even lower than in Russia and in Ukraine. The agricultural machinery fleet in Kazakhstan, although in the process of being updated, is aged. According to ARKS, a high portion of Kazakhstan‘s current fleet (including 77 percent of its tractors and 59 percent of its harvesters), was more than 15 years old at the time of the 2006 agricultural census. Improvements in crop management practices funded by expanding State subsidies have contributed to higher and more stable wheat yields. Beginning around 2002, government support for agriculture has increased significantly in the form of subsidies on the prices of fuel, seed, fertilizer, and agricultural chemicals. - 34 - Government Policy for Agriculture 1.11. The Government of the Republic of Kazakhstan (GRK) is encouraging diversification of the country’s economy to reduce its dependence on oil, whose price volatility and resulting fluctuations in revenues make budget management challenging. The GRK recognizes the importance of the agricultural sector in diversifying economic growth, reducing rural poverty and in contributing to improved food security. The GRK‘s main strategies for agriculture are set out in the Strategic Plan elaborated by the GRK every three years. The Strategic Plan 2009-2011 defines three strategic objectives for the agricultural sector in Kazakhstan. The first strategic objective is the sustainable development of the agro-industrial complex sectors, increasing their competitiveness, ensuring food security and adaptation of agrarian production to the WTO accession conditions. The second strategic objective is the preservation, rational use and rehabilitation of forest resources, fauna resources and natural reserves, as well as establishment of conditions for sustainable water supply and efficient water management. The third strategic objective for agriculture defined in the plan is to establish normal conditions for rural welfare based on optimization of rural settlements, ensuring growth of rural territories‘ capacity through integrated rural development. 1.12. The GRK has significantly increased its support to the agricultural sector since the end of the 1990s7. This is evidenced by the introduction of support measures including a significant budget allocation increase and institutional improvements, such as new laws and institutional reorganizations. The budget allocation to agriculture reached KZT 390.7 billion (around US$ 2.7 billion) in 20108, which represents 8.76 percent of the total budget allocations in Kazakhstan for 2010 and 36% of agricultural GDP (World Bank 2010). The most important budget allocations are for (i) development of rural finance associated with agricultural machinery and (ii) crop production, such as marketing, phytosanitary activities, soil fertility, and other crop production supporting activities. GRK expenditures for livestock production are much smaller than the government expenditures for crop production, even though livestock contributes about 44% of total agricultural output. The type of GRK expenditures in supporting the agricultural sector is varied and can be classified as follows: (i) direct input subsidies; (ii) subsidized credit programs (i.e. with reduced interest rates), (iii) market price support schemes (such as the one carried out by the Food Contract Corporation to stabilize grain prices and maintain state reserves), (iv) expenditure on developing public goods (such as market information, land title and registration, disease control, seeds and grain quality analysis and grading) and expenditure on forestry and fisheries. Exposure of Agriculture to Natural and Climatic Disasters 1.13. Agricultural production in Kazakhstan is an extremely risky economic endeavor. A large portion of risks associated with agricultural production in Kazakhstan is due to climate events. Drought is the most pervasive peril affecting crop production in Kazakhstan. Reasonably higher levels of agricultural productivity can be achieved during years of adequate rainfall, but the region is subject to frequent drought and is considered a zone of risky agriculture. Historically, Kazakhstan‘s agricultural production suffers from serious drought events. As a result, the agricultural value added in Kazakhstan is marked by frequent and sharp year-to-year fluctuations. Besides droughts, the occurrence of hailstorms and autumn `early frost are also 7 Kazakhstan: Public Expenditure and Institutional Review for the Agricultural Sector. World Bank (2010) 8 Economist Intelligence Unit: Current policy: The 2010 budget deficit is below the government's target. March 2nd 2011. http://country.eiu.com/article.aspx?articleid=987853083. - 35 - important perils affecting crop production in the country, and pests and disease outbreaks can also lead to severe crop losses. Figure 1.1 shows the relation between the fluctuations9 in the annual agricultural GDP growth and the occurrence of major drought years (highlighted in yellow). Figure 1.1. Kazakhstan: Historic Agricultural GDP Growth and occurrence of droughts Kazakhstan: Agriculture Value added (annual % growth) 29% 21% 17% 13% 7% 6% 9% 3% 2% 0% -1% -3% -7% -5% -6% -12% -21%-24% -19% -23% 1997 1991 1992 1993 1994 1995 1996 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Year reported with drought Source: Authors from World Bank. World Development Indicators (2010); The Economist Intelligence Unit (20011); Kazakh-Hydromet (2010) Government objectives for crop insurance 1.14. The GRK has three objectives for crop insurance. The first government objective for crop insurance is to protect the famers‘ against loss of their crop production due to the effects of adverse natural, climatic and biological hazards. Kazakhstan is one of the most exposed countries in the world to drought losses in spring wheat followed by late spring and harvest frosts, hail and crop pests and diseases – GRK seeks through crop insurance, to protect small and medium farmers from going bankrupt following major crop losses. The second government objective towards crop insurance is to assist farmers‘ to gain access to rural finance by creating the conditions for lending through providing farmers with an insurance collateral to be used to protect their loans against default in the event of major crop losses. The third objective of GRK with regards to crop insurance is to assist in enhancing the effectiveness of government support programs for crop production. 9 Caution should be exercised in interpreting the GDP shortfalls in the years 1993, 1994, and 1995 as the GDP decline in these years is not related only to drought. In the early 1990‘s, following the breakup of the Soviet Union and the loss of massive government subsidies for State and Collective farms and livestock enterprise there was a sharp decline in agricultural productivity. Local agricultural officials began to set productivity thresholds for individual fields. Fields that consistently failed to meet the threshold -- typically 0.6 to 0.7 tons per hectare against a national average of about 0.9 tons per hectare -- were taken out of grain production and converted to permanent pasture. The decline in grain area accelerated in the mid-1990s when shrinking livestock inventories caused feed-grain demand to plummet, leading to a 75- percent drop in barley area between 1993 and 1999. During these six years, total grain area in Kazakhstan contracted at the rate of nearly 2 million hectares per year. - 36 - 1.15. In 2004, the Government of the Republic of Kazakhstan enacted a Law on Compulsory Crop Insurance. In accordance with its objectives towards agricultural insurance, GRK enacted a Law on Compulsory Crop Insurance dated March 10, 2004 and this Law became effective on April 1, 2004. The law established the terms and conditions for the implementation of a mandatory salvage-based loss of investment costs crop insurance10 scheme for all farmers growing a range of strategic grain, oilseed and other field crops grown in Kazakhstan.. Under the Law, the insurance indemnity is determined in accordance with the normative production cost per hectare for each crop type, multiplied by the planted area of the insured crop. The insured loss is determined as ―a positive difference between the normative cost per one hectare of crop production, which is set at the moment of concluding an insurance contract, and the revenue from one hectare of crop production in the area affected by the adverse natural events, multiplied by the exact area on which the given crop was produced and which was affected by an adverse weather event‖. 1.16. The Kazakhstan crop insurance system is based on a public private partnership (PPP). The Law on Compulsory Crop Insurance created the institutional framework and regulated and established the legal, financial and organizational basis for the implementation of crop insurance. The implementing agencies of the current crop insurance system in Kazakhstan are MoA (through the Direction of Strategic Planning), the Fund for Financial Support for Agriculture (FFSA), the private commercial insurance companies and the Farmers‘ Mutual Crop Insurance Associations (FMCIA) and finally the local authorities in each Oblast and Rayon. Under the PPP, GRK provides financial contributions to the crop insurance scheme through an Indemnity Fund, termed the ―Fund for Financial Support for Agriculture‖ (FFSA). GRK supports agricultural insurance by paying 50% of the insurance indemnities reported by the insurance companies. This is similar to a free 50% quota-share reinsurance facility. Government payouts are made both to private insurance companies and to the FMCIAs. 1.17. Crop insurance has reached high levels of penetration in Kazakhstan which is a function of its obligatory nature; however, the system is experiencing serious drawbacks. The crop insurance penetration in Kazakhstan, averaging 74 percent of the cropped area for the period 2005-2010, is relatively high. Notwithstanding the high level of penetration reached by obligatory crop insurance in the country, the experience to date with the mandatory crop insurance highlights the operational and financial drawbacks with its implementation. From 2005 to 2010 the average annual loss ratio after 50 percent government reimbursement of claims is 75 percent, indicating that the program is currently not financially sustainable. In addition to the financial drawbacks, the crop insurance system has also experienced operational problems. These operational problems are evidenced by the low level of sum insured chosen by the farmers to insured their crops or an average of US$ 25.7 per hectare in 2010. From a financial point of view the insurers and mutuals are very exposed to catastrophe drought losses due to the fact that the obligatory crop insurance system is not currently reinsured again excess losses. Objectives and scope of the study 1.18. The overall objective of the current study is to assist Government of the Republic of Kazakhstan to improve the existing mandatory crop insurance program. The specific objectives of this World Bank study include: (i) to perform a crop and weather risk assessment for 10 The Salvage-based Loss of Investment Costs Crop Insurance policy (termed ―Loss of Investment Cost‖ policy for short) protects farmers against adverse natural, climatic and biological phenomena resulting in production shortfalls that cause farmers‘ expected revenues from the insured crop in the insured unit to fall short in respect the investments they made in growing the crop in their insured unit(s) - 37 - key crops; (ii) to review the current crop insurance scheme; (iii) to identify the potential gaps of the current crop insurance scheme and to provide recommendations for its improvement, based on international experience and best practice; (iv) to identify opportunities for the development of alternative crop insurance products including area-yield index insurance(AYII) and weather index insurance (WII), as well as hail named-peril crop insurance; and (v) to identify ways to tailor the provision of crop insurance to small farmers and the rural poor (particularly in the southern part of the country). 1.19. The study aims to identify sustainable market-based alternatives to the current crop insurance system in Kazakhstan. In this regard, all the options for improvement of the current crop insurance system that have been developed under this study are marked- based and take into account global experience and the best insurance and reinsurance industry practices for agricultural insurance. 1.20. The study follows the principles established in the agriculture risk management framework developed by the World Bank. The development of market-based agricultural insurance risk transfer solutions in the region also implies promotion of several practices, including establishment of: (i) an adequate agricultural risk management framework for the countries in the region, which would include farmers‘ segmentation, an accurate assessment of the risk faced by the agricultural sector, an adequate risk financing strategy for the agricultural risks, and proper institutional arrangements; (ii) well identified roles of the public sector, insurance industry, and the farmers with regards to agricultural insurance; and (iii) possible actions to be taken by the governments in order to support agricultural insurance. 1.21. The study is mainly focussed on spring wheat crop production in the main growing regions of Kazakhstan. Owing to the geographical size of the country, the study is limited to spring wheat grown in the eight main spring crop production Oblasts in Kazakhstan (namely: NKO, Akmola, Kostanay, Aktobe, Pavlodar, WKO, EKO, and finally Karaganda). The study also explores opportunities for crop insurance development for small farmers in Tole-Bi rayon in SKO. For those activities related to the identification of opportunities for new crop insurance product development (Crop hail, AYII and WII), the study is limited to a pre-feasibility analysis and the eventual development of prototypei products for spring wheat in a couple of selected rayons in the North and East (including Altynsarinski and Auliyekolski Rayons in Kostanay Oblast; Aktogaiski and Zhelezinski Rayons in Pavlodar Oblast; Bulandinski and Enbekshilderski Rayons in Akmola Oblast), and finally Tole-bi Rayon in SKO. The Oblasts and Rayons selected for the analysis of new crop insurance product opportunities are highlighted in red in Map 1.3. - 38 - Map 1.3. Oblasts under the scope of the Agricultural Insurance Feasibility Study and selected Rayons for Agricultural Insurance Prototype Development. Source:Authors Report Outline 1.22. The study is set out in six sections. Section two presents an overview of agricultural production systems and markets in Kazakhstan followed by an assessment of the climatic hazards and other risks affecting spring wheat in the main crop areas in Kazakhstan. Section three reviews the current mandatory crop insurance system in Kazakhstan and its performance and identifies a series of institutional, operational, technical, and financial drawbacks of the current system. Section four presents a phased strategy and a series of options and recommendations for GRK to consider for the introduction of market-based solutions that aim to strengthen the current crop insurance scheme. Section five explores the opportunities for new crop insurance product development in Kazakhstan including pre-feasibility analyses for AYII, WII and named peril hail insurance for selected Rayons in the country. Finally section six deals with challenges of tailoring crop insurance to the needs of lower income smaller farmers. . - 39 - Chapter 2: Crop and Weather Risk Assessment Objectives and Scope of Agricultural Crop and Weather Risk Assessment 2.1. To date, in Kazakhstan there has been little formal risk assessment for crop insurance purposes of the key climatic, biological, and natural perils and their impact on crop production and yields and farm incomes. The risk assessment presented in this chapter aims (i) to aid policy makers and planners in Kazakhstan in understanding the major climatic and natural-peril risk exposures in spring wheat which is the main food and export crop grown in the country; (ii) to quantify wherever possible the value of expected spring wheat crop losses in normal and catastrophe loss years, and (iii) to discuss the implications for any modification of the current crop insurance program in place in the country. The specific objectives of the agricultural crop and weather risk assessment are (a) to identify and quantify the key natural, climatic, and biological perils affecting spring wheat production in Kazakhstan; (b) to quantify the frequency and severity of the perils affecting spring wheat production; and (c) to define homogeneous spring wheat crop risk zones and to map them. 2.2. This Chapter presents a preliminary risk assessment of weather risks and their impact on spring wheat crop production and yields in north Kazakhstan Region. This chapter starts with a review of data availability in Kazakhstan for spring wheat risk assessment purposes. This is followed by an overview of climate and the agro-ecological regions and spring wheat crop production systems in the selected Oblasts of Kazakhstan, and then an analysis of spring wheat production and yields and the climatic constraints to production including an analysis of rainfall data and the relationship to national and rayon-level spring wheat crop production and yields. The final part of this chapter presents the results of a Crop Risk Assessment Model, CRAM, which uses time-series rayon-level production and yield data to estimate values of risk, expected losses and expected claims costs for spring wheat in the eight selected Oblasts in Kazakhstan. This latter analysis is very relevant to crop insurers‘ understanding of risk accumulation and maximum expected losses in spring wheat. Data Availability for Crop and Weather Risk Assessment 2.3. Data for Spring Wheat crop risk assessment. There are three types of data which are commonly used in the assessment of climatic risk in crop production (a) time series weather data (b) crop damage and/or production loss data by cause of loss for each crop and which may also include estimates of the financial value of the damage or losses and (c) time-series crop area, production and yield data: the analysis of variance in annual crop production and yield data is commonly used to design and rate multiple-peril crop insurance, MPCI, programs. 2.4. Kazakhstan has a modern and an efficient national meteorological service known as the National Hydro-meteorological Service of Republic Kazakhstan, or Kazhydromet (KHM) . The surface meteorological network of the Republic Kazakhstan managed by KHM includes 260 weather stations (50 automated) and 71 agro-meteorological measurement points. The variables recorded at most meteorological stations include air temperature, precipitation rate, snow, soil moisture, and evaporation. The data communication system is well structured but somewhat obsolete in some of the more remote measurement locations. Each set of observations undergoes a rather strict quality check process according to WMO standards. Weather data are transferred to each Oblast Center (CHM) at the appropriate frequency (3 hours for temperature to one dekad - 40 - for soil moisture). From the CHM, a communication specialist transfers individual weather messages into the KHM centralized web server in Astana. The web-server also shares weather data with other professionals in the country (synoptic specialists, agro meteorologists, climatologists and to GIS-Meteo). 2.5. Given the huge size of Kazakhstan the weather station density is currently not able to provide adequate coverage for all the country. According to the World Meteorological Organization (WMO), Kazakhstan –owing to its size - would need to have approximately 1,600 weather stations in order to achieve an optimum weather station density. KHM continuously invests in the upgrading of its weather station network. The geographic distribution of the weather station network in Kazakhstan is presented in Map 2.1. Map 2.1. Kazakhstan: Weather Station Network Source: KHM 2011 2.6. The weather risk assessment performed under this study is based on the information provided by KHM. Under this feasibility study, KHM has provided the World Bank with access to daily rainfall, and minimum and maximum daily temperature data for ten weather stations located in seven Rayons selected for the development of agricultural insurance prototypes. The selected weather stations for which the daily data were provided are located in the following seven Rayons: Auliyekolski and Altynsarinski in Kostanay Oblast; Aktogaiski and Zhelezinski in Pavlodar Oblast; Bulandinski and Enbekshilderski in Akmola Oblast; and Tolebi in SKO. In addition to the daily data, KHM also provided monthly rainfall and average minimum and maximum daily temperature data for the following twenty weather stations located in ten Oblasts under analysis: Bulaevo and Saumalkol in North Kazakhstan Oblast; Diyevskaya and Mikhailovka in Kostanay Oblast; Yegendykol and Schuchinsk in Akmola Oblast; Mikhailovka and Aktogay in Pavlodar Oblast; Kamenka and Chingirlau in WKO; Komsomolskoe and Novoalekseevka in Aktobe Oblast; Korneevka and Karaganda Agricultural Experimental Station in Karaganda Oblast; Dmitriyevka and Samarka in EKO; and Shimkent city and Kazygurt in SKO. 2.7. Kazakhstan has very good spring wheat crop production statistics records. The ARKS is responsible for the collection, recording, and management of national crop production data and - 41 - statistics in Kazakhstan. ARKS is charged with the recording of seasonal crop acreage, production and yield data for all major food crops and also for horticultural crops in Kazakhstan. In addition to ARKS, there is a Department of Statistics, under the Ministry of Agriculture and the Land Resources Management Agency of the Republic of Kazakhstan which is also involved in the collection of agricultural statistics. A third important source of agricultural statistics is through the ―household accounting‖ system which is carried out by the regional administrations (Akimats) of townships, villages (Auls) and rural counties (Rayons). There is a close interaction between all state bodies within the system of recording of agricultural statistics in Kazakhstan. The data is collected from all categories of farms engaged in agricultural production and services. Among them there are Agricultural Enterprises (Production Enterprises - PEs), Peasant Farms (Commercial Farms - CFs) and Household farms. Statistical observations are conducted on the basis of the Statistical Register of Agricultural Organizations. This register includes all these categories of agricultural producers. PEs are covered by a general statistical survey which they are required to complete and submit on a monthly, quarterly and annual reporting basis. To obtain data on CFs and Household farms, a system of general farm accounting surveys (Households accounting books and CF accounting books) is carried out by the rural county administrations (akimats). The Farm accounting surveys contain complete information on the area sown by each type of farmer. To determine the volume of agricultural production and average per hectare yields of the CFs and Household farms, sample statistical surveys (sample size 30% and 5% respectively of all farms) with semi-annual and annual intervals are conducted. 2.8. A major statistical analysis at rayon level for spring wheat has been performed under this World Bank study. Under this study the World Bank has analyzed ARKS 17-year (1994- 2010) time-series annual sown area, harvested area, crop production, and yield data for spring wheat, with a breakdown into PEs and CFs, for each of the 118 Rayons located in the 8 Oblasts selected for the analysis (Akmola, Kostanay, NKO, Aktobe, WKO, Pavlodar, EKO, and Karaganda). The quality of the data was, in general, very good. Only 4.60 percent of the data entries were missing. This time-series data has enabled a series of useful analyses to be conducted. To date, preliminary analysis have been made of (i) the Rayon-level, Oblast-level, and National-level distribution of crop exposure (value at risk) for spring wheat by type of farmer and overall, (ii) the annual variation in spring wheat crop production and yields at rayon level to identify the areas of higher yield variability risk, and (iii) to conduct simple correlation analysis with rainfall variables. Furthermore, these series have been used to develop the risk assessment models at rayon level to assess the risk exposure for spring wheat as well as to establish expected yields and illustrative premium rates for i) an individual grower multiple peril crop insurance (MPCI) program for spring wheat (See Section 4 for further details) and ii) an area AYII program for spring wheat (see Section 5). 2.9. In Kazakhstan, there is no systematic monitoring and recording of loss or damage to spring wheat production arising from natural perils including floods and droughts. In some countries, public sector organizations (Ministries of Agriculture or the Agencies responsible for Natural Disaster management) systematically record crop damage (area damaged and percentage loss of crop production and yields) arising out of major natural and or climatic events (and also record the cause of loss): usually these estimates of damages are used to make public sector compensation payments to farmers and rural households in the affected areas. Time-series crop damage data is therefore very useful for analyzing the frequency and severity of major events. In Kazakhstan, it is understood that neither MOA nor the Ministry of Emergency Situations11 11 According the Ministry of Emergency Situations of the Republic of Kazakhstan (written communiqué 5 March 2011), between 1996 and 2010 no government financial disaster relief assistance was provided to farmers affected by natural disasters in Kazakhstan. - 42 - systematically record crop damage by cause of loss and therefore, there is no national data-base of historical damages by cause of loss in crops. Owing to the non existence of a data-base of historical damages to the crops it has not been possible to access or use crop damage data in this report. In order to partially overcome this problem the World Bank Team has worked in establishing the relationships between the spring wheat yield shortfalls and rainfall. The establishment of the relationships between spring wheat yields and rainfall, allowed the study to identify drought years. However, owing to the lack of information in the country, it was not possible to identify years with losses due to other perils such as hail or freeze. Climate and Agro-ecological Regions 2.10. Kazakhstan experiences a marked continental and dry climate. Kazakhstan experiences a continental climate with long cold winters and very short hot summers with a short growing season that varies from as little as 105 days in the most northerly regions to 165 days in the south (Nomura 2008). Seasonal temperatures are polarized and vary depending on the region. Average winter temperatures during the day are -16°C to -18°C in the far north and about -6°C in the south; summer temperatures average 21°C in the north and 27°C in the south. Snow starts to fall around November and the mountain passes are snow-bound until April and sometimes even into May. Precipitation in the spring wheat crop production areas in Kazakhstan is very low. Total annual rainfall in north Kazakhstan averages from 280 millimeters per year in Aktobe Oblast to 400 millimeters in NKO. The rainfall in north Kazakhstan is distributed throughout the year with a peak during the months of June, July, and August. In South Kazakhstan the rainfall is also distributed throughout the year but with peak winter rainfall and a pronounced dry season during the months of June, July, and August. In the South Kazakhstan region the total annual rainfall is higher than in north Kazakhstan: for example in SKO the average annual rainfall for the period 1990 to 2010 is 556 millimeters. Although precipitation is higher in southern Kazakhstan on account of very high evapotranspiration levels most agriculture is dependent on irrigation. Map 2.2 shows the monthly distribution of rainfall for 10 selected Oblasts in Kazakhstan. - 43 - Map 2.2. Kazakhstan: Monthly Rainfall distribution for selected Oblasts in Kazakhstan Source: Authors based on rainfall data provided by Kaz-hydromet. 2.11. In Kazakhstan drought events are determined by two important atmospheric circulation patterns that affect inter annual rainfall variability. The so called ―Azores High‖ affects rather homogeneously the entire country by generating anticyclones (usually associated to dry conditions) that move from west to east, while the so called ―Siberian low‖ generates anomalies of opposite signs in the east and west of Kazakhstan. A consequence of the interplay of these patterns of atmospheric circulation is that west/central Kazakhstan is usually dry, whereas more favorable conditions for agriculture are found to the north-east.12 How widespread or localized a drought event may be is influenced by the interaction of such circulation patterns. 2.12. The country is divided into sixteen agro-ecological zones according to the temperature and water availability for plant growth. The flat areas of the country can be divided according to the cumulative temperature suitable for crop growing and humidity factor by hydrothermal ratio (HTR). Under this criterion the territory of Kazakhstan can be divided into 9 agro climatic zones: from poor moderate – warm (zone I) to very dry and hot (zone IX). The hilly and mountainous areas of Kazakhstan can be divided into additional 7 natural landscape zones (Zones X to XVI). Temperature increases from South to North, from cumulative temperatures of 4000оС in the south to cumulative temperatures of 2000оС in the northern parts of the country. Soil moisture content varies from north to south in Kazakhstan. During the warm summer months the territory of Kazakhstan is characterized by HTR‘s varying from 0.2 in the south and up to 1.1 in the north of the country (Image 2.2). An HTR of 1.0-1.3 indicates poor wet zone (forest-steppe), 0.7-1.0 – arid zone (steppe), 0.5-0.7 – very arid (dry steppe), 0.3-0.5 – dry (semi-desert), less than 0.3 – very dry (desert). Map 2.3 summarizes the agro-ecological zones of Kazakhstan. 12 ―Anticyclone‖ conditions imply subsidence of moist air from top layers in the atmosphere, hence less condensation (dry conditions). ―Cyclone‖ conditions imply convergence of moist air from surface atmospheric layers, hence more cloud formation (wet conditions). - 44 - Map 2.3. Kazakhstan: Agro ecological Zones. Source: ARKA Consulting from KHM Overview of Spring Wheat Crop Production in Kazakhstan 2.13. Kazakhstan is well endowed with land resources for agricultural production. The country has 76.5 million hectares of agricultural land. According to the 2006 agricultural census, 61 percent of the agricultural land in Kazakhstan is permanent pasture, and 32 percent is classified as arable land (systematically cultivated for the production of row crops). Of the remainder, 3 percent is used for hay production and 4 percent is ―long-term fallow‖ (indicating potentially arable land that has remained uncultivated for at least several consecutive years). Of the 24 million hectares of arable land, about two-thirds, approximately 18 million hectares, is devoted to grain production. Total sown area, including grains, forage crops (mostly perennial grasses), technical crops (chiefly oilseeds and cotton), and food crops (potatoes, vegetables, and melons) decreased sharply during the late 1990‘s due to the contraction of the grain and forage- crop areas.(USDA-FAS, 2010)13 2.14. Kazakhstan privatized its agricultural land in 2003 and today three types of legal entity are recognized: (i) Agricultural Enterprises (also termed Production Enterprises PEs) many of which are former state collective farms which have been privatized and which include joint stock companies, limited liability partnerships and cooperatives, and which are mainly located in northern and central Kazakhstan and are typically large-scale commercial grain producing companies (ii) Peasant or Individual farms (also termed Commercial Farms - CFs) which are generally less than 1,000 Ha and which are also involved in commercial crop and livestock production and (iii) Household Plots which are not registered and which consist of small family vegetable plots and livestock holdings which produce mainly for self consumption. Reference to Table 2.1 shows that in 2004 there were 2.2 million farms in Kazakhstan, the bulk of which or 13 http://www.pecad.fas.usda.gov/highlights/2010/01/kaz_19jan2010/ - 45 - 93% were Household Plots accounting for less than 1% of all arable land, but contributing to 50% of the value of agricultural output. At the other extreme, in 2004, the 4,600 PEs controlled nearly 13 million Ha or 59% of the total arable area with an average size of arable farm of slightly greater than 2,800 Ha and with some farms as large as 400,000 Ha. Between 2004 and 2007 the total number of farms increased to 2.4 million. In spite of privatization in 2003, a freely functioning land market has been slow to develop and much of the former state-owned arable land has not been privatized and instead is leased to private corporate farms under 49 year leases (World Bank 2010; Nomura 2008). Table 2.1. Kazakhstan Farm-Ownership Structure in 2004 (2007) Commercial Agriculture Subsistence Total Item PEs Agricultural CFs Peasant Household Plots Enterprises (Multiple Farms (Individual (Individual Ownership) ownership) Ownership) No % No % No % No % Number of Farms (2007) 7,340 0.3% 194,550 8.1% 2,206,870 91.6% 2,408,760 100% Number of Farms (2004) 4,600 0.2% 156,000 7.2% 2,000,000 92.6% 2,160,600 100% Labour force (000) 326 14% 280 12% 1,782 75% 2,388 100% Agricultural Land (000 Ha) 43,420 56% 34,228 44% 325 0.4% 77,973 100% Arable Land (000 Ha) 12,921 59% 8,816 40% 231 1% 21,968 100% Average Arable Farm Size (Ha) 2,809 57 0.1 10 Gross Agric. Output (KZT Bio) 171 24% 178 26% 349 50% 698 100% Sources: World Bank 2010; Nomura 2008 2.15. Kazakhstan is an important producer and exporter of high-quality wheat. Average annual production is about 13 million tons, but output is highly dependent on weather and in recent years has fluctuated between 10 and 17 million tons per year. Between 2 and 8 million tons is exported annually, mainly to destinations in Europe (including Russia and Ukraine), northern Africa, and Central Asia. Kazakhstan also produces around 2 million tons of barley, and a small amount of oats, corn, and rice, but wheat is by far the country‘s most important commodity. The production of oilseeds (sunflower seed and rapeseed) is increasing but total oilseed output remains well below 1.0 million tons per annum. The country also grows a small amount of cotton in southern Kazakhstan, with annual lint output at around 100,000 tons. (USDA-FAS, 2010). 2.16. The main spring wheat crop production areas are situated in Northern Kazakhstan. About 83 percent of the country‘s spring wheat is produced in four oblasts located in north- central Kazakhstan: Kostanay, Akmola, Pavlodar, and NKO; 3 percent of the country‘s wheat is produced in EKO; 4 percent is produced in Karaganda Oblast; and 7 percent in western Kazakhstan (Aktobe Oblast and WKO). Spring wheat occupies 95 percent of the total wheat area in Kazakhstan and virtually all of the wheat in the four north-central oblasts. Minor grains include spring barley and oats (which are grown in the same region as spring wheat), winter wheat (southern Kazakhstan.), and rice (southern Kazakhstan, mostly in Kzylorda oblast). Out of the total area planted with spring wheat in north Kazakhstan region, 68 percent is planted by large Agribusiness Enterprises (farms with more than 1,000 Ha of wheat), while the remaining 32 percent of the area planted with spring wheat is planted by Commercial Farmers (farmers with between 100 Ha up to 1,000 Ha). Map 2.4 shows the distribution of spring wheat planted area (in hectares) in the 8 Oblasts of northern and central Kazakhstan at rayon level. The most important - 46 - spring wheat rayons with between 320,000 Ha and 640,000 Ha of spring wheat are located in Kostanay, NKO and Akmola: conversely in EKO and WKO the planted area of spring wheat is less than 2,500 Ha in many rayons. Map 2.4. North Kazakhstan: Planted area with spring wheat at Rayon level. Source: Authors from Agency of Statistics. 2.17. In northern Kazakhstan, on account of the very extreme winter climate all wheat is spring sown and depends on a combination of snow-melt and summer rainfall. Planting of spring wheat commences in mid-May once the winter snows have melted and average soil temperatures have achieved the minimum temperatures (12-15˚ C) for seed germination and crop growth. In northern Kazakhstan there is a very narrow fortnight‘s sowing window for spring wheat of between 15 May and 30 May according to the region: wheat which is planted beyond end May is exposed to early autumn frosts from the beginning of September. Given the very low spring and summer average rainfall in much of northern Kazakhstan, the level of winter snow-fall and thus snow melt at the time of sowing is a critical factor in determining the success of the spring wheat crop: indeed farmers and scientists spoken to advised that they could predict in April/May whether the harvest would be successful according to the quantity of accumulated winter snowfall. The main varieties of spring wheat are 90 to 110 day varieties and harvesting normally starts in late August and runs through to mid-September. 2.18. Most spring wheat in Kazakhstan is grown under extensive farming systems using low levels of technology and low production costs. Spring wheat production is mechanized, adopting technical practices which date back to the Soviet Union times. Technology levels are generally low and the use of chemical fertilizers is extremely low for most farmers. In order to produce 1 MT of spring wheat recommended fertilizer rates are in the order of 35-45 Kg of nitrogen, 8-12 Kg of phosphorous and 17-27 Kg of potassium (Arka Consulting 2011): however, it is understood that average fertilizer use is much lower than recommended. In general the PEs have better access to production credit and use higher levels of purchased inputs of seeds, fertilizers and equipment compared to the CFs. In 2010 average costs of production of wheat varied widely - 47 - from a low of about KZT 5,000/Ha (US$ 35/Ha14) for the lowest technology producers to a high of about KZT 20,000/Ha (US$ 140/Ha) and occasionally as high as KZT 25,000/Ha (US$ 170/Ha) for the highest technology producers. 2.19. Wheat prices in Kazakhstan tend to be very volatile. This is evidenced by the variation in September average farm-gate prices paid to farmers between 2006 and 2010 with a low price of KZT 12,600/Mt (US$ 104/MT) in September 2006, prices as high as KZT 31,000/Mt (US$ 261/MT) in September 2008 which was a severe drought year and when total production was significantly reduced, an average of KZT 25,000/MT (US$ 172/MT) in September 2009 which was a good year for spring wheat and then again a major increase in average wheat prices in 2010 which was a very severe drought year to about KZT 35,000/MT. The Food Contract Corporation (FFC) a State Owned Enterprise is a major player in the Kazakhstan grain market and was established by government to maintain grain reserves and to stabilize grain prices for the benefit of producers and consumers: annually it purchases some 10% to 15% of the wheat market and appears to have a strong influence on farm-gate wheat prices paid by the rest of the market. In view of the wide fluctuations in wheat prices returns to wheat production are highly variable: in times of high wheat prices farmers who incur low average costs of production can make high profits if they achieve average yields of about 10 centners/Ha (1 MT/Ha 2.20. Spring Wheat average yields in Kazakhstan are low. On account of the marginal climate, long-term soil degradation and low technology use, average yields for spring wheat, are still low or about10.1 centners per hectare (1.1 MT/Ha) over the past five years. This compares unfavorably with, for instance, Canada, which has similar climate conditions and where yields reach 27 centners per hectare (2.7 MT/Ha) or Australia which has a similar extensive crop system and harvests an average of about 18 centners per hectare (1.8 MT/Ha) of wheat. Spring wheat yield performance is more variable for Commercial Farmers (CFs) than the larger Production Enterprises (PEs). While the spring wheat average yield for CFs is 9.6 centners per hectare for the most recent 5 years, the spring wheat average yield for PEs is on average for the same period 7.5 percent higher or 10.3 centners per hectare. The main reason the PEs, on average, achieve higher average yields for spring wheat than the CFs is due to their higher use of improved seeds and fertilizers and improved tillage practices. Spring wheat yield performance is also uneven throughout the northern areas of the country. The rayons situated in the northern areas of NKO and Kostanay Oblast show the best performance for spring wheat crops. In these areas average spring wheat yields are above 12 centners per hectare. The worst performance for spring wheat crops is observed in WKO, Aktobe15, southern regions of Karaganda Oblast, and south-west of Pavlodar Oblast. In these areas spring wheat rayon-level 5-year average yields are below 6 centners per hectare. Map 2.5 summarizes the geographical distribution of spring wheat yields at rayon level throughout the 8 selected Oblasts in north Kazakhstan region. Further details of this analysis are provided in Annex 1. 14 At a 2010 exchange rate of KZT 145 = US$ 1.00 15 It is noted that average spring wheat yields are somewhat higher in two of the most westerly rayons of WKO and this also applies to two Rayons located in northern Aktobe. - 48 - Map 2.5. North Kazakhstan: Average Spring Wheat Yields per Rayon from 2006-2010 (Centners/Hectare) Source: Authors based on spring wheat yield data provided by Agency of Statistics. 2.21. Spring wheat crop production and yields have improved significantly during the last decade. Average Spring Wheat yields in the main production areas in North Kazakhstan region reach 10.1 centers per hectare. Crop management practices fueled by expanding State subsidies have contributed to higher and more stable wheat yields. Beginning in 2002, government support for agriculture has increased significantly in the form of reduced (subsidized) prices for fuel, seed, fertilizer, and agricultural chemicals. The average wheat yield for 2005 through 2009 is 13 percent higher than the average yield of 1986 through 1990, which was the peak of the so-called intensive technology movement in the Soviet Union. One of the most interesting developments in Kazakhstan agriculture in recent years has been the introduction and spread of reduced-tillage technology. According to the MOA figures, reduced tillage was employed on almost 60 percent of the sown grain area in 2009, including 1.3 million hectares under zero-tillage. The sector has also increased the use of fertilizers and certified seeds. The application rates for mineral fertilizer increased nearly six-fold between 1999 and 2010, and continue to increase due, in part, to the subsidized fertilizer prices. Arguably the most important technological factor contributing to the improvement in Kazakhstan grain yield has been the increase in the use of certified seeds. The government has been providing support to agricultural research facilities, paying 40 percent of the research and development costs for breeder and foundation seeds. Most spring wheat enterprises use only first-reproduction seed. However, the machinery inventories are becoming outdated (particularly in the segment of commercial farmers) and have declined significantly over the past 20 years. Figure 2.1 shows the evolution between 1994 and 2010 of the sown area and yields of spring wheat in the 9 selected Oblasts in North Kazakhstan region. - 49 - Figure 2.1. North Kazakhstan Region: Evolution of Spring Wheat Sown Area and Yields North Kazakhstan Region: Evolution of Spring Wheat Sown Area and Yields 14.0 12.7 16.0 12.4 11.6 11.3 14.0 12.0 10.8 10.1 9.5 9.4 12.0 10.0 8.7 8.4 7.5 7.7 10.0 8.0 7.3 6.3 6.4 8.0 6.0 5.0 y = 1.6032ln(x) + 5.6195 6.0 4.0 R² = 0.252 4.0 4.0 2.0 2.0 0.0 0.0 1998 1999 1994 1995 1996 1997 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Sown Area (million hes) Yield (Centner/he) Log. (Yield (Centner/he)) Source: Authors based on Agency of Statistics data. 2.22. Spring wheat crop production is particularly risky in Kazakhstan. Kazakhstan has the highest yield variation (as expressed by the coefficient of variation, COV in national average spring wheat yields16) of 29 percent, compared to 5 percent in the European Union, or 8 percent in Canada. The northernmost areas of the country are less risky for spring wheat production than the southern, western and eastern areas of the country. In the rayons situated in NKO, north of Kostanay Oblast and northwestern areas of Akmola Oblast the coefficient of variation (CoV) of spring wheat yields is less than 40 percent. In the rayons situated in south of Kostanay Oblast and eastern parts of Akmola Oblast the CoV of spring wheat yields show values between 40 to 50 percent. In Karaganda and western areas of EKO the CoVs for spring wheat are mostly between 50 to 60 percent, except for the rayons situated in the mountainous areas of EKO where the CoVs are between 40 to 60 percent. Pavlodar shows high CoVs for spring wheat production at rayon level: on average the CoVs for spring wheat production for the rayons situated in Pavlodar Oblast are between 50 to 70 percent. The Oblasts situated in the western areas of the country show the highest level of risk for spring wheat production. For instance, in WKO the average CoV at rayon level for spring wheat is between 70 percent and 100 percent. Map 2.6 summarizes the distribution of the CoVs for spring wheat at rayon level in the 8 selected Oblasts in North/Central Kazakhstan. 16 The COV is the standard deviation (SD) about mean annual yield divided by the mean yield and expressed as a percent. A COV of > 100% shows that the SD is larger than the mean yield, or in other words crop yields are highly variable. - 50 - Map 2.6. North Kazakhstan: Coefficients of Variation of Spring Wheat Rayon-Yields Source: Authors based on Agency of Statistics spring wheat data. 2.23. Spring Wheat crops produced by Production Enterprises (PEs) are less risky than spring wheat crops produced by Commercial Farmers (CFs). The PEs through their use of higher levels of inputs and technology tend to achieve both higher and less variable (and therefore less risky) average yields in spring wheat over time than the CFs. The analysis of variance in spring wheat yields for the 17-year series from 1994 up to and including 2010 shows that the observed CoV for PEs is on average 10% lower than the spring wheat yield CoV observed for CFs. This difference is more accentuated in the main spring wheat production areas of Kostanay, NKO, and Akmola where the observed CoV of Rayon-spring wheat yields for PEs is respectively 16 percent, 27 percent, and 25 percent lower than the CVs for CFs in these same Rayons/Oblasts. The main reason for the differences in the CoVs of spring wheat yields between PEs and CFs is the introduction of soil moisture efficient technologies like zero tillage that enables the crop to perform better during the recurrent droughts that effect Kazakhstan‘s grain producing regions. Key Climatic Perils and Impact on Crop Production and Yields 2.24. In northern and central Kazakhstan spring wheat production and yields are highly influenced by climatic and biological factors. Drought is the most pervasive peril affecting rain- fed crop production in northern Kazakhstan. Spring wheat crops can also be damaged by the occurrence of hailstorms, and autumn early frost. Pest and diseases, mainly those fungal diseases like leaf blotch caused by Septoria tritici and rust caused by Puccinia tritici are also common in north Kazakhstan region. 2.25. Spring Wheat Yields in Kazakhstan are highly influenced by the occurrence of droughts17. Reasonably high yields can be achieved during years with adequate rainfall, but the 17 For a more detailed indication of how the correlation coefficients were determined see note to Box 2.2 - 51 - country is subject to frequent drought and is considered a zone of risky agriculture. Historically, Kazakhstan grain production suffers from serious drought two out of every five years. As a result, crop production and yields are marked by frequent and sharp year-to-year fluctuations. The aggregate annual average spring wheat yields for the 8 selected Oblasts in the northern and central regions of Kazakhstan is highly correlated with the cumulated rainfall/snowfall index from January to September as shown by the overall (aggregate for the 8 Oblasts) correlation coefficient18 of 73%. This strong relationship between spring wheat average yields and cumulated rainfall between January and September is also evidenced at the individual Oblast level. In this regards, all the selected Oblasts in northern/central Kazakhstan, except NKO and Karaganda Oblast, show correlation coefficients between annual average spring wheat yields and cumulated rainfall/snowfall between January and September of each year that are above 0.65 (65%). Figure 2.2 shows the relationship between spring wheat annual average yields and the total cumulated snowfall/rainfall from January to September for the period 1994 to 2010. 2.26. Spring wheat farmers in Kazakhstan have suffered severe losses due to the occurrence of drought events. Between 1994 and 2010 crop years, spring wheat has suffered significant crop losses on six occasions: 1995, 1996, 1997, 1998, 2004, and 2010. Under this study an estimate has been made of the value of the historical spring wheat production losses in each year (see summary in Figure 2.3. and Annex 1 for further details). In 1995, a drought affected Kostanay and Karaganda Oblasts causing estimated production losses in the spring wheat crop valued at KZT 74.4 billion (equivalent to a 26 percent reduction in the value of production)19. In 1996, a severe drought affected the western part of the country (WKO and Aktobe oblast) and a .moderate drought affected the eastern oblasts (Pavlodar and EKO). The total losses due to the 1996 event amounted to KZT 46.4 billion (or a 15 percent reduction in the value of production). In 1997, the spring production areas in Kazakhstan were affected by drought again. On this occasion the reduction in the total value of production was only 4 percent of the expected value for that year. The year 1998 was one of the worst years in terms of drought damage in spring wheat crop production. The 1998 drought was particularly severe in WKO, Aktobe, Kostanay, and Akmola Oblasts; but also affected NKO, Pavlodar and Karaganda Oblast. The event caused a reduction of 51 percent of the expected total spring wheat crop value of production for 1998 (a loss of KZT 179.4 billion). 2004 and 2005 were also dry years. The total estimated value of losses amounted to KZT 74 billion (18 percent) in 2004, and KZT 49.1 billion (12 percent) in 2005. The dry conditions in Pavlodar, and East Kazakhstan during 2008 also caused losses on spring crop production. Most recently, in 2010, a devastating drought affected the main spring wheat crop production areas throughout much of the country (Kostanay, Akmola, Karaganda, NKO, and Pavlodar) causing losses amounting to KZT 158.5 billion equivalent to a shortfall of 58% against the expected gross value of spring wheat production of KZT 274.4 billion. Figure 2.3 shows the spring wheat crop production losses from 1994 to crop year 2010. 18 The interpretation of the correlation coefficient or R value in this case is that 73% of the variation in Oblast annual yields is explained by the variation in total January to September rainfall each year: this means that the remaining 27% of yield variation is due to other non-rainfall variables. 19 Losses calculated in terms of loss gross value of production due to yield shortfalls in respect of the expected yield for each of the years, assuming sown area and prices as per the most recent 5-year average. - 52 - Figure 2.2. Northern/Central Kazakhstan: Relationship between Spring Wheat Yields and Total Cumulated Snowfall/Rainfall from January to September Source: Authors from Kaz-hydromet and Agency of Statistics. Note: Correlations are computed by a) cumulating for the January - September period precipitation measured at each of the weather stations for which monthly data has been made available by KHM (see Section 2.6), b) taking the simple average of each pair of stations belonging to the same Oblast, and c) relating such cumulated precipitation index with Oblast level spring wheat yield data. The ―Northern/Central Kazakhstan Rainfall Index‖ is calculated by aggregating the Oblast cumulative precipitation indexes in proportion of the share of each of each of the Oblasts on the average spring wheat planted area for the period 2006-2010. - 53 - Figure 2.3. Kazkahstan: Spring Wheat. National Losses in terms of Gross Value of Production due to Droughts (1994 – 2010) North Kazakhstan : Evolution of Spring Wheat Gross Value of Production 600 (GVP) and Yields 14.0 Yield (Centner/hectare) 500 12.0 GVP (KZT billion) 400 10.0 8.0 300 6.0 200 4.0 100 2.0 0 0.0 1996 2001 1994 1995 1997 1998 1999 2000 2002 2003 2004 2005 2006 2007 2008 2009 2010 Crop Year Actual GVP (KZT billion) GVP Losses (KZT billion) Actual Yield (Centner/he) Expected Yield (Centner/he)/ Expected GVP (KZT billion) Source: Authors, from ARKS 2.27. Hail is reported to be a moderate to severe problem in spring wheat in some parts of the country. Many parts of Kazakhstan experience hail in early and mid summer associated with major rain storms: the months of peak hail activity are May to July as shown by the data on monthly hail incidence for selected weather stations: for example at Kostanay weather station the return period for a hail event in the month of May is one in three years and in August, one in eight years. (Table 2.1). Hail is a localized peril which tends to cause severe damage in wheat at the time of crop maturity and harvest in August/early September: these months tend to have a lower hail exposure. Table 2.1. Kazakhstan: Monthly return period for the occurrence of hailstorms Recurrence Period for Hailstorms for each month throughout the year (in years) Oblast I II III IV V VI VII VIII IX X XI XII Kostanay 42 5 3 6 8 21 84 Akmola 5 8 5 13 63 NKO 7 5 11 5 7 Pavlodar 21 8 9 7 30 10 63 SKO 11 4 3 8 11 21 42 42 42 Source: KHM Hail frequency Statistics (1990-2010) 2.28. Early Autumn Frost can be a problem for spring wheat crop production in some areas of north Kazakhstan. The occurrence of early autumn frosts during late August and the beginning of September may cause damage to wheat crops that were sown late in the season. Early frost damage affects wheat crops when they are in milk grain or dough phenology stages prior to harvest. The damage occurs when the temperatures fall bellow – 2 Celsius degrees for more than two hours. Losses due to early autumn frost on wheat production can be moderate to severe. Historical records of monthly absolute minimum temperatures indicates that the return periods for frost below – 2 Celsius degrees during the last week of September is once in 20 years for some locations in Akmola, Karaganda, Kostanay, and Pavlodar. The probability of having an early frost increases dramatically for each week beyond the last week of September. - 54 - 2.29. Spring wheat is susceptible to locust attacks in northern Kazakhstan. Locust attacks are not infrequent in north Kazakhstan and the country has suffered recurrent crop and pasture damages from locusts. There are 2 main species of locusts in Kazakhstan (i) the ―Asian‖ locust which is not considered a major problem and (ii) the Italian locust ( Callitamus italicus) which is the most common and dangerous pest in nearly all of Kazakhstan: in 1999 the country experienced a severe outbreak of Italian locusts and as this species has a peak cycle of every 10 to 12 years the next outbreak is expected in 2010 or 201120. The scale of the problem increased dramatically after independence (during1996–2001), when cessation of state subsidies for wheat production in the northern steppe lands led to the abandonment of up to one third of the former wheat lands. The resulting mosaic of weedy fields, pastures, and bare ground provides ideal breeding grounds for locusts. In 1999, more than seven million hectares of land were invaded by Italian locusts and 220,000 hectares of crops were destroyed caused a total damage equivalent to US$ 15 million. In 2008, again more than 200,000 hectares of crops were destroyed in SKO21. The Government was forced to conduct a massive chemical control campaign throughout the country with assistance from Food and Agriculture Organization (FAO). 2.30. Spring wheat fungal leaf diseases are also a problem for spring wheat production in Kazakhstan. Despite the dry climate, cultivation of susceptible varieties results in epidemics of leaf rust on average in 1 year out of 4, affecting over a million hectares with yield losses of up to 25–30%. Excess moisture and high humidity during the month of July generates the conditions for the development of rust. Most of the wheat cultivars planted in the region are susceptible to leaf rust and only recently have several resistant lines and new varieties have been tested in trials22. Stripe rust, caused by Puccinia striiformis f. sp. tritici is considered the most important disease of wheat in Central Asia and the Caucasus (CAC). Although stripe rust has been present in the region for a long time, it has become a serious constraint to wheat production in the past 10 years. Rust attacks were observed in northern areas of Kazakhstan in 2007 and in 2009. Assessment of crop production risk exposures Spring Wheat Values at Risk 2.31. The total Spring Wheat values at risk (VAR) in north Kazakhstan region is valued at KZT 452 billion (about US$ 3.0 billion)23. The bulk of spring wheat production in Kazakhstan is concentrated in a relatively small area. Out of the KZT 452 billion of total spring wheat values at risk (VAR), 87 percent (KZT 392 billion) is concentrated in a relatively small area of approximately 240,000 square kilometers which comprises NKO, the northern rayons of Kostanay Oblast, and northern and western rayons of Akmola Oblast. The remaining 13 percent of the spring wheat VAR, or KZT 56 billion, is distributed throughout a vast area that comprises 20 IRIN Asia 2007. Kazakhstan: Locust invasion in west under control officials say. Almaty 10 July 2007. http://www.irinnews.org/report?reoportid=73115 21 Locust in Kazakhstan. Presentation made A. Latchininsky & R. Sivanpillai. University of Wyoming 22 Leaf rust of spring wheat in Northern Kazakhstan and Siberia: incidence, virulence, and breeding for resistance* A. Morgounov A D, L. Rosseeva B and M. Koyshibayev CAustralian Journal of Agricultural Research 58(9) 847–853 http://dx.doi.org/10.1071/AR07086 Submitted: 8 March 2007 Accepted: 8 June 2007 Published: 28 September 2007 23 For the purpose of valuate spring wheat production an average price of KZT 3,210 per center was considered. This price is the result of the average spring wheat farm gate price for the month of harvest (September) for most recent three crop seasons (2008, 2009, and 2010) - 55 - EKO, Pavlodar, Karaganda, Aktobe, WKO, rayons located in the south of Kostanay Oblast, and rayons located in the east of Akmola Oblast. Map 2.7 summarizes the geographical distribution of spring wheat VARs at rayon level throughout the 8 selected Oblasts in north Kazakhstan region. Further details of this analysis are provided in Annex 1. 2.32. The fact that 87 percent of spring wheat exposure is concentrated in a relatively small geographic area of northern Kazakhstan has severe implications for crop insurers in terms of the chance of face catastrophic losses on spring wheat crop production. This fact indicates a huge risk accumulation in the area comprised by NKO, the northern rayons of Kostanay Oblast, and northern and western rayons of Akmola Oblast and an increase of the chance of having a systemic event affecting the main part of the spring wheat crop portfolio. 2.33. The major spatial differences in spring wheat VARs will need to be addressed carefully in the re-design and strengthening of the Obligatory crop scheme for spring wheat production. The principle of any crop insurance scheme is to ensure that an optimal spread of risk is achieved both spatially and temporally. The concentration of VARs in NKO, the northern rayons of Kostanay Oblast, and northern and western rayons of Akmola Oblast will need to be assessed closely under the redesign and strengthening of the existing obligatory crop insurance scheme and possible introduction of new crop insurance products and programs. Map 2.7. North Kazakhstan: Spring Wheat Exposures in North Kazakhstan Region Source: Authors from Agency of Statistics. Expected Value of Spring Wheat Crop Losses 2.34. An estimation of the expected value of crop losses for the spring wheat crops at rayon level for each of the 9 selected Oblasts in north Kazakhstan region has been conducted. The estimation of the expected value of crop losses for spring wheat crops was based on an analysis of variance in time-series average seasonal rice yields in each zone under the CRAM model (See Annex 1 for full details of the model‘s assumptions). 2.35. The analysis of expected losses shows that spring wheat production in Kazakhstan is extremely risky with annual average expected losses valued at KZT 66.5 billion (about US$ 443 - 56 - million). The annual average expected losses for spring wheat is valued at KZT 66.5 billion per crop year (US$ 443 million), equivalent to 14.7 percent of the total spring wheat VAR of KZT 452 billion. However, within north Kazakhstan region there are some Oblasts and Rayons which are more risky than others. Map 2.8 shows the average expected losses for spring wheat crop production expressed as a percentage of the VAR in each of the rayons in northern Kazakhstan. 2.36. The highest average annual expected losses in spring wheat applies to Aktobe Oblast and WKO located in western Kazakhstan. Within these 2 Oblasts there are only 2 rayons in western WKO and two rayons in northern Aktobe which have annual average expected losses of less than 25%. In all other rayons the average annual expected losses for spring wheat are extremely high at between 25 percent to over 38 percent of the total values at risk. These annual average expected losses are equivalent to a total loss of spring wheat production one in every three to four years. This indicates that the western zone of Kazakhstan is very marginal for spring wheat crop production. 2.37. Conversely, spring wheat production is much less risky in the northern Oblasts of NKO, Kostanay and Akmole. In rayons such as those situated in NKO, north eastern and southern areas of Akmola Oblast and the northern area of Kostanay Oblast, the annual average expected loss for spring wheat ranges between 10 percent to 15 percent of the total values at risk. Pavlodar Rayon and the western areas of EKO can be considered as intermediate in terms of risks for spring wheat production. In these rayons the average expected losses that ranges between 18 percent to 27 percent. Last, there is a vast geographic area comprising center and south of Kostanay Oblast, center and east of Akmola Oblast, most of Karaganda Oblast and the eastern rayons of EKO that has acceptable levels of expected losses. In these areas the average expected loss per hectare averages from 15 percent to 18 percent of the total values at risk. Map 2.8. North Kazakhstan: Expected losses for Spring Wheat (% total values at risk) Source: Authors from CRAM. 2.38. The analysis of expected losses also shows that spring wheat produced by commercial farmers is much more risky that the spring wheat produced by production enterprises. Annual expected losses for spring wheat produced by PEs are 22 percent lower, on average for the whole north Kazakhstan region, than the annual expected losses for spring wheat produced by CFs. While the annual average expected losses for spring wheat produced by CFs is 17.24 percent of - 57 - the total spring wheat VAR produced by CFs, the average expected losses for spring wheat produced by PEs is only 13.39 percent. It is understood that the main reason for these differences in terms of spring wheat expected losses between PES and CFs is because the PEs are able to manage their production risks better (they have more working capital and better machinery) and are applying soil moisture conservation technologies to the crops (e.g. zero tillage). The fact that the expected losses of spring wheat crops grown by PEs is lower than the expected losses for CFs is an important finding that should be taken into consideration in the design of any individual grower loss of yield crop insurance scheme. The eventual modification of the obligatory crop insurance scheme should recognize the risk management efforts implemented by the different types of insured. 2.39. The analysis of 17-year (1994 -2010) rayon-level yields for spring wheat in NKO shows that 1998 was the worst loss year in this series with total production losses of 7.7 million metric tons of spring wheat, which was equivalent to a financial loss of 41.6 percent of the total expected value of spring crops in northern Kazakhstan. Although 1998 was a severe loss year for spring wheat production in northern Kazakhstan, even worse crop losses could occur in future. From an insurance view point, underwriters need to know with a high degree of confidence the maximum losses that they might incur (termed the Probable Maximum Loss, PML 24) either 1 in 100 years, or if it is necessary to be even more conservative, 1 in 250 years. This information is an invaluable aid to structuring an insurance and reinsurance program and to determining how much capital must be reserved to cover the PML loss year. 2.40. The results of the World Bank’s PML loss cost analysis at 100% yield coverage level show that the 1 in 10 year expected PML is equivalent to a loss of 34.0 percent of the total value at risk (VAR) of the north Kazakhstan region spring crop or a loss of KZT 153.6 billion (US$ 1.02 billion) , while the 1 in a hundred year PML loss is calculated at 54.61 percent of the total VAR for spring crop in north Kazakhstan region or a loss of KZT 246.8 billion (US$ 1.6 billion). These PML estimates show that the spring wheat crop in north Kazakhstan region is very exposed to catastrophic (mainly drought) losses and that these losses greatly exceed the retention capability of local insurance companies. The preliminary PML presented in this report will need to be developed in future and used to (a) establish the catastrophe loading which must prudently be added to the calculated base rates and (b) assist in the design of a risk financing and risk retention and risk transfer/reinsurance strategy for the Kazakh insurance market. 24 The Probable maximum Loss is defined as ―An estimate of the maximum l oss that is likely to arise on the occurrence of a single event considered to be within the realms of probability, remote coincidences and possible but unlikely catastrophes being ignored‖. - 58 - Figure 2.3. North Kazakhstan: Probable Maximum Losses for the Spring Wheat Portfolio. 70% Percentage of Loss over the total 60% 61% - 277 billion 60% - KZT 271 billion 57% - KZT 263 billion 50% 54% - KZT 247 billion 50% - KZT 225 billion 40% VAR 34% - KZT 154 billion 30% 20% 10% 0% 0 50 100 150 200 250 Return Period (years) Source: Authors from CRAM. 2.41. The PML loss cost analysis at 100% yield coverage level show that Commercial Farmers are more exposed to catastrophic losses in north Kazakhstan than the Production Enterprises. While the 1 in 10 year expected PML for CFs is equivalent to a loss of 38.0 percent of the total value at risk (VAR) of these types of farmers in the north Kazakhstan region, the 1 in 10 year expected PML for PEs shown by the same analysis is 33.7 percent of the VAR for the PEs. The 1 in 100 year expected PML for CFs is equivalent to a loss of 58.0 percent of the CFs‘ VAR in the north Kazakhstan region and the 1 in 100 year expected PML for PEs is 55.7 percent of the PEs‘ VAR. Conclusions to Spring Wheat Risk Assessment 2.42. The analysis of rayon level crop production and yields for spring wheat in north Kazakhstan region shows that this crop is heavily exposed to losses caused by droughts. This is evidenced by the average loss cost estimated by the CRAM for a 17-year period, 1994 up to 2010 of 14.71 percent of the total value at risk of spring wheat production and a calculated 1 in 100 year PML of 54 percent of the national spring wheat gross value of production. 2.43. The design of any crop insurance program for spring wheat in north Kazakhstan region should take into account the differences in expected yields and yield variability by rayon. The analysis of spring wheat expected yields and expected yield variability for the different rayons in north Kazakhstan region, show different levels of risk and, therefore, different rates across the different rayons in the region. 2.44. The design of any crop insurance program for spring wheat in north Kazakhstan region should also take into account the differences in terms of expected yields and risk between the Production Enterprises and the Commercial Farmers. In Kazakhstan there are important differences in technology levels used by different types of farmers to grow spring wheat and even when they are situated in the same region, they obtain different yields and/or their crops perform differently under similar crop-stress (e.g. drought) situations. In the case of spring wheat production in north Kazakhstan, PEs perform better than CFs, both in terms of yield performance and stability of such performance. These facts that differentiate the spring wheat production of PEs and CFs should be taken into consideration in the designing of future crop insurance products and schemes in Kazakhstan. - 59 - Chapter 3: Review of Kazakhstan Crop Insurance Program 3.1. This Chapter presents a review of the Compulsory Crop Insurance Program including its key features, results and financial performance and highlights the issues and drawbacks of the current program and then in Chapter 4, options for strengthening the program are considered. Policy and Regulatory Framework for Crop Insurance 3.2. Kazakhstan has a history of State-supported compulsory agricultural insurance. During the Soviet period a national compulsory crop insurance scheme operated from 1970 to 1991 and all state farms and collective farms participated in this scheme which was based on a multiple-peril crop insurance cover which provided limited indemnity for loss of production costs invested in growing the crop. Following independence in 1991 there was no crop insurance in Kazakhstan for the next decade. 3.3. The current crop insurance system in Kazakhstan is a Public-Private Partnership (PPP) that was created by Law in 2004. The Law № 533-II of March 10th, 2004 established the legal and regulatory framework and financial and organizational basis for the implementation of a national crop insurance program in Kazakhstan. (A copy of the Law is attached in Annex 2). The law sets out the terms and conditions for the implementation of a compulsory salvage-based Loss of Investment Costs (LIC) crop insurance program for farmers growing a range of strategic cereal, oilseed and other field crops, in Kazakhstan. The implementing agencies of the current crop insurance system in Kazakhstan are the Direction of Strategic Planning of the Ministry of Agriculture (MOA)), the Fund for Financial Support for Agriculture (FFSA), the private and mutual insurance companies and finally the local authorities in each Oblast and Rayon. 3.4. The crop insurance law in Kazakhstan is aimed at meeting three objectives . The first objective is to protect the famers‘ against loss of their crop production due to the effects of adverse weather events. Kazakhstan is one of the most exposed countries in the world to drought losses in spring wheat, followed by late spring and harvest frosts, hail and to a lesser extent excess rain and flood and also to crop pests and diseases. GRK seeks to protect small and medium farmers from going bankrupt following major crop losses by providing subsidized crop insurance for the main crops grown in Kazakhstan. The second objective is to assist farmers to gain access to rural finance by protecting their crop loans against default due to weather induced crop failure. The third objective is to assist in enhancing the effectiveness of government support programs for crop production. 3.5. The current crop insurance system in Kazakhstan is compulsory by law. The law states that all the farmers in Kazakhstan who produce a series of strategic crops including cereals, oilseeds, sugar beet and cotton are obliged to purchase crop insurance; otherwise, farmers cannot obtain access to other government subsidized programs supporting agriculture and are subject to the application of financial penalties by the GRK. Under the compulsory crop insurance program all private commercial insurance companies and mutual farmers‘ associations licensed to operate crop insurance are equally obliged to offer crop insurance and to insure all the insurance proposals received from the farmers regardless of the quality of the risk. The only reason that an Insurer may refuse to conclude a compulsory contract of insurance cover is if the Insured farmer has failed to submit his or her application and to complete the insurance contract including the payment of premium by the agreed policy inception cut-off date. - 60 - 3.6. The Law of Compulsory Crop Insurance is very comprehensive and specifies the terms and conditions of the standard crop insurance policy which is offered throughout Kazakhstan. The Law of 2004 prescribes the terms and conditions of the standard Loss of Investment Costs (LIC) crop insurance policy that all insurance companies are obliged to adhere to, including the insured crops and insured perils, the amount of the sum insured for each crop grown in each Oblast and Rayon and which is based on 3 optional levels of crop production investment costs, through to specification of the minimum and maximum premium rates that can be charged for each crop in each Oblast of the country. The Law also states the basis of indemnity and loss assessment procedures that apply on the standard LIC Policy. 3.7. The Law also specifies the GRK’s Financial Support to the Compulsory Crop Insurance Program. In addition to its statutory and regulatory roles in setting the terms and conditions of the compulsory crop insurance programs, GRK also provides financial support to the program in the form of a 50 percent reimbursement of paid claims to the private and mutual insurance companies each year. The settlement of the 50% claims is administered through an Agent appointed by GRK, in this case, the Fund for Financial Support for Agriculture (FFSA). Under the law, GRK is also responsible for funding the FFSA‘s operating costs. Compulsory Crop Insurance Policy Terms and Conditions 3.8. The Kazakhstan compulsory crop insurance policy is a Loss of Yield Policy which indemnifies the Insured when the value of the harvested production falls short of the costs invested in growing the crop due to the action of insured perils. The policy is sometimes referred to as a Salvage-based loss of yield because it only indemnifies crop production losses at the point when the sale value (revenue) of any residual harvestable crop production (termed the ―salvage‖) is inadequate to cover the costs of production invested in growing the crop up to the time of loss. Alternatively this policy is called a Loss of Investment Costs Crop Insurance policy (―Loss of Investment Cost‖ policy for short). 3.9. A key potential advantage of the Loss of Investment Cost (LIC) Policy is that it can be used in situations where there are no accurate historical individual farmer-level crop production and yield records. A conventional individual grower Loss of Yield Policy requires each grower to provide between 5 and 10 years of their actual crop production and yields in order to establish an average or normal yield and to then establish an Insured Yield against which to measure yield reduction or loss due to the action of insured perils. In Kazakhstan it is understood that few farmers can provide an accurate crop yield history. The Loss of Investment Cost (LIC) Policy can, however, be offered to individual growers in situations where yield history data is not available as the basis of indemnity does not depend on measuring yield loss against a pre- established Insured Yield. The LIC policy establishes a sum insured based on the production costs invested in growing the crop and in the event of a loss due to an insured peril(s) the policy makes an indemnity payment where the estimated sale value of the salvage is inadequate to cover the costs invested in the crop at the time of loss. 3.10. The main drawback of the LIC policy is that in the event of partial crop area or production losses, the actual remaining or salvageable crop yield has to be measured on a field by field basis and the value of the salvage estimated – this is a time consuming and costly loss adjusting exercise - 61 - 3.11. The Loss of Investment Cost (LIC) policy has been extensively promoted for more than 30 years in Mexico where it is usually linked to bank lending in the form of seasonal crop loans, in other parts of Central and South America and also in several Eastern European countries. 3.12. This Section presents a review of the key features of the Kazakhstan Compulsory Loss of Investment Costs (LIC) Policy. A summary of the main terms and conditions of the LIC policy wording which is termed the Standard Form of Compulsory Crop Insurance Contract) is contained in Box 3.1 and full details are presented in the Policy Wording in Annex 2. Box 3.1. Summary of Terms and Conditions of Cover of Kazakhstan Standard Compulsory Loss of Investment Costs Crop Insurance Policy Item Detail Type of Cover Loss of Investment Cost Policy. The underlying basis of insurance and indemnity is a salvage-based loss of crop yield policy Insured Interest – Crops Spring and Winter Wheat, Spring and Winter Barley, Winter Rye, Buckwheat, Oats, Millet, Maize (grain), Chick-pea, Pea, Brassica, Rice, Sunflower, Safflower, Soyabean, Sugar Beet, Cotton Location Republic of Kazakhstan: All crop growing regions Criteria for Acceptance of Crop Insurance is compulsory for all producers of the above insurable crops Risk / Compulsion of Cover throughout Kazakhstan. Insured Perils Cover is provided against loss or damage to crop production due to ―adverse weather events‖, defined as:  Natural Phenomena Long-lasting: Drought, Frost-killing, Lack of Warmth (Low Temperature), Excess Moisture in the soil, Excess Moisture in Air (Excess Humidity), Flood, Shallow Dry Wind and  Natural Phenomena Short-lasting: Hail, Excess Rain, Frost, Strong Wind, Mudflow. Cover Period From the time of Sowing of the Insured crop through to completion of harvest. Insured Unit For each farmer, the Insured Unit is defined as a ―field‖. The Insured is obliged to declare and insure each and every separate field of the Insurable Crop(s) and to submit a map of the field locations. Sum Insured The Sum Insured is based on the Normative Costs of Production for 1 hectare of the Insured Crop, multiplied by the Insured Area. The Insured is permitted to selected from 3 optional levels of Normative Costs of Production:  Science-based agricultural technology  Simplified agricultural technology  Costs of: fuel & lubricants, seeds, wages Deductible (Franchise) The Compulsory Insurance Law prohibits the use of franchises or deductibles. Basis of Indemnity and Where the action of Insured Perils causes the Actual Value of Harvested Claims Settlement Production (Salvage) to fall short of the Costs of Production Invested in growing the crop (the Sum Insured), the policy will indemnify the amount of shortfall. In the event of a Total Loss, the indemnity is the Sum Insured for the 100% damaged area. In the case of a Partial Loss of area and or yield, the salvage (harvestable production) from the affected area is estimated in field at the time of harvest and valued at the prevailing sale price of the crop. Where the Value of Salvage (Crop Revenue) falls short of the Investment Costs the shortfall is - 62 - indemnified. Exclusions Any losses which occur due to causes other than ―adverse weather events‖ Other Conditions Cover is only binding from the time of payment of premium by the Insured Source: Authors based on Standard Form of Compulsory Crop Insurance Contract (See Annex 2) Compulsion of Cover 3.13. Kazakhstan is one of very few countries in the World where crop insurance cover is compulsory for all farmers who grow cereals, oilseeds and other strategic crops. As noted above, in Kazakhstan, crop insurance cover is legally enforceable under the Law № 533-II of 2004 and where a farmer deliberately avoids purchasing crop insurance this is penalized by heavy fines which are again enacted by Law. Kazakhstan is one of a handful of countries where governments have made crop insurance compulsory. Currently there are about 100 countries offering some form of public, or private or PPP agricultural insurance and countries where agricultural insurance is compulsory include China (compulsory cover only applies to swine epidemic disease cover), Cyprus (compulsory government crop insurance for all crops), Japan (crop insurance is compulsory for wheat and rice only), Kazakhstan (crop insurance is compulsory for major strategic crops, but is voluntary for livestock) North Korea (compulsory national rice and maize insurance program), Netherlands (compulsory livestock epidemic diseases insurance), Switzerland (compulsory livestock epidemic disease cover) and the Windward Islands (compulsory windstorm cover for export bananas) (Mahul & Stutley 2010). In all other countries agricultural insurance is either purely voluntary or in some cases borrowers of credit are obliged by the lender to purchase crop-credit insurance protection. The potential advantages and disadvantages of compulsory national crop insurance schemes are reviewed later in this Chapter. Insured Crops 3.14. Under the 2004 Compulsory Crop Insurance Law insurance is mandatory for a list of 17 strategically important food and export crops including spring wheat and other grains, oilseeds, sugar beet and cotton. A full list of the compulsory insurable crops is given in Box 3.1. Spring wheat is the most important crop accounting for over 90% of the program‘s Total Sum Insured (TSI) over the past 6 years (2005 to 2010). Insured Perils 3.15. The Compulsory LIC Policy insures a wide range of adverse climatic perils . The insured perils include a broad range of climatic perils including drought, winter freeze, frost, low temperature, hail, excess rain, flood, water-logging (excess soil moisture), excess humidity, wind (hot dry winds and strong winds) and one natural peril, mud-flow which may or may not be associated with excess rain. 3.16. The Compulsory LIC Policy excludes all other causes of crop loss including the natural peril of fire and all biological perils (pests and diseases). It is understood that plant pests and especially fungal diseases in cereals are a major cause of loss in adverse climatic years in the RK. In most other countries which operate similar LIC policies, coverage is usually ―All Risk‖ and includes all natural, climatic and biological perils (unavoidable and uncontrollable pests and diseases) which result in loss of expected crop production and crop revenue. In Kazakhstan it is likely that loss assessment is more complicated because the adjustment would need to take into account losses due to insured adverse climatic events and to separate these from other uninsured causes of loss. This is a theme which is discussed further in the Chapter 4. - 63 - Definition of Insured Unit 3.17. The definition of the Insured Unit is critical for adjusting and indemnifying losses under any crop insurance scheme. In Kazakhstan the Insured Unit is the separate or individual ―Field‖. The policy requires farmers to declare and insure all their fields sown with the insurable crop(s) and to provide maps and schedules showing the location and area of each field and crop(s). With the lack of field boundary fences and or roads, the definition of what constitutes a separate field may, however, be very open to interpretation at the time of loss assessment. Cover Period 3.18. The principle of most crop insurance programs is to provide protection to the crop during the growing season and to terminate cover on completion of the harvest. There are, however, a great many variants on the inception dates of cover and some multiple-peril crop insurance (MPCI) loss of yield programs will only incept cover once the sown crop has germinated and is emerged and a full stand has been established (e.g. Spain and Portugal), while other programs will insure against loss of the sown seeds due to perils which cause germination failure. The LIC program in Mexico is an example of a program which insures crops from the time of sowing against natural and climatic events and biological perils which lead to germination failure. In a few cases, including the USA‘s Federal Crop Insurance Program, FCIP, coverage may even extend to ―prevented sowing‖, for example, due to extreme drought or excess rain preventing access to machinery to sow the crop. 3.19. The Cover period of the Compulsory LIC Policy is not clearly defined. Article 9 of the LIC policy wording states that the contract is deemed valid and binding for all Parties from the moment of payment of the Insurance Premium and is in force until an agreed termination date. The Law on Compulsory Crop Insurance, stipulates that the contract of insurance must be concluded and in place no later than 15 days after completion of sowing: it is, however, very unclear whether the intention of the policy is to provide back-dated coverage from the time of sowing of the crop, and also whether losses that occur within the 15 day period would be deemed insured or not (ARKA 2011). Under a voluntary crop insurance program it would be unacceptable to permit farmers to purchase cover up to 15 days after completion of sowing because of the potential for anti-selection and normally such polices carry a sales cut-off date of at least 30 days prior to the start of sowing25. In Kazakhstan it might be argued that anti-selection is minimized because the program is compulsory for all farmers: however, the practice of allowing farmers to purchase cover up to 15 days after completion of sowing may lead to moral hazard – for example where crop germination is poor the famer may reduce his husbandry and management standards thereby accentuating the loss of crop production in the knowledge the policy will pay out a claim. Premium Rates 3.20. Government is responsible for setting the reference premium rates on the Compulsory Crop Insurance Scheme and these rates are declared in the Law of 2004 and subsequent amendments. In 2003 GRK through the MOA commissioned a crop insurance premium rating study for the LIC crop insurance policy through the Kazakh Actuarial Center (CJSC). CJSC used 11 year (1991-2001) MOA crop-area damage data at Oblast level to design a system of regional 25 Examples where anti-selection could arise under a voluntary program include where pre-existing adverse climatic conditions are developing at the time of sowing (e.g. a major freeze or drought) and farmers purchase cover knowing there is a high probability of crop failure. - 64 - or Oblast-level technical premium rates for the major insured crops (grains, oilseeds, sugar beet and cotton). For each major crop type in each Oblast, the average loss cost rates were calculated according to the percentage of total sown area damaged and with adjustment by spreading catastrophe loss years in particular Oblasts (e.g. WKO) over the whole portfolio. The technical rates were then reduced by 50% on account of the government 50% claims subsidies and finally loaded for Insurers administration and operating expenses and profit margins by load factors of between 10% and 25% 26. Further details of the CJSC crop rating methodology are presented in Annex 2. The final commercial premium rates were published in the 2004 Compulsory Crop Insurance Law and were last updated in 2008 when a system of minimum and maximum reference rates was introduced for grain crops, since when they have remained unchanged. 3.21. For each crop and group of Oblasts, the GRK premium rates are presented in terms of a minimum and a maximum premium rate that may be charged by the Insurance Companies. A list of the current reference premium rates that apply in 2011 are contained in Table 3.1. In the higher rainfall growing areas of northern and eastern Kazakhstan the premium rates for grains are between 1.78% and 3.48%, rising to between 5.21% and 9.15% in the most drought prone areas of Western Kazakhstan including the Oblasts of Aktobe and WKO. For all other classes of crops including oil seeds, sugar beets and cotton the minimum and maximum reference rates are the same for all Oblasts or in other words there is no regional differentiation of the premium rates. 3.22. Since the GRK reimburses the insurers for 50 percent of the claims free of cost, the MOA determined technical premium rates are correspondingly adjusted downwards to reflect only 50% of the expected claims costs. These reduced technical rates are then loaded for acquisition costs and insurers‘ operating expenses and reasonable profit margins to derive the MOA approved minimum and maximum commercial premium rates or reference rates (Table 3.1). As such, farmers receive an implicit premium subsidy of 50 percent of the amount of premiums they have to pay. Table 3.1. Crop Insurance Commercial Premium Reference Rates set by Law Premium Rate (%) Crop Group Name of Oblast Minimum Maximum Аkmola, Аlmaty, EKO,Zhambyl, Kostanay, NKO 1.78 3.48 1) GRAINS Кaragandy, Kyzylorda, Pavlodar, SKO 3.17 5.83 (CEREALS) Аktobe, WKO 5.21 9.15 2) OILSEEDS National (applicable in all Oblasts where crop is grown) 2.01 3.44 3) SUGAR BEET National (applicable in all Oblasts where crop is grown) 5.76 8.39 4) COTTON National (applicable in all Oblasts where crop is grown) 0.92 Source: Compulsory Crop Insurance Law 2004 Sum Insured 3.23. In Kazakhstan the Sum Insured of the compulsory LIC policy is determined by a “Normative Cost of Production” per hectare, which is established by MoA per crop and Oblast 26 Full details of the CJSC‘s 2003 crop insurance premium rating methodology are set out in : CJSC (2003) Tariff Calculation Methodology for Comulsory Crop Insurance in the Republic of Kazakhstan. Report prepared for the Ministry of Agriculture of the Republic of Kazakhstan, Kazakh Actuarial Center, CJSC, November 18, 2003 - 65 - and agro-ecological zone in the country. The Normative Costs of Production are approved by GRK and are then published by Law. 3.24. In each Oblast Farmers are permitted to chose between 3 optional levels of Sum Insured for each insured crop according to technology levels. The three optional sum insured levels correspond to high technology/high production costs cultivation (Science-based Agro- technology), medium technology (Simplified Agro-technology) and low technology (based on three production input cost items, wages, fuel and seeds). Farmers are free to elect to insure their crop at any technology level irrespective of their actual technology levels and production cost structures. Table 3.2 presents a summary of the three levels of average normative costs or sums insured in KZT per hectare for each crop type which applied in 2009 and 2010. 3.25. The per hectare sums insured are in general very low reflecting both the low average technology levels adopted by Kazakhstan’s farmers and therefore the low costs of production for most crops and also government’s desire to manage the financial exposure on this compulsory national crop insurance program by capping the maximum permitted sums insured. Reference to Table 3.2 shows that in 2010 for the highest Science-based Agrotechnology level, the average sum insured for all crops was only KZT 9,908/Ha (US$ 68/Ha)27 with range from a low for millet of KZT 4,830/Ha (US$ 33/Ha) to a high for sugar beet of KZT 40,646/Ha (about US$ 280/Ha). For the lowest sum insured option based on the three production cost items of wages, fuel and seeds, the average sum insured for all crops was a very low TZK 3,674/Ha (US$ 25/Ha) in 2010. For spring wheat which is the most important insured crop grown in Kazakhstan, 2010 average sums insured ranged from a maximum of KZT 8,829/Ha (US$ 61/Ha) down to KZT 3,407/Ha (US$ 23/Ha) for the lowest sum insured option (Details of 2010 wheat Normative Costs by Oblast and agro-ecological risk zone are included in Annex 2). Table 3.2. 2009-2010 Average Normative Costs of Production by Crop Type (KZT/Hectare) Crop Science-based Simplified Agro- Costs based on salary, fuels Agro-technology Technology & lubricants and seeds Brassica napus 8,409 6,230 2,353 Buckwheat 8,482 5,848 3,180 Chick pea 7,990 6,101 3,699 Cotton 24,060 14,211 9,706 Grain maize 25,804 16,836 8,640 Millet 4,830 3,392 1,731 Oat 6,904 4,682 2,987 Pea 8,064 5,780 3,935 Rice 33,524 21,459 5,130 Safflower 8,323 6,407 3,311 Soya 10,624 6,822 4,803 Spring Barley 7,676 5,111 3,104 Spring Rye 6,518 3,874 2,449 Spring Wheat 8,829 5,831 3,407 Sugar Beet 40,646 23,469 7,583 Sunflower 9,026 5,661 3,824 Winter Wheat 9,361 6,143 3,699 Average all crops 9,909 6,540 3,674 Average all crops (US$) 68 45 25 Source: Authors‘ calculation of average normative costs per crop based on Regulation 25 March 2009 No 410. 3.26. The Normative Costs per Hectare are multiplied by the number of insured hectares in each field to derive the Total Sum Insured (TSI) for each crop in each Insured Unit. 27 A 2010 exchange rate of KZT 145 = US$ 1.00 has been used in this analysis. - 66 - Basis of Indemnity 3.27. The policy indemnifies the farmer when the expected revenue (actual remaining crop production times the prevailing farm-gate sale price) from the insured crop in the insured unit (which in Kazakhstan is defined as the individual “field”) falls short of the sum insured (investment cost) insured by the farmer. In such cases, the insurance policy indemnifies the famer by the difference between the sum insured (investment cost) per hectare less the expected revenue per hectare times the area of the insured unit. 3.28. Damage is assessed by in-field sampling by a committee of up to 5 persons (organizations). The field loss assessment exercise is designed to assess i) the cause of loss to verify whether this is due to an insured (uninsured) peril(s), ii) the affected / damaged area of the crop in each Insured Unit and iii) whether damage to the crop is a Partial or Total loss. The procedures for indemnifying total and partial losses are explained below. . 3.29. Where a total crop loss occurs, the loss assessment and indemnity procedures are relatively simple. In the case of total crop loss the main task at the time of in-field crop loss assessment is to measure the area (in hectares) of totally (100%) damaged crop due to insured perils. The damaged area is then multiplied by the per hectare normative costs of production (sum insured) selected by the Insured to derive the total value of the claim. No deductible is applied to the claim. The LIC Policy Wording is, however, ambiguous as to whether the total loss must apply over the entire area of the declared Insured Unit (field) in order to qualify for a claim, or whether a total loss in any hectare of the insured crop will open the policy for a claim. This lack of clarity over the definition of the Insured Unit may lead to misunderstandings between the Insurer and the Insured at the time of loss assessment. Also the policy does not clearly state what constitutes a total loss and does not appear to carry a Constructive Total Loss Clause.28 3.30. In the case of partial losses due to insured perils the loss assessment procedure and basis of indemnity is considerably more complicated. In the case of partial losses to crop production and yields, in-field assessment is required using 1 metre square crop-cut samples located at random in the field to estimate the actual remaining harvestable yield (salvage) in the affected area. Loss assessment is a costly and time consuming process and the Law stipulates that a Committee of five persons (including the local Rayon executive authority, the FFSA representative, Insurance Agent, Insurance Company and the Insured farmer) must inspect the effected field(s), and determine the damaged area and within this area the amount of harvestable yield (salvage). The value of salvage must then be calculated by one of 2 ways: (i) according to the actual sale value of the crop received by the farmer; or (ii) prior to sale of the crop by using an estimate of the local sales price for the crop. If the value of salvage is lower than the sum insured (investment costs), the difference is indemnified. A worked example is given in Box 3.2. There is, however, a major drawback of this procedure namely that because an insured price for valuing the salvage is not pre-agreed at the time of binding the insurance contract, neither the Insurer nor the Insured have any clear idea of the amount of indemnity they may receive in the event of partial crop losses (this issue is discussed further in the next section). 28 Under the Mexican LIC Policy there is a Total Constructive Loss Clause (CTL) under which a Total Loss is defined as ―where more than 90% of the sown crop stand and expected production has been lost and where it is not economic to harvest (salvage) the remaining proportion of the crop because the harvest costs would exceed the sale value of the salvage‖. It is not clear whether a CTL definition applies under the Kazakhstan Compulsory LIC Policy - 67 - Box 3.2. Example of Indemnity Calculations for Partial Crop Losses SUM INSURED DETAILS: Insured Crop: Sunflower Insured Field Number: 53 Area of Insured Field (hectares): 300 Sum Insured per hectare (KZT per hectare): 3,390 Total Sum Insured field No 53 (KZT): 1,017,000 Cause of Loss: Drought PARTIAL LOSS: CLAIMS CALCULATION: Affected area: 300 Ha, partial losses Harvestable Yield (salvage) (MTons): 5.5472 Sale price for sunflower (KZT/MTon): 55,000 Value of Salvage (KZT): 305,100 Value of Salvage per hectare (KZT/Ha): 1,107 (305,100 ÷ 300) Difference between Sum Insured and Salvage Value (KZT/Ha) -2,373 (1,107 – 3,390) Insurance Payment (KZT): 711,900 (300 Ha x KZT 2,373/Ha) Source: FFSA 2011 Government Financial Support to Crop Insurance in Kazakhstan 3.31. GRK provides financial support to the Compulsory Crop Insurance Scheme in two ways by i) compensating the Insurers for 50% of all the claims incurred and ii) funding of the administration and operating expenses of the Fund for Financial Support for Agriculture, FFSA. The current crop insurance system is financially supported by GRK. Government supports the crop insurance scheme by reimbursing 50 percent of the crop insurance claims paid by the insurance companies and mutual societies without requiring the insurance companies to pay pro rata premiums for this cover. This GRK cover is similar to a free quota-share (50%-50%) reinsurance agreement. 3.32. The 50% claims compensation fund is administered by the FFSA which is responsible for monitoring and managing the financial transactions on this insurance scheme on behalf of government and for approving the claims reimbursements to individual insurance companies . The FFSA performs a very important function on the compulsory crop insurance scheme by maintaining a data-base on each and every insured farmer since scheme inception in 2005 showing by Oblast and Rayon the crops grown and premium and claims data. Between 2004 and 2010 the FFSA received total administrative and operating (A&)) subsides from government of KTZ 319 million (US$ 2.5 million) or an average of KTZ 46 million (US$ 0.36 million) per year. Apart from the direct financial support of the FFSA‘s A&O expenses, GRK also indirectly provides subsidies towards the costs of field-level loss assessment activities (government staff, vehicles, equipment etc). 3.33. Over the past 6 years, GRK has provided KZT 4.7 billion to the FFSA of which 93% has been allocated to settling the 50% of claims and 9% to the A&O expenses of the FFSA . Reference to Table 3.3. shows that in the start-up phase of this scheme, GRK provided KZT 2 billion per year for the first two years and since then smaller payments such that by completion of 2010 the scheme has received total financial subsidies valued at KZT 4.7 billion (about US$ 31 - 68 - million)29. Over the six years to 2010, the 50% claims compensation fund has received a total of KZT 4.3 billion (US$ 28.9 million) funds from government. 3.34. Over the past 6 years (2005-2010), the FFSA has reimbursed the Insurance Companies a total of KZT 3.84 billion (US$ 25.6 million) equivalent to 46.7% of total paid claims. Over this period FFSA has settled an average of KTZ 641 million per year (US$4.3 million) in 50% claims to the insurance companies each year. It is, however, noticeable that over the past three years the trend has been for increased claims and in 2010 the cost of the 50% claims reimbursement to FFSA was KZT 1.2 billion (US$ 8.0 million). At end 2010 the claims fund had KZT 538 million in reserves which would be inadequate to cover the average claims cost of the past 3 years. The level of GRK budgetary support for crop insurance in 2011 is not known. Table 3.3 State Budget program 050 “Support for crop insurance” (million KZT) Year Budget Allocation Disbursements Allocation Agent services payment 50% Claims Actual 50% Balance on for crop insurance Compensation Claims Claims Fund system administration Fund Payments 2004 2,000.0 10.0 1,990.0 2005 2,000.0 10.0 1,990.0 520.1 3,459.9 2006 100.0 60.0 40.0 236.0 3,263.9 2007 300.0 68.4 231.6 350.2 3,145.3 2008 100.0 68.4 31.6 819.1 2,357.8 2009 100.0 49.9 50.1 693.1 1,714.7 2010 100.0 52.2 47.8 1,224.5 538.0 Total 4,700.0 318.9 4,381.1 3,843.1 Average 783 53 730 641 Source: Authors, adapted from ARKA 2011 and FFSA 2011 data 3.35. Issues relating to the whether it is more cost effective for government to provide 50% quota-share coinsurance of claims, or to switch its financial support to premium subsidies and or towards the purchasing of reinsurance and or some form of catastrophe excess of loss reinsurance protection are considered further in Chapter 4. Performance Assessment: Technical Results, Liabilities, Reinsurance 3.36. This Section presents a review of the Compulsory Crop Insurance Results from 2005 to 2010 and full details are contained in Annex 3. Coverage (Insured Crops, Insured Farmers and Insured Area) 3.37. Crop insurance has reached high levels of penetration in Kazakhstan. Owing to its compulsory nature, crop insurance has reached high levels of penetration in the country both in terms of the number of insured farmers and insured area. 3.38. The crop insurance portfolio is mainly comprised of spring sown cereals of which spring wheat is the most important insured crop. In the six years the scheme has operated from 29 At the current 2011 exchange rate of KZT 150 = US$ 1.0 - 69 - 2005 to 2010, spring crops have accounted on average for 98% of the insured area and spring wheat has accounted for 86% of the insured area of spring crops (Annex 3). 3.39. Over the period 2005 to 2010, the Compulsory Crop Insurance Scheme issued an average of 23,494 crop insurance policies per year with an average of 523 hectares of insured crops per policy. Coverage peaked in 2008 with a total of 33,957 insured policies, but in 2010 the number of policies was halved to only 16,766 polices. It is noticeable, however, that in 2010 the average size of insured farm was significantly larger than in any other year with an average of 756 Ha per policy, or 144% of the 5-year average of 523 Ha/policy. According to the insurance industry the major losses experienced in 2009 disproportionately affected smaller farmers: insurers were very reluctant to insure these small farmers in 2010 resulting in a major reduction in the number of insured polices and a shift in the portfolio towards larger insured farms in 2010. 3.40. It is difficult to report the type of beneficiary and percentage of insured farms in Kazakhstan. In 2010 there were a total of 7,441 registered Agricultural Enterprises (Production Enterprises) and 193,435 Peasant Farms (Commercial Farms) or a total of 200,876 farms (Arka 2011). With a total of 16,766 insured crop policies in 2010, this might suggest an insurance penetration level of only 8.3% of all farms. Caution must, however, be exercised in interpreting these figures for the reasons set out below. It is understood that practically all (100%) of the 7,441 Agricultural Enterprises (PEs) purchase crop insurance because of their high profile and the major financial penalties (fines) which they are likely to incur if they do not purchase compulsory cover. Conversely it is understood that crop insurance uptake by small Commercial Farms (CFs) is very low across the country. Finally the number of farm holdings data distorts the true penetration rates for crop insurance because it includes large numbers of small Peasant Livestock holdings which are outside the scope of this compulsory crop insurance scheme. 3.41. The Crop insurance scheme is highly concentrated in the main spring cereal belt of northern Kazakhstan. Over the past 6 years a total of 73.4 million hectares of crops have been insured under the Compulsory LIC Scheme. Overall, 77% of this total area has been underwritten in the 3 northern Oblasts of Kostanay (29% of insured area), Akmola (26%) and NKO (22%). Conversely crop insurance is relatively unimportant in the other 11 Oblasts none of which accounts for more than 5% of total scheme insured area to date (Figure 3.1). Figure 3.1. Kazakhstan: Percentage Distribution of Insured Area by Oblast (2005-2010) 35% % of Total Insured Area (Ha) 29% 30% 26% 25% 22% 20% 15% 10% 5% 4% 5% 5% 3% 3% 2% 1% 1% 0% 0% Source: FFSA 2011 - 70 - 3.42. The insured area has increased over time with a peak of 15.0 million insured hectares in 2009 representing 82% of the total eligible acreage of the insurable crops grown in Kazakhstan in 2009. In 2005 the first year of the compulsory scheme, the insured area was 10.5 million Ha (70% of total sown area), by 2008 this had increased to 14.5 million Ha (84% of total sown area) and in 2009 a further increase was registered to 15.0 million Ha (82% of total sown area). In 2010, however, the insured area declined significantly to 12.7 million hectares or only 68% of total eligible acreage. 3.43. Over 6 years the scheme has operated the insured area of crops has averaged 73% of the total sown area; this implies that in spite of the compulsory nature of the national crop insurance scheme that more than a quarter of the cropped area has not been insured. The fact that on average nearly one-quarter of the insurable area has remained uninsured shows that the compulsory scheme is failing to achieve one of its core objectives namely to ensure that all farmers are insured against catastrophe crop loss. There are a number of reasons for the failure to achieve a higher level of insurance penetration. Over the past 3 years most insurance companies have incurred negative underwriting results particularly in the drought prone regions of WKO and Aktobe and they are increasingly reluctant to operate in these high risk areas: also their results have on average been worse for small peasant farmers and as evidenced in 2010 they were not willing to renew cover for this group of farmers. Many crop producers interviewed under this study do not perceive any benefits from the compulsory crop insurance scheme and some prefer to pay the fines for failure to contract insurance rather than to pay the premiums. 3.44. The pattern of crop insurance penetration as measured by the ratio of insured area to total planted crop area is very uneven across the country. In the northern and eastern regions of Kazakhstan where large-scale spring wheat production is concentrated, over the 5 year period 2006 to 2010, the percentage of total insured area has exceeded 70% of total planted area in the Oblasts of Kostanay (90%), Pavlodar (79%), NKO (76%) Akmola (75%), EKO 72%. Conversely uptake has been much lower in the very drought prone region of western Kazakhstan in the Oblasts of WKO (58%) and Aktobe (52%). Even lower penetration rates have been recorded in the southern region of Kazakhstan as evidenced by the rates in Zhambyl (41%), Almaty (41%) and SKO (27%) (Figure 3.2. and Map a1 in Annex 3). Given the fact this is a compulsory insurance scheme the extremely low uptake figures in southern Kazakhstan are unexpected. The main reasons for the very low uptake in southern Kazakhstan are due to (i) Insurers‘ reluctance to insure the often very small farms and (ii) the fact that a high proportion of agricultural cropping is irrigated and farmers do not see any value in purchasing the catastrophe drought insurance cover. - 71 - Figure 3.2. Kazakhstan: Insured Area Penetration Rates by Oblast (2006-2010) 100% 90% % Total Planted Area Insured (Ha) 90% 76% 79% 80% 75% 74% 68% 70% 61% 55% 58% 60% 50% 41% 41% 40% 27% 30% 20% 10% 0% Source: World Bank analysis of FFSA data 2011 Scheme Liability (Total Sum Insured) and Levels of Protection afforded to Farmers 3.45. Since inception in 2005, the average scheme total liability has been about KZT 40 billion (US$ 267 million) per year. Over the past six years the Total Sum Insured (TSI) of the scheme has increased from KTZ 34.4 million (US$ 229 million) in 2005 to a peak of KZT 52.9 Billion (US$ 353 million) in 2009 reflecting the gradual increase in coverage over time as the program has matured in terms of numbers of insured farmers and insured area. In 2010, however, with the reductions in number of insured farmers and insured crop area, the TSI was correspondingly reduced to KZT 47.3 billion (US$ 325 million). (See Annex 3 for details). 3.46. The Compulsory Crop Insurance Scheme Financial Liability is very unevenly distributed throughout Kazakhstan. Reference to Figure 3.3 shows that over the past 6 years the scheme liability has been highly concentrated in the 3 northern Oblasts of Akmola, NKO and Kostanay accounting for 78% of total sum insured (TSI) and which is a reflection of the fact that grain production is highly concentrated in these regions. This represents a very large accumulation of risk in the event of a major drought event affecting the northern region. Conversely, no other Oblast has accounted for more than 5% of scheme liability over the past 6 years and liability is very low in the southern part of the country (SKO, Almaty, Zhambyl and Kyzylorda Oblasts). - 72 - Figure 3.3. Kazakhstan: Distribution of Total Sum Insured per Oblast (2005 to 2010) 35% 31% 30% 25% Percent of TSI (KZT) 25% 22% 20% 15% 10% 5% 4% 5% 3% 2% 3% 2% 1% 1% 1% 0% Source: FFSA 2011 Levels of Sum Insured Protection purchased by Farmers 3.47. The sum insured levels elected by the farmers to protect their crops are extremely low and over the past 6 years the average sum insured for all crops was only KZT 3289 per hectare (US$ 23/Ha30). As noted previously, MoA is responsible, in conjunction, with the local authorities in each region for the setting of the sums insured for each crop type in each Oblast and agro-ecological zone based on three levels of technology package (high, medium and low) and the associated costs of production. Farmers are then permitted to choose which level of production costs they wish to insure their crops grown in their own fields. An analysis of the actual average sums insured elected by farmers since scheme inception shows that most farmers are purchasing the lowest permitted sum insured level which only covers three production cost items (wages, fuel & lubricants and seeds): over the 6 year period the average sum insured has been KZT 3,289/Ha (about US$ 23/Ha) with a range from an average low in 2007 of KZT 2,871/Ha (US$ 20/Ha) to an average high in 2010 of KZT 3,728/Ha (US$ 26/Ha). (See Annex 5). 3.48. On the basis of this study it appears that many farmers elect the lowest permitted sum insured levels per hectare in order to minimize the amount of premium they have to pay . Farmers‘ reasons for selecting the lowest permitted normative costs of production sum insured level typically centered on the fact that the policy only insured a low proportion of their production costs and expected yield and because they would have to incur a near total loss in order to trigger an indemnity, the coverage provided was not beneficial to them: for this reason farmers generally opt for the lowest sum insured level. 3.49. The current levels of sums insured do not afford farmers adequate protection against the loss of the costs they have invested in growing the crop. For spring wheat where average total costs of production were about KZT 12,000 to KZT 17,500/Ha in 2010, the current sum insured levels of about KZT 3,500/Ha (minimum) and KZT 9,000/Ha (maximum) only represent between 20% and 50% of the actual total costs invested in producing the crop. In the event of a total loss the indemnity payment is therefore very inadequate to cover their incurred costs and may not be adequate to cover the costs of production loans from banks etc. 30 At a 2010 average exchange rate of Kazakhstan Tenge 145 = US$ 1.0. - 73 - 3.50. Over the past five years the effective amount of yield protection afforded to farmers by the Loss of Investment Cost Policy has gradually been eroded. Under the LIC Policy an indemnity is payable when the value of any harvestable production (salvage) is inadequate to cover the costs of production invested in growing the crop. The analysis presented in Figure 3.4. shows that over the past 6 years the average sum insured per hectare has only increased by 13% from KZT 3,288/Ha (2005) to KZT 3,728/Ha (2010). Over this period, however, the average September farm-gate price for wheat has increased significantly by 350% from KZT 1,149/center (2005) to KZT 3,988/center (2010)31. With the increase in wheat prices which are used to value salvage there has been a major erosion of /reduction in the underlying insured wheat yield: In 2005 with an average sum insured of KZT 3,288/Ha when the average farm gate sale price for wheat was very low at KZT 1,149/center, the policy would start to indemnify losses when the remaining yield fell below 2.85 centners per hectare: however, by 2010 with a high average sale price of KZT 3,988/center a farmer would only receive an indemnity when his actual yield fell below 0.93 centners/Ha. 3.51. The current LIC crop insurance policy provides one of the lowest levels of yield protection of any major individual grower national crop insurance scheme. The current LIC Policy with its very low sums insured equates to an insured yield coverage level of about 1 centner per hectare or on average only 14% of the national average spring wheat yield. This compares with global norms for most individual grower loss of yield insurance policies which offer minimum insured yield coverage levels of about 40% to 50% of average yield and in some countries the maximum insured yield coverage levels that farmers can purchase are 75% to 85% of the farmer‘s normal average yield. There is a need to review the insured yield coverage levels provided under the Kazakhstan compulsory crop insurance scheme and this is a subject which is discussed in Chapter 4. Figure 3.4. Evolution of Wheat Sums Insured, average wheat farm gate prices and effective yield coverage levels of LIC Policy 4500 3.50 Wheat Yield Equivalent (Centner/Ha) Average Sum Insured (KZT/Hectare) 4000 3.00 2.86 Wheat Price (KZT/Centner) 3500 2.50 3000 2.30 2500 2.00 2000 1.50 1.29 1.38 1500 1.05 0.93 1.00 1000 500 0.50 0 - 2004 2005 2006 2007 2008 2009 2010 2011 Average Sum Insured (KZT/Ha) Wheat price (KZT/Centner) Wheat Yield Equivalent (Centner/Ha) Source: Authors based on FFSA sum insured data and September Wheat Price Data from Arka Consulting Financial Performance: Premiums, Claims and Claims Ratios 31 Wheat prices based on data provided by Arka Consulting 2011 - 74 - Overall Results 3.52. The compulsory crop insurance program in Kazakhstan has experienced poor overall underwriting results over the period 2005 to 2010. The long-term average loss ratio for the 6- year period 2005-2010 is 140% and in four of the six years the scheme has operated at a financial loss in terms of the gross claims to premium or loss ratio exceeding 100%. The average net loss ratio to the insurance companies after the reimbursement of 50 percent insurance losses (claims) from the government is 75%. Assuming average administrative and acquisition expenses of 25% to 30% of gross premium for the industry, the average net loss ratio of 78% indicates that at best the insurance companies and farmers‘ mutual insurance associations are operating on a breakeven basis, but that most of the companies have been operating at a financial loss. (Table 3.4). 3.53. The results show that underwriting results have deteriorated badly over the past three years 2008 to 2010 which coincides with drought loss years, especially in 2010. Over the past three years the program has incurred negative underwriting results in all years and an average loss ratio of 182% (99% after operation of the government claims subsidies) and with a worst loss year in 2010 with 261% loss ratio: in 2010 underwriters paid out claims of KTZ 2.61 for every KTZ 1.00 they received in premiums prior to recoveries from FFSA/government. On account of the very poor results all but three of the seven registered commercial insurance companies have now declined to write the compulsory crop insurance scheme. Table 3.4. Summary of Kazakhstan Crop Insurance Financial Results 2005-2010 Item 2005 2006 2007 2008 2009 2010 Total No. of Policies (000) 19.0 13.6 25.4 34.0 32.2 16.8 141.0 Total Insured Area (Million Ha) 10.5 9.1 12.1 14.5 15.0 12.7 73.8 Sum Insured (Million KZT) 34,372 26,650 34,796 46,645 52,903 47,266 242,631 Premiums (Million KZT) 899 685 997 1,093 1,114 1,074 5,862 Average Premium Rate 2.61% 2.57% 2.87% 2.34% 2.11% 2.27% 2.42% Claim payments (Million KZT) 1,065 478 701 1,710 1,465 2,805 8,223 Loss Ratio (%) 119% 70% 70% 156% 131% 261% 140% Loss cost (%) 3.1% 1.8% 2.0% 3.7% 2.8% 5.9% 3.4% FFSA Compensation (Million KZT) 520 236 350 819 693 1,225 3,843 FFSA Compensation % of claims 49% 49% 50% 48% 47% 44% 47% Loss Ratio % (Net of FFSA support) 61% 35% 35% 81% 69% 147% 75% Source: FFSA 2011 Regional Performance (By Oblasts) 3.54. Scheme performance varies widely across different regions and the very poor underwriting results in Aktobe and WKO are making the scheme financially unviable. The pattern of claims varies widely by geographic region. The best performing Oblast is NKO which over the past 6 years has contributed 22% of total scheme liability, but only 3% of claims and has a long-term loss ratio of only 24%. The next best performing Oblast is Kostanay which has the largest share or 31% of total scheme liability but which has only accounted for only 11% of total claims and which has a long-term loss ratio of 73%. At the other extreme, Aktobe and WKO in western Kazakhstan which have collectively accounted for only 4.8% of total scheme liability over the past 6 years have incurred 41% of all claims and respectively have 6-year long-term loss ratios of 381% and 507%. These two Oblasts are severely prejudicing the financial viability of - 75 - the whole of the national crop insurance program and measures of controlling the claims costs in these two Oblasts must urgently be introduced. (Figure 3.5). (See Annex 3 for full details). Figure 3.5. Scheme Liability, Claims and Loss Ratio by Oblast (2005-2010) 35% 600% 30% 507% 500% 25% 412% 381% 400% 20% 300% 15% 199% 200% 10% 176% 132% 5% 105% 100% 100% 73% 24% 30% 0% 2% 0% % of TSI % of Claims Payments Loss Ratio Source: World Bank analysis of FFSA data Causes of Loss 3.55. Drought has been the main cause of loss under the compulsory crop insurance scheme. Drought events accounted for 91% of the total reported damaged area in the 5-year period 2006- 2010 and with the inclusion of losses due to drought in combination with other perils such as hail or freeze, total drought losses rises to 97% of the total damaged area. The second most important cause of loss has been Hailstorms, which have accounted for 2.5% of the total area lost due to insured perils. . (See Annex 3). Adequacy of Premiums Rates and Claims Costs 3.56. Over the past 6 years there has been a tendency for average premium rates to be reduced although this is not justified by the claims experience. At the time of scheme inception in 2005 a system of single of fixed premium rates for each crop in each Oblast was adopted under the compulsory crop insurance scheme and these rates applied during the three year period 2005 to 2007. During this three year period the average premium rate was 2.68% with very little variation year on year from a low average of 2.57% in 2006 and an average high of 2.87% in 2007. Prior to the 2008 renewal the rating system was reviewed and amended to introduce a system of minimum and maximum premium rates and the minimum premium rates were lower than the original single fixed premium rates. In 2008 the scheme was also opened up to competition by the newly formed mutual crop insurance associations. For the past 3 years (2008- 2010) the premium rates charged by the insurers have decreased to an average of 2.24% representing a reduction of nearly 17% on the previous three year average. According to the commercial insurance companies the Mutual Insurers generally charge their members the minimum premium rates set by government and that price competition has been the cause of the reduction in the overall average premium rates charged on the scheme. At the same time that average premium rates have been reduced, the claims on the scheme have risen significantly (Figure 3.6.). - 76 - 3.57. The past three years have seen a tendency for increased claims on the compulsory crop insurance scheme and with the reduction in average premium rates the scheme has now become seriously under-rated. In Figure 3.6 a comparison is made of the annual average premium rate charged on the scheme the annual average loss cost which is equivalent to the value of claims divided by the total sum insured and expressed as a percentage. The loss cost is a useful ratio as it shows the minimum premium rate that would need to be charged on an insurance scheme to exactly cover the paid claims. Figure 3.6 shows that in the three years to 2007 the average premium rates were adequate to cover the average loss costs, but that since 2008 there has been a major increase in claims on this scheme such that the average loss costs have exceeded the average premium rates in all three years, culminating in the peak loss cost of 5.93% of TSI in 2010 against an average premium rate of only 2.27%. This analysis shows that there is an urgent need to review and adjust the premium rates on this scheme in order to maintain its financial viability. Figure 3.6. Evolution of the Average Premium Rate and Loss Cost (2005-2010) 6.50% Average premium rate & Loss Cost (% of 6.00% 5.93% 5.50% 5.00% 4.50% 4.00% TSI) 3.50% 3.67% 3.00% 3.10% 2.87% 2.77% 2.50% 2.61% 2.57% 2.34% 2.27% 2.00% 2.01% 2.11% 1.79% 1.50% 2004 2005 2006 2007 2008 2009 2010 2011 Average Premium Rate % Loss cost (%) Source: World Bank analysis of FFSA Data 2011 3.58. The claims experience between 2005 and 2010 shows that the scheme is severely under- rated in WKO and Akmole Oblasts in Western Kazakhstan and this applies to a lesser extent in several other Oblasts. The 6-year long-term average loss cost on the scheme currently stands at 3.4%, but the pattern of losses varies widely across geographical region. Figure 3.6 shows that over this period average premium rates have been highest in WKO with an average premium rate of 7.8%: however, this premium rate has been totally inadequate to cover the actual claims as evidenced by the loss cost of 39.6%. This is followed by Aktobe with average premium rates of 5.8%, but again the program has been very under-rated in this Oblast as the break-even rate to cover actual claims should have been 22.2%. Other Oblasts where the scheme has been under- rated include Pavlodar, Zhambyl, EKO and Almaty. In contrast, the scheme has performed very well in NKO with a 6-year long term average loss cost of only 0.5% compared to an average premium rate of 2.0%: this Oblast has in effect cross-subsidised the very poor results in WKO and Aktobe over the past 6 years. This evidence shows that very substantial premium rate increases would be required in WKO and Aktobe if the scheme was actuarially rated in these Oblasts: however the minimum average rates of at least 40% (WKO) and 22% (Aktobe) would not be commercially acceptable to farmers in these Oblasts. - 77 - Figure 3.6. Analysis of Adequacy of Premium Rates by Oblast (2005-2010) Zhambyl 3.5% WKO 7.8% SKO 2.7% Pavlodar 3.6% NKO 2.0% Kyzylorda 5.2% Kostanay 1.7% Karaganda 4.7% EKO 3.4% Almaty 3.5% Aktobe 5.8% Akmola 1.9% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% Average Premium Rate and Loss Cost (% of TSI) Loss cost Avg Premium Rate Source: World Bank analysis of FFSA Data 2011 3.59. The variation in claims experience between Rayons in the same Oblast clearly indicates a need to introduce a system of Rayon-level actuarial rating. The 5-year claims experience also varies significantly between Rayons in each Oblast. For example in Rayons located in northern NKO average loss costs are less than 0.25%, but in the southern part of the Oblast which is drier, rayon-level loss costs are higher with a maximum of 0.5-1.0% loss cost. Similar differences in drought risk exposure and loss costs between rayons also apply to other Oblasts such as Pavlodar, Akmola, Kostanay and WKO. (See Map 3.1.). This analysis clearly demonstrates a need to consider introducing a system of rayon-level rates in future. Map 3.1. Variation in Loss Costs by Rayon (2006-2010) Source: World Bank analysis of FFSA Data 2011 - 78 - 3.60. The analysis of variation in yields at Rayon-level also shows that there are significant differences in the yields obtained by different types of farmer and which would need to be addressed in rating? Specifically, the analysis of time-series Rayon spring wheat yields obtained by the large-scale Production Enterprises and the smaller Commercial Farms shows that yields are much more variable for the Commercial Farmers on account of their lower levels of technology and this would require higher premium rates to be charged for this group of farmers. Financial and Reinsurance Issues 3.61. The claims results have been very volatile in the past 6 years with a trend over the past three years for increased claims. Over the past 6 years the pattern of claims on the compulsory crop insurance scheme have been very volatile with a range from a low of KZT 478 million (US$ 3.9 million) in 2006 to a high of KZT 2,805 million (US$ 8.4 million) in 201032. There has also been a marked trend towards increasing claims over the past 3 years which coincide with three very bad drought years. (Table 3.4). 3.62. Over the past 6 years, the scheme has paid total claims of KZT 8.22 billion (US$ 62 million) of which the private and mutual insurers have paid 53% of total claims and government, through the FFSA, has reimbursed the insurers for 47% of the total value of claims.(Table 3.4). The government provides 50% proportional or quota share reinsurance protection to the insurance companies free of cost. 3.63. Under the compulsory crop insurance scheme, insurance companies have little scope to managing their liability on their 50% of risk retentions and they are very exposed to major losses. The insurance companies have very little ability to manage their exposure to losses on this compulsory scheme because they are not able to exercise decisions over individual risk acceptance or to decline to underwrite the portfolio in regions which they consider to be too risky. In spite of the government‘s reinsurance protection, the insurance companies remain very exposed to loss as they retain unlimited liability on their 50% share of claims. Any company which has written an imbalanced portfolio which is overly concentrated in the western Oblasts of Kazakhstan will have incurred unsustainable underwriting losses over the past 6 years and especially in 2010 as evidenced by the gross loss ratios of 900% Aktobe (450% net of FFSA reinsurance) and 1075% WKO (537% net of FFSA reinsurance). Even those insurance companies which are fully capitalized and maintain claims reserves would not be able to sustain such losses in the medium term. 3.64. Currently, private commercial insurers are unable to obtain commercial reinsurance protection to limit their exposure to catastrophe losses on the obligatory crop insurance scheme. It is understood that several private commercial insurers have in the past sought quotations from international reinsurers to provide either quota-share or non-proportional reinsurance on their portfolios. International reinsurers, are however, reluctant to provide capacity support for the scheme for reasons including: i) the compulsory nature of the scheme which prevents them from selecting which risks they are willing to underwrite in which region(s), ii) the fact that the scheme is under-rated and iii) the fact that indemnity and loss assessment procedures would require strengthening to meet international standards. 3.65. The Farmer’s Mutual Crop Insurance Associations are not required to maintain claims reserves and they are therefore unable to meet their liabilities if claims exceed the net 32 Actual annual average exchange rates KZT and US Dollars used: 2006 KZT 122 = US$ 1.00; 2010 KZT 145 = US$ 1.00. - 79 - premiums they collect from their grower members. The private registered insurance companies are obliged by insurance legislation to both meet minimum capital requirements to underwrite any class of business, but also they retain full liability to settle any and all legitimate claims. In contrast, the farmer mutual crop insurance associations which have been formed since 2008 do not face any minimum capital requirements, or requirements to meet claims which exceed the net premiums (premium net of management charges which are typically in the order of 25% of gross premium) paid into their mutual funds by their grower members. On the basis of this study it is apparent that several of the mutuals interviewed were not able to meet their 50% share of claims liabilities in full in 2010 and these associations were therefore forced to pro-rata down the claims paid to all farmers who incurred losses. These companies were, however, able to collect the government 50% of claims payments in full and to return these amounts to their members. It is not possible to report on the extent of this problem in both 2009 and 2010 in terms of the number of mutual associations which could not meet their liabilities on their 50% retained claims. 3.66. In Kazakhstan there is a concern that if the mutual crop insurers are not able to meet their claims liabilities in full this will rapidly undermine farmer’s confidence in mutual insurance. International experience shows that where mutual crop insurers have not been able to meet their financial liabilities in catastrophe years that this rapidly undermines the confidence of their farmer members and usually leads to the collapse of the mutual(s). There is an urgent need in Kazakhstan to i) review the extent of this problem among the mutual insurers, ii) to seek ways of strengthening their claims reserves and iii) to consider alternative ways of reinsuring the mutuals in order to cap their liability in the event of catastrophe losses – for example through some form of non-proportional reinsurance on their retained risk. This is a subject which is reviewed further in Section 4. 3.67. Government is also exposed to catastrophe losses on this scheme and which may not have been budgeted for. To date government has funded the FFSA to the tune of KZT 4.4 billion and has paid out KZT 3.8 billion in 50% reinsured claims. Government‘s liabili ty over the past 6 years has been very volatile varying from total payments in 2006 of KTZ 250 million to as high as KZT 1.2 billion in 2010. In common with the problem faced by insurance companies, government‘s liability cannot be budgeted ex-ante and also the claims fund may be inadequate to cover catastrophe claims which may be incurred every hundred or more years. (Figure 3.7). Figure 3.7. Evolution of Claims and Share of Claims paid by Insurers and Government 3000 Paid Claims (KZT Million) 2500 2000 1500 y = 362.05x + 103.28 R² = 0.6515 1000 500 0 2005 2006 2007 2008 2009 2010 Total Claims Net Claims to Insurers Government Reinsurance Linear (Total Claims) Source: Authors based on FFSA information - 80 - 3.68. Due to ambiguities in the indemnity formula used to settle losses on the Loss of Investment Costs Policy, both the insurance companies and government face considerable uncertainty over their financial liability in the event of claims. The LIC policy indemnity formula for partial losses values any remaining harvestable production (salvage) according to the actual local sale price for wheat at the time of loss. This means that neither the insurance companies nor the government can predict their financial liabilities with any accuracy at the time of renewing cover each year: instead they have to wait until the time of harvest to discover their liability. This is illustrated in Table 3.5 for a 1 hectare field of wheat with an average normative costs of production sum insured of 3,500/Ha and three different wheat prices which apply at harvest ranging from a low of KZT 2,000/center (as encountered in 2007); KZT 2,500/center and finally a high price similar to the average for 2010 of 3,500 /center. When average sales prices for wheat are low (KZT 2,000/center), insurers‘ are more exposed to first loss because once the actual average yield falls below 1.75 centner/Ha the revenue from the crop will be less than the break-even sum insured value of KZT 3,500/Ha. Conversely when wheat prices are high (KZT 3,500/center) the insurers‘ are much less exposed to loss as they will only be liable for a claim once the actual remaining yield (salvage) falls below 1.0 centner/Ha. Insurers‘ and government‘s uncertainties over their financial exposure can only be removed by agreeing a valuation price for salvage at the time of policy inception. This valuation price could either be based on an average of the recent historical farm-gate prices for each crop in each Oblast or based on MOA‘s estimated average prices for the current season. Table 3.5. Examples of Indemnity Calculations according to different wheat prices at harvest Price of Wheat at Harvest (KZT/Centner) Indemnity Calculation Price 1: KZT 2000 Price 2: KZT 2500 Price 3: KZT 3500 per centner per centner per centner 3,500 3,500 3,500 Sum Insured (KZT/Ha) 1.75 1.40 1.00 Break-even yield (Centner/Ha) 2,000 2,500 3,500 Sale value wheat (KZT/centner) 3,500 3,500 3,500 Revenue Value for Salvage (KZT) Source: Authors 3.69. Insurers and government face major exposures on this scheme in the event of a catastrophe year as defined by the probable maximum loss expected 1 in 100 years. To date the worst loss on this scheme over the 6-year history was in 2010 equivalent to a loss of 5.9% of the total liability (TSI), valued at KZT 2.81 billion (US$ 19.3 million). This loss is equivalent to a return period of about 1 in 45 to 50 years, but is not the worst loss that insurers and their reinsurer (government may expect in future. Figure 3.8 and Table 3.6 show the results of a modeling exercise to calculate the probable maximum loss for the scheme for return periods of up to 250 years and for four different assumptions (i) the actual average 3-year portfolio, ii) assuming a wheat indemnity sale price of KZT 2000/center, iii) wheat indemnity sale price KZT 2,500/centner and finally iv) wheat indemnity sale price for KZT 3,500/Ha. Figure 3.8 shows that for option i) the actual portfolio the 1 in 100 year PML is 9.4% of TSI which would be equivalent to a loss of KZT 4.4 billion (US$ 29.3 million) at 2010 portfolio TSI of KZT 46.48 billion: this PML loss would equate to a gross loss ratio of nearly 396% at 2010 premium terms. The analysis also shows that if the actual average prices of wheat were to fall to levels of between KZT 2000/center and KZT 2500/center that the PML exposures would be very much higher on this scheme. - 81 - Figure 3.8. Probable Maximum Loss estimates assuming 2010 average sum insured per hectare and 3 scenarios of wheat prices 25% 20% % Loss Cost - % TSI 15% 10% 5% 0% 0 50 100 150 200 250 Return Period (years) 3-year Average Price Price= KZT 3500/centner Price= KZT 2500/centner Price = KZT 2000/center Source: Authors analysis Table 3.6. Estimates Probable maximum Loss According to different price assumptions (KZT 000)* Return Period Actual Portfolio Wheat Price 1 Wheat Price 2 Wheat Price 3 (Years) avg last 3 years KZT 3500/Centner KZT 2500/Centner KZT 2000/Centner 10 785,966 682,553 1,159,043 1,707,702 25 1,778,278 1,565,370 2,520,147 3,484,647 50 2,867,899 2,543,651 3,929,562 5,430,554 100 4,372,797 3,960,971 6,195,513 7,950,115 250 7,354,656 6,847,274 9,111,056 11,736,143 Source: Authors analysis of FFSA data. * PML values calculated based on portfolio TSI of KZT 46.48 billion. Assessment of the Technical, Operational and Institutional Features of the Compulsory Crop Insurance Program 3.70. This sub-section provides a review of the technical, operational, institutional and financial features of the compulsory crop insurance program and highlights some of the key advantages, issues and drawbacks of the current program. Technical 3.71. The salvage based loss of investment crop insurance policy implemented in Kazakhstan has several advantages, but also a series of drawbacks. The main advantages of the Loss of Investment Cost policy are (a) that it provides comprehensive multiple-peril crop insurance, MPCI, protection to the farmer again the loss of his production costs invested in growing the crop - 82 - and (b) that it can be used in situations where there is inadequate or no information on individual farmer historical crop yields. This is in contrast to a conventional loss of yield MPCI policy which relies on time-series farm-level crop production and yield data to establish a normal average yield and then an insured yield for each farmer. There are, however several potential drawbacks of the Loss of Investment Cost policy including the need for in-field yield-based loss assessment where partial losses are involved and often the difficulty of establishing objectively the salvageable amount of the crop and its sale value and whether this salvage value exceeds the insured investment costs leading to a claim. 3.72. The coverage levels provided by the crop insurance scheme in Kazakhstan are extremely low. The sum insured levels that MOA is responsible for fixing each year by crop and by Oblast according to three technology levels: high, medium and low are extremely low for most crops as shown in Table 3.2. In practice it appears that practically all farmers in Kazakhstan elect the cheapest or lowest sum insured coverage option of about KZT 3,500/Ha (slightly less than US$ 25/Ha nationally for spring wheat) because they will pay the least amount of premium for this option. However, this means that on average farmers are only insuring between 20% to 30% of their total production costs. The very low crop insurance cover levels are therefore, often inadequate to put farmers back into production in the event of major crop losses. 3.73. The crop insurance premium rates are calculated for each crop at an Oblast level and do not take into account differences in risk either at the local Rayon level or at the individual farmer level. Ideally crop insurance premium rates should be established for homogeneous agro- climatic risk zones, but in most countries such risk zoning and mapping is not available: therefore rates are typically established for administrative regions and then smoothed to ensure consistency with adjoining regions. In Kazakhstan the Oblast represents a very large geographical area and there is evidence that agro-climatic and soil conditions vary widely across most of the Oblasts and this is reflected in major crop yield differences within the Rayons in each Oblast33. This evidence suggests there is a need for a technical study to review the possibility of introducing a system of rayon-level crop insurance premium rates in the short term and while more detailed risk zoning and risk mapping is carried out. For this reason, under this World Bank Study, an actuarial rating exercise has been carried out at Rayon-level for spring wheat and the results of this analysis are presented in Chapter 4. 3.74. Crop Insurance Premium Rates are currently fixed by law for each Crop Type and group of Oblasts according to Minimum and Maximum rates and Insurers have little influence over rating decisions. The private and mutual insurance companies have no say in setting premium rates or in risk selection: they are, however, permitted to charge an individual farmer a premium rate that falls between the MOA fixed minimum and maximum premium rates. According to the Private Insurance Companies, they use the minimum and maximum premium rates to manage, in part, their risk: in the most drought prone regions of the country they charge the maximum premium rates to try to avoid writing too much crop business and also where they consider an individual farmer to be a poor risk they again quote the maximum premium rates. While private insurance companies use this difference in minimum and maximum premiums to perform some sort of risk selection, farmers‘ mutual associations advised that they nearly always 33 This is an issue which was identified by the CJSC in their original rating study when they noted that Oblasts are very large territories and agricultural conditions within the Oblast may change significantly leading to a situation that farms in low risk areas end up subsidizing farmers with high risk because they both pay the same tariff averaged for the Oblast. The CJSC, however played down the need for rayon-level and possibly farmer-level premium rates according to whether they are classified as high or low risk, because of the compulsory nature of the scheme. (CJSC 2003). - 83 - had to operate at the minimum premium rate levels, because their members were not willing to pay the higher rates permitted by law. For spring wheat grown in the more favourable / higher rainfall regions of Akmola, Almaty, Kostanay, East and North Kazakhstan, the current commercial premium rates are very low at between 1.78% (minimum) and 3.48% (maximum) and where the mutuals are forced to charge the minimum rate of 1.78% due to competition, it is unlikely that this rate will be sustainable in the medium to long-term. It appears that there is a need to reconsider the system of minimum and maximum permitted premium rates. 3.75. The current Crop insurance premium rates do not differentiate between farmers in terms of technological levels or implementation of risk management practices . The current rating system does not differentiate between farmers in terms of their technology levels and risk management practices and therefore does not provide any incentive for farmers to manage their risk exposures for example to drought by adopting snow conservation measures in winter and minimum tillage practices and soil moisture conservation practices during the growing season. Section 2 showed that the large Production Enterprises (PEs) achieve higher average yields and less variable yields in spring wheat as such they represent a lower risk exposure group that Commercial Farmers (CFs). The Private Insurers reported that they aim to incentivize the best farmers by offering them the minimum rates and conversely by penalizing low technology users in the riskier area by charging the maximum permitted rates. 3.76. It is understood that the crop insurance premium rates do not include specific allocation for the costs of in-field loss assessment/loss adjustment expenses. Under the current system each of the five organisations involved in the in-field assessment of total and partial crop losses is responsible for bearing their own costs of this exercise. In 2010 the average size of a crop insurance policy was 756 Ha, with average premium of KZT 64,037 (US$ 442) or a premium per hectare of KZT 85/Ha (US$ 0.58/Ha). Because of the combination of very low average sums insured per hectare and low premium rates, the average premium per policy is very low for a cover requiring such intensive in-field loss assessment, and it is likely that the total costs of adjusting losses will exceed the total premium generated per policy. This evidence suggests there is a need to review (a) the adequacy of the premium rating methodology for covering loss assessment expenses and (b) ways of overhauling loss assessment procedures in order to make these more efficient and to reduce loss assessment expenses (See Section 4 for recommendations regarding strengthening of loss assessment procedures). Operational 3.77. The obligatory nature of the Scheme means that there is a major potential problem of moral hazard in the most drought prone regions of the country. In some regions of Kazakhstan, seasonal rainfall is extremely marginal for spring wheat production and even in normal rainfall years is very dependent on the amount of winter snow fall and soil moisture status at the start of the cropping season. Western Kazakhstan (WKO and Akmola) experiences major droughts every 2 or 3 years and it is possible for farmers to predict a drought year according to the amount of winter snowfall: however, because of the obligatory nature of the scheme insurers cannot decline to insure farmers in these areas although they can already foresee that they will incur major drought induced crop losses. In a predicted severe drought year, the behavior of some farmers in these regions is also likely modified in two main ways (a) they elect to buy the maximum ―Normative Costs‖ Sum Insured level in the expectation of receiving a claims payment and (b) they may incur less than their normal levels of expenditure on crop husbandry and inputs because they know they are likely to lose their crops and in which case they can expect to claim on their insurance policies (this is termed moral hazard). The private insurance companies have incurred major losses in West Kazakhstan in the past and are now very reluctant to insure farmers in this - 84 - region and it appears that this is one of the major reasons that most of these companies have ceased underwriting crop insurance. 3.78. Private Insurance Companies do not have their own networks of locally based qualified agronomists to conduct pre-inspections on the Insured Farms at the time of sowing in order to confirm whether the farmer has complied with the correct sowing practices, seed rates etc. As such cover is open to moral hazard. The costs of establishing such a network and in inspecting each and every farm would be prohibitively expensive to the insurers, and under the current rating system they are not able to increase rates to cover their A&O expenses. 3.79. The Obligatory Nature of the Scheme means it is very difficult for Insurance Companies to exercise any form of Accumulation Control. Drought and frost are systemic or covariate risks that have a potential to correlate over very wide geographic areas. In the context of any MPCI crop insurance program, it is very important that insurers attempt to manage their crop insurance portfolios in extreme drought years and they mainly do this by setting underwriting limits in each region according to the relative rainfall patterns and drought exposures. In Kazakhstan the huge size of the country means that even in extreme drought years that not all the spring wheat grown in the country is equally affected by drought and if insurers were able to set underwriting capacity limits in each region this would enable them to manage their drought exposures to a major extent. This is, however, currently impossible to achieve because of the compulsory nature of the crop insurance program. 3.80. Crop insurance is currently marketed exclusively through local sales agents in each Oblast and Rayon and this represents an expensive delivery channel. All private companies market their crop insurance policies through local sales agents who receive a commission for their services which is paid out of the premium collected from the farmer. It is understood that many of the mutual insurance companies also rely on sales agents to promote and market their crop insurance policies to farmers who then sign up to the mutual as a member. Current legislation caps the commission payments to agents and brokers at 10% of the Commercial premium (also termed Original Gross Premium, OGP) 34. This represents a significant cost which is paid by the farmer. In many countries, insurers use rural service organizations such as agricultural banks, microfinance institutions (MFIs), input suppliers and or farmers associations to deliver and administer crop insurance to their clients: in such cases the rural organisation acts as an agent, but because of its much lower operating overheads and economies of scale in dealing with large numbers of clients or members, the commission payments can be significantly reduced as can the insurance company‘s own administrative expenses. This offers the potential to pass on these cost savings to the Insured in the form of lower premium rates. The subject of developing alternative and cheaper ways to deliver crop insurance to small and medium farmers in southern Kazakhstan is reviewed further in Section 6. 3.81. Loss adjustment requires the participation of several parties, it is expensive, and sometimes lacks transparency. In Kazakhstan up to 5 persons are involved in adjusting the crop losses at the local level and this is a very time consuming and costly exercise. The main drawback of any Loss of Investment Cost policy is the requirement especially where partial crop losses are involved, for individual farmer and field by field loss assessment to determine the damaged area and the remaining amount of harvestable crop (termed salvage) in the field and on which basis to calculate the expected value of the salvage. It is then necessary to determine 34 It is understood that in early years of the compulsory crop insurance program that Agent‘s fee s were as high as 30% of premium, but that the scheme regulators have subsequently imposed a 10% ceiling in order to make the scheme more acceptable to insurers. - 85 - whether the value of salvage exceeds the costs invested in the crop up to the time of the loss and also whether it is economic for the farmer to continue to manage the remaining undamaged portion of the crop through to harvest, or whether these costs will exceed the salvage value. This is both a time-consuming and costly exercise and is often subject to conflict over whether the value of salvage exceeds the sum insured investment costs (in which case there would be no claim) or not, in which case the farmer would receive a claim according to the amount by which the expected value of salvage falls short of the insured costs of production. 3.82. Crop output valuation prices are determined at the time of harvest as opposed to being pre-agreed and specified in the policy wording and this section has shown that because actual farm-gate wheat prices can vary widely according to demand and supply, this leads to major variations in the amount of physical crop production and yield which is required to cover the insured costs of production. This situation where neither the farmer nor the insurer can quantify their indemnity exposures (in terms of the amount of crop production and yield loss which needs to occur before an indemnity is paid) is very unsatisfactory and needs to be reviewed. Institutional 3.83. The Kazakhstan compulsory Crop Insurance scheme represents a public-private partnership which is highly regulated by government and underwritten by the private commercial (and mutual) insurance sectors, with financial claims subsidy support provided by government through the FFSA. The regulatory, operational and financial roles assumed by GRK include : (i) to approve the standard Loss of Investment Cost policy for crop insurance including the basis of coverage and insured perils and other policy terms and conditions; (ii) to approve each year the sums insured based on the normative costs of production of the crops that are subject to compulsory insurance; (iii) to develop the minimum and maximum commercial premium rates to be used for the different crops and regions under the crop insurance scheme; (iv) to approve the budget uses for supporting crop insurance; (vi) to establish and regulate the commissions payable to sales agents, (vii) to set the norms and procedures for the adjustment of crop losses at the individual insured farmer and field level and finally (viii) to provide financial subsidy support both through contributions to the A&O costs of the FFSA and in the form of a 50% share in the value of paid claims. In most countries where governments support agricultural insurance PPP‘s they usually assume a much lesser degree of operational and underwriting control and their main function is to provide legal and regulatory support and financial subsidies. 3.84. At the inception of the Compulsory Crop Insurance Scheme in 2005 seven private insurance companies registered their interest in underwriting this product, but on account of poor underwriting results only three companies continue to support the scheme in 2010 . The level of participation of the private insurance companies in crop insurance in Kazakhstan is very low. Only 7 out of 37 private non-life insurance companies in the market are currently licensed to operate crop insurance in the country and only three of them actively underwrote the obligatory crop insurance scheme in 2010. It is understood that the main reason companies have ceased underwriting this class of business is because it is deemed unprofitable and that crop insurance is the least profitable of 10 classes of compulsory insurance in Kazakhstan. There is a danger that unless the scheme can be returned to profitability the private insurance sector companies may cease to provide their support in future. 3.85. Due to the lack of interest of the private insurance industry in participating in the mandatory crop insurance scheme, the GRK decided in 2006 to license Farmers’ Mutual Crop Insurance Associations to write crop insurance. Under the Law No 163-III ZRK of the Republic of Kazakhstan on Mutual Insurance dated 5 July 2006, government authorized farmers - 86 - to form mutual insurance associations to underwrite the compulsory crop insurance policy. As of 2011, there are more than 38 farmers‘ mutual insurance associations offering crop insurance in Kazakhstan. 3.86. Private insurance companies and farmer’ mutual associations providing crop insurance are not equally regulated. While private insurance companies are subject to the regulation of the Agency for Financial Market and Financial Regulation and Control, the farmers‘ mutual crop insurance associations are regulated separately by the law of the Republic of Kazakhstan on Mutual Insurance. The Agency for Financial Market and Financial Regulation and Control exercises a tight control over the insurance activities in the country including crop insurance. Non-life Private insurance companies are required to have a minimum capital of KZT 1.2 billion (US$ 8.3 million) in order to operate. On top of the minimum capital requirements, private insurance companies are frequently monitored on their solvency, net retentions, and the implementation of risk management procedures. Conversely, the farmers‘ mutual associations offering crop insurance are not regulated and supervised by the Agency for Financial Market and Financial Regulation and Control. Therefore, farmers‘ mutual associations are not subject to any minimum capital requirements, or controls over net retentions and solvency requirements. In fact, currently, an association with more than 250 farmers can constitute a farmer mutual association to provide crop insurance to its associates without having to establish any capital or claims reserves. According to the interviews held with farmers‘ mutual insurance associations during the mission, several mutuals had not collected enough premiums to pay the full amount of claims incurred during the crop season 2010. Therefore, in order to remain solvent, these mutual insurers had been forced to reduce the 2010 paid claims on a proportional basis according to the total amount of premiums they had collected during the year. Evaluation of Crop Insurance Effectiveness for Key Stakeholders 3.87. This final sub-section briefly summarises the main issues and concerns which face the three main stakeholder groups which are involved in the Kazakhstan obligatory crop insurance scheme namely, farmers, insurers and government. (Box 3.2.). Farmers 3.88. It appears that the compulsory nature of the crop insurance scheme is very unpopular with many farmers. With very few exceptions farmers who were met during the conduct of this study expressed their major dissatisfaction that the crop insurance scheme is obligatory. In response some farmers refuse to purchase crop insurance and they prefer to pay the accompanying fines. In addition the majority of farmers purchase the minimum permitted level of normative costs of production protection in order to minimize the costs of their crop insurance premiums. Although most farmers appear to understand that the very low levels of sum insured coverage they are purchasing do not afford them with adequate financial risk protection in the event of a major crop loss, they have such low expectations of the scheme that they would prefer to purchase minimum cover in the knowledge that they will only receive an indemnity in the extreme case of near total crop failure. There also appears to be a problem in farmers‘ understanding of the operation of the policy and their dissatisfaction with the cumbersome indemnity and loss assessment procedures. - 87 - Insurance Companies 3.89. In recent years the private insurance companies have lost money underwriting the obligatory crop insurance scheme and they have increasing withdrawn from underwriting this class of business. Originally seven insurance companies signed up their interest to support the obligatory crop insurance scheme, but on account of severe financial underwriting losses, today most companies have ceased to support the scheme and only three companies are still actively underwriting the scheme. The insurance industry‘s main concerns are that because of the compulsory nature of the scheme they can only exert very limited control over individual risk selection and underwriting: they therefore lack any incentives to invest in crop underwriting and claims management staff and operating systems and procedures. In particular the insurance companies are concerned about their exposures in western Kazakhstan where they have incurred major losses over the past 6 years. In spite of government financial support in the form of 50% share on all claims, the insurance companies face major financial exposures because they are unable to access commercial quota share and or non-proportional reinsurance to cap/limit their liability to loss. 3.90. The private commercial insurers are concerned what they perceive to be unfair competition from the Farmer Mutual Crop Insurance Associations. The private insurers are concerned that the mutual insurers are not subject to insurance regulations and they do not need to maintain minimum capital or claims reserves. Insurers also noted that the mutuals were not subject to the minimum crop insurance premium rates and that they therefore faced unfair competition whereby the mutuals under-cut their rates. Finally the insurers expressed their concerns over the reputational risks the insurance industry faced if the mutuals were not able to meet their claims liabilities as occurred in 2010. Government 3.91. The Kazakhstan compulsory crop insurance scheme was launched with very well intentioned social objectives, but it appears that it is failing to achieve its financial and institutional objectives. The program was originally conceived as a mechanism to ensure that all farm workers and small peasant farmers would receive a minimum indemnity in the event of crop failure due to drought or other natural and climatic perils. The obligatory nature of the scheme was intended to be a short-term measure which would enable the insurance sector to develop a sound and stable crop insurance market based on a PPP between private and public sectors, while at the same time providing time to educate farmers in the benefits of crop insurance so that they would continue to purchase cover once the scheme was made voluntary. After 6 years the scheme has failed to develop a strong crop insurance market and to educate farmers. The fact that the program is very unpopular with many farmers also suggests it is failing to meet its social objectives. Finally government faces major uncertainties over its financial exposure to claims. 3.92. In summary the current obligatory crop insurance scheme is failing to meet the requirements of all three major stakeholders and for these reasons government has request The World Bank to provide technical support to review the scheme and to identify practical options for strengthening the scheme in future. - 88 - Box 3.2. Key issues facing main stakeholders in the compulsory crop insurance scheme Source: Authors - 89 - Chapter 4: Strategy and Options for Strengthening the Current Crop Insurance Program 4.1. This section presents a series of options and recommendations on the possible ways to strengthen and improve the current obligatory national crop insurance scheme for GRK to consider. These options and recommendations are made on the basis of the detailed diagnostic technical, institutional, financial and operational review of the obligatory crop insurance scheme that presented in Section 3, and where relevant, draws on international practice and experience. 4.2. A phased approach is recommended for strengthening and improving the obligatory crop insurance scheme in Kazakhstan and for gradually converting this into a fully market- based commercial crop insurance system. Agriculture in Kazakhstan has had to adjust to the major structural changes that occurred at independence 20 years ago including the switch from collective and state farming systems to individual Household farms, small to medium sized Commercial Farms (CFs) and large Production Enterprises (PEs)along with the financial challenges this brought to individual farmers in terms of the need to invest in crop production machinery and equipment and to purchase seeds and fertilisers and plant protection chemicals. Since 2004, GRK, MOA, the insurance companies and other stakeholders (including since 2008, the farmer mutual crop insurance associations) have invested in the promotion and implementation of mandatory crop insurance scheme. These investments have laid the foundations for a national crop insurance scheme and this is a considerable achievement given the logistical challenges of providing crop insurance to large numbers of often small to medium farmers who are widely scattered over very large geographical regions. The obligatory crop insurance scheme has, however, in the past three years suffered major financial losses due to very severe droughts in parts of Kazakhstan and unless the scheme is reformed and strengthened it is in danger of failing. It is also recognized that any changes to the existing crop insurance scheme will need gradually to be introduced and phased in over a three to five year period with the central aim of moving this program onto a sounder commercial basis and more market oriented approach. 4.3. Under the recommended phased approach towards a market based national crop insurance scheme, three major phases and strategy options have been identified on the basis of this study. The phased approach offers government a sequential set of measures which can be adopted over the next three to five years to move this program onto a much more sound commercial footing and which will be capable of attracting international reinsurance support. The three phases include: 1. Strengthening of current obligatory insurance scheme and achievement of financial stability (short-term, 1-3 years) 2. Transition towards a market based Crop Insurance System (short-term, 1-3 years) 3. Transformation of the Obligatory scheme into a National Commercial Crop Insurance Pool backed up by International Reinsurance under a suitable public private partnership(medium-term, 3-5 years) 4.4. Within each of these strategic options a series of legal and regulatory, technical, operation, institutional, financial and reinsurance recommendations are made. - 90 - Phase 1: Returning the Obligatory Crop Insurance Scheme to Profitability and Financial Stability 4.5. This section sets out a series of short-term measures designed to return the crop insurance scheme to profitability and to ensure the continued participation of the private commercial insurers and finally to ensure financial stability of the system. Legal and Regulatory Considerations 4.6. Review and Reform of Obligatory Crop Insurance Law. There is a clear need to review and amend the Obligatory Crop Insurance Law of 2004 to make this more market oriented and to ensure the viability of the current crop insurance scheme in Kazakhstan. International experience shows that crop insurance is most successful where this is implemented by the private insurance sector on a strictly commercial basis and that insurance companies should have responsibility for risk selection and underwriting and for setting their own terms and conditions. Under a public- private partnership (PPP), agricultural insurance legislation should define the types of public sector support to agricultural insurance and set the overall framework for agricultural insurance: an agricultural insurance law should not, however prescribe specific terms and conditions of cover including sums insured, rates, loss assessment procedures etc and which should be the responsibility of underwriters. 4.7. In the short-term, it is recommended that the crop insurance law should be amended to permit insurance companies to set their own policy terms and conditions and rates and sum insured levels. Immediate changes in legislation which should be considered include to remove the minimum and maximum premium rates from the law and to permit the insurance companies to set their own premium rates for each crop in each Oblast and Rayon. Similarly more flexibility should be given to insurers over the sum insured they offer to each farmer under the LIC policy. 4.8. Options for gradually phasing out obligatory crop insurance and replacing this by a system of voluntary crop insurance need to be addressed by crop insurance policy makers in Kazakhstan. In Kazakhstan, GRK originally planned that crop insurance would be compulsory for the first 2 or 3 years only and that once farmers had gained knowledge and experience with crop insurance that this would be moved onto a voluntary basis. To date, however, the Law has not been amended to permit voluntary insurance. Some of the major problems associated with obligatory insurance have been highlighted in Section 3 including farmer‘s dissatisfaction with crop insurance which some regard as a tax, the fact that farmers in low risk regions who adopt high levels of risk management practices and technology are currently cross-subsidising those farmers in the most risk-prone areas and whom arguably should not be producing annual crops because of the very high frequency and severity of drought and other causes of loss. Equally the obligatory nature of the program causes major problems for insurers because they are unable to exercise any form of risk selection and or accumulation control, both functions are keys for underwriting a commercially viable crop insurance program. As evidenced over the past 6 years, most private insurers have lost a lot of money writing the obligatory crop insurance scheme and it is very unlikely that any new companies will register to underwrite this business so long as cover is obligatory. While government has been very successful in promoting mutual insurers a number of the organizations have also experienced major losses in 2010 thereby threatening their future viability. Options for gradually phasing out obligatory crop insurance and replacing this by a combination of voluntary insurance and crop-credit linked insurance should be explored further by the various working groups. - 91 - 4.9. In the medium term the crop insurance Law should be amended to provide the basis for a fully market based national crop insurance scheme under an appropriate public-private partnership. Under this strategic option GRK may wish to consider the establishment of some form of crop insurance pool drawing on the experience of countries such as Spain, Turkey and South Korea which operate commercial agricultural insurance pool schemes under strong PPP relationships. (The potential benefits of coinsurance pools in agriculture are reviewed in this Section). The current crop insurance law would need to be amended to reflect these changes under a national pool program. 4.10. Any changes in the crop insurance legislation must consider both the Private Commercial Insurance sector and the Private Farmers Mutual Crop Insurance Associations. The previous section has indicated that currently the mutuals are regulated separately from the private insurance companies under the Law of The Republic of Kazakhstan on Mutual Insurance dated 07.05.2007. Under this Mutual Act, the farmers‘ mutual crop insurance associations do not have to meet any minimum capital reserves requirements to cover catastrophe losses, they have considerable flexibility over the premium rates they charge their members including the ability to reduce rates or suspend premium payments the following year if they have achieved an underwriting profit, and they are not legally obliged to settle claims which exceed their collected premiums. The Insurance Companies are regulated both by the Insurance Act and by the Obligatory Crop Insurance Law and have to meet minimum capital requirements and to meet their claims liabilities in full. In order to create a level playing field, it is important that the Mutuals are brought under the same regulations as the private commercial insurers. 4.11. The Obligatory Crop Insurance Law will need to be reviewed and redrafted to support the introduction of market-based crop insurance. The Law of 2004 will require significant modifications and amendments in order to move to a market-based crop insurance system. This task will need to be conducted by a legal expert with knowledge of international agricultural insurance legislation. Such an expert should be familiar with the legal and regulatory frameworks on major PPP agricultural insurance schemes in countries such as Spain and Turkey and which may have applications to Kazakhstan. Technical Strengthening of the Loss of Investment Costs Policy Design 4.12. On the assumption that in the short-term government and the insurance sector will need to continue to offer the uniform Loss of Investment Costs (LIC) Policy, a series of recommendations are identified for improving and strengthening this product. It is assumed that there will not be time prior to the 2012 crop insurance season to introduce any new crop insurance products or programs and therefore the standard LIC policy will continue to be marketed for all insurance crops. A series of simple practical changes and strengthening in the LIC policy design are identified in Box 4.1. and these measures are reviewed further below. There should be time to introduce these amendments to the existing LIC policy for implementation in the 2012 season starting in May 2012. 4.13. In the short term, even if government elects to maintain LIC crop insurance as an obligatory class of insurance for farmers, special consideration will need to be given to farmers located in Aktobe and WKO. The previous section has shown that Aktobe and WKO are so exposed to drought losses that conventional crop insurance cannot continue be provided to farmers in these two Oblasts and that alternative solutions need to be considered if the overall program is to be stabilized and returned to profitability. GRK will therefore need to consider short-term amendments to the crop insurance law to reflect this situation. - 92 - 4.14. There is a need to introduce a sales period and final sales cut-off date(s) at least one month prior to sowing to avoid situations of anti-selection and moral hazard. Currently the policy is very exposed to anti-selection because the final date for binding cover is 2 weeks after completion of sowing: farmers can therefore monitor the development of the season and if pre- existing drought conditions are developing they can purchase the highest level of sum insured cover in the expectation of receiving a higher indemnity. Most multiple-peril crop insurance (MPCI) programs introduce final sales cut-off dates for each crop in each region at least one month prior to the recommended sowing dates. It is recommended that policy makers in Kazakhstan introduce a similar sales cut-off date on the obligatory crop insurance scheme. 4.15. It is recommended that the cover period should be clearly stated in the Policy wording and specifically that cover should only incept following germination of the crop and full-stand establishment (defined as >10 centimeters in height for grasses ( Graminae) and two-leaf stage for dicotyledons). The reason for this recommendation is to avoid instances of moral hazard whereby farmers fail to adopt the technical recommended improved seed varieties and sowing densities and optimal sowing dates etc. 4.16. The Insured Unit for the purposes of loss adjustment should be strengthened. Currently the Insured Unit (IU) is understood to be the individual ―field‖ of each crop. In northern Kazakhstan fields may be extremely large (>1,000 Ha) and field boundaries are very poorly demarcated on the ground and in order to avoid conflict at the time of loss assessment it is recommended that this definition should be tightened-up. For small farm units of less than 100 hectares, underwriters may wish to consider redefining the IU as ―the sown area of all fields of the same crop type, gown in the same farm location‖. . 4.17. It is recommended that Crop Insurers should be given greater flexibility to set their own sum insured levels with each insured according to the Insured’s requirements for cover. It is recommended that government modify current legislation to permit insurance companies to negotiate the production cost-based sum insured levels with their clients or in other words to provide farmers with more choice. It is recognized that if sums insured are significantly increased this will have implications for premium rating under the current LIC basis of insurance and indemnity. (See rating section for further discussion). 4.18. Government should amend the Obligatory Law to permit insurance companies to set their own premium rates for each crop in each zone. Actuarial rating should be introduced to reflect differences in risk exposures between Rayons in each Oblast and possibly differences in technology levels and risk exposures between different types of farmer.35 In conjunction with revisions to the rating procedures, policy makers should study the merits of introducing an individual farmer bonus-malus system in future: farmers who have not submitted claims in the past would benefit from a premium discount at renewal, while those farmers who have high claims would have their premium rates loaded accordingly. Since the FFSA maintains a data base of claims dating back to scheme inception in 2005 it would be a relatively simple task for the insurance companies to check which farmers are eligible for a premium rate discount and which would be subject to a rate increase. 35 To adjust rates due to local farm-level factors underwriters in Argentina typically select several factors which influence risk exposures and which are very easy to verify in the event of a claim including: sowing dates, previous crop, type of tillage, soil type, texture and evidence of water-logging. These factor are duly elicited in the proposal form and each of them given a score which was then used to adjust premium rates up or down - 93 - 4.19. For the purposes of valuing actual revenue in cases of partial crop losses, it is recommended that a pre-agreed unit value be specified in the Policy Wording at the time of contracting insurance. The reason for this proposal is to overcome the drawbacks of the current scheme where partial losses are value at the current local market price of the crop at the time of loss: Section 3 highlighted the uncertainty this causes to insurers and government alike because they cannot assess their financial liability until losses are incurred. It is normal on any MPCI policy including LIC policies to pre-agree and specify in the policy special conditions the Unit Insured price which will be used to value crop output and which is usually based on historical average farm-gate prices for each crop in each zone. Similarly the current system is very unclear to farmers who do not know how much of their crop production and yield they will have to lose before the value of salvage falls short of their costs of production invested in growing the crop thereby opening the policy for a claim. A pre-agreed sales price means that the farmer can calculate exactly how much yield he needs to lose in order to trigger a claims settlement. Box 4.1. Recommendations for Strengthening Compulsory LIC Crop Insurance Policy Item Detail Criteria for Even if government decide in the short-term to maintain Compulsory crop insurance Acceptance of Risk / for all producers special consideration will need to be given to farmers located in Compulsion of Cover Aktobe and WKO. Insured Perils Maintain current coverage to include loss or damage to crop production due to: adverse weather events‖, Sales Cut-off date Introduce a policy sales cut-off date 1st April, or date tba Cover Period From the time of crop emergence and full stand establishment (e.g. wheat 10 cm stage in crop) through to completion of the crop harvest. Insured Unit Strengthen the definition of the Insured Unit which is currently defined as the ―individual field‖. For small farms of less than 250 Ha consider re-defining the IU as ―the total area of all fields of the same crop grown in the same location or farm‖. Sum Insured Government should amend the Obligatory Law to permit Insurance Companies to have the option to establish an agreed sum insured with each farmer according to farmer‘s own circumstances and crop insurance requirements. Premium Rates Government should amend the Law to permit insurance companies to set their own premium rates for each crop in each zone. Actuarial rating should be introduced to reflect differences in risk exposures between Rayons in each Oblast and possibly differences in technology and risk exposures between different types of farmer Bonus Malus System It is recommended that underwriters introduce a Bonus-Malus system on the compulsory crop insurance scheme. The objectives of the bonus-malus include:  Over time to introduce individual farmer experience rating within each Rayon. Farmers who do not submit claims would be rewarded with reduced premium rates over time: those farmers who submit frequent claims would be penalized by higher premium rates to reflect their higher risk exposure.  A bonus-malus system should reduce the tendency for farmers to submit speculative claims notices in the hope of receiving an indemnity to only those cases where a major insured cause of loss has occurred which is likely to give rise to a claims settlement. This would reduce the costs of in-field loss assessment for the Insurance companies. Basis of Indemnity In the case of a Partial Loss of area and or yield, the estimation of salvage and Claims (harvestable production) from the affected area should continue to be estimated in Settlement field immediately prior to harvest. However, the Crop Sale Price which is used to value the salvage should be pre-agreed based on an historical sales price for each crop in each Rayon and stated in the Policy Wording. Where the Value of Salvage (Crop Revenue) falls short of the Investment Costs the shortfall is indemnified. Source: Authors - 94 - Increasing the Sum Insured levels 4.20. The current Loss of Investment costs policy does not provide adequate levels of protection to the majority of farmers: in most cases farmers only purchase the cheapest “nominal costs” coverage which for spring wheat is equivalent to about 10% to 15% only of the value of the expected crop revenue. If the sum insured coverage levels were to be increased under this program, this would have implications for all stakeholders. For farmers higher coverage would mean that the likelihood of their receiving a claims payment would be significantly increased and also in the event of a claim the indemnity should be adequate to cover their production costs in full and to ensure they can repay any due credit: the downside would be that higher premium rates would have to be levied if the coverage levels were increased (See below for further discussion). For the private insurance companies and mutuals, increased coverage levels would have to be accompanied by actuarial rate increases and with the increased sum insured liability it would be essential for the insurers to have a comprehensive reinsurance protection program in place. For government the higher coverage levels would mean a correspondingly higher budgetary allocation would need to be made to cover the 50% claims reimbursements. There is a need carefully to set the sums insured in each Rayon according to: i) the actual production costs of different types of farmer in each Rayon and ii) the risk exposure in each rayon. For example a sum insured of KZT 10,000 per hectare can be easily covered in NKO at a reasonable premium rate, but to provide the same high sum insured in Pavlodar or Aktobe would require extremely high premium rates. Revising the Crop Premium Rating Methodology 4.21. There is a need to revise the Oblast-level crop premium rating methodology and to update the premium rates on an actuarial basis. The 6-year long-term gross loss ratio at end 2010 was 140% (equal to a 70% loss ratio net of the government 50% claims reimbursement), suggesting that on average the scheme has operated on a break-even basis. The premium rates were last revised by government in 2008 and since then the scheme results have deteriorated badly over the past three years during which time the average loss ratio has been 182% (91% net of the government 50% claims subsidy). In some Oblasts including NKO and Kostanay the scheme has performed very well, but in others including Aktobe, WKO, and Zhambyl the claims performance has been very poor and the scheme is now severely under-rated in these Oblasts and rates require adjusting on an actuarial basis. 4.22. The Law specifies a system of minimum and maximum premium rates per Oblast, but given the differences in risk exposures and claims experience between rayons it is also recommended that a Rayon-level rating system is now introduced on the scheme. There are now up to 6-years of actual claims experience for each crop in each Rayon/ Oblast to permit the premium rates to be revised and updated on an actuarial basis in each Rayon. A simple actuarial analysis has been conducted on the 6-year actual premiums and claims at Rayon and Oblast level to examine the rate changes that would be required to achieve a target loss ratio of 60% (after application of the government 50% claims subsidies) on the scheme. The results of this analysis at Oblast-level are summarized below and the full results are presented by Rayon in Annex 4. The analysis shows that 6-year actual results at the Oblast-level indicates that on average the current premium rates are adequate to generate a 60% loss ratio net of the government 50% claims reimbursement in Akmola, Kostanay, Kyslorda and NKO (Caution should be exercised in interpreting the calculated rates in Kyslorda because very little business has been underwritten in this Oblast to date and the results may not be representative of the real patter of claims). However, in order to achieve a 60% target loss ratio very significant rate increases would be required in Aktobe (average premium rate increased from 5.8% to 18.5%) and in WKO (average - 95 - premium rate increase from 7.8% to 33.0%) and in Zhambyl (rate increase from 3.5% to 12.1%). Smaller actuarial rate increases are also required in EKO and Pavlodar in order to achieve a net loss ratio of 60% after government 50% claims subsidy. (Table 4.1 and Figure 4.1). Table 4.1: Adjustments in Average Oblast-level Premium Rates to achieve an average 60% target Loss Ratio Adjusted Gross Adjusted Net 6-year Actual 6 Year Premium Rate for Premium Rate 60% Implied Average Average Loss Oblast 60% Target Loss Target Loss Ratio Percentage Rate Premium Rate Cost (2005- Ratio (before FFSA (after FFSA change required (2005-2010) 2010) intervention ) intervention) Akmola 1.9% 2.0% 3.38% 1.69% -13% Aktobe 5.8% 22.2% 36.97% 18.48% 217% Almaty 3.5% 4.6% 7.64% 3.82% 10% EKO 3.4% 6.0% 9.96% 4.98% 47% Karaganda 4.7% 4.7% 7.86% 3.93% -16% Kostanay 1.7% 1.3% 2.10% 1.05% -40% Kysylorda 5.2% 0.1% 0.15% 0.08% -99% NKO 2.0% 0.5% 0.79% 0.40% -80% Pavlodar 3.6% 7.1% 11.81% 5.90% 66% SKO 2.7% 0.8% 1.34% 0.67% -75% WKO 7.8% 39.6% 65.94% 32.97% 323% Zhambyl 3.5% 14.6% 24.27% 12.14% 243% Total 2.4% 3.4% 5.65% 2.82% 17% Source: Authors analysis of FFSA 6-years crop insurance results 2005-2010 Figure 4.1. Comparison of Oblast-level average crop insurance premium rates with premium rates required to achieve 60% loss ratio (net of government 50% claims subsidies) 35.00% 30.00% Premium Rate (%) 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Actual Average Premium Rate Adjusted Premium Rate 60% Target Loss Ratio (with government 50% claims subsidy) Source: Authors analysis of FFSA 6-years crop insurance results 20005-2010 - 96 - 4.23. Scheme managers should also examine the merits of introducing a Bonus-Malus System to adjust the premium rates charged to individual farmers according to their own individual claims experience for each crop over the past 6 years. Since the obligatory crop insurance system has been in operation for 6 years it would also be possible in future to introduce a Bonus-Malus system for adjusting individual farmer‘s premium rates according to their own claims experience over time: low claimants would benefit from lowered premium rates, high claimants would be penalized through higher rates. The bonus-malus system can also be effective in reducing unnecessary loss assessment expenses for insurers36. The Crop Insurance Law would need amending to permit the introduction of a bonus-malus system. 4.24. If in future farmers elect higher sum insured levels under the salvage-based Loss of Investment Cost Policy this will have important implications for rating. Under the LIC Policy if a farmer elects to insure a higher level of production costs this directly increases the threshold or break-even yield for a given sale price of the crop. The risk of incurring an insured loss also increases with higher levels of insured or break-even yield and this means that higher premium rates would have to be applied at higher sum insured levels. Under the current rating system the setting of maximum and minimum rates for each crop in each Oblast may in part reflect the range of sum insured levels from the low technology/minimum costs of production basis through to the much higher costs associated with the scientific technology package. However, the only systematic way of calculating the rate increases that would need to accompany higher sums insured is through an analysis of yield variation at Rayon and preferably individual farmer levels. Such an analysis cannot be conducted on the past 6-years claims experience. Measures to Improve Scheme Profitability 4.25. The 6-year results show that the financial viability of the crop insurance scheme is being severely prejudiced by the inclusion of Aktobe and WKO Oblasts and those options for these two Oblasts must be considered now if the scheme is to return to profitability. Section 3 showed that over the past 6 years Aktobe and WKO in western Kazakhstan have collectively accounted for only 4.8% of total scheme liability, but have contributed 41% of total claims and respectively have 6-year long-term loss ratios of 381% and 507%. These two Oblasts are severely affecting the financial viability of the whole of the national crop insurance program and measures of controlling the claims costs in these two Oblasts must urgently be introduced. Options presented below include: (i) increasing government‘s share of claims subsidies from 50% to 75% or even 100% in WKO and Aktobe and (ii) removing these 2 Oblasts from the Crop Insurance scheme and establishing a separate government compensation mechanism for farmers in these 2 Oblasts. As If Analysis With and Without Aktobe and WKO 4.26. An “As If” analysis of the 6-year results shows that the actual long term loss ratio would be reduced from 140% to 96% with the elimination of Aktobe and WKO and that following the government’s 50% participation in claims the program would have generated 36 The bonus malus system can be also implemented to reduce the number of false claims declarations and to reduce the costs of unnecessary loss assessment visits. In Brazil, Allianca do Brasil has successfully implemented a bonus malus system to avoid this problem. Farmers automatically lose a 5% bonus at the renewal of the policy if they submit a loss notice on their policy (this is regardless of whether or not the loss qualifies for a claims settlement). Claims personnel must be trained to explain this to the insured - 97 - sound underwriting profits for the insurance industry. If Aktobe and WKO had not been included in the scheme since 2005, this would have led to a reduction in total premium earnings of 13% only, but would have led to a huge saving in claims costs for insurers of 40% of actual claims (or a reduction in claims of KZT 3.3 billion – US$ 22.0 million) and a reduction in the 6- year long-term loss ratio from 140% to 96%. Following application of the government 50% reimbursement of claims the As If loss ratio for the insurance would have been reduced to about 48% (Table 4.2). Reference to Figure 4.2 shows that if Aktobe and WKO had not been included in the scheme, the performance of the scheme as measured by the reduction in the annual loss ratio would have been significantly improved in 2005 (loss ratio reduced from 119% to 71%), in 2009 (loss ratio reduced from 131% to 46%) and finally in 2010 (loss ratio halved from 261% to 136%). Notably in 2008 the As If loss ratio would have slightly increased from 156% to 180% because in this year the actual claims in Aktobe and WKO were very low. These two Oblasts (in common with much of the northern and central regions of Kazakhstan) were traditionally livestock grazing regions which were converted to cereal (mainly spring wheat) production, during Soviet times. The underlying problem, however, in most of Aktobe and WKO Oblasts is that soils are mostly poor, average annual rainfall is very low and in a normal year is barely adequate for growing spring wheat in most of the rayons located in these two Oblasts. As such these are very marginal Oblasts for spring wheat crop production and no matter how a crop insurance policy is structured in these two Oblasts, it will always be very heavily exposed to loss and the technical and commercial premium rates that would have to be charged would be so high (an average commercial rate prior to application of government claims subsidies of between 37% Aktobe and 66% WKO) as to make the scheme commercially unviable in these two Oblasts. There are, however, a few areas in these Oblasts where spring wheat production is more stable and where crop insurance might possibly be considered in future, albeit at very high rates, including Taskalisky and Kaztalisky Rayons located in the south west of WKO and Karagaly and Martuk Rayons in northern Aktobe Oblast Table 4.2. Comparison of Crop Insurance Accumulated 6-Year Results With and Without Aktobe and SKO (2005-2010) Item Whole Portfolio Modified Portfolio Percent (With Aktobe/WKO) (Without Aktobe/WKO) Reduction No. of Policies 140,961 134,193 -5% Total Insured Area (Ha) 73,770,915 69,453,803 -6% Sum Insured ('000 TH) 242,631,438 231,048,598 -5% Premiums ('000 TH) 5,861,958 5,102,670 -13% Average Premium Rate % 2% 2% -9% Claim payments ('000 TH) 8,222,776 4,910,314 -40% Loss Ratio % 140% 96% -31% Loss cost % 3.39% 2.13% -37% Source: Authors analysis of FFSA crop insurance data 2005-2010 - 98 - Figure 4.2. Comparison of Annual Loss Ratio (%) with and without Aktobe and WKO 300% 261% 250% 200% 180% Loss Ratio % 156% 150% 131% 136% 119% 100% 71% 70% 68% 70% 65% 46% 50% 0% 2005 2006 2007 2008 2009 2010 WITH Aktobe / WKO WITHOUT Aktobe / WKO Source: Authors analysis of FFSA crop insurance data 2005-2010 Future Options for Aktobe and WKO 4.27. The first option which is being examined by the insurance companies is to continue underwriting the LIC scheme in Aktobe and WKO on the understanding that GRK will agree to reimburse these insurers for a higher percentage of claims in these two Oblasts. Currently government reimburses insurers for 50% of the value of all paid claims. Under this option, GRK would agree to indemnify a higher percentage of claims incurred in Aktobe and WKO only: according to the insurance industry they have already requested government to consider raising their share to 70% or 75% of total claims in these 2 Oblasts in 2012. Under the current study the additional costs to government of financing i) 75% and ii) 100% of claims in Aktobe and WKO have been calculated based on an analysis of the past 6-year actual claims and the assumption that government reimburses exactly 50% of all claims in the other Oblasts. Over the past 6 years, government‘s 50% contribution to claims costs would have amounted to GTZ 4.1 billion or an annual average of KZT 0.685 billion per year. If Government agreed to increase their share of claims in Aktobe and WKO, the As If annual cost to government would increase to i) Option A 75% claims share in Aktobe and WKO: KZT 0.823 billion per year or an increase of 20% and ii) Option B 100% claims share in Aktobe and WKO: KZT 0.961 billion per year or an increase of 40% (this latter 100% claims subsidy option shows the maximum expected costs to government if it wishes to continue including Aktobe and WKO in the LIC scheme and assuming the insurance companies act purely as administrators in these 2 Oblasts, but do not accept any claims liability: the ―As if‖ 100% claims cost to government in these 2 Oblasts would be an average of about KZT 550 million or US$ 3.7 million per year). (Table 4.3). While this option may provide government a short-term solution to retaining the support and underwriting capacity of the insurance industry and in increasing the profitability of the program after the government‘s incr eased claims reimbursements, in the medium to long-term this does not tackle the fundamental issue that commercial spring wheat crop insurance is not financially viable in most of the rayons located in Aktobe and WKO Oblasts. - 99 - Table 4.3. Comparison of Budgetary implications to Government of increasing the claims compensation level from 50% to 75% and 100% of claims in Aktobe and WKO (KZT '000) Option A. As If Option B. As If Cost to Cost to Total Paid Government Government with % Government with % Year Claims 50% share share of 75% increase share of 100% increase claims claims in Aktobe claims in Aktobe and WKO and WKO 2005 1,064,870 532,435 656,273 23% 780,110 47% 2006 477,670 238,835 244,427 2% 250,018 5% 2007 700,538 350,269 377,709 8% 405,149 16% 2008 1,709,623 854,812 869,927 2% 885,042 4% 2009 1,465,129 732,565 996,650 36% 1,260,735 72% 2010 2,804,945 1,402,473 1,794,519 28% 2,186,565 56% Total 8,222,776 4,111,388 4,939,504 20% 5,767,619 40% Average 1,370,463 685,231 823,251 20% 961,270 40% Source: Authors analysis of FFSA crop insurance data 2005-2010 Notes: * This analysis is based on the assumption that government‘s share of total paid claims is exactly 50%. Reference to Section 3, shows that over the past 6 years government‘s share has been slightly lower at 47% of total claims. 4.28. The second option for government to consider would be to take Aktobe and WKO out of the crop insurance scheme altogether and to establish a separate disaster relief scheme for producers of annual crops in these 2 Oblasts. The argument for taking Aktobe and WKO altogether out of the scheme centre on: i) these are very marginal rainfall areas for growing annual grain crops and ii) from a crop insurance viewpoint these areas are effectively uninsurable with 6 year long-term loss costs of 22% and 40% respectively and catastrophe losses every other years with loss ratios as high as 1000% in major drought years. It is therefore necessary for government to consider whether to continue to promote annual crops in these Oblasts and to use a separate disaster relief fund to compensate farmers in severe drought years. The cost of this program for Aktobe and WKO assuming the same 100% original claims costs and compensation levels as under the current scheme would average about KZT 550 million (US$ 3.7 million) per year and with a peak of about KZT 1.6 billion (US$ 10.7 million) in a very severe drought year such as 2010. Strengthening the Scheme‟s Operating Systems and Procedures 4.29. In conjunction with the design changes identified to strengthen the LIC policy, there are a series of potential measures scheme administration should consider in order to improve underwriting and claims operating systems and procedures and to reduce the costs of these operations. Box 4.2 summarises some of the key identified operational areas which require strengthening. 4.30. Loss notification procedures are cumbersome and require the Insured to obtain a weather report from the nearest meteorological station to validate that an insured event has occurred. While an event report might be used to prove that a specific event peril such as excess rain leading to flooding or frost has occurred (but not hail as this is usually very localised), such a report is usually irrelevant in the case of a progressive peril such as drought and which is the major cause of loss in Kazakhstan accounting for more than 95% of claims. It is recommended that the loss notification procedure must be continued but that the requirement for the insured to - 100 - submit a meteorological report be dropped in future (This will require an amendment to the crop insurance Law). The introduction of a Bonus-Malus system is also recommended to reduce the propensity of farmers to declare small frictional losses and to only declare major losses which are likely to give rise to a claims settlement: this measure should hopefully reduce the number of in in-field loss assessment exercises that need to be conducted and lead to cost savings. 4.31. There is a need to review the in-field loss assessment procedures on the obligatory crop insurance scheme with a view to streamlining this activity and making it more cost-effective. Currently the Law stipulates that once a loss has been notified by the Insured Farmer, representatives from 5 organisations must form a committee to assess the loss in-field. This committee comprises 1 representative from the: local administration, FFSA, local Agent, Insurer and finally the Insured farmer. Each party is responsible for covering their own costs of attending and assessing the loss and the process is very time consuming and costly. Furthermore the process is weighted against the Insurer. In most countries, loss assessment functions are carried out exclusively by the Insurer‘s own trained loss assessors or by a third party qualified loss adjuster and the farmer attends the loss assessment exercise. No other parties are required in the in-field loss assessment procedure. The loss assessor adopts standardized procedures for estimating the amount of yield reduction or loss and the farmer is required to approve or refute the assessed loss, in which case a second assessment will be made and if the Insured is still not satisfied with the outcome of the loss assessment, he is entitled to request this passes to arbitration. It is recommended that in Kazakhstan scheme management review the in-field loss assessment procedures and reduce the number of persons required to attend the loss assessment. In some countries, crop loss assessment functions are contracted out to specialist companies of loss adjusters. Although Kazakhstan does have certified and approved companies of loss adjusters, at present none of them specialize in crop loss assessment, but in the medium term options for creating such specialist entities should be explored. 4.32. In future it may be possible to use remote sensing as an aid to the crop loss assessment process. The NSA has a very well developed remote sensing capability to provide crop monitoring at a resolution of the ―individual field‖ in northern Kazakhstan where field size tends to be much larger than in the south. There are many potential applications of the NSA‘s crop monitoring services to the existing crop insurance scheme, especially in the area of yield determination at the farm level. Going forward it is recommended that crop insurance scheme management should meet with the NSA to review potential remote sensing services that could be provided to monitoring major loss events on this scheme. 4.33. It is also recommended that a review of policy marketing and distribution channels be conducted with a view to seeking to reduce administrative costs. Currently most policy promotion and sales is handled by individual sales agents in each Oblast and Rayon, but potentially costs be used if these functions were channeled through rural aggregators such as farmer cooperatives, input dealers and rural banks, including MFIs. 4.34. It would appear that there is a need to accompany improvements in the crop insurance policy design and scheme implementation with a much higher level of farmer awareness and education and training programs on the role of crop insurance. Currently it appears that there is both a lack of understanding on the part of many farmers on how the LIC policy works and also a very high level of skepticism of the benefits of crop insurance. In parallel to the improvements in policy design and rating and operational procedures it is recommended that government provides support to the commercial insurance companies and mutual insurers to conduct crop insurance education and training programs. - 101 - Box 4.2. Options for streamlining and reducing the costs of policy marketing, pre- inspections and loss assessment systems and procedures Requirement Detail Pre-Inspections need to  Since the scheme is compulsory this should reduce potential anti- be introduced section and moral hazard which are often associated with multiple-peril crop insurance schemes.  Currently few of the insurance companies have regional networks of trained field agronomists to conduct pre-inspections and loss assessment and this task is likely to be prohibitively costly on small farm units.  A system of sample pre-inspections should be considered for the large farms in northern Kazakhstan Loss Notification and  The Law currently stipulates that 5 individuals (organizations) must be Loss Assessment involved in field-level loss assessment and this is a very costly exercise Procedures need to be for each party and requires revision. The loss assessment committee is streamlined and made comprised of 5 representatives from the: local administration, FFSA, more cost-effective local Agent, Insurance Company and finally the Insured. The composition of this committee is weighted against the Insurer.  In most countries, loss assessment functions are carried out exclusively by the Insurer‘s own trained loss assessors or by a third party qualified loss adjuster and the farmer attends the loss assessment exercise. The loss assessor adopts standardized procedures for estimating the amount of yield reduction or loss and the farmer is required to approve or refute the assessed loss, in which case a second assessment will be made. No other parties are required.  In Kazakhstan farmers are required by law to obtain a meteorological report from the nearest meteorological weather station to prove an insured event has occurred. This requirement does not seem relevant in the case of drought (which accounts for 95% of all losses to date) which is a progressive peril as opposed to being a specific event. It is recommended that this requirement is discontinued. Farmers should, however be required to continue to submit a loss notification report to the insurer within a specified period e.g. 48 hours for specific events (e.g. hail, flood, autumn frost) and also to notify drought losses when these become apparent.  A Bonus-Malus system should be introduced to dissuade farmers from submitting claims save where a major loss has occurred and which is likely to lead to an indemnity. Use of Remote Sensing to The NSA is already involved in applications of remote sensing to estimate support in-field loss crop sown area and production and yield estimates for the MOA and to assessment monitor crop status during the growing season. It is recommended that the crop insurance scheme managers should review potential supporting roles by the NSA at the time of loss assessment. Review Policy Marketing Currently most policy sales are made through local Agents located in each and Distribution Oblast and sub-region. The Agents are currently paid 10% brokerage by law. Channels with a view to Alternative crop insurance marketing and distribution channels should be reducing costs promoted in order to reduce costs including sales though cooperatives and farmer associations, input dealers, rural banks, grain merchants, etc. Farmer Education and In conjunction with the proposed improvements in operating systems and Training Programs on procedures, greater emphasis needs to be paid on farmer education and Crop Insurance training programs about the basis of insurance and indemnity on the current LIC program Source: Authors 2011 - 102 - Institutional Strengthening 4.35. In the short term there is a major challenge to find ways of encouraging more private commercial insurance companies to support the crop insurance scheme. Several measures have been identified in this section which will hopefully be attractive to local insurers including i) the introduction of actuarially determined premium rates , ii) the measures to reduce insurers‘ liability in Aktobe and WKO and iii) recommendations for strengthening loss assessment procedures and for giving insurers more direct control over this important function. However, in the medium term it is also probable that Insurers will insist of being given greater control over risk selection and underwriting function if they are to join the scheme. There is also a need to create a level playing field for commercial insurers and the farmer mutual insurance associations. Mutual insurance associations should in future be regulated under the Agency for Financial Market and Financial Institutions Regulation and Control and be required to follow the same guidelines in terms of the capital requirements and in terms of constitution of insurance reserves as per the commercial insurers. Failure to do so creates biased competition and the possibilities for the mutuals to fail to meet their financial obligations. Financial and Reinsurance Considerations 4.36. Crop insurance companies in Kazakhstan are very exposed to Catastrophe losses on their retentions and options for enhanced reinsurance protection need to be considered. GRK currently provides free of cost proportional reinsurance protection equal to 50% of the claims to the private insurance companies and mutuals in Kazakhstan. Currently, however, neither the private insurers nor the mutuals have any reinsurance protection on their 50% retentions and they are therefore very exposed to major systemic drought losses. In the case of the private insurance companies their ability to absorb and settle catastrophe losses is much higher than the mutuals because of their much larger size, formal requirements for capital and claims reserves and their diversified non-life insurance portfolios under which crop insurance only represents a very small fraction of their overall premium earnings and overall liability. However, there is still a need to design affordable reinsurance protection for the private companies on their 50% retentions through some form of Stop Loss Reinsurance protection (See Figure 4.3 with illustration of current GRK 50% quota share reinsurance protection and an alternative Non Proportional Stop loss reinsurance protection structure). 4.37. The Farmer Mutual Crop Insurance Associations are very exposed to losses which exceed their members premium contributions and ways of capping their exposure to catastrophe losses through some form on non-proportional or stop loss reinsurance protection need to be developed. The mutuals are usually small, do not have reserves and only underwrite crop insurance business: they have to settle claims out of the net premiums they receive from their members (these net premiums are typically in the order of 70% to 75% of gross premium after payment of acquisition costs and the mutual‘s own administration and operating expenses) and in any years where the actual claims exceed the average they are unlikely to be able to meet their liabilities in full. If the mutuals are to form a financially stable and viable alternative to private commercial crop insurance in future, it will be essential to put in place a formal risk layering and proportional and or non-proportional risk transfer (reinsurance) program for the mutuals in order to cap their losses at somewhere between 70% to 100% of their premiums 37. 37 In insurance terms this is referred to as Gross Net Premium Income (GNPI) which is the original gross premium the company has earned net of any policy cancellations and returns of premium. - 103 - Figure 4.3. Comparison of Proportional (Quota Share) and Non-Proportional (Stop Loss) Reinsurance Treaty Structures Proportional Quota Share Non Proportional Stop Loss Loss Ratio = 50% Insurance Company Retention 50% Government Reinsurance Stop Loss Reinsurance Layer Loss Ratio =100% Insurer Primary Retention Loss Ratio =0% Source: Authors 4.38. In the short-term it is unlikely that the Kazakhstan obligatory crop insurance scheme will be able to meet the standards required by international reinsurers to attract their capacity and therefore any non-proportional reinsurance solutions will probably have to be provided by GRK. To date GRK has paid out a total of KZT 3.8 billion in 50% quota share reinsured claims or an average of KZT 630 million per year to the crop insurers. This section briefly reviews whether it is may be more cost effective for the government to use this funding to support an excess of loss program and or to also finance premium subsidies. Simple Burning Cost Analysis for Government Excess of Loss Claims Cover 4.39. An as if analysis has been carried out on the 6-year actual claims under the assumption that instead of providing 50% quota share reinsurance protection, government would provide aggregate non-proportional excess of loss claims compensation for the crop insurance industry. This analysis assumes that if government had not provided free 50% proportional claims compensation over the past 6n years, the insurers would have had to double the premium rates they charged on this scheme to cover their 100% claims liability. Therefore the actual 6-year annual average premiums have been doubled to represent the original gross premiums the insured would have had to have charged. Two As If Claims scenarios are analysed: 1) Insurers retain claims up to 100% of the value of the original gross premiums (i.e. up to 100% loss ratio), and government provides unlimited protection for any losses excess of 100% of gross net premium income (GNPI) and 2) in recognition that insurers have to pay business acquisition costs (brokerage) and administration and operating costs charges out of their original gross premium thereby reducing the premium they retain to cover claims, the government claims - 104 - compensation program cuts in for any losses excess 70% of GNPI. A further assumption is that initially government excess of loss claims protection would be provided free of charge to the insurers. Finally the As If analysis has been carried out with and without Aktobe and WKO. The main caveat of this analysis is that the government excess of loss compensation cover can only be analysed on an annual aggregate industry-level basis and not by individual private commercial insurance company or farmers‘ mutual crop insurance association – in practice individual companies underwrite different portfolios in different regions of the country and would incur different premium to claims ratios and the claims to the government stop loss program would inevitably be higher than can be modelled under this simple As If analysis applied to the historical claims. 4.40. The results of this simple As If analysis are presented in Table 4.4 and show that with Aktobe and WKO included the government funded Aggregate Industry Excess of Loss Compensation with a priority of 100% of GNPI would have incurred one claim only in the 6- year period of KZT 0.66 billion in 2010 (average of 110 million per year) and with the reduced 70% of GNPI priority, two claims a small one in 2008 and then again in 2010 with total stop loss claims of KZT 1.5 billion (average of 347 million per year). If Aktobe and WKO had been excluded, the 100% of premium ―aggregate‖ excess of loss compensation cover would not have incurred any claims in the past 6 years and with reduced 75% priority only one small claim in 2008 of KZT 0.27 billion. Great caution must, however, be exercised in interpreting this simple 6-year aggregate As If analysis (a) because it is based on only six years data and although it includes the 2010 losses which was a very severe 1 in 50 drought year and which will bias the results towards this bad year, even more severe losses could occur say 1 in 100 years and (b) the aggregate analysis excluding Aktobe and WKO is unrealistic and masks the reality that individual insurance companies and mutuals would have incurred losses in excess of 100% of their premium on their own regional crop insurance portfolios over the past 6-years. The analysis is presented purely to illustrate the option of capping industry losses in catastrophe years through some form of government catastrophe claims compensation mechanism. Table 4.4. As If Analysis of Costs to Government of Aggregate Insurance Industry Excess of Loss Protection and Full Premium Rates charged to farmers (KZT „000) A) Including AKTOBE and WKO Actual Premium As If 100% Gross As If Cost to As If Cost to (with government Premium (without Actual Total Government for Government for Year 50% claims government 50% Paid Claims losses XS 100% losses XS 70% reinsurance) claims reinsurance) Gross Premium Gross Premium 2005 898,607 1,797,214 1,064,870 0 0 2006 684,722 1,369,444 477,670 0 0 2007 997,392 1,994,785 700,538 0 0 2008 1,093,232 2,186,463 1,709,623 0 179,099 2009 1,114,366 2,228,732 1,465,129 0 0 2010 1,073,639 2,147,278 2,804,945 657,668 1,301,851 Total 5,861,958 11,723,915 8,222,776 657,668 1,480,950 Average 976,993 1,953,986 1,370,463 109,611 246,825 B) Excluding AKTOBE and WKO Actual Premium As If 100% Gross As If Cost to As If Cost to Actual Total Year (with government Premium (without Government for Government for Paid Claims 50% claims government 50% losses XS 100% losses XS 70% - 105 - reinsurance) claims reinsurance) Gross Premium Gross Premium 2005 801,657 1,603,315 569,520 0 0 2006 673,999 1,347,999 455,304 0 0 2007 902,616 1,805,232 590,778 0 0 2008 916,678 1,833,355 1,649,162 0 365,814 2009 897,503 1,795,005 408,788 0 0 2010 910,218 1,820,435 1,236,762 0 0 Total 5,102,670 10,205,340 4,910,314 0 365,814 Average 850,445 1,700,890 818,386 0 60,969 Source: Authors analysis of FFSA Insurance data Indicative Rating for Aggregate Stop Loss Reinsurance protection for the Obligatory LIC Crop Insurance Scheme (Spring Wheat) 4.41. As an extension of the above simple 6-year As If analysis, some preliminary modelling has been conducted for spring wheat (which accounts for over 95% of the obligatory crop insurance portfolio) to estimate the layering and pricing on an Aggregate Stop Loss protection cover for the existing loss of investment cost scheme. This analysis using simulation techniques to calculate over 5000 iterations the expected aggregate losses (in terms of loss costs and or loss ratios) over the entire scheme. The results of the simulation exercise are then used to estimate the value of claims excess of the specified priorities (in this case 70% and 100% priorities assumed and full value reinsurance protection is provided excess of these priorities). Appropriate loading is then added to the calculated technical rates to cover reinsurers‘ costs and profit expectations. The results of this stop loss rating analysis are presented in Table 4.5. for the 2 priorities and with and without Aktobe and WKO. While caution must be exercised as this is a preliminary rating analysis, it is considered robust and suggests that the indicative costs of providing full value (i.e. up to 100% of TSI) Aggregate stop loss protection excess of 100% and 70% priorities would be in the order of KZT 0.27 billion to KZT 0.31 billion per year. Table 4.5. Illustrative Rating Analysis for Aggregate Stop Loss Protection for Obligatory Loss of Investment Costs Crop Insurance Scheme: Spring Wheat cover only. A. With Aktobe and WKO Total Sum Insured (KZT Billion) 41.80 Stop Loss Premium Limit Priority % Option XS. Amount (KZT (%TSI) (TSI) (%TSI) billion) 96.56% xs 3.44% 0.75% 0.313 Priority 70% of GNPI 95.09% xs 4.91% 0.64% 0.266 Priority 100% of GNPI B. Without Aktobe and WKO Total Sum Insured (KZT Billion) 39.20 Stop Loss Premium Limit Priority % Option XS Amount (KZT (%TSI) (TSI) (%TSI) billion) 96.65% xs 3.35% 0.66% 0.271 Priority 70% of GNPI 95.22% Xs 4.78% 0.57% 0.225 Priority 100% of GNPI Source: Authors - 106 - Premium Subsidy Considerations 4.42. With the switch from 50% proportional reinsurance to non proportional reinsurance and the parallel need for insurers to double their premium rates, GRK may also need to consider whether farmers can afford these rate increases or if it will be necessary to introduce premium subsidies. In order to illustrate the potential costs of premium subsidies, the above 6- year analysis has been expended to include both the aggregate industry-level stop loss reinsurance protection and 50% premium subsidies. Over the past 6 years, government‘s 50% of claims liability has been an average of 685 million per year, but in severe drought years such as 2010, government‘s 50% claims share was much higher at KZT 1.4 billion. If Aktobe and WKO had been excluded from the scheme, government‘s 50% quota share claims reimbursement would have been reduced to an average of KZT 409 million per year: this compares with the As If average annual costs to GRK of stop loss protection and 50% premium subsidies of between 850 million and 911 million per year. (Table 4.6.). Table 4.6. As If Analysis of Costs to Government of Aggregate Insurance Industry Excess of Loss Compensation scheme and assumed 50% Premium Subsidies (KZT „000) A) Including AKTOBE and WKO As If 100% Total Cost to As If Cost to As If Cost to Total Cost to Gross Premium Government Government for Government Government Government for (without 50% Percent losses XS 70% Year for losses XS for losses XS losses XS 100% government Premium Premium + 50% 100% Gross 70% Gross Premium + 50% 50% claims Subsidies Premium Premium Premium Premium Subsidies reinsurance) Subsidies 2005 1,797,214 0 0 898,607 898,607 898,607 2006 1,369,444 0 0 684,722 684,722 684,722 2007 1,994,785 0 0 997,392 997,392 997,392 2008 2,186,463 0 179,099 1,093,232 1,093,232 1,272,331 2009 2,228,732 0 0 1,114,366 1,114,366 1,114,366 2010 2,147,278 657,668 1,301,851 1,073,639 1,731,307 2,375,490 Total 11,723,915 657,668 1,480,950 5,861,958 6,519,625 7,342,908 Average 1,953,986 109,611 246,825 976,993 1,086,604 1,223,818 B) Excluding AKTOBE and WKO As If 100% As If Cost to As If Cost to Total Cost to Total Cost to Gross Premium Government Government for (without Government Government 50% Percent Government for losses XS 70% Year for losses XS for losses XS losses XS 100% government Premium Premium + 50% 100% Gross 75% Gross Premium + 50% 50% claims Subsidies Premium Premium Premium Premium Subsidies reinsurance) Subsidies 2005 1,603,315 0 0 801,657 801,657 801,657 2006 1,347,999 0 0 673,999 673,999 673,999 2007 1,805,232 0 0 902,616 902,616 902,616 2008 1,833,355 0 365,814 916,678 916,678 1,282,491 2009 1,795,005 0 0 897,503 897,503 897,503 2010 1,820,435 0 0 910,218 910,218 910,218 Total 10,205,340 0 365,814 5,102,670 5,102,670 5,468,484 Average 1,700,890 0 60,969 850,445 850,445 911,414 - 107 - Source: Authors analysis of FFSA Insurance data 4.43. It is recommended that GRK should study very carefully the issues surrounding premium subsidies before deciding whether to switch from the current system of claims subsidies to premium subsidies. The current system whereby government compensates 50% of the claims costs and then caps premium rates at approximately 50% of the technically required rates: in some regions of the country current premium rates are above the actuarially required rates and in other parts of the country actual rates are far too low. On the one hand this results in distorted crop insurance price signals in the market and on the other hand the 50% claims compensation does not provide the local insurers with the catastrophe protection on their retained claims that they require. Finally international reinsurers are not willing to support an under-priced scheme. While the authors are very cautious about recommending premium subsidies, it would be preferable in Kazakhstan to have an actuarially rated and commercially priced program and for government to then decide whether to provide financial support in the form of premium subsidies and also in the short term in the form of non-proportional stop-loss reinsurance protection and then in the medium term to promote the participation by international reinsurers. 4.44. In conclusion while further analysis is required to study the options for introducing a government of Kazakhstan financed non-proportional stop loss reinsurance program for individual private and mutual insurers, there are major potential benefits of this option of which the most important would be the guarantee that if a mutual or indeed a private commercial insurer incurred catastrophe losses in excess of its premium and reserves that government would indemnify these losses rather than the insurer defaulting on its obligations to farmers. However, the implications for insurers of moving to a non-proportional reinsurance program would be that they would have to increase their rates to actuarially determined levels (rates would have to be roughly doubled on average) and it is likely this move would prove very unpopular with farmers unless government were to provide premium subsidies. Issues relating to the provision of premium subsidies are reviewed further at the end of this section. Phase 2: Transition towards a Market-based Crop Insurance System 4.45. This section explores a set of issues and options for GRK to consider during the transition over the next few years to a market-based crop insurance system and which centre on the introduction of individual grower multiple-peril crop insurance MPCI either as a complement to, or as a substitute for the current Loss of Investment Cost Policy and the introduction of commercial international reinsurance. 4.46. It is assumed that crop insurance would probably continue to be obligatory for farmers during this interim phase. This is a decision which GRK will have to review with the local stakeholders 4.47. The proposals set out in this section for a Spring Wheat MPCI Program assume that Aktobe and WKO will no longer be included in the crop insurance scheme because this crop cannot be commercially insured in these two Oblasts and in any of the individual rayons: as previously noted the drought risk exposure in Aktobe and WKO is so high that it is impossible to offer crop insurance at commercially acceptable rates. It is intended that farmers in these two Oblasts are protected by a separate disaster-relief mechanism, but details of this separate program have not been considered further under this crop insurance feasibility study for Kazakhstan. - 108 - Interim Measures to Crowd-in the Private Commercial Insurance Companies 4.48. Government is keen to study ways of reforming the compulsory crop insurance program which will encourage greater participation by the private commercial insurers . The past 6 years have seen an exodus of the private commercial insurers for the Kazakhstan Obligatory Crop Insurance scheme and if the scheme is to remain viable this tendency must be halted now. Indeed measures need to be taken to try to crowd-in as many of the private commercial insurers as possible if the scheme is to diversify its crop insurance product range in future and to also start offering other classes of agricultural insurance cover including for livestock, forestry and possibly aquaculture. In the short term it is predicted that if Aktobe and WKO are excluded from the commercial crop insurance scheme this single amendment will encourage non-life insurance companies to join up to the crop insurance scheme. It is also expected that the market oriented measures outlined in this section including introduction of new crop insurance products, starting with individual grower MPCI and actuarially determined premium rates and support in the form of premium subsidies and non-proportional reinsurance by government (and possibly in some cases by international reinsurers), will encourage insurers to sign up for the crop insurance program. 4.49. In addition measures must be taken to ensure that the Farmers’ Mutual Crop Insurance Associations are able to continue to remain in the Kazakhstan crop insurance scheme, but only if they are able to meet their financial liabilities in full. As noted in the previous section in 2010 some mutual crop insurers were not able to meet their financial liabilities in full and therefore where claims exceeded their net premiums they were forced to pro rata down each claimants paid claim. Such actions contravene the purpose of contracting crop insurance and may rapidly erode farmers‘ confidence in the crop insurance program. Therefore measures must be sought to ensure that in future suitable excess of loss reinsurance is in place to Introduction of New Crop Insurance Products and Programs 4.50. There is a need to offer more choice of crop insurance products in order to meet the risk transfer requirements of different types of producers growing different types of crops located in different agro-climatic regions of Kazakhstan. For the past 6 years, Kazakhstan has implemented a single Loss of Investment Costs (LIC) multiple-peril crop insurance policy for grains, oilseeds, fibres (cotton) and some leguminous and root crops. While the LIC product is suitable for grains, oil seeds and other field row crops, it is not suitable for most horticultural and fruit crops and especially for crops which have multiple or staggered harvests. Furthermore the product is not best suited to irrigated crops. If Kazakhstan is to develop a voluntary crop insurance market which is demand driven and which meets farmer‘s risk management and risk transfer needs it will be necessary to start investing in the design and rating of new crop insurance product types. 4.51. Under this World Bank feasibility study risk assessment and product design and rating analyses have been conducted to assess the potential for introducing four new crop insurance product types into Kazakhstan including both traditional indemnity based products and new index-based products . These product types include: i) Individual grower multiple-peril crop insurance (MPCI) which is reviewed in this section, and then in Section 5, three products are covered including ii) traditional Crop-hail insurance, iii) Area-Yield Index-based crop insurance (AYI) and finally iv) Weather Index-based crop insurance (WII). - 109 - Individual Grower Multiple-Peril Crop Insurance for Kazakhstan Features and Challenges for Introducing MPCI 4.52. There are a number of drawbacks of the Loss of Investment Costs multiple peril yield- based crop insurance product which Kazakhstan’s insurers have now underwritten for 6 full years. Although one of the principle attractions of the LIC policy is that it does not require accurate individual grower yield data, there are also several drawbacks of the product which were highlighted in Section 3 and centre on: 1) farmer‘s dislike of the product because it does not provide them with a clearly established loss of yield guarantee they can understand, 2) the fact that the basis of indemnity is complicated and highly dependent on the prevailing market price for the crop at the time of loss and 3) the complications of accurately rating this product for different LIC sum insured levels. 4.53. Given the insurance industry’s experience with underwriting the LIC yield -based policy for the past 6 years, it would be relatively simple to make the transition from this cover to a more standardized individual grower Multiple-Peril Crop insurance (MPCI) Policy. The crop insurers of the LIC policy have over the past 6 years gained considerable experience in underwriting loss of yield-based multiple crop insurance and in conducting in-field loss assessment to establish actual yields and the amount of loss. This experience would enable them relatively easily to design, rate and implement individual grower MPCI. 4.54. The main differences of an individual grower MPCI product to the LIC Policy include i) the establishment of a pre-agreed Insured Yield at the time of policy subscription (the Insured Yield is usually calculated as a percentage of the individual farmer‘s historical average or normal crop yield or the local area average yield), ii) a pre-agreed unit Valuation Price which is applied to the Insured Yield to calculate the sum insured and iii) loss assessment involves the measurement of the actual yield which is compared to the Insured Yield and the amount of shortfall is indemnified accordingly at the pre-agreed valuation price. This basis of insurance and indemnity which is based on loss of yield is potentially much more transparent and understandable for farmers and loss assessment is also much more objective as yield loss is measured rather than expected shortfall in production costs compared to the estimated value of the remaining crop (salvage revenue). 4.55. The international experience of MPCI is that the product is very popular with farmers, but on account of the high premiums associated with this product, most schemes are dependent on government support in the form of premium subsidies. The international literature on MPCI often highlights the drawbacks encountered under voluntary schemes of anti-selection and moral hazard, or the difficulties of establishing average farmer yields and corresponding premium rates, through to high premium costs requiring government support in the form of premium subsidies and the often very high costs of individual grower in-field loss inspection and loss assessment. While these arguments are indeed very valid, they apply as equally to the existing Loss of Investment Cost policy in Kazakhstan. Currently issues of anti-selection are less of a problem because the scheme is obligatory for all farmers, but because it is a loss of yield multiple peril scheme it shares the drawbacks of other MPCI schemes. 4.56. In Kazakhstan the main challenge for introducing individual grower MPCI centres on the procedures for establishing an individual grower “normal average yield” for a given crop and on which basis to then establish an insured yield as a percentage of the average yield. With the possible exception of the USA, very few countries maintain accurate time series data- bases on individual farmer‘s crop production and yields. MPCI is, however, the most popular - 110 - crop insurance product with farmers and it is widely underwritten by commercial insurance companies throughout the world, both in developed and developing countries. In most countries insurers use regional (county, district etc) historical crop production and yield data to establish an average yield and to then establish an insured yield or optional insured yields as a percentage of the regional average yield. Where high levels of insured yield coverage are offered for example 75% or greater of the regional yield this may lead to major anti-section by those farmers whose average yields are normally below the regional average usually on account of their below average technology and lower use of inputs. In Kazakhstan, most farmers save for the very big production enterprises are unlikely to be able to provide their historical individual crop yields by field and by farm for the past 10 years or more. It is therefore proposed to develop MPCI on the historical Rayon-level yield data which in the case of spring wheat is available for up to 17 years and which can be disaggregated by type of farmer into production enterprises and commercial farmers, if required. In Kazakhstan anti-selection will be less of an issue because in most instances the maximum Insured Yields that are offered will not exceed 30% to 40% of the Rayon average yield: only in exceptional cases where average yields are very stable within a Rayon might 50% coverage be provided. 4.57. In Kazakhstan the methodology adopted by the Statistics Agency for sampling and recording crop sown and harvest area, production yields at the Rayon-level is considered to be accurate and to form the basis on which to design either an individual grower MPCI cover and or AYII, at least in the case of Spring Wheat. Under this study a detailed review has been conducted in the 8 main grain producing Oblasts of the Rayon-level crop area, production and yield data for spring wheat for the past 16 years and it has been concluded that this data is accurate and can form the basis of the design and rating of either an individual grower MPCI cover or an Area-yield index cover. 4.58. A further challenge is the rating procedures for MPCI cover when individual grower yield data are not available and rayon-level yield data has to be used. MPCI rating is based on a statistical analysis of variation in time series yields. The individual grower MPCI rating exercise for Kazakhstan has been based on 17-years of rayon-level spring wheat annual yields for i) production enterprises, ii) commercial farmers and iii) the overall rayon yields for both groups of farmers. The rating methodology involved the cleaning and de-trending of the rayon average yields and estimation of the average n expected yield for each Rayon based on the average of the last 5 years yield. Insured yield coverage levels of between 10% and a maximum of 40% of the rayon level expected yield were then selected and the losses for each coverage level simulated 5000 times to establish the pure loss cost rates. In order to estimate the higher yield variability between farmers the coefficients of variation around mean yield were increased by a factor of 15%. These were then adjusted to include catastrophe loadings and to derive the technical rates and then finally the indicative commercial premium rates for a target 60% loss ratio that would need to be charged to individual spring wheat farmers in each Rayon. Insured Yields and Sum Insured 4.59. Under an MPCI Policy the sum insured contains two elements, an insured yield and an agreed unit valuation price. The analysis of variation in 17-year spring wheat yields show these to be so variable that the maximum insured yields that could be offered to farmers in individual rayons are no more than about 40% of the Expected Yield. For the purpose of this exercise, spring wheat has been valued on the basis of the average national wheat prices for the period - 111 - 2008 to 2010 of KZT 3,120/Centner. It is stressed, however that any unit valuation price could be used based on the costs of production per center through to the final sales price. 4.60. Under the MPCI Policy there is more flexibility in offering farmers optional coverage levels according their circumstances and also to offer higher levels of coverage than under the current LIC Policy. The previous section showed that under the LIC, farmers are electing the lowest normative cost level of sum insured of about KZT 3,500 /Ha for spring wheat in 2010, equivalent at the prevailing sale price of wheat to an underlying insured yield of about 1 Centner/Ha or 10% to 15% of the average yield of the crop. One of the major objectives of the MPCI policy would be to offer farmers the option to purchase higher levels of coverage up to 40% of the expected Rayon yield. An example of the calculation of the insured yield and sum insured is given in Table 4.7 for spring wheat in Akkol Rayon which has an overall expected yield of 8.7 Centners/Ha: at the maximum 40% Insured Yield cover level the farmer would have an insured yield of 3.5 Centners/Ha and at the given unit insured price of KZT 3,120/Centner, his sum insured would be equivalent to KZT 10,808/Ha. The 40% coverage level would afford the famer about 3 times the current LIC protection. Conversely in those Rayons where farmers achieve much higher average spring wheat yields, the insured yield coverage levels and sums insured area correspondingly higher. The maximum expected yield in any one Rayon is 17.29 centners for commercial producers in Mendikara Rayon, Kostanay Oblast and in this case the 40% coverage yield is equivalent to 6.9 Centners/Ha with a sum insured of KZT 21,578/Ha. The spring wheat insured yields and sums insured have been calculated separately for Production Enterprises and Commercial Farmers and overall in all the Rayons and are attached in Annex 4. Table 4.7. Example of the Calculation of the Sum Insured for an MPCI Policy: Spring Wheat grown in Akkol Rayon, Akmole Oblast Production Commercial Item / Type Farmer Overall Enterprise Farmer 8.9 8.1 8.7 Expected yield (Centner/Ha) Insured Yield Coverage Options 0.9 0.8 0.9 10% Coverage (Centner/Ha) 1.8 1.6 1.7 20% Coverage (Centner/Ha) 2.7 2.4 2.6 30% Coverage (Centner/Ha) 3.5 3.2 3.5 40% Coverage (Centner/Ha) 4.5 4..0 4.3 50% Coverage (Centner/Ha) 3,120 3,120 3,120 Unit Insured Value (KZT/Centner) Sum Insured (KZT/Ha) 2,761 2,515 2,702 10% Coverage (KZT/Ha) 5,522 5,029 5,404 20% Coverage (KZT/Ha) 8,284 7,544 8,106 30% Coverage (KZT/Ha) 11,045 10,059 10,808 40% Coverage (KZT/Ha) 13,806 12,574 13,510 50% Coverage (KZT/Ha) Source: Authors 4.61. The Total Sum Insured of the scheme would be increased significantly if these proposals were adopted and under the assumption that all farmers continue to purchase MPCI cover. If it is assumed that the individual grower MPCI program continued to be obligatory for - 112 - all spring wheat farmers Table 4.8 provides some illustrative estimates of the total sums insured that would apply across the 6 eligible Oblasts (Aktobe and WKO being eliminated from the scheme)38. The actual 2010 expiring TSI for the obligatory scheme was KZT 47.3 billion (US$ 326 million): under an MPCI program, at 20% coverage level across all Rayons and the average unit valuation price of KZT 3,120 / center, the TSI would increase to KZT 84.8 billion (US$ 585 million) and at the highest 50% coverage level, the total scheme liability would rise further to KTZ 212 billion (US$ 1.46 billion). It is important to recognize that with the proposed increases in sum insured levels and in total scheme liability that the expected claims will also rise and therefore it is very important to examine the expected losses that might occur (See later section). Table 4.8. Estimated Total Sum Insured by Oblast for Compulsory MPCI Crop Insurance scheme according to coverage levels (KZT million) Insured Yield Coverage Level Oblast 10% 20% 30% 40% 50% Akmola 10,985 21,970 32,956 43,941 54,926 EKO 1,181 2,361 3,542 4,723 5,904 Karaganda 1,346 2,691 4,037 5,383 6,729 Kostanay 14,786 29,573 44,359 59,145 73,931 NKO 13,216 26,432 39,648 52,864 66,080 Pavlodar 910 1,821 2,731 3,641 4,552 Total 42,424 84,849 127,273 169,697 212,121 US$ (million) 293 585 878 1,170 1,463 Source: Authors based on Statistics Agency Rayon spring wheat production data MPCI Rating Methodology and Indicative Premium Rates 4.62. The statistical rating methodology used in this study to establish individual grower MPCI rates conforms to the MPCI rating procedures which are adopted by the insurance industry. The rating procedures are based on an analysis of variance in the rayon-level 17 year yields and which have been adjusted to reflect the higher yield variation of disaggregated individual farmer-level yields in spring wheat by multiplying the coefficient of variation x 1.15. Full details of the individual grower MPCI rating procedures are contained in Annex 2 and Annex 4. 4.63. The rates presented in this report are indicative commercial premium rates for a 60% target loss ratio, but it is stressed that final decisions over rates will be taken by insurers and their reinsurers. The burning cost or pure premium rates were calculated for each insured yield coverage level and then a very conservative catastrophe or reserve load added to derive the technical rates for each coverage level from 10% to 50% of expected yield. Finally the technical rates have been grossed up by 40% to achieve a conservative target loss ratio of 60%. The indicative commercial rates for spring wheat by Rayon and coverage level are presented in Annex 4 and the MPCI rating tool and MPCI rating data-base can be made available to the key stakeholders on request. Finally, it is important to note that the final commercial premium rates derived under this study are high not only because of the underlying yield variability, but also because of the conservative loadings applied and if this scheme passes to the detailed design and planning stage there should be room to analyse the actual insurers A&O Costs for example and if 38 For the purposes of this analysis the average sown area of spring wheat per Rayon for the past 3 years (2008 to 2010) has been used to establish the TSI in each Rayon and Oblast and in total. - 113 - these are lower than assumed under this study, it would be possible to reduce the 40% gross-up to a more reasonable level. 4.64. A comparison has been conducted of the average actual premium rates charged on the obligatory loss of investment cost (LIC) scheme and the new proposed MPCI program indicative commercial premium rates. The LIC premium rates are in effect subsidized by a factor of 50% because of the government 50% claims reinsurance program over the past 6 years. These rates have therefore been doubled to reflect the 6-year average full (100%) commercial premium rates the insurers would have had to have charged if there had been no government financial support on claims. The average full rates per Oblast would therefore range from a low of 3.5% in Kostanay to a high average of 9.4% in Karaganda (Table 4.9). The Rayon-level MPCI indicative commercial premium rates for spring wheat with 60% target loss ratio for 20% coverage range from an average low of 3.6% for NKO (and are in fact lower than the existing LIC average rates in this Oblast), to a high average of 7.1% in Pavlodar Oblast on account of the much higher yield variability in the Rayons in this Oblast. At the higher 30% insured yield coverage levels the commercial premium rates in most rayons exceed 8.0%. The one exception is NKO where 40% coverage could be offered at relatively affordable commercial premium rates of less than 10%. At the highest 50% insured yield coverage levels the commercial premium rates in most rayons exceed 15.0%, being in some cases (such as for Karaganda, EKO, and Pavlodar) above 20% of TSI. Table 4.9. Comparison of Loss of Investment Cost Scheme Average Rates with New MPCI Indicative Average Commercial Premium Rates for 60% Target Loss Ratio Existing LIC Scheme Rates New MPCI Commercial Rates for Spring Wheat 6-year Insured Yield Coverage Level average Oblast As if 100% LIC Full LIC Premium Premium 10% 20% 30% 40% 50% Rates Rates * (50% rates) Akmola 1.9% 3.9% 1.5% 4.8% 8.9% 13.6% 18.4% EKO 3.4% 6.8% 2.5% 6.9% 12.0% 17.3% 22.7% Karaganda 4.7% 9.4% 1.8% 5.6% 10.2% 15.2% 20.4% Kostanay 1.7% 3.5% 1.4% 4.6% 8.7% 13.2% 17.9% NKO 2.0% 4.0% 1.1% 3.6% 6.7% 10.3% 14.1% Pavlodar 3.6% 7.1% 2.1% 6.2% 11.1% 16.3% 21.5% Total 2.2% 4.4% 1.4% 4.5% 8.3% 12.6% 17.1% Source: Authors based on FFSA Insurance Data and Statistics Agency Rayon Wheat Yield data Notes: * As If Full Premium Rates assuming no government 50% claims payment subsidies 4.65. If the scheme continues to be obligatory for all farmers, the estimated total commercial premium income generated by the individual grower MPCI cover for spring wheat would increase significantly. Application of the commercial premium rates to the TSI corresponding to each coverage level indicates that 100% premium for a scheme which insures all eligible spring wheat grown in the 6 Oblasts may be in the order of KZT 3.8 billion (US$ 26.1 million) at the 20% coverage level, rising to a total premium of KZT 36 billion (US$ 251 million) at 50% coverage level. (Table 4.10). - 114 - Table 4.10. Spring Wheat MPCI Estimated Commercial Premium for Coverage Levels 10% to 50% for compulsory scheme covering all eligible insured acreage (KZT million) Insured Yield Coverage Level Oblast 10% 20% 30% 40% 50% Akmola 165 1,056 2,949 5,956 10,088 EKO 30 163 424 819 1,341 Karaganda 24 149 411 819 1,370 Kostanay 210 1,366 3,844 7,803 13,267 NKO 142 943 2,659 5,420 9,290 Pavlodar 19 113 302 593 978 Total 589 3,790 10,590 21,410 36,334 US$ equivalents 3.9 25.3 70.6 142.7 242.2 Source: Authors based on Statistics Agency Rayon spring wheat production data 4.66. This analysis also suggests that government of Kazakhstan may need to consider introducing crop insurance premium subsidies as an alternative to its current practice of subsidizing 50% of the claims in order to make crop insurance more affordable to farmers. The potential costs to government of switching its current 50% of claims reinsurance to 50% premium subsidies are shown in Table 4.11 assuming the scheme continues to be underwritten on a 100% of spring wheat area obligatory basis. The fiscal implications for government are if the levels of insured yield protection and sums insured are increased in Kazakhstan the corresponding premium subsidies will also increase significantly as shown by the illustrative cost of KTZ 1.9 billion for 50% premium subsides at 20% coverage level. (See further discussion at end of Section 4). Table 4.11. Fiscal Cost to Government Kazakhstan of 50% Premium subsidies on MPCI program (KZT million) Insured Yield Coverage Level Oblast 10% 20% 30% 40% 50% Akmola 82 528 1,475 2,978 5,044 EKO 15 81 212 410 670 Karaganda 12 75 205 409 685 Kostanay 105 683 1,922 3,902 6,634 NKO 71 472 1,330 2,710 4,645 Pavlodar 10 56 151 296 489 Total 295 1,895 5,295 10,705 18,167 US$ equivalents 2.0 12.6 35.3 71.4 121.1 Source: Authors based on Statistics Agency Rayon spring wheat production data MPCI Probable Maximum Loss Estimates 4.67. Spring Wheat Production in Kazakhstan is very exposed to drought losses and with the increase in Insured Yield Coverage Levels and Sums Insured of the proposed individual grower MPCI scheme for spring wheat it is very important that the Probable Maximum Loss (PML) exposures be quantified and a risk financing and reinsurance strategy is identified. Under the current risk modelling exercise for spring wheat MPCI cover the PMLs‘ associated with return periods of up to 1 in 250 years have been calculated for each coverage option from - 115 - 10% to 50% of average rayon yield and the results are summarized for an obligatory scheme in Table 4.12 and Figure 4.4). The analysis suggests that for a maximum 50% coverage level the expected losses that might occur every 10 years could be in the order of KZT 36.3 billion (17.13% of the value of the TSI), and for the 1 in 100 year PML a loss of KZT 99.7 billion (47% of TS), equivalent to a loss ratio of about 274%. Table 4.12. Estimated Probable Maximum Loss for MPCI Wheat Scheme (KZT million) Coverage Item 10% 20% 30% 40% 50% Total Sum Insured (KZT mio) 42,424 84,849 127,273 169,697 212,121 Probable Maximum Loss: 1 in 10 years (% of TSI) 0.41% 2.09% 5.62% 10.85% 17.13% 1 in 50 years (% of TSI) 3.13% 10.83% 19.79% 29.14% 37.93% 1 in 100 years (% of TSI) 5.43% 16.80% 28.70% 39.17% 46.98% 1 in 250 Years (% of TSI) 10.80% 25.99% 38.71% 46.96% 54.42% Probable Maximum Loss: 1 in 10 years (KZT) 175 1,777 7,153 18,414 36,336 1 in 50 years (KZT) 1,329 9,191 25,191 49,444 80,456 1 in 100 years (KZT) 2,304 14,251 36,524 66,463 99,652 1 in 250 Years (KZT) 4,581 22,051 49,268 79,688 115,428 100 years PML /Loss Ratio (%) 391% 376% 345% 310% 274% PML 1 in 100 years (US$) 15.4 95.0 243.5 443.1 664.3 Source: Authors Figure 4.4. Estimated Probable Maximum Loss for MPCI Wheat Scheme PML Loss of Investment coverage under three different price 60.00% scenarios (in Loss Cost) 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% 1 50 100 150 200 250 10% Coverage 20% Coverage 30% Coverage 40% Coverage 50% Coverage Source: Authors Implications for MPCI Reinsurance for Spring Wheat 4.68. Given the catastrophe risk exposures of spring wheat production to drought in Kazakhstan, it is extremely unlikely that the insurance sector would be willing to assume the increased liabilities implied under the proposed MPCI program unless government is willing to provide reinsurance support for this initiative. Currently government provides 50:50 quota share reinsurance protection to the private commercial crop insurers and mutual crop insurers but as previously noted this protection does not cap their exposure to catastrophe losses. Therefore - 116 - for the purposes of this study some preliminary analyses have been conducted for non- proportional stop loss reinsurance protection for the spring wheat MPCI program. 4.69. Some preliminary modeling has been conducted to establish the indicative reinsurance pricing for an Aggregate Stop Loss Reinsurance Protection for the spring wheat MPCI program. The modeling has been conducted assuming full value protection and priority levels of 70%, 100% and 150% of GNPI for the four MPCI insured yield coverage levels. The results of this analysis are summarized in Table 4.13 and further details of the stop loss reinsurance pricing methodology which draws on accepted reinsurance-industry stop loss pricing methods are presented in Annex 4. The analysis shows that for a full value Aggregate stop loss reinsurance protection for losses excess of 100% GNPI, the stop loss reinsurance pricing would be in the order of KZT 991 million (38.6% of GNPI) for 20% coverage level, rising to KZT 5.8 billion (16% of GNPI) for 50% coverage. If the lower 70% of GNPI priority was to be adopted the Aggregate Stop loss pricing would be correspondingly higher. 4.70. In practice, it should be recognized that unless the crop insurers form a single pool entity the pricing of stop loss reinsurance cover would have to be conducted for each individual insurance company and mutual insurer according to their own portfolio size and distribution and insured yield coverage levels in each region. Table 4.13. Indicative Pricing for Aggregate Stop Loss Reinsurance for Kazakhstan Spring Wheat MPCI program Insured Yield Coverage Level Item 10% 20% 30% 40% 50% Stop Loss Reinsurance Premium (KZT Million) Priority % of GNPI 70% 242 1115 2675 4948 7930 100% 227 991 2243 3895 5845 150% 208 829 1711 2664 3515 Stop Loss Reinsurance Premium (as Percent of Commercial Premium, GNPI Priority % of GNPI 70% 41.14% 29.42% 25.26% 23.11% 21.82% 100% 38.59% 26.14% 21.18% 18.19% 16.09% 150% 35.24% 21.88% 16.15% 12.44% 9.67% Source: Authors 4.71. During the transition phase to a more market-based crop insurance program, it is possible that the larger commercial crop insurance companies may be able to seek reinsurance support from international reinsurers either on a non-proportional stop-loss basis, or a combination of proportional and non proportional reinsurance. In order to gain access to international reinsurance capacity, the individual insurance companies would need to demonstrate: i) their financial ability and adequate solvency margins to sustain losses, ii) a core crop insurance underwriting and claims management team, iii) a balanced MPCI crop portfolio with geographic spread to ensure risk spread and indeed reinsurers may insist on sum insured limits per Oblast or region to ensure accumulation control; iv) evidence that the commercial premium rates for each insured yield coverage level are technically derived and contain both catastrophe loading and margin to cover A&O Costs and profit margin; and v) evidence that the insurer has in place the necessary in-field pre-inspection and crop loss assessment staffing systems and procedures to control risk and adjust losses in a timely and accurate fashion. - 117 - 4.72. While some commercial insurers may be able to meet the requirements of international reinsurers, it is unlikely that at present any of the farmer mutual crop insurance companies would be able to meet these requirements and therefore in the short term it is likely that non- proportional reinsurance cover would need to be provided by government. 4.73. Finally, it should be noted that while the non proportional stop loss reinsurance model illustrated in Table 4.12. assumes a single excess layer providing full value cover, in practice very few international reinsurers are willing to offer unlimited protection and that they will normally insist on a layered reinsurance program with capped liability and for higher levels of catastrophe risk it may be necessary for government to assume liability (See end of Section 4 for further discussion). Phase 3- Transformation into a fully Commercial Crop Insurance Scheme for Kazakhstan 4.74. This Section presents proposals for a voluntary commercial public-private agricultural insurance partnership for Kazakhstan, a central feature of which would be the formation of a crop co-insurance Pool system, backed up by international reinsurance. This section presents a series of options and recommendations for GRK and key stakeholders to consider for the introduction of voluntary agricultural insurance in Kazakhstan, the creation of a crop insurance Pool system which would be designed to encourage participation by the private commercial insurance companies and to introduce a specialist agricultural insurance entity to act on behalf of the Pool and capable of developing and implementing new classes of agricultural insurance. The pool would be supported by a commercial reinsurance program which would be placed with international reinsurers. Government and the FFSR‘s roles are also explored and it is likely that this will centre on financial support in the form of premium subsides and possibly catastrophe reinsurance protection. Transition from Obligatory to Voluntary Crop Insurance 4.75. As part of the transition to a market based crop insurance system, policy makers in Kazakhstan will need to consider making crop insurance voluntary. As previously noted Kazakhstan is one of a small minority of countries to adopt obligatory crop insurance and almost unique in trying to implement obligatory crop insurance through the private commercial insurance sector. 4.76. In the short-term if crop insurance is made voluntary it is likely that there will be a major reduction in the demand for crop insurance while the farming sectors adjust to the realities of a demand-driven voluntary crop insurance system. At this stage it is not possible to predict how great the contraction in demand by farmers for voluntary crop insurance may be, but this is likely to be significant. Under a voluntary system, crop insurers would be free to select which types of farmer, which crops and which regions they are willing to underwrite under the crop insurance program. 4.77. The Obligatory Crop Insurance law of 2004 will need amending to reflect the new requirements of a voluntary Public-Private Partnership crop insurance scheme in Kazakhstan. According to the roles government play in supporting this scheme the Law will need amending, - 118 - for example to focus more specifically on government financial support in the form of proportional and or non-proportional reinsurance and also potentially premium subsidy support. 4.78. Under a future voluntary crop insurance program Kazakhstan insurers would need to seek to link crop insurance with other rural services for example input supplies (seeds, fertilizers etc) and with seasonal crop-credit provided through the banking system. In many countries, there are very close linkages between public sector crop-credit provision and crop insurance and banks make their lending conditional on the farmer having crop insurance in place at the time of receiving his loan. In other words, crop insurance is mandatory for credit recipients and voluntary for non-credit recipients. Normally such crop-credit linked programs make the bank the first beneficiary on the crop insurance policy in order to ensure that any indemnity is used to repay the outstanding amount of loan: any balance on the insurance settlement is then paid to the farmer. Countries which operate compulsory crop-credit insurance programs include Mexico (both commercial crop loans and small farmer group loans are insured on a compulsory basis), India (which has the world‘s largest national area-yield crop-credit insurance scheme implemented through the public-sector Agricultural Insurance Company of India, AIC, covering about 25 million smallholder farmers per year), The Philippines (where public sector crop credit provision is again conditional on the farmer having an individual grower MPCI cover in place through the parastatal agricultural insurer, the Philippines Crop Insurance Corporation, PCIC) (See Mahul & Stutley 2010 for further details of countries with linkages between crop credit and crop insurance). Rationale for Creating an Agricultural Insurance Pool in Kazakhstan 4.79. In Kazakhstan under the third phase of transformation to a market-based crop insurance system, it is recommended that policy makers review the potential to form a Coinsurance Pool which is specifically dedicated to underwriting a national agricultural insurance portfolio under a suitable private public partnership. In countries where insurance markets are developing and there is little or no tradition of crop or livestock insurance or rural insurance infrastructure, a pool coinsurance program may be a much more attractive and cost- effective proposition for commercial insurance companies than if they were to try to operate independently. 4.80. The potential benefits of an Insurance Pool include the ability to underwrite a much broader and larger book of business and the potential to achieve a much better geographical spread of risk, than if each company were operating independently. Other benefits economies of scale in the costs of developing new products and programs and in underwriting risks and in adjusting claims where a single lead coinsurer is appointed (or a separate Managing Underwriting Unit is created) to implement the business on behalf of the pool members. There are also major potential cost savings in the purchasing of reinsurance protection for a pooled coinsurance program. It is recognized that there are potential drawbacks of Pools including the possible reduction in competition on the range of products and services provided and in the premium rates offered by the pool. Details on the benefits and limitations of Pools are contained in Box 4.3. 4.81. Coinsurance Pools for agricultural insurance have proved to be very popular with private and mutual insurers in many countries including most notably, the Agroseguro39 Pool in Spain, the Tarsim Pool in Turkey, the Philippines livestock insurance pool, the Austrian Hail Insurance pool and various other pool arrangements in China, Malawi, Mongolia and Ukraine. 39 AGROSEGURO stands for the Agrupación Española de Entidades Aseguradoras de los Seguros Agrarios Combinados (Spanish Group of Insurance Entities of the Combined Agrarian Insurance) - 119 - 4.82. The rationale for forming a coinsurance pool in Kazakhstan centres on a number of key factors which include: i) The very small number of private commercial companies which are currently supporting this scheme and the need to crowd in commercial insurers if crop insurance is to remain a viable proposition in Kazakhstan. LIC/MPCI crop insurance is a catastrophe class of business and many insurance companies are reluctant to risk their capital on this risky class of business: however, under a Pool agreement individual companies can participate with very small shares of the overall risk if they wish. ii) The prohibitively high start-up investment costs for individual insurance companies in creating their own internal crop underwriting and claims departments and then in developing regional networks of marketing and sales agents and trained crop inspectors and loss assessors to administer the scheme implementation. Pools offer the opportunity to create a single centralised insurance underwriting and claims management and loss assessment capability (often termed a Managing Underwriter, MU) and for individual Pool members to contribute towards the running cost of the MU whilst benefiting from the advantages of economies of scale in the fixed and variable costs of the single MU. iii) The lack of common standards at regional level in the underwriting of crop risks and especially in the in-field loss assessment capabilities of individual insurance companies and the farmer mutual insurance associations. Under a Pool Agreement the MU would be responsible for coordinating all underwriting and loss adjustment activities and in ensuring that common standards are adopted throughout the country. iv) A lack of consistency in crop rating and competition which is driving down the crop insurance premium rates to unsustainable levels. Under a Pool Agreement all insurers would issue standard crop insurance policies and they would all adopt the same premium rates for each crop in each zone and region; v) The difficulties of arranging commercial international reinsurance protection for individual Kazakhstani insurance companies with very different underwriting standards and portfolios. Under a Pool agreement, a single reinsurance program would be purchased by the MU and it would be much cheaper to transact a single reinsurance contract for the Pool. Box 4.3. Benefits and Limitations of Coinsurance Pool Arrangements Benefits Economies of scale through operating as a single entity with shared (pooled) administration and operating functions leading to costs savings due to: * Reduced staffing requirements (fixed costs); * Shared costs of product research and development, actuarial and rating; * Reduced costs of underwriting and claims control and loss adjustment. Cost advantages in purchasing common account (pooled) reinsurance protection rather than each company trying to place its own reinsurance program. Advantages due to: * Stronger negotiating position with reinsurers; * Larger and more balanced portfolio and better spread of risk; * Reduced costs of reinsurance due to pooled risk exposure; * Reduced transaction costs (reinsurance brokerage, etc). - 120 - No competition on rates in a soft market and ability to maintain technically set rates. Most pools operate as the sole insurance provided or monopoly (e.g. Austria, Senegal, Spain, Turkey), and there is therefore no competition on pricing. Ability to maintain underwriting and loss adjustment standards. Under a pool monopoly arrangement, the pool manager can ensure that common and high standards are maintained in the underwriting of crop and livestock insurance and in the adjusting of claims. Where companies are competing against each other for standard crop insurance business, there is often a problem of varying loss adjustment standards between companies. Limitations Pool may act as the sole agricultural insurer, resulting in lack of competition in the market in terms of the: * Range of products and services offered by the monopoly pool underwriter; * Restrictions on the range of perils which are insured; * Restrictions on the regions where agricultural insurance is offered and/or the type of farmer insured; * Lack of competitiveness in premium rates charged by the pool. Source: Authors. Institutional Framework for a PPP Agricultural Insurance Pool in Kazakhstan. 4.83. The proposed Kazakhstan Agricultural Insurance Pool would involve the active participation of the public and private sectors. An outline institutional framework for the Agricultural Insurance Pool is shown in Figure 4.5 which draws on the experiences of the organizational structures of the Spanish and Turkish agricultural insurance pools. (See Annex 5). The functions of each of the key stakeholders are discussed below. 4.84. The Agency for Financial Market and Financial Institutions Regulation and Control would play an important legal and regulatory role on this scheme. The Agency would be responsible for amending the current Law to make crop insurance a voluntary class of business and to permit insurance companies to design and rate their own agricultural insurance policies. The law should be amended in such a way to create an enhancing framework for agricultural insurance in Kazakhstan and one which permits new classes and products such as livestock and forestry insurance to be added in future. The legislation should also be modified to reflect government‘s supporting roles including possibly financial subsidies on premiums and catastrophe claims reinsurance. 4.85. The Ministry of Agriculture would continue to play a central role in the policy and planning for agricultural insurance in Kazakhstan and would represent government’s fiscal interests in supporting the scheme. It is proposed that MOA through its Department of Strategic Planning would represent GRK‘s interests in the agricultural insurance scheme and it would be responsible for planning the three yearly and annual agricultural insurance plan and budget in conjunction with the FFSA and private insurance industry. MOA could also provide a useful supporting role in the conduct of research and development into new agricultural insurance products and programs. 4.86. The Fund for Financial Support to Agricultural Insurance, FFSA would continue to act as the main public sector implementing agency on this pool agricultural insurance scheme. Currently the FFSA has been responsible for participating in the adjusting of field-level crop insurance claims, in maintaining an individual grower crop insurance and claims data-base, and in settling the government‘s 50% share of crop claims to the insurance companies. In future it is recommended that FFSA‘s roles would no longer include participation in field-level loss assessment, but that its functions would be expanded to include: - 121 -  Coordination with the Crop Insurance Pool‘s Managing Underwriting Company (MUC) in the development of the technical studies required for the design of new crop, livestock, forestry and aquaculture insurance policies and programs into the system.  Management of the government‘s financial fund for the support of agricultural insurance and disbursement of funds (including as appropriate, premium subsidies and catastrophe reinsured claims payments) to the MUC on behalf of the Pool Coinsurers.  Maintenance of crop insurance underwriting and claims data-bases  Information and advice to farmers 4.87. The central feature of the agricultural insurance system would be the creation by the Non-Life Insurance Companies of an Agricultural Coinsurance Pool which would be designed to underwrite all classes of agricultural insurance business. The agricultural insurance pool would be a legally constituted insurance company with paid up capital contributed by each participating non-life insurer and which would report to and be supervised by a Board of Directors which would coordinate the activities of the Pool Coinsurers with Government departments. 4.88. It is proposed that the coinsurers also create a separate Managing Underwriting Company which would be responsible for underwriting the scheme on behalf of the coinsurers and which would handle premiums and claims on behalf of the pool coinsurers and which would also negotiate reinsurance on behalf of the Pool Coinsurers. Figure 4.5. Organisational Framework for Kazakhstan Agricultural Insurance Pool Fund for Financial Agency for Financial Market and Ministry of Support to Agriculture Financial Institutions Regulation Agriculture and Control (FFSA) Financial support: Policy, planning, premium subsidies, research and Insurance legal & catastrophe development regulatory reinsurance Pool Management Board Kazakhstan Non-Life International AGRICULTURAL Insurance Companies Reinsurers INSURANCE POOL Farmer Mutual Crop Insurance Associations Managing Underwriting Company Farmer Associations, Cooperatives, Large Farmer Production Rural Banks and other Aggregators Enterprises Small and Medium Farmers Source: Authors Functions of the Pool Coinsurers and the Managing Underwriting Company Agricultural Insurance Pool - 122 - 4.89. It is proposed that participation in the Agricultural Insurance Pool would be open to all the Non-life private insurance industry in Kazakhstan. The local stakeholders would also need to decide whether to open up participation to the farmers‘ mutual crop insurance associations, so long as these mutuals could meet the minimum capital requirements of the Pool and comply with other norms and conditions of the general insurance act and amended Agricultural Insurance Law. 4.90. In Spain which is one of Europe’s oldest and largest Agricultural Insurance Pool formed in 1980, there are currently 28 coinsurers in the AGROSEGURO Pool which includes both private and mutual insurance companies including Mapfre Insurance & Reinsurance Company, Spain‘s largest insurance company and the Spanish public-sector catastrophe reinsurer, Consorcio de Compensacion de Seguros. The largest shareholder in the pool is Mapfre with a shareholding of more than 15%; the smallest coinsurer has less than a 1% share in the Pool. Size of shareholding is therefore not a bar to participation in the Pool. Each company‘s share of annual agricultural insurance premiums and liability is determined according to its percentage share in the Pool during that underwriting year. Participation in the pool is completely voluntary and insurance companies are permitted to join and leave the Pool after completion of an underwriting campaign (year), but in order to maintain continuity, in practice companies agree to sign up to the pool for a three-year period. In Turkey, the TARSIM Agricultural Insurance Pool has 22 insurance company members, but in this instance each insurance company has equal shares in the scheme. 4.91. It is recommended that the Kazakhstan Agricultural Insurance Pool is constituted on similar lines to the Spanish model namely that participation in the pool is voluntary, that individual insurance companies are permitted to take up different size of share-holding in the company and that the principle that each companies‘ share of premiums and claims is proportional to its shareholding as opposed to any other criteria (for example, the amount of premium the companies individually collect and cede to the pool). Finally the concept of requesting companies to sign-up for a period of 3 years to ensure stability should also be considered. 4.92. Under the market-based Pool crop insurance system it is hoped that the participation by private insurance companies will be high and the shares in the pool will be fully subscribed. The principle of a coinsurance pool is that 100% of the liability is divided up between its subscribing members according to pre-agreed underwriting limits and agreed risk retention limits of each company. If, however, the scheme cannot be fully subscribed in the start-up phase, government could consider becoming a coinsurer for a limited period of time. Pool Managing Underwriting Company 4.93. There are two main options for managing the Agricultural Insurance Pool, including i) for the coinsurers to appoint a lead coinsurer to manage the business on behalf of the pool members, or ii) to create a separate limited company owned and funded by the Pool members and which would have its own core team of underwriting and claims management staff, an equipped office and regional capability to administer and implement loss assessment etc. 4.94. In Spain and in Turkey the national Pool agricultural insurance programs are implemented on behalf of the Pool coinsurers by specialist Managing Companies, established by and reporting to the Pool Coinsurers. In Spain AGROSEGURO is a limited company formed by the pool insurers in 1980 to transact agricultural insurance and to manage claims and to indemnify losses on the coinsurers‘ behalves. AGROSEGURO started with a very small team of - 123 - agricultural underwriters, claims managers and loss assessors and office support staff, and today has grown into Europe‘s largest agricultural insurance managing company underwriting over 260,000 agricultural insurance policies and a further 30,000 livestock forestry and aquaculture policies generating total premiums of US$ 864 million in 2010. In 2010 AGROSEGURO had a full time complement of about 75 permanent staff based in its headquarters in Madrid and in each of the 14 autonomous regions: it has a General Management Unit, Legal Department and Regional Branches and core operational departments responsible for i) product research and development, ii) production and communication (underwriting), iii) claims administration and loss assessment, iv) administration and accounting and v) organization and information technology systems. As such it functions as a very professional commercial managing company on behalf of its coinsurers. AGROSEGURO‘s internal A&O costs are financed out of earned premiums on the agricultural insurance business in writes on behalf of the pool: in 2010 its internal A&O expenses amounted to 3.55% of total earned premiums (AGROSEGURO 2011)40. In Turkey, the agricultural pool insurance scheme is managed by TARSIM, a private corporation established in 2005 by the insurance companies to carry out all the underwriting and claims adjusting and settlement tasks of the Pool in accordance with the Agricultural Insurance Act of 2005. (Bora 201041; Uçak & Berk, 200942) . 4.95. In Kazakhstan it is recommended that in conjunction with the formation of an agricultural insurance pool scheme that the participating coinsurers also approve the establishment of a Managing Underwriting Company (MUC). Initially this company would only require a very small full time staff complement including a General Manager who should be an experienced non-life insurance expert, an agricultural insurance underwriter who should have experience under the obligatory crop insurance scheme, a claims manager, a crop loss assessment manager who would be responsible for coordinating the field-based loss assessment activities, a small number of junior underwriters, data analyst and data-base specialist, accounting and other back-up support staff. The MUC should also have a small permanent staff capability to manage the crop insurance program and especially claims notification and loss assessment procedures in each region/Oblast of Kazakhstan. It is recommended that the MUC assume responsibility for all product design and rating, crop underwriting and policy issuance and management of premiums, through to being responsible for claims, loss assessment and claims settlement. The MUC should report regularly and submit premium and claims bordereau accounts to the Pool Coinsurers Management Committee. Pool Operating Systems and Procedures 4.96. The proposed operating systems and procedures for the Kazakhstan Agricultural Insurance Pool are shown in Figure 4.6. 4.97. Under the proposed Agricultural Insurance Pool the participating insurance companies (coinsurers) would be responsible for marketing the MUC approved crop insurance policies at agreed premium rates. Both in Spain and in Turkey AGROSEGURO and TARSIM respectively are responsible for the design and rating of crop (and livestock and other classes) insurance policies and in setting commercial premium rates for each region. The Pool coinsurers 40 AGROSEGURO (2011). Spain: The multiperil crop insurance 2010. 41 Bora, B. (2010) Subsidised Agricultural Insurance in Turkey (TARSIM). Available from: http://www.mapfre.com/mapfrere/docs/html/revistas/trebol/n54/docs/Articulo2en.pdf 42 Uçak H., and A. Berk (2009). Structural change in Turkish agricultural insurance policy and recent developments. http://www.piu.org.pl/public/upload/ibrowser/ Wiadomosci%20Ubezpieczeniowe /numer2/WU2_09ucakberk.pdf - 124 - are exclusively responsible for all policy promotion, marketing and sales using either their own sales agent networks and or through farmer associations, cooperatives, individual brokers, banks and other sales distribution channels. In Spain the companies market the AGROSEGURO policies on their own paper and they receive a pre-agreed commission to cover policy marketing. All policies are placed with AGROSEGURO and all premiums net of commissions is pooled by AGROSEGURO. A key principle of the Spanish system is that the shares of premium and liability are determined by each company‘s shareholding in the Pool at the start of the underwriting campaign and not by its actual share of policy sales and generated premium volume. It is proposed that the same procedures would apply to the Kazakhstan Pool namely that the Pool member would be responsible for all policy marketing and sales, that they would receive an agreed commission for transacting this business on behalf of the Pool, that premiums net of commission would be paid over to the MUC and that each member‘s share of liability would be based on their share in the coinsurance pool and not on the basis of their volume of sales. In this way all pool members will share equally in the pool of risks from different regions of Kazakhstan. 4.98. The MUC would be responsible for the functions of product design and rating, underwriting and risk acceptance, claims administration and loss assessment and for negotiating common account reinsurance protection on behalf of and reporting to the pool coinsurers. The MUC would design and rate standard crop insurance products and policies which would then be approved by the Insurance Regulator. Once approved the policies would be marketed by the Pool coinsurers at the MUC agreed premium rates for each crop and risk zone. As such there would be no competition between insurance companies over premium rates, but rather according to the quality of their services to farmers. The MUC would be responsible for handling premiums and for settling claims on behalf of the pool. 4.99. Loss Notification and Loss Assessment Procedures would be streamlined and strengthened under the proposed system and fall under the direct control of the MUC. The MUC would be responsible for establishing a regional network of part-time crop loss assessors in each Oblast and Rayon. These part-time loss assessors would receive intensive training in crop loss assessment procedures and in the use of standardized loss assessment procedural manuals which would be designed by the MUC (possibly with support from MOA and FFSA). In the start-up phase of the Pool program, it is likely that a national; network of about 40 to 50 trained43 loss assessors would be adequate to manage the expected volume of crop losses, so long as these teams were properly funded and had access to transport and communications. The loss assessors would report to and be controlled by the MUC regional representative. 4.100. Farmers and or their agents would be responsible for communicating losses (potential claims) to the MUC local or regional office and wherever possible would use web-based or cell- phone reporting systems. In order to reduce the propensity for farmers to submit claims for minor losses which fall within the farmer‘s self-retention, it is recommended that a bonus-malus system be introduced (See previous section for further details). 4.101. The adjusting of losses would be conducted by the MUC approved loss assessors and would no longer involve the formation of a 5-person Rayon-level committee. The only persons who would be required to attend the loss assessment would be the approved loss assessor, the farmer (Insured) and at the farmer‘s request, his local agent. The loss assessor would be required to complete a standard loss assessment form in the field and to estimate the amount of crop loss 43 This recommendation is based on the Spanish system where AGROSEGUO in 2010 maintained a network of about 350 crop loss assessors: in 2010 the crop loss assessors adjusted a total of nearly 95,000 claims or an average of about 275 claims per adjuster (AGROSEGURO Annual Report 2010). - 125 - or yield reduction and to confirm whether this loss is above or below the threshold which would give rise to a claim. The farmer would then be required to counter-sign the field loss assessment form and to confirm his agreement or not with the assessed result: where the farmer does not agree with the assessment he would in the first instance be entitled to a second assessment to be conducted in the presence of a third party and if at this stage there is still no agreement, arbitration procedures would apply. If the farmer agrees with the assessed results the loss assessment would then be forwarded to the MUC for processing: either to closed (losses below the threshold for a claim) or adjusted and settled (losses above the threshold giving rise to a claim). The MUC would be responsible for settling claims to each claimant either directly or through local distributers as agreed. Figure 4.6. Kazakhstan Crop Insurance Pool Operating Systems and Procedures EXPERIENCIA DE ESPAÑA EN EL DESARROLLO DE LOS SEGUROS AGRICOLAS Kazakhstan: Proposed Operating Procedures Flow Chart Payment of Premium Managing Underwriting Insured Farmers, Company (MUC) Pool Insurance Cooperatives Associations, Intermediaries Companies Policy promotion marketing & sales Communication of Losses Loss Assessment and Payment of Claims Source: Author‘s adapted from Burgas 2007 Government Support to the Kazakhstan Agricultural Insurance Pool 4.102. As part of the switch to a market-based Pool Agricultural Insurance System in Kazakhstan, the role of Public-sector support to this scheme should be reviewed. To date GRK‘s main support to the obligatory crop insurance scheme has been in the form of free Quota Share reinsurance of 50% of the incurred claims, but this is likely to change under the transition to a market-based agricultural insurance scheme. There is now a considerable body of literature on the different ways governments around the world support crop insurance and this section briefly reviews some of the roles GRK might play in future44. A summary of the potential roles government may play in supporting agricultural insurance is presented in Box 4.4 and these roles are considered below in the context of Kazakhstan. 44 For a recent review of government support to agricultural insurance in more than 65 countries see Mahul and Stutley 2010. - 126 - Legal and regulatory 4.103. GRK can facilitate a review and amendments of the Crop Insurance Law to support the introduction of the voluntary Pool Agricultural Insurance system into Kazakhstan. As previously reported the Obligatory Law of 2004 will require a comprehensive review and amendments to facilitate the introduction of voluntary commercial crop insurance. In this process GRK may wish to review the Turkish Agricultural Insurance Act of 2005 which lead to the introduction of the subsidized TARSIM Pool Agricultural Insurance system. Enhancing Data and Information Systems 4.104. There may be important roles for GRK in enhancing data and information systems for crop insurance in Kazakhstan. In Kazakhstan if the insurance sector is to design new crop insurance products and programs, it will increasingly need access to time series crop production and yield data, to weather data and to remote sensing data. Government can facilitate the access to this data. In addition, GRK has already identified a need in increase the density of weather stations if it is to introduce WII and the APPAP II project contains a sub-component with funding to invest in weather stations . Furthermore if an AYII is to be introduced in future, the Insurance sector will need to collaborate closely with MOA and the Agency of Statistics to obtain accurate estimates of area (Rayon) yields at the time of harvest. (See Section 5). Product Research and Development and Access to Data 4.105. Governments can usefully assist the private insurance sector to conduct research into the demand for new agricultural insurance products and to then provide access to crop and weather data and statistics which are essential to the insurers if they are to design and rate new products and policies. In Spain, ENESA/MOA plays a central role in assisting AGROSEGURO with the development of new products and programs. In Kazakhstan, MOA/FFRS could perform a similar and very important role going forward. Also KHM and the NSA are likely to have very important roles in provision of needed data and information and to assist in the future development of WII and remote sensing based insurance products and programs. Education, Training and Capacity Building 4.106. On the basis of this study there would appear to be an important role for GRK to support Pool management to strengthen farmer insurance awareness and education training programs. Governments can play a key role in supporting farmer awareness and education programs and capacity building and workshops and technical training programs for key agricultural insurance staff. In Kazakhstan it appears that relatively little priority has been given by crop insurers to date on explaining the role of insurance and the operation of the LIC policy and government through MOA and other rural institutions could provide financial and logistical support for farmer-level crop insurance education. Insurance Company staff will also need specialist training in product design, actuarial and rating, underwriting and claims administration and loss assessment systems and procedures. Similar training also needs to be provided to staff in the banks, MFIs, and suppliers if these organizations get involved as delivery channels/agents. Catastrophe Risk Financing 4.107. In Kazakhstan under the PPP, it is likely that government will continue to play an important role in supporting the Crop Insurance Pool risk financing insurance and reinsurance program. In many countries government is actively involved in the reinsurance of - 127 - agriculture: key territories where government acts as a catastrophe reinsurer (either directly or indirectly through a national reinsurance company) include the USA and Canada, Spain, Brazil, India, South Korea, China and Kazakhstan. In Kazakhstan, it is proposed to introduce commercial international reinsurance of the Pool scheme: it is, however, likely that GRK will need to continue supporting the reinsurance program on a catastrophe basis at least in the short term (This subject is reviewed further below). Premium Subsidies 4.108. Premium subsidies are the most widely practiced form of government support to agricultural insurance practiced by over two thirds of countries which have some form of agricultural insurance. Globally average premium subsidy levels are in the order of 50% of the full cost of premium, but in some countries (including USA, Spain, Portugal and Italy) governments provide subsidies as high as 75% to 80% of the premium. Premium subsidies are, however, very controversial for a number of reasons. The provision of non-discriminatory premium subsidies is regressive because subsidies disproportionately benefit the larger farmers to the detriment of small and marginal farmers. Also subsidies that cover a large part of the overall premium tend to promote moral hazard whereby farmers grow high risk crops which attract high premium subsidies in regions which are not technically suited to the crop. Once premium subsidies have been introduced by governments it is politically very difficult to reduce or to withdraw these subsidies and in many of the countries which operate non-discriminatory premium subsidies the fiscal costs to government are extremely high. 4.109. In Kazakhstan it is unlikely that voluntary crop insurance will take-off without government support to premium subsidies. Since the introduction of obligatory crop insurance in 2005, the insurance companies have in effect received 50% premium subsides through the government 50% share in claims. Under the proposal to increase crop insurance coverage levels and the need to introduce actuarially determined rates, the crop insurance premiums which will need to be charged to farmers will have to increase significantly going forward. It is likely, however, that under a voluntary crop insurance program that very farmers will continue to buy crop insurance unless this is accompanied by premium subsidies. It is therefore suggested that government should switch the bulk of its financial support out of 50% Quota Share reinsurance into providing crop insurance premium subsidies. (The cost implications are considered below). Box 4.6. Roles for Government in Supporting Agricultural Insurance Legal and Regulatory Framework. One of the most important functions for government in facilitating agricultural insurance markets is the establishment of an appropriate legal and regulatory framework and where necessary to enact specific agricultural insurance legislation. Enhancing Data and Information Systems. Time-series data and information on crop production and yields and climate are essential for the design and rating of any traditional crop insurance product or new weather index product. Governments can provide an invaluable service by creating national data bases and to then make these data bases available to all interested private commercial insurers either free of cost, or at concessionary rates. Product Research and Development. Among the major start-up costs for any new crop or livestock insurance program is the design (including the design of loss assessment procedures) and rating of new products, and then in the pilot testing of the new products and programs. Such costs may be prohibitive for individual private commercial insurers especially in developing countries. In such situations there is justification for government to provide financial support to product design and rating, especially where the products and rates are then made available to all interested insurers. Education, Training and Capacity Building. Governments can play an important role on new agricultural insurance programs by supporting (a) farmer awareness and education programs and (b) capacity building and workshops and technical training programs for key agricultural insurance staff. - 128 - Catastrophe Risk Financing. Agricultural insurance often has to protect against catastrophe perils of flood, drought, and wind storm in crops and epidemic disease outbreak in livestock. Most insurance companies do not have adequate capital to retain their catastrophe risk exposures and they typically purchase some form of contingency financing and or reinsurance protection. For new companies which do not have large amounts of capital and have not yet built up claims reserves, the ability to retain risk is usually low and they typically need to purchase quota share treaty reinsurance and to then seek non- proportional reinsurance protection on their retention. In start-up situations where the insurance company does not have an established track record and loss history the costs of reinsurance protection may be very high. In such situations, government support to the reinsurance program may be highly cost effective. Public Sector Premium Subsidies. Premium subsidies are the most widely practiced form of government support to agricultural insurance practiced by over two thirds of countries which have some form of agricultural insurance. Governments justify the provision of agricultural insurance premium subsidies on the grounds that they make insurance more affordable for farmers particularly small and marginal farmers thereby increasing the rate of adoption and uptake of agricultural insurance. There are, however, major drawbacks of premium subsidies including the disproportionately benefit larger farmers to the detriment of small and marginal farmers, they tend to promote moral hazard namely to encourage crop production in high risk regions, once premium subsidies once introduced are very difficult to reduce or to withdraw and they represent a major cost to government. Source: Authors Agricultural Insurance Pool: Financial and Reinsurance 4.110. The possible financial and reinsurance options for the Kazakhstan Agricultural Insurance Pool are presented in Figure 4.7. The proposal assumes that the Pool managers will purchase international reinsurance protection (non-proportional and possibly proportional as required) at fully commercial reinsurance rates, and that GRK will provide financial support in the form of crop insurance premium subsidies and possibly through involvement in the financing of catastrophe reinsurance layers. Some preliminary estimates are made for the voluntary MPCI program of the potential demand for cover and the associated premiums and liability based on the assumptions presented in this section, along with provisional estimates of the costs of premium subsides and finally some comments on a structured risk financing and risk transfer program for Kazakhstan. - 129 - Figure 4.7. Kazakhstan Agricultural Insurance Pool Financial Flows Kazakhstan Financial Flows Catastrophe Fund for Financial Support to Reinsurance International Agriculture, FFSA Reinsurers Premium subsidies Commercial Reinsurance KAZAKHSTAN Farmers Payment of Premium AGRICULTURAL INSURANCE POOL Payment of Claims Source: Authors Voluntary Crop Insurance Portfolio Financial Estimates 4.111. It is currently very difficult to estimate the future demand for crop insurance under a voluntary commercial crop insurance scheme. There are considerable uncertainties over the future demand for voluntary commercial crop insurance. In practice demand will be influenced by farmer‘s demand for and acceptance of the existing and new crop (and livestock) insurance products that are introduced, the commercial premium costs and farmer‘s affordability or willingness to pay for crop insurance, the levels of premium subsidies offered by government and finally decisions over whether to link crop insurance to crop credit provision. 4.112. Some provisional financial estimates of Total Sum Insured, premium and costs of premium subsidies to government have been calculated for the spring wheat MPCI program over the next 5 years. The model assumes an average 40% coverage level and demand projection uptake rates over the next 5 years under a voluntary program starting in year 1 with 10% uptake and rising to 50% uptake after 5 years. It is recognized that these are optimistic uptake assumptions. On this basis, total scheme liability in year 1 might be in the order of KZT 17 billion (US$ 113 million) rising by year 5 to KZT 85 billion (US$ 567 million) with corresponding year 1 commercial premium of nearly KZT 2.1 billion (US$ 14.0 million) rising to KZT 10.7 billion (US$ 71.3 million) by year 5. (Table 4.14). 4.113. The provision estimates of the costs to government of premium subsidies assuming 50% premium subsidy level would be nearly KZT 1.07 billion (US$ 7.1 million) in year 1 rising to KZT 5.35 billion (US$ 35.7 million) by year 5. Estimates are also provided for premium subsidy levels of 25% and 65% of premium. (Table 4.14). - 130 - Table 4.14. 5-Year Estimates of Voluntary MPCI Uptake and Total Sum Insured and Premium Income and Costs of Premium Subsidies (40% Coverage level) MPCI 40% MPCI Crop Insurance Uptake Scenarios next 5-years (KZT Million) Item Coverage Level Year 1 Year 2 Year 3 Year 4 Year 5 100% Basis 10% 20% 30% 40% 50% Sum Insured 169,697 16,970 33,939 50,909 67,879 84,849 Commercial Premium 21,410 2,141 4,282 6,423 8,564 10,705 Cost of GRK Premium Subsidies: 25% of premium 5,353 535 1,071 1,606 2,141 2,676 50% of premium 10,705 1,071 2,141 3,212 4,282 5,353 65% of premium 13,917 1,392 2,783 4,175 5,567 6,958 Probable Maximum Loss: PML 1 in 100 years 66,470 6,647 13,294 19,941 26,588 33,235 PML 1 in 250 years 79,688 7,969 15,938 23,906 31,875 39,844 Source: Authors Risk Financing and Reinsurance 4.114. The associated financial liability on the MPCI program that the Pool co-insurers would need to protect through a combination of insurance and reinsurance are also modeled for 1 in 100 year and 1in 250 year probable maximum loss scenarios. In Year 1 for 10% MPCI scheme uptake, the 1 in 100 year PML liability is estimated at KZT 6.6 billion (US$ 44 million) rising to KZT 33.2 billion (US$ 221 million) by year 5 (assuming 50% uptake). It is recommended that this would be the minimum level of protection which the Pool scheme management should protect through their risk financing and reinsurance program and if a more conservative level of protection is required they could consider the 1 in 250 year PML estimates. (Table 4.14). 4.115. It is assumed that in future the Agricultural Insurance Pool will purchase common account stop loss protection from international reinsurers in order to protect the program against catastrophe losses. It is likely in the initial stages that international reinsurers will only be willing to provide layered stop loss reinsurance protection in order to limit their liability to catastrophe claims and that the Government of Kazakhstan may therefore need to also participate in the structured risk financing program by providing catastrophe reinsurance for low frequency but high severity losses. An example of layered insurance and reinsurance is presented in Figure 4.8 below. - 131 - Figure 4.8. Example of Agricultural Risk Layering 50 Size of the Loss Government 40 Risk Reinsurers Transfer 30 Insurance Companies 20 Cooperatives Risk & Mutuals Pooling 10 Agricultural Risk Producers Retention 0 Minor Small Medium Large Catastrophic Type of Event: Mahul & Stutley 2010 4.116. For the proposed spring wheat MPCI program and 30% coverage level, figure 4.9. presents an illustrative commercial insurance and stop loss reinsurance program which is intended to provide layered protection to the Pool coinsurers for losses excess 100% priority up to a maximum 300% loss ratio (300% of Gross Net Premium Income, GNPI). The 300% loss ratio is equivalent to a loss of about 35% of total sum insured and is equivalent to an expected PML of 1 in 200 years. Under this scenario the international stop loss reinsurers would provide layered protection for two layers: Layer 1, 50% excess 100% GNPI and Layer 2, 50% excess 150% GNPI, while government would come in with catastrophe reinsurance protection for losses excess 200% up to 300% of GNPI. In year 1 of the program assuming 10% uptake of the MPCI spring wheat program and 30% coverage level, the indicative pricing for these three layers would be: Layer 1 3.9% of GNPI; Layer 2, 2.7% of GNPI and Layer 3, 3.4% of GNPI45. 45 GNPI 10% uptake would be equivalent to GTZ 1.492 billion (US$ 10 million - 132 - Figure 4.9. Illustrative Stop Loss Reinsurance Program for Kazakhstan wheat MPCI Program and assumed 30% Coverage level Loss Ratio % GNPI 300% Government of Kazakhstan Layer 3 100% xs 200% GNPI 200% International Reinsurers 150% Layer 2 50% xs 150% GNPI International Reinsurers 100% Layer 1 50% xs 100% GNPI Agricultural Insurance Pool Primary Retention 0% Source: Authors 4.117. It is unlikely that international reinsurers would agree to reduce the Pool’s priority to less than 100% GNPI. In this instance the pool may also elect to reduce its exposure by reinsuring some of its primary retention on a proportional or quota share treaty basis. These and other risk layering options would be developed further in the design and planning stages of a commercial crop insurance program for Kazakhstan - 133 - Chapter 5: Opportunities for New Crop Insurance Products 5.1. In Kazakhstan there appears to be both a need for and opportunities for Risk Differentiation and Product Development. Under the major mature agricultural insurance programs found in countries such as USA, Canada, Spain, Australia, the markets are highly developed and differentiated in terms of the range of crop, livestock, forestry, and aquaculture insurance products they provide to different segments of the farming community and in terms of the perils that are underwritten. In the case of crop insurance there are usually a range of product types ranging from simple named peril crop hail and or frost covers through to individual grower MPCI policies and new index-based products. Currently in Kazakhstan a single Loss of Investment Cost Policy providing a low level of costs-based protection is provided on a compulsory basis to all peasant farmers and agri-business enterprises, irrespective of whether the product meets their risk management needs and to a major extent this probably explains many farmers‘ dissatisfaction with the existing scheme. 5.2. As part of this World Bank study an assessment has been conducted of the potential to design and implement new crop insurance products including named peril frost and hail cover, AYII and WII for specific types of Kazakh farmer and for different regions according to the key risk exposures. This section presents the findings of this assessment for these new crop insurance products including a description of the cover design features and indicative sums insured and premium rates for each product and where possible some provisional estimates of the potential demand for each product. Named Peril Crop Insurance 5.3. Crop hail has been the second most important cause of insured claims on the Obligatory crop insurance scheme over the past 6 years. Hail is a moderate to severe problem in many parts of Kazakhstan with peak months of hail exposure occulting between May and July (Section 2). Over the past 6 years of operation of the Obligatory LIC scheme, hail has been the second most important cause of loss after drought, accounting for about 2.5% of the total are lost due to insured perils. On the basis of the field visits it appears that there may be demand under a voluntary scheme by farmers in some regions for a hail-only insurance policy. This section presents some preliminary recommendations for the design and rating of a spring wheat hail policy for Kazakhstan. Features, Advantaged and Disadvantages of Named Peril Damage based Crop Insurance 5.4. Crop Hail insurance is the world’s oldest form of crop insurance, the product is very standard and well understood and is extensively applied to the insurance of wheat and a wide range of cereals, horticultural and tree fruit crops. Single peril hail insurance is the simplest and best known type of indemnity-based crop insurance which has operated for more than 100 years in Europe, North America, Argentina, Australia and New Zealand. Today there is a large body of accumulated experience with crop- hail damage-based insurance & indemnity products and wordings are readily accessible through international associations of hail insurers, premium rates can initially be set based on transferred international experience and so long as suitably high each and every loss deductibles (or franchises) are maintained, the rates are generally not high, and finally standardised damage-based loss assessment procedures can be accessed from the international hail associations and training provided to local staff. 5.5. Crop-hail insurance is distinguished from the Kazakhstan Loss of Investment Costs Policy, in that the insurance and indemnity system is not based on “loss of the crop yield”, but rather on the “percentage hail damage” caused to the crop. Under a damage based indemnity system, physical loss or - 134 - damage to the crop is measured in the field soon after a specific loss event to an insured peril and the claim is usually settled shortly after the time of loss. Normally the damage is measured as a percentage loss, and this percentage is applied to an agreed sum insured e.g.: incurred production costs or other agreed value) for the crop. The sum insured may be adjusted downwards if the actual crop is found to be below the normal production potential for uninsured reasons, for example poor crop establishment. A deductible is usually applied to the loss expressed as ―percentage damage‖, although this can be a fixed value. This method is most applicable to programmes with single or a limited number of discrete event perils (e.g. hail, windstorm and frost). An example of the basis of insurance and indemnity for hail insurance is given in Annex 6 with a worked example. 5.6. The key advantages of a named-peril (hail) damage-based indemnity policy include: (a) there is no need to collect time-series individual grower production and yield data on which basis to establish a normal average yield and then an Insured Yield because the policy uses a damage based indemnity procedure rather than loss of yield; (b) the sum insured can be set according to an agreed value per acre, based either on production costs, or production costs plus an element of the expected gross margin profit, or finally a revenue valuation based on the farm-gate sale price of the crop times the expected output; and finally (c) loss adjustment is based on percentage damage estimation to the crop according to its growth stage and this procedure is usually easier and cheaper to implement than yield-based loss assessment. 5.7. Insurers are generally willing to insure hail damage because it is considered as a non-systemic or catastrophe class of crop insurance business and is not subject to anti-selection by farmers. Many crop insurers are very reluctant to offer MPCI loss of yield cover against drought, flood, windstorm and flood because of the systemic or catastrophe nature of these perils. Conversely hail is usually a relatively high frequency but low severity peril and so long as the insurer can achieve a geographical spread of risk, is not subject to catastrophe losses. Hail is also usually an unforeseeable and unpredictable event and it is therefore unlike drought cover, it is not subject to anti-selection and or moral hazard. 5.8. The drawbacks of damage-based crop insurance and indemnity policies include: (a) the product is best suited to specific event perils that cause obvious and easily measured damage to the crop such as hail or wind and sometimes frost or excess rain, but it is not suitable for progressive perils which impact over time on the crop such as drought and where losses can only be objectively measured in terms of yield reduction or loss; and (b) the product is not suitable for other perils such as flood. Indeed flood is not offered by insurers as a single-peril on traditional indemnity-based crop insurance policies because of the problems of anti-selection. Cover Design Considerations for Spring Wheat Cover in Kazakhstan 5.9. Hail insurance is a very flexible class of insurance which can be designed to provide farmers with a wide range of options and choice. (Key features of a hail policy for spring wheat are reviewed below and a summary of these cover design options is presented in Box 5.1) 5.10. Hail Insurance can be designed to prove a very high level of crop revenue protection if requested by the insured. The sum insured for crop hail insurance is very flexible as it is not dependent on an insured yield per se. Insurers will generally permit farmers to insure their crops against a minimum level equivalent to the costs of production per hectare or amount of crop loan per hectare through to a maximum valuation based on the gross revenue (yield x expected market sale price). In Kazakhstan, in principle, a spring wheat grower achieving an average yield of about 15 centners/Ha (1.5 metric tons per hectare) could insure 100% of his expected wheat revenue against hail, or a value of about KZT 45,000/Ha (at current sales prices for wheat of about KZT 3,000/Ha). - 135 - 5.11. Hail insurance can be marketed both pre-season and during the crop growing season right up to the time of harvest. Because hail is considered an unforeseeable and unpredictable event, many hail insurers will permit farmers to purchase cover at any time during the growing season subject to a waiting period of 24 or 48 hours. This is in contrast to loss of yield based policies which insurer against drought where a sales cut-off period of usually a month prior to sowing is required to avoid pre-existing conditions and potential anti-selection. 5.12. Hail insurers may offer the Farmer choice to insure only a part of his total fields and total cultivated area. In the start-up phase of a new crop hail insurance program for spring wheat in Kazakhstan it is, however, recommended that in order to achieve volume and a spread of risk, farmers should be required to declare and insure all their fields and farms of the same crop located in any one Rayon. 5.13. The Insured Unit for the purposes of Loss Adjustment can also be very flexible under a crop hail policy. There are many options for defining the Insured Unit (IU) for a crop hail policy namely the area of the insured crop over which the damage is estimated and the policy excess applied. At the highest level of aggregation the IU can be defined on a whole farm-basis as the entire area of the same crop grown in the same farm or location. Conversely insurers may agree to offer insurance on a Field by Field basis in which case the IU is the individual field. Finally some insurers agree to offer crop insurance on an acre by acre basis such that any area within a field which is subject to damage which exceeds the policy excess is eligible for an indemnity. For Kazakhstan spring wheat it is recommended that insurers offer optional covers for Whole-Farm insurance and on a field by field basis. 5.14. It is conventional on a hail policy to apply a first loss excess (deductible) which is retained by the insured. The objective of the policy excess is to eliminate small frictional hail losses which are very time consuming and costly for the insurer to adjust and which would rapidly erode the premium reserved to pay hail claims and to only indemnify the more severe hail losses which are of economic consequence to the Insured. The excess is usually applied on an each and every hail loss event basis and can take several forms including: i) a percentage damage deductible which is deducted for the gross assessed percentage damage. An example would be a 6% deductible which is applied on an eel basis. If the assessed damage amounted to 5% this would fall below the deductible and there would be no claim. If the assessed damage was 15%, the net damage would be 9% (15% - 6% deductible) which would be applied to the sum insured. ii) a percentage franchise (which is sometime termed a ―qualifying franchise). If the assessed damage amounted to 5% and a 6% franchise applied, there would be no indemnity as the damage would be below the 6% franchise. However for the 15% damage example, as this exceeds the 6% qualifying franchise, the damage would be indemnified from the ground-up or in full and the Insured would receive an indemnity of 15% damage applied to his sum insured. iii) on some policies rather than apply a percentage damage excess, a fixed value excess (for example KZT 10,000 eel) is applied either as a franchise or as a deductible. In this case the gross value of the assessed hail damage would have to exceed KZT 10,000 to give rise to a claim. 5.15. From a loss assessment perspective some hail insurers prefer to apply a deductible as opposed to a franchise. The operation of a franchise places very high demands on the need for accurate in-field percentage damage assessment. For example with a 6% franchise, where the percentage hail damage is light and where sampling error means that the assessed damage could be anywhere from 4% to 8% over the IU, the tendency is for the farmer to dispute any assessment which falls below 6% and which would - 136 - give rise to a claim in full. Conversely with a 5% deductible which is deducted from the assessed damage the level of precision required when assessing low levels of hail damage is less demanding. 5.16. Crop hail deductibles or franchises are commonly in the order of 3% to 6% for cereals, but is areas of high hail risk exposure may need to be correspondingly higher. In much of Europe, North America and in Latin America hail deductibles are between 3% to 5% eel. In Argentina, the industry has followed a standard hail policy for wheat with 6% franchise for many years. In some parts of the word with a high hail risk exposure, deductibles may, however, need to be higher and in the order of 10% or greater per event. 5.17. Crop hail damage assessment procedures for wheat are very well developed. Standard hail loss assessment procedures have been developed by the European, South African, Argentinean and US crop hail insurers and these manuals of procedures should be readily accessible by crop hail insurers in Kazakhstan to adapt to their own conditions. Box 5.1. Cover Design Features for Named Peril Hail Policy in Kazakhstan Insured Crop: Spring wheat during the cover period against physical loss or damage to the crop due to the action of direct hail damage. This policy does not insure against loss of wheat grain quality (price- downgrading) in spring wheat. Basis of Insurance and Indemnity. Percentage Damage-based Policy Insured Perils: Hail: direct physical damage or loss to the Insured crop. In addition, physical loss or damage caused by wind associated with hail may be considered as an optional peril, subject to the payment of an additional premium. Cover Period: From the time of crop emergence and full stand establishment (defined as 10cm stage in wheat) through to the completion of harvest of the crop. Cover may be purchased at any time between the opening and completion dates of cover subject to a waiting period of 24 hours from the time of payment of premium up to the time of inception of hail cover. Insured Unit: The Insured is obliged to declare and insure all his/her fields of spring wheat grown in the same Rayon. The Insured may select to insure all their fields of spring wheat as: i) a single Insured Unit (whole farm-basis), or ii) separately by field (field by field basis). Basis of Sum Insured: A fixed amount in KZT per hectare which the Insured may elect based on his/her coverage requirements and which may range from a costs of production per hectare through to a maximum level based on the average crop revenue value for spring wheat in that Rayon as specified by the local department of the MOA. The total sum insured will be calculated by multiplying the per hectare sum insured by the area (in hectares) of each and every insured field declared by the Insured and summing the total. Basis of Indemnity: (i) Gross Value of Loss = Sum Insured x Percentage Hail Damage. Where the percentage damage exceeds the Policy Excess, giving rise to a claim, the Indemnity (I) = Sum Insured x (Percentage Damage – Policy Excess) = Net value of claim (ii) Policy Excess: The policy excess will be applied on an each and every loss (eel) basis in each Insured Unit as defined. Excess options include: i) Percentage damage deductible (e.g. 6% eel) or ii) Percentage damage franchise (e.g. 6% eel), or iii) A fixed value amount (deductible or franchise), for example KZT 10,000 eel. Loss Notification procedure: The Insured is responsible for notifying the Insurer of a hail event which is expected to exceed the policy Excess (deductible or franchise) within 48 hours of the loss event occurrence. Loss Assessment procedure: In-field loss assessment will be adopted using standard procedures to measure the area damaged by hail and to assess the average percentage hail damage in each Insured Unit. - 137 - Under and Over Insurance (of Spring Wheat Area) : In the event of a loss, if it is discovered that the Insured has under-declared his/her spring wheat sown area by more than 5% of the total area, the Insurer retains the right to apply the Law of Average to any claims settlement. In the case of Over-Insurance the maximum amount payable in the event of a total loss will be the actual assessed cultivated area times the agreed per hectare sum insured and subject to the policy excess. Exclusions: All perils apart from Hail (and wind- optional) are excluded. Other Key Conditions: The Insured must declare and insure all his/her spring wheat grown in the same Rayon; Premium is payable prior to inception. Source: Authors Hail Exposure and Preliminary Rating Considerations 5.18. Hail is usually a localised phenomenon which seldom accumulates over wide areas and as such it does not pose the same catastrophe exposure as systemic perils such as drought. Hail is often a relatively high frequency (i.e. it occurs every season), but low severity (hails tends to be localized) class of crop insurance business. 5.19. In order to design and rate a crop hail insurance program it is necessary to obtain data on three key parameters, hail exposure data, hail hazard data and hail severity data. In some countries hail exposure data is available from meteorological stations in the form of the number of recorded hail days per month, but because hail is a localized phenomenon, this may not be representative of hail occurrence at the regional level. In Kazakhstan information is available from KHM for selected stations on the frequency of occurrence of hail by month for the period 1990 to 2010 which shows that the peak hail months are from May to July (Figure 5.1.). While exposure data is readily available for spring wheat as per the cultivated area and value of spring wheat per Rayon and Oblast there is very little recorded information in Kazakhstan regarding hail severity and damage. For these reasons a simple hail damage simulation model has been developed for spring wheat in Kazakhstan and in order to generate indicative hail rates for each Oblast. Figure 5.1. Kazakhstan Frequency of Occurrence of Hail by Oblast 1990 to 2010 (21 years) 45% Frequency of Occurence (% of years) 40% 35% 30% 25% 20% 15% 10% 5% 0% Kostanay Average Akmola Average Pavlodar Average SKO Average Source: Authors from KHM (Figures taken from Table 2.1.) - 138 - 5.20. A simple hail rating simulation model has been developed for spring wheat in Kazakhstan using industry approved hail rating methodology. The model has been developed for 13 sample weather stations in selected Rayons of Akmola, Kostanay, NKO, Pavlodar and SKO Oblasts. The model combines exposure data for each Rayon during each month of the growing season with the hazard model (frequency of occurrence of hail by month), a hail severity index from Low to Very severe and a hail vulnerability model which are then combined to simulate hail percentage damage distributions with 10,000 iterations and to estimate the average burning cost or pure loss cost for each station/Rayon. The average loss costs are then adjusted for a 6% hail franchise and grossed up by an assumed factor of 20% to derive indicative average commercial hail rates for each sample Rayon. These rates are summarised in Figure 5.2 and Annex 6. It is stressed that these rates are indicative and would require further analysis if a commercial crop hail scheme is to be launched in Kazakhstan in future. 5.21. The preliminary hail rating for spring wheat in selected rayons in Kazakhstan suggests that it should be possible to design hail cover at affordable rates to producers. For the 6% franchise average rates vary from about 2.5% to 3.5% in the lowest hail risk regions, up to 5% or 6% in the medium hail risk regions and as high as 7.5% in Mikhailovka Rayon Kostanay Oblast and 12.9% in Kazgurt Rayon, SKO. In the next stage it would be useful to validate these relative hail rates with local agricultural specialists in each Oblast and Rayon and who are familiar with hail exposure and hail damage in wheat. Figure 5.2. Indicative Average Rayon-level Commercial Hail Rates for Spring Wheat in a sample of Oblasts and Rayons 14.00% 12.9% Average Premium Rate (%) 12.00% 10.00% 7.5% 8.00% 6.7% 6.00% 5.6% 5.5% 5.8% 4.1% 3.9% 4.4% 4.00% 3.1% 2.8% 2.6% 2.0% 2.00% 0.00% Source: Authors analysis of crop hail incidence data provided by Arka consulting Conclusions on Crop Hail 5.22. Crop hail insurance should be relatively easy to design and implement in Kazakhstan as a commercial crop insurance product. Since this is a non-catastrophe crop insurance product, it should be relatively easier for the crop insurance companies and possibly the Farmer‘s Mutual Insurance Associations to underwrite this product with limited access to reinsurance protection. Because the average rates for single-peril hail insurance are potentially considerably lower than the individual grower loss of yield covers (LIC Policy or MPCI and the AYII product), there is also more potential to market this cover to Kazakhstan farmers on a voluntary basis and without the need for government premium subsidies. It is likely that the demand for single-peril crop hail insurance for spring wheat will be quite - 139 - low in the initial stages of implementation of this product as hail exposure is not as widespread as the drought risk exposure. There would be an important start-up cost namely to design the suitable crop hail loss assessment procedures for Kazakhstan and to then identify a core of loss assessors who would receive specialist training in hail loss assessment procedures in wheat. Finally, it is likely that there will be demand for crop hail in other crops, for example cotton and horticultural crops grown in southern Kazakhstan and over time there should be potential to develop and expand a crop hail portfolio in Kazakhstan. Area-Yield Index Crop Insurance 5.23. On the basis of this feasibility study it appears that there may be considerable potential in Kazakhstan to design and implement AYII as an alternative to or as a complement to the existing individual grower LIC and new proposed individual grower MPCI crop insurance programs. Outline proposals are presented below for a prototype Area-Yield Index product and program for spring wheat grown in Kazakhstan, but it is stressed that further design work will be required if the insurance companies decide to proceed with the pilot testing and implementation of this product. It is assumed that AYII would be introduced as a voluntary commercial crop insurance program. Further details of the AYII product are contained in Annex 7. 5.24. The prototype spring wheat AYII product presented in this section excludes Aktobe and WKO Oblasts and their Rayons on account of the commercially uninsurable risk exposures in these two Rayons. The AYII program presented in this section therefore relates only to the 6 main wheat growing Oblasts of Akmola, EKO, Karaganda, Kostanay, NKO and Pavlodar. Aktobe and WKO are excluded from the AYII analysis as they could not be included in a commercial AYII scheme. However, it is possible that government may wish to use an AYII approach to providing disaster compensation to farmers in these two Oblasts and for this reason in Annex 7, coverage levels, insured yields and indicative premium rates are also presented separately for Aktobe and WKO. Features, Advantages and Disadvantages of AYII 5.25. AYII represents an alternative approach to MPCI insurance which aims to overcome many of the drawbacks of traditional individual grower MPCI insurance. The key feature of this product is that it does not indemnify crop yield losses at the individual field or grower level; rather, an Area-Yield-Index product makes indemnity payments to growers according to yield loss or shortfall against an average area yield (the index) in a defined geographical area (e.g., the total sown area of spring wheat grown in a single Rayon). An area-yield index policy establishes an Insured Yield which is expressed as a percentage (termed the ―Coverage Level‖) of the historical average yield for each crop in the defined geographical area such as a Rayon and which forms the Insured Unit (IU). Farmers whose fields are located within the IU may purchase optional Coverage Levels which typically vary between a minimum of 50 percent and a maximum of 80 percent of the historical average yield. The actual average yield for the insured crop is established by sample field measurement (usually involving crop cutting) in the IU and an indemnity is paid by the amount that the actual average yield falls short of the Coverage level purchased by each farmer. 5.26. An example of the basis of insurance and indemnity for an AYII cover for spring wheat is shown in Figure 5.3 for spring wheat. For this example Bulandinski rayon in Akmola was selected. For this example it is assumed that the actual 5-year area average yield of spring wheat is 10 centners/Ha in Bulandinski Rayon and all framers are offered the same coverage level of 70% of the average yield or 7 centners per hectare. The unit sum insured is KZT 3,210 per center giving a standard sum insured of - 140 - KZT 22,470/Ha. Three farmers each with total planted area of 1,000 Ha purchase cover. It is a moderate drought season and farmer A achieves an actual average yield on his land of 7 centners/Ha; Farmer B, incurred more severe losses and achieved 5 centners/Ha and finally farmer C achieved only 3 centners/Ha on average. However under the AYII cover, the policy does not indemnify each farmer according to his own losses, but rather according to the actual average yield reduction at the Rayon level. In this example the Rayon actual average yield was only 5 centners/Ha and therefore the yield loss was 2 centners/Ha (7 centners – 5 centners) with an indemnity payment of KZT 6,420/Ha Each of the insured farmers therefore received the 2 centner/Ha indemnity over their 1,000 Ha farms valued at KZT 642,000 per farmer, in spite of the fact that Farmer A did not suffer any yield shortfall against the Rayon 70% cover level, while farmer C incurred more than a 2 centner/ha shortfall on his own land.. Figure 5.3. Illustrative example of Basis of Insurance and Indemnity for AYII cover in wheat Insurance Contract Conditions: Insured Peril: All Risk Policy Area Yield Index Insurance Payout Crop: Spring wheat Examples 11 PE Rayon Level Expected Yield Rayon: Bulandinski in Akmola Oblast Type of Farmer: Production Enterprise (PE) 10 PE‘s 5-year rayon-level Average Expected Yield for PEs (EY): 10 centner/ha 9 PE‘s rayon –Level Guaranteed Yield at 70% coverage (GY): 7 centner./ha Agreed price (AP): KZT 3210 /center 8 Yiel;d (Mt./ha) Sum Insured KZT 22,470/Ha 7 Insured Unit Area (IUA): Farm A, B, and C, each, with 1,000 hectares Insurance Payouts = 2 centner/he. 6 And identical TSI of KZT 2.25 million 5 PE‘s rayon-level actual average yield (AY)= 5 centner/ha, but: 4 PE Rayon Level Actual Yield PE-Farmer‘s A actual yield (AYA) = 7 centner/ha; crop loss 0 centners/ha. PE-Farmer‘s B actual yield (AYB) = 5 centner/ha; crop loss 2 centner/ha. 3 PE-Farmer‘s C actual yield (AYC) = 3 centner/ha; crop loss 4 centner/ha 2 Insurance Payout Calculation: (IPC) 1 If AY < GY, then: 0 IPC = (GY – AY) * AP* IUA Farm A Actual Farm A Actual Farm A Actual IPC = (7 centners/he - 5 centners/he) * KZT 3,210 /he.* 1,000 hectares = Yield = 7 Yield = 5 Yield = 3 KZT 6,420/Ha or KZT 642,000 per farmer centner/ha centner/ha centner/ha AYII provides payouts to all the farmers situated in the area selected as IU Actual Yield Farmer Actual Loss Area-yield Index Payouts Source: Authors 5.27. AYII works best where farmers’ crop production systems, technology levels and crop output and yields in the defined IU are relatively homogeneous and furthermore responds best to systemic risks such as drought which tend to affect farmer‘s production and yields in the same way across wide geographic areas. In Kazakhstan spring wheat crop production systems are understood to be relatively homogeneous at the Rayon-level, but as Section 4 has shown there are differences in technology levels and the average production and yields obtained by the 2 categories of wheat producers, i) production enterprises and ii) commercial farmers. As AYII is intended as a catastrophe product, in principle, these technology and yield differences between these two categories of farmers should not, however, negate the use of this crop insurance tool. (See below for further discussion of homogeneity of spring wheat production at Rayon level in Kazakhstan and the issue of basis risk). 5.28. In the context of Kazakhstan a key potential advantage of AYII over individual grower MPCI is the ability to offer higher levels of insured yield coverage at lower rates because losses are adjusted against an area yield index and not at the individual farmer level. Section 4 showed that although individual grower MPCI is technically feasible in Kazakhstan, on account of the very high variability in crop production and yields, the maximum coverage level that can be offered to growers at affordable premium rates in most Rayons is only between 20% and 50% of average yield. AYII indemnifies losses according to yield variation or loss at the area-level, in this case the Rayon and because of the - 141 - considerably lower variation in aggregate rayon level crop production and yields year on year, it should be possible to offer higher coverage levels of 30% to 70% and even higher in some regions and at more affordable rates than for individual grower MPCI. 5.29. Other key advantages of the AYII approach are that moral hazard and anti-selection are minimized, and as the costs of administering such a policy are much reduced, this offers the potential to market this product at lower premium costs to farmers. As the policy responds to yield loss at the county or Rayon area-level and not at the level of the individual farmer, no farmer can influence the yield indemnity payments and as such anti-selection and moral hazard are minimized. Administration costs are also greatly reduced because there is no need for pre-inspections on individual farms and loss assessment is not conducted on an individual farmer and field by field basis, but rather according to a pre-agreed random sampling of crop yields on plots within the IU. These costs savings can be passed on to farmers in the form of lower crop insurance premiums. 5.30. The main drawback of an AYII policy is “Basis Risk” or the potential difference between the insured area-yield outcome and the actual yields achieved by individual insured farmers within the insured area. Basis risk arises where an individual grower may incur severe crop yield losses due to a localized peril e.g. hail, or flooding by a nearby river, but because these localized losses do not impact on the district or rayon average yield, the farmer who has incurred severe crop damage does not receive an indemnity. In addition, basis risk may arise where individual farmer crop production and yields are highly heterogeneous (different) in the same Rayon, which will invalidate using and Area-based approach. (See below for further discussion of Basis Risk if farming systems, production and yields within the Insured Unit are not relatively homogeneous). 5.31. AYII is potentially a flexible crop insurance product that can be implemented at the micro-level for individual farmers, or alternatively as a meso-level product that is designed to protect the credit portfolio of a regional financial institution. In Kazakhstan there may be scope to design AYII both as a micro-level individual grower product for medium to large wheat (or other cereal) producers and to then design it as a meso-level product to protect the cooperative or MFI loan portfolios to large numbers of small rural households in individual Rayons in Southern Kazakhstan (further discussion in Chapter 6). International experience with AYII 5.32. AYII has been widely adopted for smallholder rice and wheat cropping in India and where crop insurance in linked to seasonal crop credit. India has operated a public-sector AYII program for more than 30 years under its public-sector National Crop Insurance Scheme, NAIS, which is implemented by the Agricultural Insurance Company, AIC, India Limited, a publicly owned specialist agricultural insurer. Crop insurance is compulsory for farmers who borrow seasonal production credit. Currently this program insures about 25 million Indian farmers each year. In order to make crop insurance widely available at affordable prices to India‘s small and marginal farmers, the government has capped premium rates at about one third of the actuarially required levels and then provided AIC with free stop loss reinsurance protection: in the case of food crops, reinsurance cuts in at a 100% loss ratio, while for commercial and horticultural crops, the priority is higher at 150% loss ratio (Figure 5.4). The program therefore shows several common features with the Kazakhstan LIC scheme including compulsion of cover, capped premium rates and government support for reinsurance, although in Kazakhstan this is based on a 50% quota share as opposed to the non-proportional stop loss protection provided in India. Although the NAIS has achieved very high levels of insurance uptake by Indian farmers, the scheme incurs major delays (often of up to 6 months post harvest) in arriving at the estimates of actual area-yield and in settling claims and this delay is very unpopular with farmers. For the past five years the World Bank has been working with AIC to strengthen and reform the NAIS scheme into a market based system including most importantly, (i) introduction of actuarial rating, (ii) switch of government financial support - 142 - from claims compensation to crop insurance premium subsidies and (iii) opening up of the market to competition by local and international reinsurers. This modified NAIS program was formally launched in the Rabi Season 2010/11 in about 10% of the total NAIS command area and if successful this market- based PPP system will gradually be introduced into all states of India. (See Box 5.2. for further details). Figure 5.4. Government Stop Loss Protection for NAIS scheme, India Loss Ratio (%) FOOD COMMERCIAL CROPS & HORTIC. (70% of premium) CROPS (30% of premium) GOVERNMENT OF INDIA & STATE GOVERNMENT STOP LOSS PROTECTION 150% AICI 100% RETENTION AICI RETENTION Source: Authors Box 5.2. Main features of India‟s mNAIS scheme for Rabi 2010/11 Actuarial regime. The mNAIS scheme operates on an ―actuarial regime‖ in which the government‘s financial liability is predominantly in the form of premium subsidies given to AICI and funded ex-ante, thereby reducing the contingent and uncertain ex-post fiscal exposure currently faced by the government under NAIS and reducing delays in claims settlement.   Up-front premium subsidies. AICI receives premiums (farmer collections plus premium subsidies from the government) and is responsible for managing the liability of the mNAIS through risk transfer to private reinsurance markets and risk retention through its reserves. It is financially able to operate on a sustainable basis.   On-account partial payment. The mNAIS product continues to be based on an area yield-based approach, with a provision for an early part payment to farmers (in season) based on weather indices.  Small insurance units. Crop-cutting experiments to assess crop yield estimates are lowered from block level to village level to reduce basis risk (i.e. the mismatch between the actual, individual crop yield losses and the insurance indemnity).  Cutoff dates. Adverse selection is reduced through the enforcement of early purchase deadlines ahead of - 143 - the crop season.  Additional benefits. Additional benefits are offered for prevention of sowing, replanting, post-harvest losses, and localized risk, such as hail losses or landslides. Early settlement of claims. mNAIS combines weather based indices for on-account partial payment of claims in case of adverse mid-season conditions, whereas area-yield indices are used for final payment of claims. The final estimation of loss is based on area-yield measurement at the time of harvest using crop cutting experiments. Source: GFDRR, 2010. 5.33. Other countries which are operating AYII include most notably the USA (where the product is termed Group Risk Plan, GRP) and this product is being researched in parts of Eastern Europe (Ukraine), Africa (Senegal, Ghana and Burkina Faso), South America (Brazil and Peru) and Asia (Bangladesh and Nepal). 5.34. In the Ukraine, a subsidised AYII scheme was launched for winter wheat in 2003, but this scheme was poorly designed and implemented and and has not taken off. In Ukraine as in Kazakhstan, agriculture is an important sector, especially for the production of winter wheat which is the leading export crop. Agriculture is very exposed to drought (e.g. in 2003, 2005, 2007), spring frosts, strong winds, hail and for winter crops, the additional risk of ―winterkill‖ (low temperature and freeze damage to the crop) which in 2003 caused catastrophe losses in more than 70% of the winter crops. During the Soviet period, Ukrainian state and collective farms were insured under the standard multiple-peril crop insurance program that operated throughout the Soviet Union. Following independence there was no crop insurance until the early 2000s. Between 2001 and 2003, a local insurer with the assistance of an international agricultural reinsurer analysed the possibilities for designing and implementing both individual grower MPCI cover for cereals and area-yield index insurance. There were major difficulties in designing and rating individual grower MPCI because following the break up of the state and collective farms it was impossible to establish normal average or expected yields for the newly formed large-scale commercial enterprises and or small to medium producers. Similarly there were difficulties in designing and rating AYII because although relatively reliable Oblast-level historical data were available, it was difficult to obtain accurate time-series information at the Rayon level. Further difficulties included the fact that post-independence average crop yields at Rayon and Oblast level showed a major declining trend due to reduced fertilizer use and the time-series data had to be carefully detrended. A voluntary pilot AYII scheme was launched in 2003 with a pool of local insurers, backed by European reinsures, but the scheme was not popular with farmers and suffered from poor implementation especially in the adjusting of losses and the scheme has now been discontinued. See Box 5.3. for further details of the problems of the Ukraine AYII pilot scheme. In the design of any AYII scheme for Kazakhstan, planners should learn from the lessons and experiences of Ukraine.46 Box 5.3. AYII Pilot Crop Insurance Scheme in Ukraine Crop Area Yield Insurance has been implemented poorly in Ukraine. A pilot hybrid multiple-peril crop insurance (MPCI) and area-yield index (AYII) scheme was launched in 2003 for all major field crops. The Rayon was the Insured Unit for the AYII cover. Indemnities were paid based on regional (Rayon) statistical records (but apparently 46 In this context, it should also be noted that the private insurance sector in Ukraine has also tried between 2003 and 2005 to develop crop weather index insurance (WII) with limited success. For a very good review of the issues and challenges faced in Ukraine see WFP & IFAD 2010. - 144 - not an official statistical report) and a farm-level inspection of actual yield and the farmer had to provide proof that the crop yield reduction was due to an insured peril – this meant farmers had to obtain reports from a local meteorological station. Complicated and unclear loss assessment procedures meant that payouts were usually delayed for up to 6 months. Recently producers have lost interest in the area-based yield index crop insurance product, and insurers have been looking for effective ways to insure crops. Source: WFP and IFAD 2010. Pre-conditions for the Design of AYII for Spring Wheat in Kazakhstan 5.35. There are several pre-conditions for the operation of AYII for spring wheat (and any other crops) in Kazakhstan including: (a) homogeneous spring wheat producing regions or zones (the Insured Unit) with low yield variation between farmers in the Insured Unit (UI), (b) for the defined Insured Unit, historical spring wheat sown area, production and average yield data for the past 15 years or more on which basis to establish the Insured Yield and technical rates, and (c) an independent and statistically accurate system of measuring average spring wheat yields in the defined region or zone and on which basis to trigger claims payments. 5.36. AYII is only effective if individual farmer-level spring wheat production systems and yields within a defined insured unit (the Rayon) are relatively homogeneous. Basis risk will arise under an AYII program if spring wheat farming systems are highly heterogeneous within the defined IU (the Rayon) in terms of soil types and fertility and soil moisture retention and in farmers‘ use of technology and inputs and where this translates into highly variable spring wheat yields between farmers in that Rayon. Under an AYII cover where all farmers in the IU are treated the same and losses are paid against shortfall on the area-average yield the danger is that farmers located on the best soils and who use high technology levels may receive an indemnity although their actual yields are well above the Rayon Insured Yield: conversely farmers using very low levels of technology and whose average normal yields are well below the Rayon average yield may also receive an indemnity even when they have not incurred any yield losses. Under this study, it had been intended to access individual farmer yield data for the six selected Rayons in Akmola, Kostanay and Pavlodar to check for yield variability and basis risk under an AYII program operating at a Rayon level. To date, however, it has only been possible to analyse spring wheat data for small samples of individual farmers in two Rayons in Pavlodar and one in Kostanay and this analysis suggests that yield variability between farmers in the same year is usually low enough to ensure that an AYII product could operate effectively with an acceptable level of basis risk caused by between farmer yield-variability. It is also notable from this analysis that in a good year (e.g. 2009), average yields are more similar between farmers in the same Rayon with COV‘s of about 20% to 25%: however, in drought years (2008, 2010), individual grower yields are more variable (COV‘s as high as 40% or greater) and this may be explained by differences between farmers in terms of soil moisture management through minimum tillage etc. (See Annex 7 for full details of this analysis of individual farmers‘ yields and issues of basis risk in Kazakhstan). 5.37. In Kazakhstan, the Kazakhstan Statistics Agency (KSA) has an accurate system for measuring and reporting spring wheat sown area, harvested area, production and average yields at a regional (Oblast) and zonal (Rayon) level and 17 years of historical spring wheat data from 1994 to 2010 are available for the purposes of this study. Section 2 of this report noted that Kazakhstan has a comprehensive system for measuring crop production and yields and in the case of spring wheat it has - 145 - been possible to access 17-years of crop production and yield data at Rayon level for production enterprises, commercial farmers and in total. This time-series production and yield data has enabled the design and rating of an AYII product for spring wheat which operates at a Rayon level. KSA reports spring wheat sown area, production and yields separately for production enterprises and for commercial farmers and in total. The prototype AYII product presented in this section of the report is based on a single Rayon-level AYII product (i.e. it is designed on combined crop production and yield data of both types of farmer), but if required separate covers could be designed for production enterprises and for commercial farmers as the AYII rating tool has been programmed to output insured yields and premium rates for both types of farmer in all Rayons and Oblasts. 5.38. For the operation of AYII it is necessary to have an independent, accurate and timely system of measuring and reporting of actual average yields in each IU at the time of harvest and on which basis to indemnify insured yield shortfall below the area actual average yield. In India which has the oldest AYII program in implementation, a national system of sample crop cutting in randomly chosen fields is used to calculate the actual average yield for each insured crop in each Insured Unit (IU). While crop- cutting is potentially a very accurate method of determining average yield in the IU this is dependent on having a statistically adequate number of sample crop-cuts, which must be randomly sited in selected fields and which if properly conducted is a very time consuming and costly exercise. In India the system only works because the state governments subsidize the costs of crop-cutting for the insurance sector. In the USA, the GRP area-yield plan operates at a county-level as the IU: in the USA county-level actual average yields are estimated by the National Agricultural Statistics Service (NASS) from grain-elevator area, production and yield reports and no in-field crop cutting is conducted. This county yield estimation procedure is considered accurate and impartial by both the insurance sector and by the Insured Farmers. 5.39. Kazakhstan adopts systematic sampling procedures for measuring actual average yields at Rayon level at the time of harvest and this could form the basis of indemnity under any Rayon-level AYII scheme in future. In Kazakhstan, KSA uses a multi-stage crop yield estimation procedure first to select villages and farmers who are differentiated into production enterprises and commercial farmers and then secondly to select fields for in-field randomly selected crop-cutting using 1 m x 1 m squares. Crop- cutting is conducted immediately prior to harvest for all major grain crops including spring wheat and the weight of the yield from each sample is adjusted according to its moisture content and the average yield in center per hectare calculated. The crop-cut yields for each type of producer are then averaged to produce the estimates of average yields per Rayon separately for production enterprises and commercial farmers and then in aggregate for both types of farmer at whole Rayon. Full details of the yield estimation procedures are contained in KSA‘s Guideline on Crop Yield Inspection Arrangements47. This yield estimation procedure at Rayon level is considered to be impartial and accurate and could form the basis of indemnifying yield loss under a Rayon-level AYII program for spring wheat. Spring Wheat: Insured Area, Yield Coverage Levels, Sums Insured and Calculated Indicative Premium Rates Area Insured Unit 5.40. For the operation of an AYII cover it is necessary to have a minimum sown area of the insured crop in each Insured Unit (Rayon). The setting of a minimum sown area is to avoid moral hazard, or in other words, to ensure that individual farmers are not able to influence the area yield outcomes in the IU. 47 KSA (2004) Guideline on Crop Yield Inspection Arrangement. Department for Industrial Statistics, Kazakhstan Statistics Agency, Almaty 2004 - 146 - In the USA the minimum area is 15,000 acres (about 6,000 Ha) per county (Skees et al 1997)48. In northern Kazakhstan where there are some very large spring wheat production enterprises the criterion used under the current study was a minimum of 10,000 Ha per Rayon based on the past 5-year average sown area from 2006 to 2010 (combined sown area for production enterprises and commercial producers). The size of spring wheat Insured Unit varies hugely across the 6 Oblasts and the 73 (84% of total) qualifying rayons with > 10,000 Ha of spring wheat. Overall, EKO has the smallest size of Insured Unit (IU) with an average area per rayon of spring wheat of slightly less 25,000 Ha, while NKO has the largest IUs with an average of about 235,000 Ha/IU (Table 5.1.). Table 5.1. Average Area of Spring Wheat per Rayon (Insured Unit) Average Area Minimum crop Maximum Rayons per No Rayons % of total Oblast per Rayon area per crop area per Oblast >10,000 Ha Rayons (Ha) Rayon (Ha) Rayon (Ha) Commercial Farmers Akmola 19 45,846 675 151,900 17 89% EKO 17 13,853 140 47,080 9 53% Karaganda 9 40,224 3,400 103,680 7 78% Kostanay 18 69,723 8,620 167,900 17 94% NKO 13 56,522 26,780 107,780 13 100% Pavlodar 11 21,664 520 108,560 7 64% Sub-Total 87 41,305 140 167,900 70 80% Production Enterprises Akmola 19 150,300 520 317,600 17 89% EKO 17 10,856 160 27,780 7 41% Karaganda 9 31,364 80 130,220 4 44% Kostanay 18 142,661 900 369,120 16 89% NKO 13 178,658 86,480 430,400 13 100% Pavlodar 11 16,850 520 67,500 5 45% Sub-Total 87 96,533 80 430,400 62 71% Overall (Commercial Farmers + Production Enterprises) Akmola 19 196,146 1,195 408,240 17 89% EKO 17 24,709 300 74,860 12 71% Karaganda 9 71,589 3,480 190,760 8 89% Kostanay 18 212,384 9,520 537,020 17 94% NKO 13 235,180 116,160 517,280 13 100% Pavlodar 11 38,513 1,300 176,060 6 55% Source: Authors Insured Yield Coverage levels 5.41. AYII policies typically offer optional Insured Yield Coverage levels of between a maximum of 90% and a minimum of 50% of the average area-yield. In India, the NAIS has traditionally offered 3 coverage levels, 60%, 80% or a maximum of 90% of the past 5-year average yield in the IU: the decision 48 Sees, J.R., R Black and B.J.Barnett (1997). ―Designing and Rating and Area Yield Crop Insurance Contract. American Journal of Agricultural Economics 79 (May 1997). - 147 - over which coverage level will apply in an IU is based on the coefficient of variation around mean yield such that in IU‘s with low yield COV‘s the maximum 90% coverage level will be applied and in IUs with a high COV only 60% coverage is offered. Under the US Group Risk Plan, farmers may select from optional coverage levels of between 50% and 90% of the county average yield. In Kazakhstan it is proposed that the insured yield for the operation of an AYII program is set as a percentage of the most recent 5-year actual average yields from 2006 to 2010. On account of the very high yield variability in some Oblasts and Rayons it will, however, be necessary to offer coverage levels as low as 20% or 80% of the Rayon 5-year actual average yield in some Oblasts and Rayons. For this reason, coverage levels and rates have been calculated for a wider range of coverage levels from 10% through to 80% of the Rayon 5- year average yield (termed the expected yield). The Rayon spring-wheat insured yield coverage levels from a minimum of 10% up to a maximum of 80% of the past 5-year average yield are shown by Oblast and by Rayon in Annex 7. Insured Values and Sum Insured 5.42. Under an AYII policy, the insured crop yields can be valued either on a costs of production basis or on a farm-gate sale price basis. In India the NAIS commonly sets the sum insured according to the amount of credit provided to the farmers. In the USA, the GRP permits farmers to insure their selected coverage level yield at up to 150% of the sales‘ reference price. In Kazakhstan the sum insured could be based on any valuation criteria requested by farmers from a costs of production valuation through to an expected revenue based valuation: however, unlike the USA it is recommended that the maximum unit sum insured value should not exceed 100% of the expected farm-gate sales price for the crop. For the purposes of this AYII prototype policy design exercise, and to maintain consistency with the MPCI policy outlines in the previous section, the past three year average September sale price of spring wheat of KZT 3,210/Centner has been used. 5.43. The estimated sums insured for a Spring wheat AYII program assuming 100% insurance uptake in the 6 Oblasts and an average 50% coverage level would be in the order of about KZT 212 billion (US$ 1.41 billion). The estimated sums insured for the spring wheat AYII program by Oblast and in Total for coverage levels of between 10% and a maximum of 80% are shown in Table 5.2. Under the assumption of 100% uptake of the AYII product for spring wheat, at the 10% coverage level the TSI would be about KZT 42.4 billion (US$ 283 million), rising to a very significant KZT 339 billion (US$ 226 billion) at the maximum 80% coverage level. It is important to recognize, however, that under a voluntary AYII program the actual take-up rates would be much lower and the TSI would correspondingly be much lower than the 100% modeled figures. Rating Methodology and Indicative Premium Rates 5.44. A preliminary estimation of the technical and commercial premium rates for an AYII program for spring wheat is presented in this report, using internationally accepted AYII crop-rating methodology. Annex 7 presents full details of the rating methodology used in this report for establishing the Rayon-level technical rates for an Area-Yield Index policy for spring wheat based on an analysis of variance in the National Statistical Agency‘s 17-year Rayon-level annual sown area, production and yield data and for coverage levels from a minimum of 10% up to a maximum of 80% of the Rayon 5 year average yields (2006-2010). The procedure adopted involves de-trending the 17-year spring wheat actual yields in each rayon and simulation of the estimated yield shortfall at each insured yield coverage level from 10% to 80% over 5000 iterations (years) to derive the average pure loss cost rates for each Rayon. The pure loss cost rates have then been smoothed and a ―security load‖ added based on the calculated PML‘s to derive the technical rates for each rayon and each coverage level. Finally for the purposes of this rating exercise, the technical rates have been grossed-up by 30% to derive indicative commercial premium rates for a target 70% loss ratio. The 70% target loss ratio is designed to allow the participating - 148 - insurers to cover their acquisition costs and administration and operating expenses and to provide a reasonable profit margin: in practice the insurers and their reinsurers will be responsible for estimating their costs and profit margins and in setting their target loss ratios accordingly. Since the costs of loss assessment for AYII are considerably lower than for individual grower MPCI, the gross-up used in this rating exercise is correspondingly lower at 30% for AYII compared to 40% for MPCI (See Section 4). 5.45. Indicative Commercial Premium Rates for AYII cover are presented in this report, but it is stressed that these rates are illustrative and that final rating decisions will be taken by insurers and their reinsurers. The AYII commercial premium rates for spring wheat for 70% target loss ratio are presented in Annex 8 by Rayon for each insured yield coverage level and a summary of the average Rayon rates in each of the 6 Oblasts is given in Table 5.2. The average percentage premium rates and corresponding value of the premiums are also shown in Table 5.2. The analysis shows that the average rates for spring wheat vary from a low in NKO to a high in Pavlodar and EKO and that in most Oblasts, premium rates start to become very expensive for coverage levels of greater than 50% of the Rayon 5- year average yield. Assuming 100% AYII scheme uptake the corresponding commercial premium for an average 50% coverage level would be about KZT 14 billion (US$ 95 million). 5.46. In order to interpret these rates it is useful to consider a maximum premium rate that farmers might be willing to pay assuming there are no premium subsidies of about 10%. The analysis clearly shows in NKO where crop yields are generally very stable that AYII insurance could be offered for high coverage levels of up to 70% at affordable premium rates of less than 10%. In Akmola and Kostanay, farmers could be offered coverage levels up to about 60% of average Rayon yields under the 10% maximum premium rate assumption. However, in Pavlodar where spring wheat yields are more variable a maximum of 50% coverage only could be provided and in EKO as low as 40% coverage to avoid exceeding the 10% commercial premium rate. Reference to Annex 7 shows exactly which maximum cover levels could be offered for less than 10% commercial premium in each Rayon in each Oblast Table 5.2. Estimated Total Sum Insured and indicative Commercial Premium for Spring Wheat AYII Program and Coverage Levels 10% to 80% of Average Rayon Yield Total Sum Insured (KZT Million) Insured Yield Coverage level (% of Rayon average expected yield) Oblast 10% 20% 30% 40% 50% 60% 70% 80% 10,985 21,970 32,956 43,941 54,926 65,911 76,897 87,882 Akmola 1,181 2,361 3,542 4,723 5,904 7,084 8,265 9,446 EKO 1,346 2,691 4,037 5,383 6,729 8,074 9,420 10,766 Karaganda 14,786 29,573 44,359 59,145 73,931 88,718 103,504 118,290 Kostanay 13,216 26,432 39,648 52,864 66,080 79,296 92,512 105,728 NKO 910 1,821 2,731 3,641 4,552 5,462 6,372 7,283 Pavlodar 42,424 84,849 127,273 169,697 212,121 254,546 296,970 339,394 Total Indicative Commercial Premium Rates (for target 70% loss ratio) Insured Yield Coverage level (% of Rayon average expected yield) Oblast 10% 20% 30% 40% 50% 60% 70% 80% 0.24% 1.03% 2.43% 4.52% 7.07% 10.34% 14.26% 18.69% Akmola 0.74% 2.25% 4.50% 7.41% 10.72% 14.64% 19.03% 23.67% EKO 0.33% 1.31% 3.05% 5.58% 8.62% 12.28% 16.48% 21.09% Karaganda - 149 - 0.24% 1.01% 2.38% 4.42% 6.93% 10.17% 14.04% 18.44% Kostanay 0.17% 0.73% 1.70% 3.14% 4.85% 7.17% 10.15% 13.80% NKO 0.38% 1.50% 3.36% 5.96% 8.97% 12.62% 16.75% 21.14% Pavlodar 0.24% 0.98% 2.28% 4.20% 6.52% 9.52% 13.16% 17.35% Total Indicative Commercial Premium (70% loss ratio) - KZT (Million) Insured Yield Coverage level (% of Rayon average expected yield) Oblast 10% 20% 30% 40% 50% 60% 70% 80% 26 226 799 1,987 3,885 6,813 10,968 16,429 Akmola 9 53 159 350 633 1,037 1,573 2,236 EKO 4 35 123 301 580 991 1,552 2,271 Karaganda 36 299 1,055 2,615 5,124 9,021 14,537 21,817 Kostanay 23 192 675 1,662 3,206 5,682 9,390 14,594 NKO 3 27 92 217 408 689 1,068 1,539 Pavlodar 101 833 2,903 7,131 13,837 24,234 39,087 58,885 Total Source: Authors. See Annex 7 for full details Coverage levels, Affordable Premium Rates and Demand for Voluntary AYII Insurance 5.47. Under a spring wheat AYII program for Kazakhstan the coverage level in each Rayon should be set in accordance with i) the underlying risk exposure and frequency, and ii) the commercial premium rate that can be afforded by the targeted farmers. In order for a crop insurance scheme to be both affordable to farmers (premium rates of no more than 5% to 10%) and sustainable, the Insured Yield Coverage level should be set at a level where payouts are no more frequent than about 1 in every 7 to 1 in 10 years. Where AYII coverage levels are set too high and therefore the commercial premium rates are also too high this discourages farmers from purchasing crop insurance and does not permit the scheme to achieve over time the economies of scale and premium volume which is necessary for the scheme to be sustainable. On the basis of the feedback from farmers met in Kazakhstan during the focus group meetings it appears that most consider the current premium rates charged on the obligatory LIC scheme (average of about 2.4% subsidized rate and average of about 4.4% for the full rate assuming no government 50% claims subsidies) as being too expensive. It was not possible in the limited time available in these panel meetings to conduct any formal demand assessment for AYII insurance. 5.48. If a voluntary AYII program is to be launched in Kazakhstan, scheme management will first need to conduct a detailed demand study in order to quantify farmer’s interest in purchasing voluntary AYII and the coverage levels and premiums they are willing to pay. This demand assessment study should be designed to provide a clear picture of farmers‘ ability and willingness to pay for voluntary AYII insurance and the coverage levels and sum insured values they wish to insure. This study may also provide useful feedback on the need for government premium subsidy support. Government Support to Premium Subsidies 5.49. If government were to elect to provide premium subsidy support this would enable farmers to either chose to insure at higher coverage levels and or to reduce their costs of purchasing AYII cover . On the assumption that government were to provide 50% premium subsidies, this would make the AYII premium rates much more affordable to farmers as shown in Table 5.3. The recommended AYII demand study would also enable estimates to be made of the TSI for a pilot AYII scheme, the estimated premium - 150 - income and therefore the costs to government of a 50% premium subsidy (or other level of premium subsidy to be determined). Table 5.3. Average Indicative Commercial Premium Rates for Spring Wheat AYII Cover net of Government 50% Premium Subsidies. Oblast Coverage level 10% 20% 30% 40% 50% 60% 70% 80% Akmola 0.12% 0.51% 1.21% 2.26% 3.54% 5.17% 7.13% 9.35% EKO 0.37% 1.12% 2.25% 3.71% 5.36% 7.32% 9.51% 11.83% Karaganda 0.16% 0.65% 1.53% 2.79% 4.31% 6.14% 8.24% 10.55% Kostanay 0.12% 0.51% 1.19% 2.21% 3.47% 5.08% 7.02% 9.22% NKO 0.09% 0.36% 0.85% 1.57% 2.43% 3.58% 5.07% 6.90% Pavlodar 0.19% 0.75% 1.68% 2.98% 4.48% 6.31% 8.38% 10.57% Total 0.12% 0.49% 1.14% 2.10% 3.26% 4.76% 6.58% 8.68% Source: Authors. Operational Considerations 5.50. For the operation of a spring wheat AYII cover, the key requirement would be to formalize the procedures for estimating actual average yield in each Rayon where insurance is offered. It is recommended that the insurance companies enter into a formal agreement with the National Statistical Agency of Kazakhstan to provide the results of their crop-cutting yield estimates for each Rayon. It is essential to minimize the time delay post harvest for the publication of these Rayon-level area yield estimates in order to settle claims to those farmers in those Rayons where the actual average yield falls short of the Insured Yield. A common criticism of the AYII program in India is that it often takes between 6 and 12 months for the state-level governments to publish the results of their annual crop- cutting experiments: farmers in the meantime need to purchase seeds and fertilizers for the next season and to pay for land preparation and sowing activities etc and they cannot wait 12 months to receive their indemnities. In Kazakhstan it is understood that the Statistics Agency should be able to make the results of their crop-cutting surveys prior to the start of the next spring wheat season. It is also likely that under an AYII scheme, insurers and their reinsurers would wish to put in place independent monitoring systems to verify the actual average Rayon crop yields in the insured rayons. 5.51. It will also be necessary to provide farmer education and training in the operation of AYII. Given the fact that AYII does not provide insurance and indemnity at the individual farmer level, but according to losses at the Rayon level, it is essential that the principles of this cover are clearly explained to farmers under a program of farmer-education and training programs. 5.52. Distribution Channels should also be investigated under any Pilot AYII program in future. Currently in northern Kazakhstan it is understood that most crop insurance is marketed through sales agents located in each Oblast and Rayon. Alternative channels for marketing and administering AYII crop insurance should be invested including through farmers‘ associations and cooperatives, through rural banks and input suppliers. The potential for crop-credit linked insurance merits investigation. 5.53. There may also be opportunities to market AYII not only to individual farmers and joint stock companies and cooperatives, but also as a Meso-level financial protection cover for banks and input suppliers operating in each Rayon. In this instance the cover would be designed to provide business interruption cover to the lending institution in the event that a catastrophe drought in a named Rayon(s) - 151 - prevents the farmers from repaying their loans therefore forcing the lender to reschedule or write-off its loans in that Rayon(s). In addition such a cover might be attractive to regional input suppliers who provide seeds, fertilizers and plant protection chemicals to farmers on credit against repayment at the time of the wheat harvest. AYII Estimates of Probable Maximum Loss and Implications for Reinsurance 5.54. Some preliminary estimates of the Probable Maximum Losses which might be expected under an Area Yield Index program for spring wheat in Kazakhstan are presented in Figure 5.5 and Table 5.4 below. For a 50% coverage level the 1-in-a-100 year PML would be about PML would be about 24% of TSI or a loss of KZT 51 billion (US$ 343 million) assuming 100% scheme uptake. This would be equivalent to a loss ratio of about 370%. For the highest 80% coverage level the 1-in-a-100 year PML would be about 45% of TSI or a loss of KZT 152.7 billion (US$ 1.02 billion) assuming 100% scheme uptake. This would be equivalent to a loss ratio of about 259%. Figure 5.5. Spring Wheat AYII scheme PML estimates by coverage level from 10% to 80% 60.00% 50.00% "10% Coverage" 40.00% "20% Coverage" Loss Cost % "30% Coverage" 30.00% "40% Coverage" 20.00% "50% Coverage" "60% Coverage" 10.00% "70% Coverage" 0.00% "80% Coverage" 1 50 100 150 200 250 Return Period (years) Table 5.4. Estimated 1 in 100 Year PML for AYII Wheat Scheme (100% figures) Coverage level Item 10% 20% 30% 40% 50% 60% 70% 80% PML (% of TSI) 1.11% 4.53% 10.04% 17.40% 24.29% 31.84% 39.15% 44.99% PML (KZT Billion ) 0.5 3.8 12.8 29.5 51.5 81.0 116.3 152.7 PML (US$ Million) 3 25 85 197 343 540 775 1,018 1 in 100 year PML 466% 461% 440% 414% 372% 334% 297% 259% Loss Ratio % Source: Authors 5.55. Some indicative reinsurance pricing has been conducted for the spring wheat AYII Program. This analysis has been conducted assuming i) aggregate reinsurance protection over the entire spring wheat insured area assuming 100% uptake, ii) three priority levels of 70%, 100% and 150% of GNPI, iii) that reinsurance would provide full-value protection up to 100% of TSI excess of these priority levels, and iv) insured yield coverage levels of 10% up to 80% maximum of the Rayon 5-year average yield. Under these assumptions Table 5.5. shows for the 100% of GNPI stop loss reinsurance option and full - 152 - value protection and 50% coverage level, that the stop loss pricing would be in the order of 26.09% of GNPI or KZT 3.6 billion (US$ 24.0 million). The 100% of GNPI stop loss reinsurance option and full value protection for the highest 80% coverage level would cost approximately 18.01% of GNPI or KZT 10.06 billion (US$ 67 million). It is worth to note that if the stop loss protection priority were lowered to 70% of GNPI the aggregate stop loss pricing would be considerably higher. 5.56. A layered risk financing program could also be developed for the AYII program involving both private insurers and reinsurers and government of Kazakhstan as a catastrophe reinsurer. Such a risk financing program could be structured along exactly the same lines as the stop loss reinsurance cover illustrated in Section 4 for individual grower MPCI. Table 5.5. Indicative Pricing for Aggregate Stop Loss Reinsurance for Kazakhstan Spring Wheat AYII Program (Full value basis) Insured Yield Coverage Level (% of Rayon average expected yield) Item 10% 20% 30% 40% 50% 60% 70% 80% Stop Loss Reinsurance Premium (KZT Million) Priority (% of GNPI) 70% 54 349 1,046 2,295 4,297 7,112 10,733 15,157 100% 52 326 948 2,005 3,609 5,693 8,084 10,606 150% 49 295 820 1,634 2,764 4,008 5,078 5,731 Stop Loss Reinsurance Premium (%GNPI49) Priority (% of GNPI) 70% 53.90% 41.86% 36.04% 32.18% 31.05% 29.35% 27.46% 25.74% 100% 51.43% 39.13% 32.66% 28.12% 26.09% 23.49% 20.68% 18.01% 150% 48.19% 35.48% 28.25% 22.91% 19.98% 16.54% 12.99% 9.73% Source: Authors Voluntary AYII Crop Insurance Portfolio Financial Estimates 5.57. Some provisional portfolio financial estimates have been calculated for AYII cover for spring wheat assuming a voluntary program and 5% incremental uptake rates per year over the next 5 years. Under the assumptions of a 50% Insured Yield coverage level and 5% uptake rate of AYII insurance per year over the next 5 years, the total sum insured might rise from KZT 10.6 billion in year 1 with corresponding premium income of KZT 692 million rising after 5 years to KZT 53 billion with premium income of KZT 3.5 billion. The costs to government of different levels of premium subsidies (ranging from 25% to 65% of premium) are also shown (Table 5.6). It is recognized that these uptake estimates are extremely ambitious for a voluntary insurance scheme and will need refinement following the recommended AYII demand study. Table 5.6. 5-Year Estimates of Voluntary AYII Uptake and Total Sum Insured and Premium Income and Costs of Premium Subsidies (50% Coverage level) AYII 50% Item Coverage AYII Crop Insurance Uptake Scenarios next 5 Years (KZT Million) Level 100% Basis 5% 10% 15% 20% 25% Total Sum Insured 212,121 10,606 21,212 31,818 42,424 53,030 49 GNPI: Gross Net Premium Income - 153 - Commercial Premium 13,837 692 1384 2076 2767 3459 Cost of GRK Premium Subsidies 25% of Premium 3,459 173 346 519 692 865 25% of Premium 6,918 346 692 1,038 1,384 1,730 65% of Premium 8,994 450 899 1,349 1,799 2,248 Probable Maximum Loss PML 1 in 100 Years 51,516 2,576 5,152 7,727 10,303 12,879 PML 1 in250 Years 70,634 3,532 7,063 10,595 14,127 17,659 Source: Authors Conclusions on AYII for Spring Wheat in Kazakhstan 5.58. AYII for spring wheat is technically and operationally feasible in Kazakhstan, but until further research has been conducted into the potential demand for this cover it is very difficult to predict likely uptake rates under a voluntary crop insurance program. 5.59. AYII for spring wheat could be underwritten either as a micro-level individual farmer cover or as a meso level product designed to protect the season loan portfolio of agencies which are lending to cereal producers (banks, input suppliers or MFIs) in Kazakhstan. There would be two advantages in offering area-yield index-insurance at a meso-level or aggregate product. The first advantage would be that basis risk would be much less of a concern than under an individual grower program. The second advantage would be that the transaction costs involved in this coverage would be lower than for individual farmer micro-level insurance. 5.60. Meso-level AYII cover may be an effective tool for governments to operate for small family farm in southern Kazakhstan. This option is reviewed further in Chapter 6. 5.61. Farmers’ demand for and willingness to pay for AYII crop insurance will also have to be studied further before any decisions are made to proceed with the design of an AYII program. This feasibility study has identified a very low level of interest in the obligatory LIC crop insurance scheme by farmers and it is probable that this would apply equally to voluntary crop insurance in future. It is recommended that a formal AYII crop-insurance demand assessment study should be implemented by the key stakeholders in Kazakhstan. India, the USA and Ukraine may provide useful lessons and experience for the design of AYII in Kazakhstan. Crop Weather Index Insurance 5.62. The analysis carried out in this feasibility study shows that developing WII contracts for hedging the drought exposure of spring wheat in the North of Kazakhstan is technically feasible. However, challenges in the possible scale of implementation, in the commercial viability and in farmers’ interest for WII may limit the scope of application of this class of insurance products. The details of the analysis are presented in the following sections but it is stressed that further research will be required if the insurance companies decide to proceed with the pilot testing and implementation of this product. It is assumed that WII would be introduced as a voluntary commercial crop insurance program. Further details of the WII product are contained in Annex 8. - 154 - 50 Features, Advantages and Disadvantages of WII 5.63. The essential feature of weather index-based insurance (WII) is that the insurance contract responds to an objective parameter (e.g., measurement of rainfall or temperature) at a defined weather station during an agreed-upon time period. The parameters of the contract are set so as to correlate, as accurately as possible, with the loss of a specific crop type suffered by the policyholder. All policyholders within a defined area receive payouts based on the same contract and measurement at the same station, eliminating the need for field loss assessment. 5.64. WII is best suited to weather hazards that are well-correlated over a widespread area and where there is a close correlation between weather and crop yield. The strongest relationships typically involve a single crop, a marked rainy season, and no irrigation. 5.65. WII is less useful where more complex conditions exist. Localized risks, such as hail, or where microclimates exist (for example, in mountainous areas) are not suitable for WII. Similarly, the scope for WII is limited where crop production is impacted by many or complex causes of loss or where pest and disease are major influences on yields. For a given environment, other insurance products may be more appropriate (such as area-yield index insurance or named-peril crop insurance). 5.66. The above mentioned conditions apply to both the climate environment and spring wheat production of the North of Kazakhstan, making the feasibility analysis of the case in object worth pursuing. 5.67. As for AYII, Basis Risk is the key constraint of WII. ―Basis‖ can be defined as the difference between the loss experienced by the farmer and the payout triggered by the weather index. It could result in a farmer experiencing a yield loss, but not receiving a payout or also in a payout being triggered without any loss being experienced. WII works best where losses are homogenous in the defined area, and highly correlated with the weather peril.51 5.68. WII can be retailed at different business levels. At the micro level, the policyholders (the insurer‘s customers) are individual farmers, households, or small business owners who purchase insurance to protect themselves from potential losses caused by adverse weather events. At the meso level, WII can be used to cover the exposure of entities such as financial service providers, farmer associations, input suppliers and processors from potential losses caused by adverse weather events. At the macro level, WII can aid governments and relief agencies in development and disaster management. International Experience with WII 5.69. The majority of WII experience has been with micro-level applications and rainfall deficit (drought). To date, many initiatives have been piloted, but only in India has a market-based scale-up of WII taken place. Table 5.7 provides a synoptic summary of the countries in which WII has been piloted. 50 The illustrative material presented from Sections 5.61 to 5.71 has been adapted from Weather Index-based Insurance in Agricultural Development: A Technical Guide , IFAD, Rome, Forthcoming. 51 A more detailed description of the strengths and weaknesses of WII is provided in Annex 8. - 155 - Table 5.7 International Experience with Weather Index Insurance at Different Levels of Aggregation Micro level Weather-indexed insurance for smallholder farmers: Examples: India, Nicaragua, Malawi, Ukraine, Thailand, Ethiopia, Kenya, Ghana, The Philippines, China. Over 30 projects in about 25 countries. Scale-up only in India Meso level Weather-indexed portfolio hedge for rural financial institutions that lend to poor farmers Examples: Peru, Ghana, Vietnam (under development) Programs are too new to assess scale-up and sustainability Macro level Weather insurance or weather-indexed contingent credit line for governments or international organizations Examples: Ethiopia, Malawi, Mexico (both AYII and WII), Caribbean States (CCRIF) risk pool for hurricanes & earthquake) Mexico has achieved major scale-up across most states in the past decade. CCRIF is insuring 16 Caribbean states Source: Dick, W. (2009) Weather and Agricultural Data for WII 5.70. WII relies on historical and current weather data that should adhere to specific quality requirements. In order to meet requirements for a commercial WII insurance and reinsurance transaction, it is recommended as a guideline that there be at least 20 years worth of historical daily data and that missing data should not exceed 3 per cent of the total daily data set. In addition, reliable and trustworthy on-going daily collection and reporting procedures should be assured. 5.71. Beyond the quality of data, it is critical to define the boundaries of the area(s) covered by the weather station(s) so that WII contracts can be written for specific areas tied to a specific station. A general rule of thumb is to consider a specific WII contract marketable within a 20 km radius of the weather station; but the applicable area may be smaller or larger and case-specific evaluation must be carried out. In general terms, the more the terrain varies, the more the acceptable distance from a station decreases. 5.72. Agricultural data and information is the second part of the WII contract design equation. The most relevant information to be collected is data on productivity (yield), but a clear description of the agricultural production practices carried out in the areas of interest is also necessary. Unfortunately, the availability of quality yield data series of adequate length and at the appropriate level of disaggregation is not common. However, lack of quality yield data does not pose as large a problem as lack of good weather data, since it is still possible to find alternative approaches to estimating yield variability. One possibility is to simulate synthetic yield data series through plant-growth models. WII Contract Design 5.73. The objective of contract design is to define a structure that effectively captures the relationship between the weather variable and the potential crop loss and to select the index that is most effective in providing payouts when losses are experienced, reducing basis risk as much as possible. The set of possible index combinations is unlimited, and numerous structures have been developed in the relatively short history of WII. One of the most commonly adopted structures is that of a - 156 - continuous payout triggered and limited by a cumulative measure of the weather variable (e.g., rainfall) for each of the different crop growth stages. (See Box 5.4 and Figure 5.6). Box 5.4: Payout Parameters in a WII Contract Using the drought coverage case represented in Figure 5.5 as an example, the parameters that characterize an incremental payout structure can be defined as follows:  Trigger: Threshold above or below which payouts are due. Payments are due when the calculated value of the index is below the trigger level (300mm).  Exit: Threshold above or below which no additional incremental payout will be applied. The maximum payout is paid if the calculated value of the index is equal to or below the exit threshold (100mm).  Tick: Incremental payout value per unit deviation increase from the trigger. With a maximum payout (the insured sum) of $200, a trigger of 300mm, and an exit of 100 mm, the monetary value of each deficit mm of rainfall below the trigger is: $200 /(300 mm-100 mm) or $1 per mm. Figure 5.6: Payout Structure of a WII Drought Contract 250 Exit: 100mm 200 Payout ($) 150 100 Trigger: 300mm 50 0 1 51 101 151 201 251 301 351 401 Source: Authors Weather data and infrastructure in Kazakhstan 5.74. As highlighted in Chapter 2, the World Bank feasibility study team has carried out a detailed review of the meteorological network managed by KHM. The weather data provided by KHM has enabled an in-depth assessment to be conducted of weather risk in the selected spring wheat production areas and to develop WII contract structures. 5.75. The continuity of historical data provided by KHM is generally of good quality. For the purpose of this feasibility study the World Bank has gained access to daily rainfall and minimum and maximum temperatures from ten weather stations located in seven Rayons listed in Table 5.8 and portrayed in Map 5.1. Table 5.8 shows that the rate of missing data is generally very low (less than 1%). - 157 - In particular, the amount of missing rainfall observation is in line with the record of missing temperature, and this seems an indication of an overall good management of weather sensors, as events leading to missing observations may be caused by general problems at weather station level. Table 5.8. Selected Weather Stations and Statistics on Available Weather Data % of Missing % of Missing Daily Station Rayon Oblast No. of Years Daily Rainfall Temperature Observations Observations Diyevskaya Auliyekolski Kostanay 26 0% 0% Kushmurun Auliyekolski Kostanay 26 0.6% 0.6% Kostanay Kostanayski Kostanay 26 0% 0% Zholboldy Aktogaiski Pavlodar 26 1.2% 1.2% Aktogai Aktogaiski Pavlodar 26 0.2% 0.2% Mikhailovka Zhelezinski Pavlodar 26 0% 0% Bolkashino Sanytausky Akmola 26 0% 0% Schuchinskoye Enbekshildersky Akmola 26 0% 0% Stepnogorsk Enbekshildersky Akmola 26 0.3% 0.3% Tolebiysky Tasarik Sko 29 0% 0% Source: Authors on KHM data Map 5.1. Weather stations analyzed in the Feasibility Study. Source: Authors‘ elaborations on Google Earth Note: Weather stations are identified by the red markers. Rayons selected for the analysis of the feasibility study are highlighted in light grey shading (Rayon names in yellow). 5.76. Over most of the country, the seasonal precipitation cycle is of mid-range rainfall intensity with a rather uniformly distributed cycle throughout the year. Therefore, rather than marked seasonal distribution and erratic patterns as in many other regions of the world, average humidity conditions and total seasonal cumulated precipitation levels seem to be the main concerns for the farming environment. Given this hydrological pre-condition, the thermal regime also plays a relevant role. Combined hydro- - 158 - thermal indices are of common use in agrometeorological monitoring in Kazakhstan and have been adopted in this study for the design of WII prototypes52. 5.77. An analysis of the last twenty-five years’ data shows that no significant rainfall trend is observed while a weak but statistically significant temperature trend has been identified53. An important step in the design of weather index insurance programs is to check for the existence of long term tendencies in key atmospheric parameters that would affect the price and the effectiveness of the contracts. A detailed trend analysis for ten weather stations in the target rayons was carried out . The results of this analysis are presented in Appendix C of Annex 8 and shows that no significant trends exist in annual rainfall. Instead, average annual temperature is slightly increasing in almost all of the analyzed weather stations. Since all indices adopted for the contract design depend mainly on cumulated rainfall and temperature trends do not seem to be affecting the developed structures, temperature de-trending was not carried out. Nevertheless, careful handling of issues related to weather trends would be recommended for the design of contracts for commercial use. 5.78. Drought exhibits a systemic pattern across the North of Kazakhstan. The detailed correlation analysis carried out at station level indicates that, in the test areas selected, weather patterns may be comparable at low scale (Rayon level) and become more heterogeneous at Oblast level (see Table 5.9). As expected, higher levels of correlations are noted for weather stations that are closer in distance and, roughly, in the same Rayon. In general, intra Oblast variability of weather patterns seems to be more pronounced. Table 5.9. Correlation Matrix for Cumulated Precipitation (January to August) in Selected Weather Stations Kostanai Pavlodar Akmolinsk Schuchinskoye Mikhailovka Stepnogorsk Baklkashino Kushmurun Diyevskaya Zholboldy Kostanay Aktogai Diyevskaya - 72% 78% Zholboldy - 54% 27% Baklkashino - 78% 59% Kushmurun 66% - 66% Aktogai 49% - 70% Schuchinskoye 61% - 59% Kostanay 60% 59% - Mikhailovka 43% 60% - Stepnogorsk 46% 51% - Source: Authors‘ analysis on KHM data. NOTE: In each Oblast section of the table the ―shaded values‖ (top-right corner) correspond to correlations between cumulated seasonal precipitation, while ―non-shaded values‖ (bottom-left corner) correspond to correlations between deviations of 10-day cumulated rainfall with respect to the average seasonal cycle. Subtracting the average seasonal cycle from the cumulated time series helps to more accurately identify the degree of homogeneity among weather patterns in an area. Similar seasonal cycles, typical of any region, may lead to overestimating weather patterns similarities. 52 See following sections and also Annex 8, Appendix C. 53 The analytical results of the trend analysis are presented in Annex 8, Appendix D - 159 - 5.79. While from a climatological point of view correlations between weather data collected at Rayon level distance are very high, when referring to the potential coverage of the territory for WII purposes different evaluation criteria must be applied. Even though Kazakhstan‘s climatic conditions seem appropriate for relaxing the 20-25 km radius rule as the criterion for identifying the scale at which WII can be retailed, the average distance between weather stations seems too high for assuming that the current network could provide a comprehensive coverage of the North Kazakhstan‘s territory. As illustrated in the maps presented in Appendix A of Annex 8, distances between contiguous weather stations start from a minimum of 70 km, which is probably too large for implementing a full micro WII coverage of the entire territory. 5.80. In conclusion, the excellent quality of weather data provided by KHM seems to fully comply with the requirements that must be met for designing and operating WII. However, while this applies for the areas surrounding the specific weather stations analyzed, an actual full scale implementation of a micro level (farm level) WII program may be hindered by the relatively low density of the weather network. While in the medium term, through the improvement of the weather network density, it may be possible to overcome these structural constraints, in the short term a widespread full scale implementation of WII at farm level does not seem to be realistic. Potential implementation of WII contracts at meso or macro levels, that may less influenced by the density of the weather network, could have greater chances of being rapidly implemented. Agricultural Data 5.81. As discussed in Chapter 2, the ARKS collects annual average spring wheat yield data at Rayon level. Given the excellent quality of the historical yield reports it was possible to use the actual official ARKS yield data for developing WII prototypes for the selected case studies.54 The Rayon annual average yields series that were provided span the period from 1990 (or 1991 in some cases) up to 2010. 5.82. Rayon yield reports provide average yield series for two categories of farm typology: Production Enterprises and Commercial farms. While in the case of AYII the analysis was carried out for the two separate classes of farms, in the case of WII, in order to identify one specific relationship between weather and yield, the two types of records were aggregated in a single Rayon yield series. 5.83. Drastic reductions in yield performance were recorded during the early years of Kazakhstan independence. At the end of the 1990‘s, more intensive production technology became again available and production levels improved significantly. In order to avoid overestimating the impact of weather elements on yield losses in the low production performance periods, data from the 1990 (or 1991) to 1998 was recentered at the mean level of the 1999-2010 period.55 5.84. As discussed in preceding sections, in order to capture the impact of weather variability on spring wheat production in the six selected Rayons of the North Kazakhstan region, KHM provided data from nine weather stations that were considered representative of the selected production environments. Since not all of the selected spring wheat producing Rayons host a main weather station56, weather measurement points from neighboring Rayons were adopted. Table 5.10 provides a list of all production Rayons and the weather stations selected to represent their weather patterns. The maps presented in Appendix A of Annex 8 provide a geographical representation of such information. 54 In WII this is not a common feature as many times the developing countries in which these kind of insurance products are tested do not have sufficiently good agricultural statistics. In such cases, information collected in the field and/or crop modeling approaches are used to reconstruct the required crop loss history. 55 See Annex 8, Appendix D for more details. 56 See Annex 8, Appendix B for a full list of weather stations available in the North of Kazakhstan. - 160 - Table 5.10: Selected Production Rayons and Representative Weather Stations Oblast Selected Rayon Station Selected to Rayon in Which Represent the Weather Weather Stations Are Pattern of the Selected Actually Located Rayon Bulandinsky WS Bolkashino Sandytausky Akmola WS Schuchinsk Burabaisky Enbekshildersky WS Stepnogorsk Stepnogorsk WS Diyevskaya Auliekolsky Auliekolsky Kostanay WS Kushmurun Auliekolsky Altynsarinsky WS Kostanay Kostanay WS Zholboldy Aktogay Aktogay Pavlodar WS Aktogai Aktogay Zhelezinsky WS Mihailovka Zhelesinsky Source: Authors Contract Design 5.85. The design of WII contract structures for covering the drought exposure of spring wheat production in the North of Kazakhstan has been carried out on the basis of the methodology developed by the Agriculture Risk Management Team (ARMT) of the Agricultural and Rural Development (ARD) department of the World Bank57. Specific additional indexing procedures were developed for the spring wheat environment of North of Kazakhstan. 5.86. In the contract design activities three weather indexes have been adopted: Cumulative Precipitation, Hydrothermal Ratio (HTR), and Humidity Factor (K). 5.87. The “Cumulative Precipitation Index” can be defined as the sum of all recorded precipitation across a specific time period and is one of the most common indexes adopted in the index approach to insuring agricultural drought. In many cereal crops contract structures the crop life is divided in phases (usually three) on the basis of the different water requirements of the different phenological phases. Accordingly, specific indexes are developed for each of the crop growth phases. Given the specific agroclimatic conditions of Kazakhstan, phase breakup did not seem to add value to the contract performance and, therefore, a single cumulative precipitation index was adopted .58 An additional feature of the North of Kazakhstan production environment is that precipitation preceding the sowing period (mainly in the form of snow) may have a relevant role in crop growth59. 5.88. The two other indexes adopted in the design of contract structures for North of Kazakhstan (HTR and K) take into account both precipitation and the contribution of temperature to plant growth. 57 The concepts and operational details of such a methodology are described in the web-based training material available at www.agrisktaining.org. 58 The rationale behind this finding is quite intuitive and is related to the fact that in the North of Kazakhstan the demand for water evaporation from the soil and plant transpiration (together defined as ―evapotranspiration‖) is lower than in many of the semiarid countries where WII is usually implemented. Hence, water can be stored in the soil for longer periods of time making crop growth less dependent on the conjunctural precipitation level. 59 This is again different from the case of semi-arid countries where at the start of the crop life the soil water balance is approximately zero and the onset of the rainfall season allows soil water reserves to be replenished. In Kazakhstan snow melt provides significant water resources that can be stored by the soil. - 161 - They are commonly used in Kazakhstan for identifying drought conditions and have therefore been selected for drought indexing purposes60. 5.89. The two indexes are both structured as a ratio between cumulated daily rainfall and cumulated average daily temperature. The differences in the two indexes are related to the periods across which the weather parameters are cumulated and to the weighting factors applied to the cumulated measures. The exact specification of the HTR and K indexes are provided in Appendix C of Annex 8. 5.90. Table 5.11 summarizes the findings of the contract design activities carried out for developing drought protection products for spring wheat in the North of Kazakhstan. As indicated in preceding sections, daily precipitation and temperature data from nine weather stations was made available for capturing the spring wheat yield variability of six Rayons in three Oblasts of the North of Kazakhstan. Accordingly, nine WII prototypes, one for each of the available Rayon-weather station combinations, were developed. Table 5.11 Performance of Drought Index Contract Prototypes Average Maximum Indicative Farm Location (Oblast- Reference Weather Station Overall Selected Basis Risk Yield Payout Premium Rayon) (Rayon-WS Name) Performance Index Index (centner/he) (KZT/he) Rate Kostanay-Auliekoski Auliekoski-Kushumurum High 9.2 HTR 19,000 13.7% 18% Kostanay-Altynsarin Kostanay-Kostanay High 11.6 Cumul Precip 24,000 16.8% 41% Akmola-Enbelshiderski Stepnogorsk-Stepnogosk High 10.8 K 23,000 15.2% 50% Akmola-Enbelshiderski Schuchinsk-Schuchinsk Medium 10.8 K 23,000 14.9% 86% Pavlodar-Aktogay Aktogay-Aktogay Medium 6.1 Cumul Precip 13,000 20.9% 90% Pavlodar-Aktogay Aktogay-Zholbody Medium 6.1 Cumul Precip 13,000 24.2% 86% Akmola-Bualindiski Sandytauski-Bodkashino Medium-Low 9.1 K 13,000 3.1% 67% Pavlodar-Zhelesinski Zhelenovsky-Mikhailovka Low 8.4 NA Kostanay-Auliekoski Diyevkaya-Diyevkaya Low 9.2 NA Source: Authors Color coding: Modelling succesful and high to medium prototype performance Modelling succesful and medium to low prototype performance Modelling not successful Note: The prototypes have been calibrated in order to trigger payouts for yield observations that fall below 65% of the average Rayon yield level. For this reason Maximum Payout levels vary according to the different Rayon average yields. As for the other simulations carried out in the Study, the price reference used for converting yield into revenue losses is KZT 3,210, corresponding to the average of the reference wheat prices in 2008, 2009 and 2010. 5.91. In terms of assessing the technical feasibility of WII for spring wheat in KZ, the selected case studies provide interesting indications. The first main finding of the contract design activity is that in seven of the nine cases it was possible to model a drought-yield relationship. Out of these seven cases, 60 The analysis of hydrothermal indexes draws extensively on the detailed study of S. Baisholanov (2011), An agro- meteorological risk assessment for cereal crop production in Kazakhstan (prepared by Saken Baisholanov, former Head of Agro Meteorology Department, Kazhydromet). - 162 - three structures show an excellent performance, three other structures are less performing but can still be considered acceptable, and one is not acceptable. In two of the nine cases it was not possible to define a meaningful drought-yield index. 5.92. The fact that for the majority of the Rayon yield-weather station combinations it was possible to develop meaningful index structures indicates that covering spring wheat’s drought exposure in the North of Kazakhstan through WII may be technically feasible. 5.93. Although it would be necessary to carry out specific ground work to verify the dynamics of each specific yield-drought index relationship, the diverse performances recorded in the various cases reflect the fact that the different weather stations may be more or less representative of the average meteorological conditions of the wheat production areas. Should the actual location of the weather measurement point with respect to the selected production area be the actual reason for low contract performance, specific corrective actions can be undertaken. 5.94. In particular, developing synthetic rainfall and temperature data sets through a gridding analysis may provide the necessary additional data reference points for carrying out the risk assessment and contract rating procedures.61 In combination with the gridding analysis, targeted network upgrading activities, like the installation of new automatic weather measurement equipment, could then provide the means for settling WII contracts. 5.95. In order to provide more specific indications on the assessment of the WII contract structure developed, two of the nine cases are presented in Figures 5.6 and 5.7.62 The two contract structures presented can be identified as the two extremes in the range of the technically acceptable structures. All contract structures have been developed and calibrated for a coverage level of 65% of the average Rayon yield level.63 5.96. The contract presented in Figure 5.7 refers to the drought index structure developed for the Altinsarin Rayon on the basis of data collected at Kostanay Weather Station. The graph shows that the potential payouts that the index would have provided in the last twenty-five years match very closely with the average spring wheat revenue losses suffered in the Altinsarin Rayon. 5.97. The contract presented in Figure 5.8 refers to the case of Aktogayskiy Rayon with data collected at Aktogai Weather Station and shows a relatively less performing match between losses and payouts. While all major drought losses are matched by a payout (the 1999 revenue loss, in yellow in the graph, was mainly attributable to a locust attack), the size of the payouts are not exactly covering the corresponding losses and other unnecessary payouts would have also been triggered by the contract. Accordingly, the lower performance of the contract for the Aktogayiski case is indicated by the significantly higher value of the ―Basis Risk Index‖ (90% vs 43% of the Altinsarin contract).64 61 Such an approach has been initially piloted in Mexico where Agroasemex, a parastatal insurance and reinsurance company, has developed a methodology using reanalysis techniques to obtain a simulated series of weather variables. Similar activities are under development in various other countries. 62 Graphical representations of the other structures are presented in Annex 8, Appendix F. 63 It should be clarified that in WII the yield coverage level is a purely indicative measure used only for calibration purposes. In fact, payouts are triggered only by specific realizations of the weather indexes, regardless of the actual recorded yield level. 64 The ―Basis Risk Index‖ was developed in order to provide an approximate indication of the mismatch between the indicative loss suffered and the potential payouts. It is defined as the ratio between a) the sum of the absolute values - 163 - 5.98. As indicated in Table 5.11, indicative Gross Premium Rates of the technically acceptable contract structures range from 13.7% to 24.2%. This is a clear indication of the fact that the potential cost of using WII contract for insuring against drought may be very high, and also that the different case study areas have different exposure to drought risk.65 Figure 5.7: Historical Performance of Drought Index Contract for Spring Wheat in Altinsarin 25,000 20,000 15,000 10,000 KZT/Hectare 5,000 0 -5,000 -10,000 -15,000 Payout: K Payout: GTK Rayon: Altinsarin Weather Station: Kostanai Indicative Yield Level Coverage: 65% Index: Cumulative precipitation, 1 March - 20 July Trigger: 95 mm Exit: 30 mm Contract Maximum Payout: KZT 24,000 Reference Gross Premium: 16.8% Basis Risk Index: 41% of the difference between payouts and revenue losses, and b) the sum of payouts. It is important to highlight that such an index has no actuarial value or rating function, and should be considered just as an approximate reference for comparing the performance of different index contracts. 65 The premium rates seem to suggest a progressively increasing exposure to drought moving from the Kostanay Oblast, to the Akmola Oblast, and to the Pavlodar Oblast. While this seems to be in line with the other findings of the study, it should be also considered that contracts developed for the Pavlodar Oblast cases (Aktogaiskiy) have a relatively lower performance and that, consequently, the higher premium rates may be also factoring in the need to provide higher payouts in order to capture all potential losses. - 164 - Figure 5.8: Historical Performance of Drought Index Contract for Spring Wheat in Aktogayskiy 15,000 10,000 5,000 KZT/Hectare 0 -5,000 -10,000 Yield reduction due to locust attack -15,000 Cumulated Rain Payout Revenue Loss Rayon: Aktogayskiy Weather Station: Aktogai Indicative Yield Level Coverage: 65% Index: Cumulative precipitation, 1 March – 31 August Trigger: 150 mm Exit: 50 mm Contract Maximum Payout: KZT 13,000 Reference Gross Premium: 20.9% Basis Risk Index: 90% 5.99. While the Basis Risk Index calculated for each of the contract structures provides an indication of the potential match between payouts and historical average yield losses in a specific area, it should be kept in mind that basis risk has also a spatial component, generated by the deviations of individual farm yields from the aggregate average yield on which the weather index is calibrated. This dimension influences the ability of the contract to compensate the potential policy holder in the appropriate measure. In this respect, in order to provide some empirical indications on the degree of ―spatial basis risk‖, the local project team collected detailed information on the 2008, 2009 and 2010 yield performance of Commercial farms and Production Enterprises located within a radius of approximately 20 to 25 km from the Kostanay, Aktogaisky and Mikhailovka weather stations.66 5.100. Figures 5.9 and 5.10 provide an “As If” scenario in which mismatches between 2010 farm level losses and the 2010 WII payouts are calculated for both Commercial Farms and Production Enterprises operating in the surroundings of Kostanay weather station. Deviations for both farm 66 The data is presented in Annex 8, Appendix G. - 165 - typologies oscillate between + 14% and – 30% of average yields indicating a relevant variability in the effectiveness of compensating farm level losses.67 5.101. While it is clearly not possible to draw any statistically significant conclusion from this simple elaboration, the exercise indicates the potential presence of a relevant basis risk dimension and it highlights the importance of carefully evaluating of the spatial basis risk component embedded in farm level WII products.68 Figure 5.9: Mismatch for the 2010 Payout for Commercial Farms Operating in the Range of 20 to 25 km from Kostanay Weather Station. Commercial Farms 20% 10% 0% 1 2 3 4 5 6 7 8 9 10 -10% -20% -30% Source: Authors Note: The bars represent the positive or negative deviation between the 2010 actual loss and the 2010 WII payout divided by the average 2008-2010 yield of each each commercial farm. The total 2010 wheat grown area covered by the ten Commercial Farms in object amounted to 14,520 hectares. 67 It would have been interesting to replicate a similar analysis for the Aktogay and Mikhailovka weather stations, but the data collected in the surroundings of Aktogay is somewhat incomplete, and it has not been possible to develop an acceptable WII contract structure for Mikhailovka. It is interesting to note that the farm level data for Mikhailovka, also presented in Figure A7. of Annex 7, seems to be less variable than that of the Kostanay farm data. 68 As an additional caveat, it may be useful to highlight that what may be considered acceptable in terms of basis risk may also depend upon the actual farmers‘ experience with agricultural insurance. Farmers that have been used to being compensated on the basis of in-field loss adjustment procedures may more sensitive to the potential impact of basis risk. - 166 - Figure 5.10: Mismatch for the 2010 Payout for Individual Production Enterprises Operating in the Range of 20-25 km from Kostanay Weather Station. Production Enterprises 15% 10% 5% 0% -5% 1 2 3 4 5 6 7 -10% -15% -20% -25% -30% Source: Authors Note: The bars represent the positive or negative deviation between the 2010 actual loss and the 2010 WII payout divided by the average 2008-2010 yield of each production enterprise. The total 2010 wheat grown area covered by the seven Production Enterprises in object amounted to 104,424 hectares 5.102. With respect to the pricing of the contract prototypes, interesting indications stem from comparing Indicative Gross Premium Rates (also referred to as “Indicative Commercial Rates”) of AYII and WII contract structures. Although the pricing methodologies adopted for the two different class of insurance products are not directly comparable, some adjustments to the WII pricing parameters may reduce the discrepancies. Hence, in the areas for which both AYII and WII products have been developed, WII premium rates were recomputed on the basis of a 60% coverage yield scenario and a 25% administration and expenses grossing up factor (previous WII scenarios were based on a 65% coverage level and a 15% grossing up factor). 5.103. Table 5.12 presents the two sets of Indicative Commercial Rates for the areas covered by both index insurance approaches. After having again highlighted that the two pricing methodologies are not directly comparable, and that the actual commercial price of any insurance product is a function of many business variables that can significantly modify its cost, a first element to be noted is that WII commercial rates are higher than AYII. It is also interesting to note that differences in premium rates are roughly related to the level of the Basis Risk Index that reflects the mismatch between WII payouts and losses: in essence, the larger the Basis Risk Index the higher the difference between the rates, suggesting that the cost of not being able to perfectly capture yield reductions is factored in the price of the WII contract. - 167 - 5.104. From a conceptual and operational point of view the findings highlighted in the preceding paragraph are quite relevant. Since WII products are structured to provide a single peril coverage against drought, and AYII policies have a larger coverage spectrum, it is clear that in order to be competitive the cost of the drought-index products should be lower than AYII policies. 5.105. The comparison between the Commercial Rate levels of WII and AYII may be one of the evaluation points to be considered by policymakers that need to select between the two classes of products. Additional elements to be taken into account in such a selection process would be the spectrum of coverage offered by the two different index products, their scale-up potential, the operational challenges that each of the scheme faces, and the expected amount of time needed to settle payouts after the end of the crop season. Table 5.12 Comparative Analysis of Indicative Commercial Rates for AYII and WII AYII WII Reference Weather Indicative Indicative Rayon Basis Risk Index Station Commercial Rate Commercial Rate Auliekolskiy WS Kushmurun 11.9% 12.6% 24% Altynsarinskiy WS Kostanay 10.5% 17.5% 50% Enbekshilderskiy WS Stepnogorsk 11.8% 14.4% 62% Aktogayskiy WS Aktogai 16.4% 27.0% 82% Source: Authors Meso, Macro and Reinsurance applications of WII in Northern Kazakhstan 5.106. As discussed above, WII can be retailed at different business levels apart from at the individual farmer or micro level, including: a) meso level, to cover the exposure of entities such as financial service providers, farmer associations, input suppliers and processors from potential losses caused by adverse weather events; b) macro level, to aid governments and relief agencies in development and disaster management.69 5.107. One specific form of meso level application of WII could be the use of index contracts as a reinsurance coverage for insurance companies involved in agricultural insurance programs. As indicated in Chapter 4, crop insurance companies in Kazakhstan are very exposed to catastrophe losses on their retentions and options for enhanced reinsurance protection need to be considered. GRK currently provides free of cost proportional reinsurance protection equal to 50% of the claims to the private insurance companies and mutuals in Kazakhstan. Currently, however, neither the private insurers nor the mutuals have any reinsurance protection on their 50% retentions and they are therefore very exposed to major systemic drought losses. 5.108. In this respect, WII structures similar to the ones designed in this Feasibility Study could be used by insurance companies to hedge the drought exposure of their agricultural underwriting portfolios. The analysis carried out in this Study shows that it is actually possible to structure drought 69 The use of on index insurance by National or Regional governments (macro level) is discussed in Chapter 6. - 168 - insurance products for spring wheat production by calibrating weather indexes on Rayon level yield records. These structures could technically form the basis of a reinsurance transaction. 5.109. From the reinsurance industry point of view, provided that the usual assurances on the reliability and independence of weather measurement are available, WII contracts are usually considered attractive as they are transparent, free of asymmetric information problems and easy to manage. Hence, if appropriately structured, there should be no problem for the insurance companies of Kazakhstan to access international reinsurance markets for WII protection. 5.110. In order to test the performance of reinsurance applications of the WII contracts developed in this Study, a simulation scenario was carried out in which hypothetical WII reinsurance payouts and actual indemnities paid by the current LIC scheme are compared.70 Two relevant caveats for putting this exercises in the right perspective are the following: i) Given the remarkable difference in coverage levels, in order to provide comparable payouts only a fraction of the LIC area covered should be reinsured with WII contracts. In this respect, the share of area covered under the LIC scheme to be reinsured was set at 5%.71 ii) It should be remembered that indemnities provided by the LIC scheme are influenced by wheat price levels, and this clearly weakens the link between drought impact on crop production and LIC indemnities.72 5.111. Figure 5.11 provides a graphical representation of LIC indemnities and potential corresponding WII reinsurance payouts. The graphical analysis shows that, although payouts and indemnities do not match perfectly, WII structures would have provided relevant payments in the main drought driven loss events.73 In addition, Table 5.13 shows that the total amount of WII hypothetical payouts across the four Rayons and over the five years history are in a comparable range, indicating that, for the examined area and time frame, the reinsurance coverage of WII would have yielded reasonable results. 5.112. Despite the reasonable performance in terms of covering the drought exposure of insurance companies, the financial dimension of the WII approach to reinsuring the LIC scheme seems to be out of scale and outside any acceptable insurance logic. Table 5.14 compares the amount of premium collected in the history of the LIC scheme in the selected Rayons and the hypothetical cost of a WII based reinsurance coverage. It is easy to note that the cost of the WII coverage is significantly higher than the amount of premium collected by the LIC scheme. Despite any potential difference in the pricing methodologies adopted (including savings generated by adopting a portfolio rating approach), and any discount offered by willing reinsurance partners, the order of magnitude of the two variables seems difficult to reconcile. 70 Only drought indemnities were considered. 71 In the large majority of policies sold in the LIC scheme coverage is roughly equivalent to 10% of the yield level while the drought index structures were calibrate to cover losses below 65% of the average yield level. 72 When wheat price levels are high the yield threshold falls and, conversely, when wheat prices are low the yield threshold raises. Hence, depending on the price level, a different aggregate indemnity amount for any specific drought event may be achieved. See Chapter 3 for more details. 73 In this framework it should be considered that, within acceptable boundaries, insurance companies may be less financially sensitive than individual farms to loss-payouts mismatches. - 169 - 5.113. In conclusion, the findings of this analysis show that adopting WII reinsurance in the context of the current LIC framework would not make commercial sense and, at the same time, reinforce the finding of Chapter 3 where it is shown that LIC is severely underrated. 5.114. Should the current LIC system be revised into an actuarially sound crop insurance scheme, a new evaluation of the use of WII contracts for reinsurance purposes may lead to different conclusions. Figure 5.11: LIC Indemnities and Potential Corresponding WII Reinsurance Payouts Aktogayskiy - WS Aktogai 2010 2008 2006 Enbekshilderskiy - WS Stepnogorsk… 2008 2006 Auliekolsky - WS Kushmurun 2010 2008 2006 Altinsarin - WS Kostanai 2010 2008 2006 - 20,000,000 40,000,000 60,000,000 Indemnities paid under current LIC scheme (KZT) Hypotetical WII payouts with 5% of LIC area reinsured (KZT) Source: Authors on FFAS data and own elaborations - 170 - Table 5.13. Aggregate LIC Indemnities and Hypothetical WII Reinsurance Payouts Hypothetical Indemnities paid WII payouts under current with 5% of LIC LIC scheme area reinsured (KZT) (KZT) 2006 4,555,320 - 2007 - - 2008 78,545,211 70,351,578 2009 13,999,618 - 2010 98,577,822 88,897,508 Total 2006 - 2010 195,677,971 159,249,086 Source: Authors on FFAS data and own elaborations Table 5.14. Indicative cost of WII reinsurance of LIC drought exposure Cost of 2.5% Indicative Gross Area insured of LIC area Premium 2.5% of LIC Premium for WII under current WII collected in area insured at 65% coverage LIC scheme reinsurance LIC scheme (Ha) level (KZT/ Ha) (Ha) coverage (KZT) (KZT) Altinsarin - WS Kostanai 4,042 174,922 4,373 17,675,868 10,930,646 Auliekolskiy - WS Kushmurun 2,637 161,783 4,045 10,665,544 6,252,665 Enbekshilderskiy - WS Stepnogorsk 3,439 47,811 1,195 4,110,551 3,657,782 2,678 Aktogayskiy - WS Aktogai 740 19 49,543 253,448 Source: Authors on FFSA data and own elaborations Note: Given that 50% of LIC indemnities are paid by the GRK, the percentage of the LIC area to be covered by WII reinsurance can be reduced by the same proportion and set at 2.5%. Remote Sensing approaches to Developing Index Insurance 5.115. In order to find new approaches to offering crop insurance in Kazakhstan, solutions based on satellite measured data could be also investigated. The use of remote sensing approaches in agricultural risk management has become more popular in recent years. In fact, satellite data can be used to estimate precipitation levels and reflections indexes, to measure energy balances and even to run crop models.74 5.116. The bulk of the current commercial applications of remote sensing indexes target pasture growth through reflection indexes such as the Normalized Difference Vegetation index (NDVI) and related transformations. However, research on the use of satellite information to estimate crop yield is underway and could soon provide interesting results.75 74 For more details see Hellmut et al., 2009 and Rosema et al., 2010 75 See, for example, Rosema et al., 2010 - 171 - 5.117. Kazakhstan capacity in remote sensing technology is remarkable and specific agricultural applications have already been developed. The entity responsible for such activities is The National Centre of Space Research and Technologies (NCSRT) of the NSA. Among its various activities, NCSRT operates a satellite-based crop monitoring system for the Ministry of Agriculture. The activities of the NCRST in this field include estimation of spring soil water content; spring crop acreage estimation; cereals sowing date control; empirical models for crop state assessment and grain production forecasting. 5.118. So far the work carried out by NCSRT has mainly had production forecasting objectives. However, the Centre holds the expertise needed to explore the feasibility of remote sensing technology for insurance applications that are currently adopted or under development in insurance schemes of other countries. Hence, specific research activities aimed at exploring the potential role of NCRST in supporting index insurance development would be recommended. Operational considerations 5.119. Whether the GRK or the insurance industry decide to consider the adoption of WII contracts in the agricultural insurance program of Kazakhstan, additional research to investigate the technical and commercial specificities of WII would be recommended. Depending on the targeted application level, the research agenda may have different objectives. 5.120. For micro level applications, a deeper understanding of the level of spatial basis risk in spring wheat production would be required. This would provide indications on the effectiveness of WII insurance products in compensating farm level losses, thus helping to understand the level to which farmers would be prepared to accept payouts that are not exactly matching with losses experienced in the field. 5.121. The most effective way to learn about the actual potential of WII products is to develop specific pilot tests in which a WII supply chain is set up and prototype contracts are designed and retailed. Although, as mentioned above, in the short term micro level WII is not a realistic alternative to the current agricultural insurance scheme, developing pilot tests could be a way of providing significant insights on possible future uses of WII. Even if not implemented at full scale, WII may still prove as an additional risk management tool to be adopted in a case by case approach where required conditions apply.76 5.122. For a meso level application, a useful investigation would be to carry out a larger scale analysis aimed at determining in how many Rayons the proposed cover would be technically feasible. In fact, the analysis carried out in the Feasibility Study showed that out of the nine cases examined in the North of Kazakhstan six resulted acceptable, one not acceptable and two not modelable. 5.123. The result of such analysis may highlight the need of generating synthetic historical data sets for rating purposes, and of installing additional weather measurement devices for contract monitoring and settling (see above). 5.124. In general terms, should Kazakhstan’s stakeholders decide to promote the use of WII, a suggested development action would be to set up a technical unit or a working group to address the technical dimension of WII program development, usually difficult to handle by individual insurance companies. Such a working group would need to combine the expertise of agronomists, agrometeorologists and insurance experts and would greatly benefit from orientation and support of the 76 For example in cases of large operations or cooperative units that have good yield history and can access to representative weather data. - 172 - GRK. Should the development of the coinsurance pool discussed in Chapter 4 take place, it would be the ideal seat for a technical unit on WII. Potential roles of the Government in enabling WII 5.125. From a WII perspective, the following list summarizes what the development community has come to consider as useful public actions that governments and international donors can carry out to support the development of WII programs77:  Improve weather station infrastructure and data systems  Fund agro-meteorological research leading to product design  Provide technical assistance for training and product development  Facilitate the development of an enabling legal and regulatory environment  Facilitate access to reinsurance  Support regular monitoring, evaluation, and impact studies Conclusions on WII 5.126. The analysis carried out in this feasibility study indicates that WII for spring wheat in the North of Kazakhstan is technically feasible. 5.127. Despite the positive technical findings, the actual density of the weather measurement network does not seem to make a full scale implementation of farm level WII a realistic option in the short term. While potential action to address this constraint may be undertaken, for the time being WII should not be considered a readily implementable alternative to the current Loss of Investment Cost insurance scheme. 5.128. In addition, the high potential cost of WII products developed in the analysis, together with a preliminary analysis of “basis risk” patterns, suggest the need to carry out further research to assess the potential interest of famers for such alternative insurance approach. 107. WII could also find application at a reinsurance level by covering the drought exposure of insurance companies. However, the findings of the analysis carried out in this study show that adopting WII reinsurance in the context of the current LIC framework would not make commercial sense. Should the current LIC system be revised into an actuarially sound crop insurance scheme, a new evaluation of the use of WII contracts for reinsurance purposes may lead to different conclusions. 77 See IFAD – WFP, 2010 - 173 - Chapter 6: Tailoring Crop Insurance to the Needs of Lower Income Smaller Farmers 6.1. The final section of this report presents some of the international lessons and experience on strategies and programs to address the agricultural risk transfer and insurance needs of small farmers and which may be applicable to the small household mixed crop and livestock farming sectors which are mainly located in southern Kazakhstan. This section also draws where possible on the findings of a short visit to one wheat growing area in Tole-bi Rayon in SKO and where many farmers are either small peasant farms or semi-subsistence household plots. Identification of Appropriate Crop Insurance Products Appropriate Crop Insurance Products 6.2. To date, Crop Insurers in Kazakhstan have offered a single Loss of Investment Cost (LIC) crop insurance product to mainly medium and large cereal producers located in northern Kazakhstan and there has been very little debate about the appropriateness of this product to small and marginal farmer in southern Kazakhstan. While the LIC product is suitable for grains, oil seeds and other field row crops, it is not suitable for most horticultural and tree fruit crops and especially for crops which have multiple or staggered harvests. Furthermore the product is designed to protect against catastrophe losses equivalent to more than 75% of 80% of expected crop revenue and while this type of cover might be of interest to large cereal producers it is not necessarily appropriate to small subsistence farmers‘ food security requirements. 6.3. There is a wide range of crop insurance product types available internationally and this report has recommended that in future Kazakhstan’s crop insurers should aim to develop and introduce several of these alternative crop insurance products. The products that have been recommended for individual farmers include named peril (hail) crop insurance, MPCI loss of yield insurance and new non- traditional index products, AYII and crop WII (Table 6.1). 6.4. Named peril crop hail insurance may be a product which could be offered in cash crops such as cotton if there is an appreciable hail exposure. (See also Section 5 for comments on crop hail opportunities in Kazakhstan). 6.5. Individual grower MPCI is often not considered suitable for small and marginal farmers because of the very high costs of administering such policies and the associated high premium costs. The very high cost of administering MPCI policies is associated with the need to conduct pre-inspections in order to minimize anti-selection and moral hazard and the requirement to conduct time-consuming and costly in-field loss assessments. While economies of scale can be achieved on large insured units with several thousand hectares of insured crop, for small and marginal farmers the A&O costs become prohibitively expensive for the insurance company78. In addition the very high premium rates on MPCI policies which are often in the order of 7.5% to 15% make this cover unattractive to small farmers unless accompanied by very heavy premium subsidies by government. 78 This point was noted by the CEO of Kazakhstan‘s largest crop insurance company who noted that because of the very high costs of administering crop insurance, his company could only insure large Production Enterprises. - 174 - 6.6. AYII is considered a more appropriate product for small and marginal farmers because of the reduced costs of administering and adjusting this type of cover. AYII does not require pre-inspections or individual farmer and field loss assessment and is therefore a less costly product for an insurer to offer to small and marginal farmers. India is the main country to date which has implemented AYII on a massive scale linked to crop credit for small and marginal farmers and is currently benefitting more than 25 million producers each year. The Indian AYII program is, however, more of a socio-economic support program: it incurs very high financial losses and is only sustainable because of very high levels of government subsidies. 6.7. The suitability of AYII for small farmers located in southern Kazakhstan will need careful consideration. As noted in Section 4, one of the preconditions of AYII is that farming systems in the defined area (Insured Unit) must be fairly homogeneous, conditions which are found in northern Kazakhstan where rain-fed spring wheat is the main crop which is cultivated with very stand planting and harvest dates and relatively homogeneous yields. Agricultural production in south Kazakhstan is, however, much more diversified within each Rayon with a mix of rain-fed and irrigated agriculture and much smaller and more scattered plots of mainly winter sown cereals, horticultural crops and cotton. AYII for winter wheat may not be ideally suited in these regions due to the much less homogeneous cropping systems and mix of rain-fed and irrigated agriculture. 6.8. Crop weather-index insurance was first introduced into India 10 years ago where it was hailed as a micro insurance product ideally suited to the needs of India’s poorest and most marginalized farmers. WII has been promoted for small and marginal farmers in developing countries because it has the potential to address correlated risk affordably and is operationally less challenging than traditional individual farmer indemnity-based crop insurance products. WII has achieved some degree of scale-up by the private insurance sector in India, and has also been introduced on a large scale as a compulsory crop-credit insurance cover by AIC, the public sector crop insurer. WII is now being pilot tested in about 25 countries in Asia, Africa and Latin America but to date none of the micro-level individual farmer crop insurance programs have achieved sustainability and scale-up. 6.9. In southern Kazakhstan, WII will only be suitable for rain-fed agriculture. Given the fact that much of agriculture is South Kazakhstan is irrigated WII policies which insure against excess or deficit rainfall will not be suitable. For example, in Tole-bi Rayon nearly 60% of the annual sown area of 758,000 Ha is irrigated. Irrigated crops tend to be high value horticultural and vegetable crops and cotton while cereals are gown on non-irrigated land. A further factor which will need to be studied carefully is the drought risk exposure: in Tole-bi Rayon technical staff noted that drought was not an economic risk exposure in rainfed crops; other factors including pests and diseases which are not indexable under a WII policy were more important79. 79 In this respect, the feasibility study team assessed the possibility of structuring a WII contract for the case of winter wheat production in the neighborhood of the Tasaryk weather station in the Tole-bi Rayon. However, information provided by local stakeholders seemed to overestimate the potential impact of drought on wheat production in the area. On the basis of the data analyzed, the lowest yield recorded (drought related, according to wheat producers), resulted higher than 70% of the average yield reference. With a drought risk of such a small magnitude, drought-index insurance may not be performing significant protection tasks. - 175 - Table 6.1. Classification of Agricultural Crop Insurance Products Type of Agricultural Insurance Product Payouts Availability a) Indemnity Based Agricultural Insurance (insurance payouts based on the actual loss at the insured unit level) 1.Named Peril Percentage of Damage Widespread 2. Multiple Peril Loss of Investment Costs Yield Loss/loss of investment costs Mexico, Kazakhstan, E. Europe 3. Multiple Peril Loss of Yield Yield Loss guarantee Widespread b) Index based Agricultural Insurance (insurance payouts based on an index measurement) 4. Area-Yield Index Area-yield Loss USA, India, and Brazil 5.Crop Weather Index Insurance Weather Index payout scale India, México, Malawi, Canadá, USA c) Crop Revenue Insurance (insurance payouts based on yield measurement and crop prices) 6. Crop Revenue Insurance (CRI) Yield and Price Loss Limited to USA Source: World Bank, 2009 Farmer Segmentation and Crop Insurance 6.10. Individual farmer crop insurance is a tool that is most effective when the farmer is a commercial producer who grows a crop(s) for sale and where the farmer invests in purchased inputs and services often using formal credit. Where the farmer is a commercial producer and uses credit to purchase seeds and fertilizers and plant protection chemicals, the farmer faces a financial risk in the event of adverse climatic conditions leading to crop failure and risk transfer through the purchasing of crop insurance is often justified. For large commercial crop producers, individual grower MPCI and or named peril crop insurance may be a suitable product (Figure 6.1). 6.11. Individual farmer crop insurance is a risk management tool that is often appropriate for commercial and semi-commercial farmers. In Kazakhstan the private insurers are targeting the larger commercial agribusiness enterprises because of the lowered unit costs of dealing with these farmers and the higher premium volume generated by each bound risk. 6.12. The insurance companies, however, face major challenges in implementing crop insurance to the smaller semi-commercial farmers on an individual farmer by farmer basis because of the very high costs of delivery and administration of insurance on small farm units . In Kazakhstan the farmers‘ mutual crop insurance associations‘ main client base is the small to medium farmers and given the mutual‘s low overhead costs and proximity to their members they are much better placed than the private companies to implement crop insurance with the smaller commercial and semi-commercial farmers. For these smaller semi commercial farmers, index insurance (both AYII and WII) may be a more viable solution than traditional MPCI (Figure 6.1.) 6.13. Traditional Individual farmer crop insurance is a risk management tool that is often appropriate for commercial and semi-commercial farmers. Traditional individual farmer crop insurance cannot provide solutions for subsistence farmers. There is much evidence today that traditional individual farmer multiple peril crop insurance (MPCI) does not work for small and marginal farmers and usually ends up being heavily subsidized by governments. For most small subsistence farmers producing food crops for on-farm family consumption, crop insurance is a luxury few of them can afford, hence governments‘ intervention to make crop insurance more affordable through premium subsidies. In Kazakhstan crop insurance is unlikely to be a useful intervention for the very small rural households were these are mainly subsistence producers. - 176 - 6.14. For subsistence farmers it may be much more cost-effective for governments to examine alternative food security mechanisms and social safety nets or, where they elect to use insurance, to consider some form of macro-level weather index programme to permit early payments to be made in the event of a major natural disaster. To date, several countries including Ethiopia, Malawi and Mexico have designed macro-level rainfall deficit index covers that have been designed to provide national and or regional governments with immediate cash liquidity following a natural disaster and to enable the government to provide an early response. Figure 6.1. Suitability of Crop Insurance Products for Different Types of Farmer •Medium/Large Farm units •Mechanised production Commercial •Access to credit MPCI Named Farmers •High levels input use •Produce for sale Peril Index Insurance •Smallholder farmers •Some assets Semi-Commercial •Some access to credit Farmers •Part consumption/part sale •Very small / no land Subsistence •Very few assets Social Programs •Subsistence farming Farmers Source: Adapted from Skees 200980 Tailoring Crop Insurance for Different Client Levels Different Levels of Aggregation 6.15. Crop Index Insurance (including both AYII and WII) is potentially a very flexible instrument which can be designed to provide risk transfer solutions either at the individual farmer-level (termed Micro-level insurance) , or at an intermediate level of aggregation as a financial business interruption protection for banks and other lending organizations such as cooperatives and MFI‘s (Meso -level insurance) and finally this is an instrument that regional and or national governments can use to insure against major systemic perils such as drought, windstorm, freeze and flood (Macro-level insurance). 6.16. In the past decade WII has been heavily promoted as an individual farmer micro-level insurance product but relatively little research and development has been conducted into meso-level and macro-level solutions. Micro-level CWII has been heavily promoted in developing countries as a low cost (in terms of administrative costs) product which is ideally suited to the needs of resource poor farmers and it is being implemented or pilot tested in about 25 countries in Asia, Africa and Latin 80 Skees, J., (2009) A Policy Vision for Developing Agricultural Insurance in Vietnam. - 177 - America81. Given the fact this product is typically designed to insure against systemic perils of ―too much rainfall‖ (excess rainfall/flood) and ―too little rainfall‖ (drought) the premium rates associated with WII are, however, typically in the order of 7.5% to 10% or higher and these premium rates are generally not affordable to the targeted resource poor farmers without government premium subsidy support. The scale-up of private-sector WII in India has in part been achieved because of premium subsidies, while the public sector WII program has achieved several million of policy sales today because it is linked on a compulsory basis to seasonal credit. No other country which is operating voluntary micro-level WII has, however, achieved commercial scale up to date82. 6.17. Crop Index Insurance can also be used as a meso-level instrument to protect rural bank lending (loan portfolio or business interruption protection). From the bank‘s perspective, farmers who have crop insurance protection are less likely to default on their loans in the event of major weather induced crop failure. It also means that in the event of a major regional loss event the bank‘s loan portfolio is protected against loss, enabling the bank to remain solvent and to reschedule farmers‘ loans and to continue lending. Claiming on a crop insurance policy and rescheduling loans are generally much more acceptable to a bank than having to resort to the courts to recover their debts. Since 2010, pilot meso-level WII initiatives have been launched in Peru and Ghana which are linked to regional bank lending to agriculture. The Peruvian product83 is a catastrophe El Nino excess rain leading to flood policy which uses an ENSO sea surface temperature index to trigger indemnities while the Ghana product 84 for rural banks is linked to rainfall deficit and or excess rainfall. In addition, in Vietnam85, a new innovate flood index product linked to regional bank lending to rice farmers is awaiting launch. 6.18. At a regional or national level there also appears to be an important role for linking disaster risk management with an ex-ante macro-level weather index or AYII insurance policy.86 To date, several countries including Ethiopia, Malawi and Mexico have designed macro-level rainfall deficit index covers that have been designed to provide national and or regional governments with immediate cash liquidity following a natural disaster and to enable the government to provide an early response. Mexico is the only country which to date has used AYII as a macro or state level crop insurance product for small and subsistence farmers (See below for further discussion). There appears to be considerable scope for using macro index products as a social safety net product for small subsistence farmers for whom commercial crop insurance is not necessarily an appropriate or cost-effective mechanism. Key Organisational and Operating Differences between Micro Insurance and Meso/Macro Insurance 6.19. There are some key structural and operating differences between a micro-level individual farmer crop insurance schemes and a meso-level crop insurance scheme purchased by a financial 81 See Table 5.7 in the preceding Chapter. 82 For an up to date review of the current status of crop and livestock WII projects around the world see WFP/IFAD (2010). The Potential for Scale and Sustainability in Weather Index Insurance for Agriculture and Rural Livelihoods, by P Hazell, J. Anderson, N. Balzer, A. Hastrup Clemennsen, U. Hess and F. Rispoli. Rome March 2010. 83 Skees, J.R., and A.G. Murphy (2009). ENSO Business Interruption Index Insurance for Catastrophic Flooding in Piura, Peru. GlobalAgRisk Inc, Lexington, KY, February 2009 84 Stutley C. (2010). Crop Insurance Feasibility Study 2010. Innovative Insurance Products for the Adaptation to Climate Change Project (IIPACC) Project Ghana 85 Skees, J., and J. Hartell (2008). Developing Index Based Agricultural Insurance to Enhance Financial Markets for Poverty Reduction in Vietnam. Vietnam: Agricultural Insurance Product Briefing Note. Ford Foundation Project. Prepared for VIBARD Agricultural Insurance Working Group, Hanoi, Vietnam, November 2008 86 For a comprehensive review of linkages between disaster risk reduction and index insurance see Warner et al, 2009. - 178 - lending institution and finally a macro-level regional or countrywide program purchased by a state or national government. Under a micro-level program the individual insured is the farmer who pays a premium (which may or may not receive government subsidy support) and in return receives a claims payment in the event of a loss. Where micro-level crop insurance is linked to crop credit the lending institution is usually listed as the first beneficiary in order to ensure that any claims settlement is used to repay the amount of loan and then the farmer receives any balance on the claim. (Figure 6.2). 6.20. Under a Meso-level crop insurance program a financial lending institution would typically be the insured policy holder and would purchase crop insurance to protect its crop loan portfolio against catastrophe climatic losses. (Figure 6.2). Under the meso-level option, the lending institution (e.g. rural a bank or MFI) would declare its total agricultural-loan acreage in each region (municipality) and the coverage level it wished to insure under the crop insurance policy in each municipality. The bank/MFI lending to agriculture would purchase a single aggregate policy to protect its loans and would be responsible for premium payment. In the event of a claim in any or all insured regions, the losses would be computed and the bank/MFI would receive an aggregate claims indemnity. The bank/MFI would then decide whether or not it wished to pass on any of this indemnity payment to the individual farmers it lends seasonal crop production credit to. If the bank/MFI elects to pass on some of the indemnities to farmers it may also require them to share in the crop insurance premium costs. The major advantage from an insurance viewpoint is that bank/MFI would be able to accept a higher degree of basis risk associated with the proposed Area-Yield Index or CWII products than individual farmers. 6.21. Under the operation of a Macro-level Policy the insured Policy Holder is typically the state (regional) government or national government. Under the Ethiopian and Malawian macro-level WII rainfall deficit products the national governments were the policy holders and the covers were designed to trigger ex-ante disaster relief payments to the national government. In Mexico the state-level governments have since 2003 purchased state-level catastrophe drought WII cover and or state-level AYII based Catastrophe Insurance (Seguros Catastrophicos) protection for specific target audiences of small subsistence farmers located in specific regions and who grow specific crops: these farmers are too small to participate in the private commercial crop insurance programs and or public-sector small farmer group loan and crop insurance programs. The state governments are the policyholders and they pay for 100% of the costs of premiums: in the event of claims the state governments have pre-agreed indemnity payment formula to distribute the lump sum indemnity payments from insurers and reinsurers to the target beneficiary families. (See below for further discussion of Mexican macro-level crop index insurance programs). - 179 - Figure 6.2. Organisational Structures for Micro and Meso/Macro-level Crop Insurance Source: Dick, W. (2009) Macro-Level Crop Index Insurance in Mexico 6.22. In 2003 Agroasemex the Mexican specialist parastatal crop reinsurer launched the first Macro-level catastrophe drought crop insurance index insurance program. Mexico is very exposed to catastrophe risk in agriculture including drought (80% of natural catastrophes), hurricanes (17%), excess rain/flood (2%) and Frost (1%). Since 1995 the Federal and State governments have operated a national natural disaster scheme under the FONDEN program which is designed to provide financial compensation to small rural farming families who are not eligible for private crop and livestock insurance. Between 1995 and 2003 the Federal Government and State governments paid out US$ 212 million and US$ 74 million respectively to small rural farmers under the FONDEN program. In 2003 as part of the FAPRAC (Fund for the Care of Rural Population Affected by Weather Contingencies), government contracted Agroasemex the parastatal agricultural insurer and reinsurer to substitute the ex- post disaster compensation programs with an ex-ante macro-level index insurance for catastrophe climatic perils (Agroasemex 2007). 6.23. The first Agroasemex pilot macro-index program was launched in 2003 as a drought insurance index cover for maize and sorghum grown in Guanajuato state. The first product was based on a drought index for 5 weather stations covering an insured area of 75,000 Ha of rainfed maize and sorghum produced by several thousand small subsistence farmers located in Guanajuato state. In recognition of the difficulties for insurance companies to provide insurance cover to individual farmers, Agroasemex specifically designed an aggregate or macro-level policy which was sold to the state government which was then responsible for setting the indemnity rules for each farmer in the insured command area. Since 2003 the catastrophe climate contingency insurance program has been massively scaled-up with the development of index products based on i) WII covers against drought, hurricane and frost, ii) AYII covers providing all risk loss of yield protection at a macro-level, and iii) NDVI-pasture index covers for livestock producers. Currently 30 of the 38 state governments in Mexico purchase climate contingency protection; for crops, 8 million hectares are insured with 3.2 million small subsistence farmers protected under the crop insurance programs and about 4.4 million head of livestock are insured under the NDVI - 180 - pasture drought index program. Further information on the Mexican Macro-level index programs are presented in Box 6.1. Box 6.1. Macro-level Catastrophic Climate Contingency Index-based Insurance for State Governments in Mexico Objectives: • To provide disaster relief assistance to small and marginal farmers affected by natural calamities Legal Framework: • National Program for the Attention of Climatic Contingencies (PACC in Spanish) managed by MoA. Main Features: • The Program provides disaster relief assistance to the farmers from a fund that is financed through the payouts generated by a Macro –Level agricultural insurance policy issued to the Federal and State Governments. Target Farmers: • Small and marginal farmers (according to the definition provided by Ministry of Agriculture) that suffer from the occurrence of non-recurrent and unforeseen and unpredictable climatic losses and have no access to any other type of financial insurance Covered Perils:  Drought; frost; hail; excessive rain; flood; cyclone and tropical storms, and tornadoes. Covered Items:  Crops: Rice, maize, beans, sorghum, barley, oats, wheat, rape, soybeans, peanuts, sugarcane, avocadoes, cacao, citric, coco, passion fruit, coffee, apples, mangoes, nuts, pineapple, peach, and plantain. Mechanism of Operation: a) Catastrophe insurance: o Federal and state Governments liabilities arisen out from the application of the disaster relief assistance to the farmers are transferred to the market through a macro level index based insurance policy. o The macro level index based insurance policy can be based on a weather index based insurance type, an AYII type, or an NDVI index insurance type. o The municipalities (rayons), the insured crops and covered perils are selected by the Federal and Provincial Government ex-ante of the inception of the coverage and the insurance premiums are paid o The premiums are co-financed by the Federal (75% - 90%) and the State government (25% - 10%). o Assistance equivalent to US$ 75 per hectare up to a maximum of US$ 1,500 for annual rain fed crop (20 hectares) and US$ 750 for other crops (10 hectares) o 1-2 months waiting period to receive the indemnity. o The program is insured by commercial reinsurers and reinsured in the international market b) Direct Support: o This type of support is provided by the Federal and State Government when catastrophic insurance is not available. o The occurrence of the loss event is determined by the National Commission of Water (CONAGUA in Spanish) o The Direct Support is co-financed in equal shares by the Federal and State governments. o Assistance equivalent to US$ 75 per hectare up to a maximum of US$ 375 per farmer.(5 hectares maximum) o 3-5 months waiting period to receive the indemnity. Achievements:  Total Insured Area: 8 million hectares  Total number of heads insured: 4.2 million heads - 181 -  3.2 million farmers protected by the insurance program  Participant States: 30/38 states participating on the catastrophe program. Source: Authors from SAGARPA . Source : SAGARPA 2010. A potential disaster relief contingency fund for WKO and Aktobe. 6.24. Given the significant financial challenges of the current LIC insurance scheme in WKO and Aktobe, an alternative way for the GRK to provide spring wheat producers with disaster relief assistance could be to set up a publicly supported contingency fund that would be used to compensate farmers in case of extreme loss events. Drawing from the recent international experience, Mexico in particular, such a fund could be ideally protected from being exhausted through a specific reinsurance coverage that, on the basis of customized weather indexes, would trigger payouts in adverse weather conditions and, therefore, replenish the fund. Contingency fund resources would be also used to pay for the cost of the reinsurance coverage. 6.25. The reference weather indexes could be developed at Rayon level, so that payouts are triggered independently by weather measurement carried out in each Rayon, or at Oblast level, with an index based on a basket of stations that would trigger a payout for the entire Oblasts (hence, less frequently). 6.26. A critical element to be assessed would be related to the actual cost of reinsuring such a contingency fund. Given the very high drought exposure in the Aktobe and WKO, reinsurance payouts would also need to be frequent and therefore expensive. 6.27. Depending on the estimated overall financial exposure of this protection scheme for the WKO and Aktobe Oblasts (see Chapter 4), and on the projected cost of a dedicated reinsurance coverage, the GRK should evaluate whether it would be more cost-effective to purchase reinsurance on the international markets or to guarantee the replenishment of the fund with its own resources. - 182 - Organisational and Operational Systems for Small Farmer Crop Insurance Low Cost Delivery Models for Small Farmers 6.28. Insurance companies throughout the World face major challenges in trying to identify cost- effective ways of delivering and administering agricultural crop and livestock insurance programs for small famers. In Kazakhstan this problem is accentuated because of the very low sums insured adopted under the Obligatory Crop Insurance scheme with an average sum insured over the past five years of only KZT 3,287/Ha (US$ 22/Ha) and an average premium rate of only 2.42% generating an average premium of about KZT 80 per insured hectare (US$ 0.53/Ha) (Table 6.3.). Insurance companies not only need to reserve premium to pay for normal expected claims, but also to pay for business acq uisition costs/sales‘ agents fees and their own administration and operating expenses. In Kazakhstan sales‘ agents fees are typically 10% of original premium and the Insurer‘s own operating and administration expenses may be a further 10% to 15% of premium. Table 6.2 shows than on very small farm units of up to 100 Ha the total average premium earnings generated by a single policy are only about 53 Ha on average is far too low to cover the fees of a sales agent let alone enable a commercial insurer to cover its policy A&O costs. This problem also applies on small and medium commercial farms from 100 Ha to 1000 Ha where the total premium of about KTZ 80,000 per policy (US$ 533/policy) would not be adequate to service the account on an individual farmer basis. In order to offer crop insurance to small farmers, Insurance Companies therefore need to seek alternative and less costly ways of marketing and administering their policies to this group of farmers. Table 6.2. Kazakhstan Average premium/Policy according to size of Insured Farm (KZT/Ha) Item Farm Insured area (Hectares) 1 100 Farm size (Ha) 500 1,000 2,500 10,000 80 8,000 40,000 80,000 200,000 800,000 Premium (KZT) Premium (US$) 0.53 53 267 533 1,333 5,333 Source: Authors based on average premium rate Obligatory crop insurance scheme 2005-10 6.29. In South Kazakhstan many of the small Household farmers are organised into Production Cooperatives and rent out their land to the cooperative management to farm on a joint stock basis and this brings economies of scale in purchasing crop insurance. Features of the joint-stock approach are described in Box 6.2. for Berli Producers Cooperative. The land belonging to the 150 members is farmed as a single unit and a single crop insurance policy is also purchased by the cooperative farm managers. The Joint Stock approach to farming is apparently adopted by between 50% to 60% of all the farmer cooperatives in Southern Kazakhstan Oblast. The extent to which this joint stock approach, to overcoming the scale-problems for small farmer crop production and for contracting crop insurance, could be replicated in other parts of Kazakhstan is not known. - 183 - Box 6.2. Small Farmer Production Cooperatives in Southern Kazakhstan In Tole-Bi Rayon small farmers in one Producer cooperative have devised their own solutions to crop production and purchase of crop insurance. The Berli Producers Cooperative in Tole-bi Rayon was formed in 1993, it has 150 members most of whom are active farmers owning a total of 2,100 Ha of which 600 Ha are cultivated with winter wheat. The average size of farm is about 5 ha to 10 Ha per member. Each farmer is too small to produce wheat on their own land. The cooperative has therefore formed a joint stock company which rents the land from all the cooperative members and then farms this on a collective basis on behalf of the members thereby enjoying economies of scale in cultivating the land as a single aggregate farm. The members also receive a share of the profits at the end of the season following the harvest and sale of the crops. The cooperative contracts obligatory LIC crop insurance each year through one of the Mutual Farmer Crop Insurance Associations on its total cultivated area of wheat: for the Mutual Insurer there are major advantages of dealing with a single entity rather than trying to insure 150 smallholder farmers with each with 5 to 10 hectares. 6.30. There are several ways in which Insurance Companies can deliver insurance products and services to small rural households and farmers which are listed in Box 6.3. Box 6.3. Distributional Models for Small Farmer Insurance (Microinsurance)  Full-service model: Commercial or public insurers provide the full range of insurance services from initial development of the product, through distribution, to absorbing risk.  Partner-agent model: Commercial or public insurers, together with microfinance institutions or nongovernmental and other organisations, collaboratively develop the product. The insurer absorbs the risk and the agent markets the product through its established distribution network. This lowers the cost of distribution and thus promotes affordability.  Community-based model: Local communities, MFIs, NGOs, and/or cooperatives develop and distribute the product, manage the risk pool, and absorb the risk. As with insurance mutuals, there is no involvement on the part of commercial insurers.  Provider model: Banks and other providers of microfinance can directly offer or require insurance contracts. These are usually coupled with credit, for example, to insure against the risk of default. Source: ProVention 2006 (Cohen and McCord 2003) Full Service Model 6.31. The traditional method termed the “Full-Service” model the Insurance Company assumes full responsibility for all insurance functions would not be cost-effective for Insurers of small farmers in Kazakhstan. Under the full service model the Insurer bears the full costs of delivering insurance to farmers including insurance awareness and education and policy sales and marketing (either through its own network of sales agents or commission brokers) and relies predominantly on individual client sales. Premiums are collected individually by the company for the insured and claims notification and settlement is managed directly by the insurer. This is the model which is typically adopted by the private commercial crop insurers in Kazakhstan today under the Obligatory crop insurance scheme. With the average crop insurance premiums of about KZT 80 per hectare this model would only be feasible for large agribusiness enterprises with several thousand insured hectares. It would not be cost effective for commercial companies to adopt the Full-Service model for smallholder agricultural insurance in Kazakhstan. Partner-Agent Model 6.32. In Kazakhstan there may be considerable potential for Commercial Insurers to enter into a “Partner-Agent” relationship with rural organizations (e.g. the Cooperatives or MFIs) which have an existing rural distribution network and a large farmer membership. Under a Partner-Agent Model, the - 184 - Insurance company enters into a formal contractual agreement with the Agent under which the Agent assumes responsibility for marketing and promoting the Insurer‘s policies to its membership, for collecting premiums from Insured‘s and paying these over to the Insurer, for notifying claims to the Insurer, and finally in some cases for distributing claims settlement payments to the Insured‘s. Usually the Insurer will agree to pay the Agent a commission for its services. This model would potentially enable the private commercial insurers in Kazakhstan to deliver crop insurance more cost-effectively to large numbers of small and medium farmers. Such a model could also be used for delivering livestock insurance to the small mixed cropping and livestock families predominantly located in southern Kazakhstan. Figure 6.3. The Partner-Agent Delivery Model for Crop (and Livestock) Microinsurance Partner-Agent Model for Microinsurance Provide Develop product, Potential Mainstream product NGO/MFIs claims & Micro- Insurance as Insurance product education insurance Companies Agent Clients as Partner Underwriting Premium Source: Al Hasan (2007) 6.33. The Partner-Agent approach has been successfully promoted for smallholder agricultural insurance in recent years in Africa, Asia and Latin America and offers a potential win-win for both parties. For the insurance company the distribution of its products through a rural institution (Agent) offers the potential to reach large number of small farmers at low cost; for the rural institution, the agreement enables them to expand the range of products and services they offer to their membership and where the organization is involved in agricultural credit provision, the potential to protect their loan portfolio with crop and livestock insurance. Community Based Insurance Model 6.34. The Community-based Model includes all forms of informal agricultural insurance which is underwritten by local communities, MFIs, NGO’s. This model has proved very popular in southern Asia (Bangladesh, India, Nepal) where NGOs and MFIs have actively promoted livestock credit insurance for members over the past 20 years. This model is in fact very similar to the Provider model identified by Provention 2006. 6.35. Farmers’ Mutual Insurance Associations are a specific type of community-based private agricultural insurance organization and have been heavily promoted by MOA in Kazakhstan since 2008 to underwrite the Obligatory Crop Insurance scheme. This model could potentially be very relevant to developing smallholder agricultural crop and livestock insurance, but only if mechanisms of providing catastrophe reinsurance can be designed. This theme is explored further below using the Mexican Agroasemex Fondos program as a possible model which could be adapted for Kazakhstan. - 185 - Provider Model for Microinsurance 6.36. In many parts of Asia the Microfinance institutions (MFIs) are involved in providing their micro-lenders with microinsurance products and services which are normally linked to credit. In Bangladesh, India and Nepal MFIs are the main source of loans for resource poor urban and rural households. Many of the MFI‘s have also introduced their own internal insurance funds providing limited life insurance cover, health cover, maternity cover, and in some cases crop and livestock insurance. Most of these internal microinsurance programs are linked to credit on a compulsory basis – in other words, members who borrow microfinance are required to purchase life insurance cover for the duration of the loan. Farmers who borrow livestock investment loans to purchase milk cows or buffalos are also required to insure the animals again death of the livestock due to named perils or under all risk individual animal insurance covers. In limited cases the MFIs also offer crop insurance. 6.37. In Bangladesh the Grameen Fisheries and Livestock Foundation (GMFP) which is part of the Grameen Bank (one of the oldest NGO/MFIs) has since 1999 operated a livestock mortality compensation scheme termed the Livestock Insurance Fund (LIF).87. The LIF program insures against death of the dairy cow where this is ―outside the control of the owner‖, and in effect it is an all -risks livestock mortality policy. Insurance is provided as part of an integrated package under which Grameen veterinary and extension staff assist in the pre inspection of the dairy cow or heifer and certify its health status. The animal is then routinely inspected and vaccinated by Grameen-trained veterinary staff and in the event of death the cause of loss is verified by the veterinary staff. These measures lead to greatly reduced livestock mortality rates and the ability to levy very low premium rates for individual animal mortality cover. The sum insured is equivalent to the amount of loan taken out to purchase the cow and premium is currently charged at a rate of 3 percent of the value of the loan. Coverage terminates once the loan has been repaid (usually over a maximum of two years). In addition, a fee of 2.5 percent of the value of the loan is levied to cover the cost of veterinary services, vaccinations, and technical assistance. The program has now operated for eight complete years during which a total of slightly over 7,000 dairy cows have been insured with an average mortality rate of 2.8 percent. The LIF liability is totally retained within GMPF, and the program does not carry any form of catastrophe reinsurance protection – it is therefore very exposed to catastrophe losses (for example due to flood and or epidemic diseases. (See Box 6.4. for further details of the GMPF Livestock-credit insurance scheme) Box 6.4. GMPF Community Livestock Development Project Livestock Insurance Fund Scope • The Livestock Insurance Fund is a component of CLDDP Livestock Development Program (1999) and compensates dairy cattle owners against mortality of their cows. • Livestock mortality insurance is compulsory for dairy farmers who purchase cows/heifers on credit using CLDDP microloans. • Insured animals: heifers, dairy cows, beef cattle (> 70 percent dairy cows) • Territorial scope: mainly northwestern Bangladesh Features • Community-based program • Coverage: animal mortality due to disease, accident, and any cause outside the control of the owner • Insurance is provided as part of an integrated package which includes, credit, technical assistance, vaccines and veterinary services, concentrate feeds and fodder, and milk marketing services. • Guarantee amount (sum insured): loan amount /replacement cost • Premium rate: 3 percent (previously 2.5 percent) of the loan money deducted at source • Service fee of 2.5 percent of value of loan is charged to Livestock Development Fund (LDF) in 87 See http://www.grameen-info.org/grameen/GrameenMotsho/index.html - 186 - order to contribute toward veterinary inputs (animal inspections, vaccinations etc) and to cover salaries of veterinary staff. Results C) 4,250 animals insured between 2001 and 2005 with a mortality rate of 3.8% and associated loss ratio of 75%. D) 7015 animals insured between 2006 and 2008 with a mortality rate of 1.1%. Overall loss ratio about 45%. Key Challenges • The Grameen livestock mortality product is not recognized under the Insurance Act 1938(2008). • The Grameen livestock mortality product is NOT REINSURED and is exposed to catastrophe claims (flood, cyclone, epidemic disease). Source: Authors, based on information provided by Grameen Bank March 2009. 6.38. In southern Kazakhstan MFIs are very active in lending to small urban and rural households and there may be opportunities to promote crop and livestock-credit insurance through the MFIs. In 2011 there are 1,769 registered MFIs in Kazakhstan of which 711 are actively lending to individuals and legal entities. The total loan portfolio of the MFIs amounts to: KZT 14.894 billion personal loans to 85,379 persons and legal loans which account for a much smaller amount of the total MFI loans equivalent to KZT 425 million to 110 legal entitles. Of the total of 711 active MFIs 570 (80%) are classified as urban MFIs and 141 MFIs (20% of total) are rural based and their main client base is small farmers. It is not known what proportion of the MFIs total loan portfolio is to farmers. There may, however, be potential to examine the role of protecting the MFIs loans to small and marginal crop and possibly livestock producers through similar credit-insurance covers to those developed by the MPFIs in southern Asia. Mutual Agricultural Insurance as a Solution for Small and Marginal Farmers in Kazakhstan 6.39. The Farmer Mutual Insurance Associations have been heavily promoted by GRK since 2008 and if their financial status could be strengthened the Mutuals might be the ideal institutional vehicle to underwrite Kazakhstan’s small and marginal crop and livestock producers. Sections 3 and 4 showed that there are currently about 38 Farmer Mutual Crop Insurance Associations (FMCIAs) underwriting the Obligatory LIC crop insurance scheme in Kazakhstan. Currently these Mutuals have very limited financial reserves and none are formally protected by any form of insurance or reinsurance protection. The individual Mutuals are therefore very exposed to catastrophe losses which exceed their reserves. In the event that claims exceed their reserves such as happened in 2010, Section 4 reported that the Mutuals had to pro rata down each claim settlement made to their members who incurred losses. International experience shows that when catastrophe claims occur which cannot be paid by the Mutual this often leads to the collapse of the Mutual. 6.40. If the Mutuals are to remain solvent and to underwrite crop and or livestock insurance for small and marginal farmers in Kazakhstan, ways of providing some form of catastrophe reinsurance protection must be developed. In the short-term it is unlikely that the private commercial insurance sector in Kazakhstan or international reinsurers would be willing to provide excess of loss reinsurance protection to the Mutuals. It is likely therefore that such a program would have to be offered through the Public Sector and given the FFSA‘s experience with administering the financial claims subsidies on the Obligatory LIC program, the FFSA would be best placed to administer some form of excess of loss program for the Mutual Crop Insurance Associations in Kazakhstan. - 187 - 6.41. Mexico has many years of years of experience with the operation of small farmer mutual crop and livestock insurance through the “Fondos de Aseguramiento” (Self Insurance Funds, SIFs) program which is reinsured by Agroasemex, the national Agricultural Reinsurance Company. The SIFs are legally registered small-scale crop and livestock producer mutual companies whose primary function is to access group crop and livestock credit. The Fondos program was originally conceived as a vehicle to provide small and marginal farmers access to credit: individually these farmers were too small to be eligible for credit, but collectively they could access group credit. As part of the agreement with the lending banks, crop and livestock credit provision was linked on a mandatory basis to crop and livestock insurance. 6.42. Since 1990 Agroasemex has provided advisory support and training to SIF members to form and register SIFs. Agroasemex also assists the SIFs to access short-term and medium-term production and investment credit, and provides technical assistance and training on crop and livestock insurance policy design and rating and in loss assessment procedures. Agroasemex closely monitors the activities of the SIFs on a seasonal basis. 6.43. The Fondos agricultural credit and mutual agricultural insurance program has proved extremely popular with farmers in Mexico. In 2005 there were 176 functioning SIFs in 24 Mexican states, of which 159 were crop producer SIFs and 17 livestock SIFs. Table 6.3 shows that in 2007 the SIFs insured more than 1 million hectares of crops and more than 4 million head of livestock (mainly cattle and pigs), generating MXN 647 million of premium (US$ 60 million)88. For crops, the basis cover is an individual grower MPCI cover which either insures against loss of the costs invested in the crop or against loss of yield: for livestock the policy is a herd-based catastrophe mortality and disease cover and which carries very low average rates. The crop and livestock products underwritten by the SIFs are eligible for federal government premium subsidies which average about 33% of full premium. Today after 18 years of operation, the Mexican SIF program is a major agricultural insurance program for small and marginal farmers. Table 6.3. 2007 Coverage by Self Insurance Funds, Mexico Insured Area (000Ha) / No. Total Sum Total Premium Premium Item Insured Animals Insured Premium Average Subsidy Subsidy (000) (MXN mio) (MXN Mio) Rate (%) (MXN Mio) (%) 1,034.3 8,927.8 588.2 6.6% 194.9 33% Crop Livestock 4,106.6 15,154.9 59.0 0.4% 19.0 32% Total 24,082.7 647.2 2.7% 213.9 33% Source: Agroasemex 2008 6.44. A key feature of the Mexican SIFs is the Stop Loss Reinsurance protection provided by Agroasemex to each SIF. Under the agreement between Agroasemex and the SIFs, Agroasemex is responsible for setting the premium rates for each crop and livestock program and then for the provision of Stop Loss Reinsurance protection. The SIF is entitled to deduct 25% of original premium to cover its administration and operating, A&O expenses. The SIF then retains an average 70% of the premium net of A&O expenses, equivalent to 52.5% of gross premium to settle retained claims. The remaining average 22.5% of gross premium is paid to Agroasemex as stop loss reinsurance premium (Figure 6.4). 88 Agroasemex 2008: Sistema Nacional de Aseguramiento al Medio Rural: Informe Final de Operaciones al cierre del ejercicio 2007. - 188 - 6.45. In the event of a claim, the individual SIF is responsible for settling claims up to a loss ratio of 52.5% plus any claims reserves held over from previous years. Any claim excess of this level is reinsured by Agroasemex. Agroasemex, in turn, purchases Stop Loss Retrocession protection on the Fondos program from international reinsurers. In any underwriting year if the SIF generates an underwriting surplus (profit), 30% of the surplus must be added to a Special Claims Reserve for Catastrophe events and the remaining 70% is allocated to a Social Fund which may be divided among the SIF members to invest in income generating activities or to contribute towards crop and livestock insurance premiums. Figure 6.4. Agroasemex Stop Loss Reinsurance of Self Insured Funds Allocation of Original Gross Premium 100% Agroasemex Stop Loss Reinsurance (22.5% of Gross Premium) 77.50% Allocation of Underwriting Surplus SIF Retained Claims  Special Claims Reserve, Catastrophe Losses 30% Surplus for the Social Fund 70% (52.5% of Gross Premium) 25% SIF Adminstrative Expenses (25% of Gross Premium) Source: Authors 6.46. Many of the features of the Mexican Self-Insurance Funds program and the lessons learned over the past 20 years are relevant to the future expansion of Mutual Agricultural Insurance in Kazakhstan. One of the key differences is that Kazakhstan does not have a public-sector organisation like Agroasemex to provide a combination of technical assistance and most importantly, reinsurance protection, to the Mutual insurers. This is a major issue which will need to be addressed under the proposed changes to introduce a more market-based agricultural insurance system based on a public- private partnership into Kazakhstan. One option for GRK to consider would be to use the FFSA as its implementing agency to provide a combination of technical support to the Mutual crop insurance associations and most importantly to provide formal excess of loss compensation protection to the mutuals along similar lines to the Agroasemex model. As a starting point to introducing such a program in Kazakhstan the Mutuals would need to be effectively regulated and would need to adopt actuarially determined rating and standardized loss assessment systems and procedures etc. - 189 - Identification of Operational Linkages to Bundle Programs 6.47. Agricultural insurance on its own is not a solution. Agricultural insurance can contribute towards the stabilization of agricultural production and farm incomes in times of major production loss and also to the modernization of agriculture through its ability to leverage access to credit thereby enabling farmers to purchase production-enhancing technology. However, agricultural insurance cannot be effective if it is provided in isolation. It should be promoted only when other essential agricultural services, including training and extension, the timely availability of inputs (seeds, fertilizers, pesticides) and efficient marketing channels for agricultural outputs, are in place. (Mahul and Stutley, 2010). 6.48. Agricultural insurance is only one tool to mitigate the risks of agricultural production, finance, and supply chain relationships. Therefore, other measures and complementary investments are needed to ensure risk is comprehensively managed and the value of insurance realized. In addition, without linking these insurance programs explicitly to finance, such as bundling the insurance with agricultural production loans or inputs, many farmers will lack both the capital to pay the insurance premium and sufficient incentive to use scarce resources on risk management. Placing these products within complementary systems with broader linkages can also facilitate simpler contract design, as other mechanisms can deal more efficiently with the non insurable risks. 6.49. Experience shows that bundling agricultural insurance with rural credit provision and input supplies could offer major advantages. The bundling of crop insurance with credit and input supplies has been shown to provide a win-win situation for farmers, lending institutions and insurers alike. The farmer gains access to seasonal crop credit, lending institutions are more willing to lend to small farmers because their loans are protected by crop insurance and the insurer benefits from: (a) reduced anti- selection, which in turn reduces the need for pre-inspections; (b) the reduced costs of marketing crop insurance; and (c) the insurance uptake and spread of risk is much better than would normally be achieved under a purely voluntary programme. Malawi is an example of a bundled WII and credit and input supply and output marketing programme that is showing early promise. 6.50. In many parts of the world public or private sector credit provision to agriculture is protected by a compulsory crop insurance policy (crop-credit insurance). The 2009 World Bank survey of agricultural insurance provision identified that in 11 percent of the surveyed countries public- or private- sector credit to agriculture is protected by compulsory insurance cover (Mahul and Stutley 2010). Examples of compulsory crop-credit or livestock-credit insurance schemes in Asia include India, the Philippines, Nepal, and Bangladesh. From an insurer‘s viewpoint there are major advantages to automatic or compulsory crop-credit insurance: (i) anti-selection is reduced, (ii) there is less need for pre- inspections, (iii) the costs of promoting and marketing the agricultural insurance program are reduced, and (iv) the insurance uptake and spread of risk and premium volume is generally much higher than under a purely voluntary program. 6.51. There are advantages for a scheme involving small farmers to be compulsory rather than voluntary, unless other circumstances allow the insurer to avoid adverse selection and high administrative costs. Even with a compulsory scheme there must be worthwhile incentives built in to counter moral hazard. Clearly operating an insurance scheme together with a credit programme can offer the level of control required by insurers reflecting the common interests of banks and insurers – if insurance is not taken out by the farmer then he will not be eligible for a loan. Loan applicants would also normally go through an initial appraisal procedure which will assist in evaluating the management potential of the farmer (Dick, 1999). 6.52. Where agricultural credit and insurance are linked there is a potential for the bank or MFI to reduce its interest rates to the extent that climatic or natural risk exposures have been transferred to - 190 - the insurance policy. The Malawi weather based crop insurance program and the Mongolia livestock index based insurance program are examples where the lending banks have reduced their interest rates to those producers who agree to purchase drought index insurance. 6.53. Wherever possible agricultural insurance should be demand-led and the ideal situation is for voluntary agreement to be reached by farmers and service providers to bundle input supply, credit and agricultural (crop) insurance. The bundled approach is much more acceptable to farmers than compulsory linkage of credit provision and insurance and offers a potential win-win situation for all parties. The farmer has timely and easy access to inputs of seeds and fertilizers and credit while the input supplier‘s and credit provider‘s financial exposures to climatic-induced crop failure and potential non- repayment is protected. 6.54. In Kazakhstan one of the key objectives of the obligatory crop insurance law of 2004 was to link crop insurance with the supply of credit, subsidized seeds, fertilizers and other government support programs. The obligatory approach to crop insurance is, however, somewhat different to the voluntary bundled approach. It is not possible to report on whether the obligatory scheme in Kazakhstan has been successful in improving farmers‘ access to seasonal crop production credit and to the subsidized government support programs. Conclusions 6.55. There is no single crop insurance product that is best suited to the needs of individual small and marginal farmers in Southern Kazakhstan. If crop insurance is to be developed for this category of farmers, it will be necessary to study the risk exposures on a crop by crop and a location by location basis and to select the most suitable product for that situation. For example crop-hail insurance may be suitable for small cotton producers while AYII might work best with rain-fed wheat. 6.56. For subsistence farmers in Southern Kazakhstan, there may be potential to design macro-level (per Rayon or Oblast) catastrophe index products along similar lines to the Mexican FAPRACC program. Such programs could be developed both for crops and livestock. 6.57. A more detailed feasibility study will be required with the small-holder farming sector in Southern Kazakhstan in order to provide guidelines and recommendations for the future development of appropriate crop and livestock risk transfer solutions. - 191 - Bibliography Agricultural Insurance in Developing Countries: Challenges and Opportunities. Brazilian Experience in the Field of. (2010). GFDDR: 9th Meeting of the Consultative Group (p. Ministerio de Desenvolvimento Agrario do Brasil). Washington, DC: GFDDR. Agroasemex 2008: Sistema Nacional de Aseguramiento al Medio Rural: Informe Final de Operaciones al cierre del ejercicio 2007 AGROASEMEX. (2009). PROGRAMA DEL SEGURO PARA CONTINGENCIAS CLIMATOLOGICAS: INFORME DE AVANCE AL CIERRE DEL EJERCICIO FISCAL 2009. Querétaro, México: AGROASEMEX. S. Baisholanov (2011), An agro-meterolological risk assessment for cereal crop production in Kazakhstan (prepared by Saken Baisholanov, former Head of Agro Meteorology Department, Kazhydromet) Bora, B. (2010) Subsidised Agricultural Insurance in Turkey (TARSIM). Available from: http://www.mapfre.com/mapfrere/docs/html/revistas/trebol/n54/docs/Articulo2en.pdf Burt, C. C. (2007). Extreme Weather; A Guide and Record Book. New York: Norton. Celaya del Toro, V. (2010). El uso de los seguros catastróficos para atender daños por contingencias climatológicas en el sector agropecuario de México. XI Congreso de ALASA:“El SEGURO AGRÍCOLA EN AMÉRICA. LOGROS Y DESAFÍOS”. Cartagena de Indias, Colombia: ALASA. CJSC (2003) Tariff Calculation Methodology for Compulsory Crop Insurance in the Republic of Kazakhstan. Report prepared for the Ministry of Agriculture of the Republic of Kazakhstan, Kazakh Actuarial Center, CJSC, November 18, 2003 FAO and World Bank. (2001). Farming Systems and Poverty: Improving Farmers' Livelihoods in a Changing World. Rome and Washington, DC: FAO. GFDRR. 2010. National agricultural insurance scheme in India: Making insurance markets work for farmers. Global Fund for Disaster Risk Management (GFDRR), The World Bank, Washington, DC Hartell, J., H. Ibarra, J.R. Skees, and J. Syroka, (2006). Risk Management in Agriculture for Natural Hazards. Rome: ISMEA, May 2006. Hazell P. B. R.(1992). The appropriate role of agricultural insurance in developing countries. Journal of International Development: Vol. 4, No.6, 567-581. Hazell, P., Pomareda, C., & Valdez, A. (1986). Crop Insurance for Agricultural Development. Baltimore MD: The John Hopkins University Press. Iturrioz, R. (2009). Primer Series on Insurance No.12: Agricultural Insurance. Washington, DC: The World Bank. Iturrioz, R. (2010). Seguro Agropecuario en América Latina : Situación Actual, Desafíos y Oportunidades. XI Congreso de ALASA –El seguro agrícola en América. Logros y desafíos. Cartagena de Indias, Colombia: ALASA. - 192 - Mahul, O., & Stutley, C. (2010). Government Support to Agricultural Insurance: Challenges and Options for Developing Countries. Washington DC: The World Bank. Munich Re Group. (2009). World Map of Natural Hazards. Munich: Münchener Rückversicherungs- Gesellschaft. Rosema, A., M. de Weirdt, S. Foppes, and C. Wilczok, (2010). FESA Micro-Insurance - Methodology, Validation, Contract Design. Delft, The Netherlands: EARS. Skees, J.R., J.R. Black and B.J. Barnet (1997). Designing and Rating an Area Yield Crop Insurance Contract. American Journal of Agricultural Economics 79 (May 1997) . Skees, J., (2009) A Policy Vision for Developing Agricultural Insurance in Vietnam The Economist Intelligence Unit Limited. (2011). Country Data- Annual Time Series. London UK: The Economist. Uçak H., and A. Berk (2009). Structural change in Turkish agricultural insurance policy and recent developments. http://www.piu.org.pl/public/upload/ibrowser/ Wiadomosci%20Ubezpieczeniowe /numer2/WU2_09ucakberk.pdf WFP/IFAD (2010). The Potential for Scale and Sustainability in Weather Index Insurance for Agriculture and Rural Livelihoods, by P Hazell, J. Anderson, N. Balzer, A. Hastrup Clemennsen, U. Hess and F. Rispoli. Rome March 2010 The World Bank. (2005). Disaster Risk Management Series 5. Natural Disaster Hotspots: A Global Risk Analysis. Washington D.C: The World Bank. World Bank, (2005). Managing agricultural production risk –Innovations in Developing Countries. Agriculture and Rural Development Department, ARD, The World Bank, Washington, D.C The World Bank. (2010). Kazakhstan: Public Expenditure and Institutional Review for the Agricultural Sector. Washington DC: The World Bank. The World Bank. (2010). World Development Indicators . Washington D.C.: The International Bank for Reconstruction and Development / The World Bank. USDA-FAS. (2010). Kazakhstan: Conmodity Inteligence Report. - 193 - Annexes - 194 - Annex 1: Spring Wheat Production Systems in Kazakhstan and Crop Risk Assessment 1. This annex provides a brief overview of Kazakhstan crop production (focusing mainly on spring wheat, the country‘s most important agricultural commodity). The annex is based on information from a variety of sources, including the Ministry of Agriculture, the National Statistics Agency, Kaz- Hydromet, Commodity Intelligence Reports, the Foreign Agriculture Service of the United States Department of Agriculture (FDAS-USDA), and interviews with farmers, farm directors and agricultural officials in Kazakhstan. 2. The annex provides a comprehensive assessment of the production risks affecting spring wheat in 8 selected Oblast in North Kazakhstan. The first section of the annex starts with the description of the main features of agricultural production in Kazakhstan and analyzes the evolution of crop production and yields for spring wheat crops in the country in the crop years from 1994 to 2010. Then, it assesses the key risk exposures and their impact on spring wheat crop production. The annex then describes the crop portfolio risk model used for the risk assessment and explains the main findings of this analysis, including the calculations of the average loss cost and the maximum probable losses. Features of agricultural production in Kazakhstan 3. Kazakhstan is an important producer and exporter of high-quality wheat. Average annual production is about 13 million tons, but output is highly dependent on weather and in recent years has fluctuated between 10 and 17 million tons. Between 2 and 8 million tons is exported annually, mainly to destinations in Europe (including Russia and Ukraine), Northern Africa, and Central Asia. Kazakhstan also produces around 2 million tons of barley and a small amount of oats, corn and rice, but wheat is by far the country‘s most important agricultural commodity. The production of oilseeds (sunflower seed and rapeseed) is increasing but total oilseed output remains well below 1.0 million tons. The country also grows a small amount of cotton in Southern Kazakhstan, with annual lint output at around 100,000 tons. 4. The country is endowed with land resources suitable for agricultural production. According to the agricultural census conducted in 2006, 61 percent of Kazakhstan‘s agricultural land (estimated in 76.5 million hectares) is devoted to permanent pasture, and 32 percent (24 million hectares) is classified as arable land which is systematically devoted to the production of field row crops. Of the remaining 7 percent of agricultural land, 3 percent is used for hay production and 4 percent is ―long fallow‖ (indicating potentially arable land that has remained uncultivated for at least several consecutive years). Two-thirds of the arable land in the country (approximately 18 million hectares) is devoted to grain production. The total sown area, including grains, forage crops, oilseeds and cotton crops, and food crops decreased sharply during the late 1990‘s due to the contraction of grain and forage-crop areas. Grain area began to rebound in 2000, and by 2010 had grown by 40 percent from the 1999 level, while forage area essentially stabilized in the early 2000‘s after a 10-year reduction. 5. The agricultural sector is highly heterogeneous in terms of farm structure and productivity . The Northern region of Kazakhstan is dominated by larger farms specialized in crop production, whereas smaller mixed farms, which include substantial meat and dairy production for the domestic market, predominate in the South. The total agricultural land area increased by about 6% between 2003 and 2007. The share of agricultural land farmed by agricultural enterprises has declined from 59% to 50% during the period 2003-2007, while land farmed by commercial farms has increased from 41% to 49%. Individual farmers have reduced their arable land slightly (by 0.14 million ha) while increasing their hayfields and pastures (by 8.47 million ha) between 2003 and 2007. In contrast, agricultural enterprises have increased their arable area by 0.92 million Ha and decreased their hayfields and pastures by 5.14 million Ha. Household plots, accounting only for 1 percent of the total agricultural land, produced a consistent and high percentage of total agricultural output - 50% between 2003 and 2007. This was mainly in the form of meat, of which they produced over 80% of output, and milk, of which they produced over 90% of output. - 195 - Agro-holding companies play a large role in Kazakhstan agriculture. In Kostanay Oblast over 40 percent of the agricultural area is held by the four largest holding companies. The largest holding company controls 900,000 hectares in Kostanay– 20 percent of the total sown area in the oblast – and owns 70 percent of the grain elevators. Table A1.1 shows the Farm Structure as of 2004. Table A1. 1 Farm Structure (2004) Commercial Subsistence Agric. Enterprises Individual Farms Household Plots Number 4,600 (0.2 %) 156,000 (7.2 %) 2,000,000 (92.6%) Labor Force (‗000) 326 (14%) 280 (11.7 %) 1,782 (75%) Agricultural Land (‗000 ha) 43,420 (56 %) 34,228 (44.3 %) 325 (0.4 %) Arable Land (‗000 ha) 12,921 (59 %) 8,816 (41.2 %) 231 (1 %) Arable land area/ farm 2,808.9 ha 56.5 ha 0.1ha No. employees/farm 70.9 1.8 0.9 Gross Agricultural Output (bn. 171 (24 %) 178 (25 %) 349 (50 %) Source: KZT) National Statistics Committee. 6. The main agricultural production areas are situated in the northern parts of the country. Kazakhstan consists of 14 administrative territories, or oblasts. About 75 percent of the country‘s wheat is produced in three oblasts in north-central Kazakhstan: Kostanay, Akmola, and North Kazakhstan. Spring wheat occupies 95 percent of the total wheat area in Kazakhstan and virtually all of the wheat in the three north-central oblasts. Minor grains include spring barley and oats (which are grown in the same region as spring wheat), winter wheat (in southern Kazakhstan areas.), and rice (southern Kazakhstan, mostly in Kzyl-Orda oblast). 7. Wheat is the main crop in the country. The total area planted with wheat according to the National Statistics Service (NAS) is 14.7 million hectares. Oilseed area has nearly doubled in the past five years but still accounts for only about 6 percent of the country‘s total crop area. For instance, the area planted with sunflower has increased almost 90 percent from 450,000 hectares in 2004 to over 850,000 hectares in 2010. Sunflowers are grown mostly in eastern Kazakhstan. Rapeseed area has increased as well, from only 15,000 hectares in 2004 to about 200,000 hectares in 2009. Rapeseed is grown in north- central Kazakhstan. Yields for both sunflower and rapeseed are consistently low, typically between 0.5 and 0.7 tons per hectare. Cotton is grown exclusively in SKO. Cotton yields are lower than in neighboring countries and the Kazakh cotton production is hampered by the deterioration of the irrigation infrastructure. The area planted with cotton doubled between 1997 and 2004, reaching a record level of 224,000 hectares and then contracted nearly as sharply over the following six years. Estimated total area planted with cotton in Kazakhstan in 2010 is 137,000 hectares, the lowest in over ten years. 8. Kazakhstan is a zone which is risky for agriculture. The soils of north-central Kazakhstan are highly variable. Fertile chernozem and kashtan soils lie adjacent to highly salty solonchak soils, which are totally unsuitable for grain production. The flat, open land lends itself to large-scale agriculture. Individual fields frequently measure over 400 hectares (1,000 acres). Annual precipitation in the main crop production areas of north Kazakhstan is very low. Total annual rainfall in north Kazakhstan averages from 280 millimeters per year in Aktobe Oblast to 400 millimeters in NKO. The rainfall in North Kazakhstan is distributed throughout the year with a peak during the months of June, July, and August. Historically, Kazakhstan grain production has suffered from serious drought two out of every five crop seasons. As a result, yields and production are marked by frequent and sharp year-to-year fluctuations Map A1.1 shows the monthly distribution of rainfall for 9 selected Oblast in Kazakhstan. - 196 - Map A1.1. Kazakhstan: Monthly Rainfall distribution for selected Oblasts in Kazakhstan Source: Authors based on rainfall data provided by Kaz-hydromet. 9. Owing to the country’s dry climate, the quality of Kazakhstan wheat is high. In years with reasonably favorable weather and average yield, about 75 percent of the wheat crop will probably qualify as milling quality. In general, grain quality tends to be higher in drought years; quality typically increases as yield decreases. For instance, during the drought year 2004, 90 percent of the wheat qualified as milling grade. Wheat quality is highest in the more southern (and drier) production regions of the main production zone in North Kazakhstan. Protein content typically reaches 14 percent in Akmola and southern Kostanay oblasts. In the northern tier of NKO, protein content rarely exceeds 11 percent. 10. The cereal crop season in Kazakhstan is from May to September. Spring grain planting typically begins in mid-May. Of the spring grains, oats are sown first, followed by wheat, then barley. Planting typically is finished by early June. The crops advance through the reproductive stage during mid-July, when temperatures climb to their highest levels and grains are most vulnerable to heat stress. Although barley is planted later than wheat, it is harvested earlier. The cereal harvest begins in late August and continues through October. Box A1.1 presents the crop calendar for the three main cereals planted in the main grain production areas in Kazakhstan. - 197 - Box A1.1. Kazakhstan: Crop Calendar for North Central Kazakhstan Region Source: Foreign Agricultural Service – United States Department of Agriculture (FAS-USDA). 11. The four field rotation is the most popular form of rotation in Kazakhstan. The dominant crop rotations for fields under conventional tillage systems are essentially unchanged from Soviet times, except that the share of wheat relative to barley and oats has expanded as planting decisions have become increasingly market-driven. The four-field rotation is the most popular, and typically includes two consecutive years of wheat followed by one year of barley, oats or sometimes an oilseed crop depending on the location. Some enterprises have eliminated barley and oats from the rotation and plant nothing but wheat. Conventional crop rotations include a year of fallow land. The purpose of the fallow year is to preserve soil moisture. Two consecutive years of wheat almost always follow the fallow year, and the first wheat crop enjoys the benefit of increased subsoil moisture. Assuming normal weather, yield drops by 15 to 20 percent for the second wheat crop. 12. The introduction of moisture saving technologies, such as the reduced tillage technology, is becoming very popular in Kazakhstan. One of the most interesting developments in Kazakhstan agriculture in recent years has been the emergence and growth of reduced-tillage technology. There is no strict definition of what constitutes reduced tillage or minimum tillage, but the term typically implies the elimination of moldboard plowing and an increased reliance on chemical weed control. Under a strict no- tillage system, neither plowing nor secondary tillage is used. According to the MoA, reduced tillage is employed on almost 60 percent of the sown grain area in 2010, including 1.3 million hectares under no- tillage. . The advantages of the introduction of reduced tillage from the farmers‗perspective are twofold. First, reduced tillage preserves soil moisture and reduces (but certainly does not eliminate) the risk of yield loss in the event of drought. Second, the adoption of reduced-tillage enables grain farmers to eliminate the fallow year from the typical four-year crop rotation and to plant a crop every year, which substantially increases the productivity of the field. For enterprises with old machinery the adoption of a reduced-tillage system typically entails the replacement of outdated seeders with newer units designed to accommodate the technology. This machinery upgrade can be prohibitively expensive, especially for small enterprises or family farms. The MoA is interested in the promotion of the reduced tillage practices and has set a nation-wide goal that up to 80 percent of the cropped areas should be produced under reduced tillage practices within five years. In order to encourage farmers to increase the use of reduced tillage practices, the MoA offers a higher direct subsidy for no-tillage wheat than for conventional-tillage wheat. 13. The Agricultural Machinery fleet is becoming outdated. Data from the National Statistics Agency indicate that inventories of agricultural machinery have declined significantly over the past 20 years. Furthermore, a high portion of Kazakhstan‘s current fleet – including 77 percent of its tractors and 59 percent of its combines – was over 15 years old at the time of the 2006 agricultural census. The statistics are somewhat misleading, however, because the data certainly include machines that are no longer in use. As is the case in other ex-Soviet Union Republics, the overall efficiency of Kazakhstan‘s machinery fleet is improving due chiefly to the replacement of aging grain-harvesting combines with new - 198 - equipment. This process of improvement is being led by the Agribusiness enterprises (also termed Production Enterprises, PE in this report). Agribusiness enterprises have the financial capacity to purchase new machinery and, in general, have very updated tractors and combine harvesting fleets. Conversely, for most of the Commercial Farmers (CE) who are mainly small enterprises or family farms and who are not sufficiently financially strong, to upgrade the machinery fleet can be prohibitively expensive. The existing gap in terms of affordability of agricultural machinery will contribute to an increase of the existing differences in terms of crop productivity between the agribusiness enterprises (PE) and the commercial farmers (CF). Spring Wheat Crop Yields in Kazakhstan 14. According to the ARKS, during the crop year 2010, the spring wheat cultivated area in Kazakhstan almost reached 13.6 million hectares. Out of the 13.6 million hectares planted with spring wheat in Kazakhstan, 80 percent – approximately 11 million hectares- were planted in central North Kazakhstan (Akmola, Kostanay, and NKO). The remaining 20% of the area planted with spring wheat is distributed through North Kazakhstan (i.e. WKO, Aktobe Oblast, Karaganda Oblast, EKO, and Pavlodar Oblast). Map A1.2 shows the main spring wheat crop production areas in the country. Map A1. 2: Main Spring Wheat Crop Production areas in Kazakhstan Source: Authors from National Institute of Statistics 15. Spring wheat experienced a decline in the cropped area during the1990s and a rebound during the 2000s. The reason for the contraction in the spring wheat crop area during the 1990s was the loss of subsidies for crop production caused by the collapse of the ex-Soviet Union. During that period much land that used to be planted with spring wheat was set aside and converted to pasture production. Government subsidies were re-directed only for farms that reached certain thresholds of spring wheat yields. For instance, spring wheat fields that consistently failed to meet the threshold - typically - 199 - established at 6 to 7 centners89 per hectare against a national average of about 9 centners per hectare -- were taken out of grain production and converted to permanent pasture. The decline in grain area accelerated in the mid-1990 when shrinking livestock inventories caused feed-grain demand to plummet. During the period from 1994 to 1999, the total area with spring wheat shrank by 24 percent, or a reduction of approximately 2.5 million hectares. The area planted with wheat started to rebound in the 2000s. By 2010, the area planted with spring wheat had increased by 65 percent compared against the year 1999, an increase of 5.3 million hectares. The reason for the rebound in the spring wheat crop area was the expansion of government subsidies to agriculture. Figure A1.1 shows the evolution of the sown area of spring wheat in the 9 selected Oblasts in North Kazakhstan region. Figure A1.1. North Kazakhstan Region: Evolution of Spring Wheat Sown Area North Kazakhstan Region: Evolution of Spring Wheat Sown Area (in miillion hectares) 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 2004 2006 2008 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2005 2007 2009 2010 Source: Authors based on Agency of Statistics data. 16. Spring wheat crop yield performance has improved significantly in recent years. The spring wheat national average yield for the period 2006-2010 was 10.1 centners per hectare, or 69 percent higher than the spring wheat national average yield for the period 1994-1998, or 6.0 centners per hectare. The reasons behind such improvement in spring wheat yields are linked to the general improvement in spring wheat crop management practices. The improvement in the spring wheat crop management practices was fueled by the expansion of government subsidies that have contributed to higher and more stable wheat yields90. Three main spring wheat crop management improvements have driven the increase on spring wheat crop yields in Kazakhstan in the recent years. First, the introduction and spread of reduced-tillage technology which, according to MoA, is currently employed on more than 60 percent of the sown grain, including 1.3 million hectares under no-tillage. Second, the crop sector has also increased its use of 89 1 Centner equals 100 kilograms. 90 Beginning around 2002, government support for agriculture has increased significantly in the form of reduced prices for fuel, seed, fertilizer, and agricultural chemicals - 200 - fertilizers and agrochemicals. The application rates for mineral fertilizer increased nearly six-fold between 1999 and 2010, and continues to increase due in part to the subsidized prices. Last, but not least, the increase in the use of certified seeds has certainly contributed to the increase in spring wheat yields. Most of the farms and enterprises in Kazakhstan use only first-reproduction seed. Crop yields have increased faster in Agribusiness Enterprises than in Commercial Farms. While the share of agricultural land area farmed by Agribusiness Enterprises has declined, their share of output increased slightly from 23% to 27% between 2003 and 2007. In contrast, while the share of agricultural land farmed by Commercial Farms has increased, their share of total output declined slightly from 27% to 24%. This suggests a higher rate of improvement in productivity in agricultural enterprises than in commercial farms. Figure A1.2 shows the evolution of the sown area and yields of spring wheat in the 9 selected Oblasts in North Kazakhstan region. Figure A1.2 North Kazakhstan Region: Evolution of Spring Wheat Sown Area and Yields.. North Kazakhstan Region: Evolution of Spring Wheat Sown Area and Yields (in centner/hectare) 16.0 14.0 12.7 12.4 11.6 11.3 12.0 10.8 10.1 9.5 9.4 10.0 8.7 8.4 7.3 7.5 7.7 8.0 6.3 6.4 6.0 5.0 4.0 4.0 y = 1.6032ln(x) + 5.6195 R² = 0.252 2.0 0.0 2004 2006 2008 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2005 2007 2009 2010 Yield (Centner/he) Yield Commercial Farmers (Centner/he) Yield Agribusiness Enterprises (Centner/he) Log. (Yield (Centner/he)) Source: Authors based on Agency of Statistics data. 17. Spring wheat yield performance is uneven throughout the northern areas of the country . The rayons situated in the northern parts of North Kazakhstan Oblast (NKO) and Kostanay Oblast show the best performance in terms of spring wheat crops. In these areas, spring wheat yields, are on average higher than 12 centners per hectare and in certain rayons can yield above 14 centners per hectare. A second tier, comprising the rayons situated in east NKO, the northern parts of Akmola Oblast, and central Kostanay Oblast, shows average spring wheat yields that range from 8 to 12 centners per hectare. A third tier in terms of spring wheat productivity can be determined by grouping the rayons situated in western areas of Pavlodar Oblast, southern areas of Akmola Oblast, north of Karaganda Oblast, south of Kostanay Oblast, some rayons situated in north of Aktobe Oblast and north of WKO. In this third group of rayons/oblasts, the spring wheat crops average yield at rayon level is from 4 to 8 centners per hectare. The worst performance for spring wheat crops is observed in the southern rayons of West Kazakhstan Oblast (WKO), Aktobe, and Karaganda Oblasts, and the south-west of Pavlodar Oblast. In the rayons situated in - 201 - these areas, the spring wheat rayon average yields are below 4 centners per hectare. Map A1.2 summarizes the geographical distribution of spring wheat yields at rayon level throughout the 8 selected Oblasts in north Kazakhstan region. Map A1.2. North Kazakhstan: Average Spring Wheat Yields per Rayon from 2006-2010 (Centners/Hectare) Source: Authors based on spring wheat yield data provided by Agency of Statistics. 18. Spring wheat crop production is particularly risky in Kazakhstan. Kazakhstan has the highest yield variation (as expressed by the coefficient of variation, COV, in national average spring wheat yields91) of 29 percent of any major wheat producing country in the World, compared with 5 percent in the European Union, or 8 percent in Canada. The northernmost areas of the country are less risky for spring wheat production than the southern, western and eastern areas of the country. In the rayons situated in NKO, north of Kostanay Oblast and northwestern areas of Akmola Oblast, the coefficient of variation (CoV) of average annual spring wheat yields is less than 40 percent. In the rayons situated in south of Kostanay Oblast and eastern parts of Akmola Oblast, the CoVs of spring wheat yields show values between 40 to 50 percent. In Karaganda and western areas of East Kazakhstan Oblast (EKO) the CoVs for spring wheat are mostly between 50 to 60 percent, except for the rayons situated in the mountainous areas of EKO where the CoVs are between 40 to 60 percent. Pavlodar shows high CoVs for spring wheat production at rayon level: on average the CoVs for spring wheat production for the rayons situated in Pavlodar Oblast are between 50 to 70 percent. The oblasts situated in the western areas of the country show the highest level of risks for spring wheat production. For instance, in WKO the average CoV at rayon level for spring wheat is between 70 and 100 percent. Map A1.3 shows the distribution of the CoVs for spring wheat at rayon level in the 9 selected Oblasts in North Kazakhstan. 91 The COV is the standard deviation (SD) about mean annual yield divided by the mean yield and expressed as a percent. A COV of > 100% shows that the SD is larger than the mean yield, or in other words crop yields are highly variable. - 202 - Map A1.3. North Kazakhstan: Coefficients of Variation of Spring Wheat Yields at Rayon Level Source: Authors based on Agency of Statistics spring wheat data. 19. Spring Wheat yields are less volatile for Production Enterprises (PEs) than for Commercial Farmers (CFs). The less risky feature of spring wheat crops produced by PEs versus the risky conditions of the spring wheat crops produced by CFs is evidenced by the lower CoVs observed for spring wheat yields produced by PEs. The CoV analysis of spring wheat yields for the 17-year series from 1994 up to and including 2010 shows that the observed CoV for PEs is on average 10% lower than the spring wheat yield CoV observed for CFs. This fact is more noticeable in the main areas of production of Kostanay, NKO, and Akmola where the observed CoVs of spring wheat yields for PEs are 16, 27, and 25 percent, respectively, lower than those observed for CFs in these rayons. The main reason for the differences in terms of CoVs of spring wheat yields between PEs and CFs farmers is the introduction by the PEs of water efficient technologies like zero tillage that makes the crop perform better during the recurrent droughts in the region. Key risk exposures and their impact on spring wheat crop production. 20. Agricultural production in Kazakhstan is an extremely risky economic endeavor. A large portion of the risk associated with agricultural production in Kazakhstan is due to climate events. Drought is the most pervasive peril affecting crop production in Kazakhstan. Reasonably higher levels of agricultural productivity can be achieved during years of adequate rainfall, but the region is subject to frequent drought and is considered a zone of risky agriculture. Historically, Kazakhstan‘s agricultural production suffers from drought two out of every five crop seasons and the country is hit by severe droughts one in every six years. As a result, the agricultural value added in Kazakhstan is marked by frequent and sharp year-to-year fluctuations. Besides droughts, the occurrence of hailstorms and early - 203 - frost are also important perils affecting crop production in the country. Figure A1.3 shows the relation between the fluctuations92 in the annual agricultural GDP growth and the occurrence of drought events. Figure A1.3. Kazakhstan: Historic Agricultural GDP Growth and occurrence of Droughts Kazakhstan: Agriculture Value added (annual % growth) 29% 21% 17% 13% 7% 6% 9% 3% 2% -1% 0% -5% -3% -7% -6% -12% -21% -19% -23% -24% 1995 1991 1992 1993 1994 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Year reported with drought Source: Authors from World Bank. World Development Indicators (2010); The Economist Intelligence Unit (20011); Kazakh-Hydromet (2010) 21. Natural calamities have a strong impact on spring wheat production in Kazakhstan. Drought is the most pervasive peril affecting rain-fed crop production in northern Kazakhstan. Early frosts at the end of the cropping period are, for all regions considered, also an important risk. Other risks such as pests, diseases and hail play a lesser role in yield deviation. A detailed description of each of the risks affecting spring wheat crop production is presented below. 22. Spring wheat yields in Kazakhstan are highly influenced by the occurrence of droughts. Reasonably high yields can be achieved during years with adequate rainfall, but the country is subject to frequent drought and is considered a zone of risky agriculture. Historically, Kazakhstan grain production suffers from serious drought two out of every five or six crop seasons. As a result, yield and production are marked by frequent and sharp year-to-year fluctuations. The aggregate annual average spring wheat yields for the 8 selected Oblasts in Kazakhstan is highly correlated with the cumulated rainfall/snowfall index from January to September as shown by the correlation coefficient (r) of 0.73. The strong relationship between spring wheat average yields and cumulated rainfall between January and September 92 Caution should be exercised at the time of the analysis of the GDP shortfalls in the years 1993, 1994, and 1995. In the early 1990‘s, following the breakup of the Sovie t Union and the loss of massive government subsidies for State and collective farms and livestock enterprises, there was a sharp declination in agricultural output and productivity. Local agricultural officials began to set productivity thresholds for individual fields. Fields that consistently failed to meet the threshold -- typically 0.6 to 0.7 tons per hectare against a national average of about 0.9 tons per hectare -- were taken out of grain production and converted to permanent pasture. The decline in grain area accelerated in the mid-1990 when shrinking livestock inventories caused feed-grain demand to plummet, leading to a 75-percent drop in the planted area of barley between 1993 and 1999. During these six years, total grain area in Kazakhstan contracted at the rate of nearly 2 million hectares per year. - 204 - is also evidenced at Oblast level. In this regard, all the selected Oblasts in North Kazakhstan, except NKO and Karaganda Oblast, show correlation coefficients between annual average spring wheat yields and cumulated rainfall/snowfall between January and September of each year that are above 0.65. Figure A1.4 shows the relationship between spring wheat annual average yields and the total cumulated snowfall/rainfall from January to September for the period 1994 to 2010. The yellow arrows indicate the years with cumulated precipitation between January and September which are below the first tertile. Figure A1.4. North Kazakhstan: Relationship between Spring Wheat Yields and Total Cumulated Snowfall/Rainfall from January to September Source: Authors from Kaz-hydromet and Agency of Statistics. - 205 - 23. Spring wheat farmers in Kazakhstan have suffered severe losses due to the occurrence of drought events. Between 1994 and 2010, spring wheat farmers in Kazakhstan suffered significant crop losses on six occasions namely in: 1995, 1996, 1997, 1998, 2004, and 2010. During the crop year 1995, a drought affected the whole spring wheat crop production areas in north Kazakhstan, being particularly severe in Kostanay and Karaganda Oblasts. The losses due to this event in terms of gross value of production assuming sown area and prices as per the most recent 5-year average were KZT 74.4 billion (26 percent reduction with respect to the expected value). In the next year, 1996, drought also affected the country. On this occasion, a severe drought affected the western part of the country, particularly WKO and Aktobe oblasts. At the same time, a moderate drought affected the eastern areas of the spring wheat production zone, particularly in Pavlodar and EKO. The losses due to the occurrence of these drought events during 1996 amounted, in terms of gross value of production assuming sown area and prices as per the most recent 5-year average , to KZT 46.4 billion (or 15 percent reduction with respect to the expected value of production for spring wheat for 1996). In 1997, the spring wheat production areas in Kazakhstan were affected by drought again. This time the reduction in the total value of production was only 4 percent of the expected value (KZT 13.4 billion). In 1997 the epicenter of the drought was located in Karaganda Oblast but also affected areas of Akmola and Pavlodar Oblast. 24. The year 1998 was one of the worst years in terms of drought incidence in spring wheat crop production. During this year, all the spring wheat crop production areas in the country, except for EKO, were affected by drought. This drought was particularly severe in WKO, Aktobe, Kostanay, and Akmola Oblast; but also affected NKO, Pavlodar and Karaganda oblasts. This event caused a major reduction in the total spring wheat crop value of production, which suffered a reduction of 51 percent. The shortfall in terms of value of production, assuming spring wheat sown area and prices as per the most recent 5-year average, amounted to KZT 179.4 billion. 2004 and 2005 were also dry years. During those years the drought affected Kostanay, NKO, EKO Akmola, (only in 2004), WKO (only in 2005), and Aktobe (only in 2005). The total amount of losses in terms of spring wheat total value of production, assuming spring wheat sown area and prices as per the most recent 5-year average, amounted to KZT 74 billion for 2004 (18 percent value of production shortfall with respect to the expected value for this year) and KZT 49.1 billion for 2005 (12 percent value of production shortfall with respect to the expected value for this year). 2008 was very unfavorable for spring wheat production due to autumn, spring and summer dryness.. 25. A devastating drought affected the main spring wheat crop production areas in Kazakhstan during 2010. This drought affected mostly the oblasts situated in central north Kazakhstan (i.e.: Kostanay, Akmola Karaganda, NKO, and Pavlodar), while the western oblasts (WKO and Aktobe) and the easternmost Oblast (East Kazakhstan) were not so severely affected by this event. Since this drought affected the heart of the spring wheat crop production area in Kazakhstan, it caused a huge loss in terms of gross value of production for the crop. The actual gross value of production for spring wheat for 2010, assuming spring wheat sown area and prices as per the most recent 5-year average, was KZT 274.4. This value was 37%, or KZT 158.5 billion, lower than the expected gross value of production (based on an expected yield of 10.2 centners per hectare) of KZT 432.7 billion. Figure A1.5 shows the national spring wheat production losses due to flood from 1994 to 2010. The details for this estimation are presented in Appendix 1.B. - 206 - Figure A1.5: Kazkahstan: Spring Wheat. National Losses in terms of Gross Value of Production due to Droughts (1994 – 2010) North Kazakhstan : Evolution of Spring Wheat Gross Value of Production (GVP) and Yields 600 14.0 500 12.0 Yield (Centner/hectare) 10.0 GVP (KZT billion) 400 8.0 300 6.0 200 4.0 100 2.0 0 0.0 2006 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2007 2008 2009 2010 Crop Year Actual GVP (KZT billion) GVP Losses (KZT billion) Actual Yield (Centner/he) Expected Yield (Centner/he)/ Expected GVP (KZT billion) Source: Authors, from ARKS 26. Early autumn frost can be a problem for spring wheat crop production in some areas of Northern Kazakhstan. The occurrence of early autumn frosts during late August and the beginning of September may cause damage to wheat crops that were sown late in the season. Early autumn frost damage affects wheat crops when they are in milk grain or dough phenology stages prior to harvest. The damage occurs when the temperatures fall below – 2 Celsius degrees for more than two hours. Losses due to early autumn frost on wheat production can be moderate to severe. Historical records of monthly absolute minimum temperatures indicate that the return periods for frost below – 2 Celsius degrees during the last week of September is once in 20 years for some locations in Akmola, Karaganda, Kostanay, and Pavlodar. The probability of having an early frost increases dramatically for each week beyond the last week of September. 27. Hail is reported to be a moderate to severe problem in spring wheat in some parts of the country. Many parts of Kazakhstan experience hail in early and mid summer. According to the map of world hail incidence (Burt, 2007), Kazakhstan can be divided in three areas based on the hail occurrence (See Map A1.4). The first area is the most hail prone in the country. The average annual number of hailstorms for this area is estimated in the range of 4 hailstorms to 6 hailstorms per year. The area comprises the central-east rayons of Kostanay Oblast, the southern rayons of NKO, all the rayons in Akmola Oblast, and the northwestern rayons of Pavlodar Oblast. The second area is on the mezzanine level in terms of hail frequency. The average annual number of hailstorms for the second area is estimated in the range of 1 hailstorm to 3 hailstorms per year. The second area comprises Kostanay Oblast (except for the central east rayons), northern rayons of NKO, southern rayons of Pavlodar oblast; all the rayons in Karaganda oblast; and the eastern rayons of Aktobe oblast. The remaining areas of the country show low frequency of occurrence of hailstorms. In such areas, the average annual number of hailstorms is less than one. Hail in Kazakhstan is associated with major rain storms. The hail season in Kazakhstan goes from - 207 - May to September, with a peak during the months from May to July. Hail is a localized peril which tends to cause severe damage in wheat from the stage of stem elongation in mid-July until crop maturity and harvest in late August/early September. The period in which spring wheat is most vulnerable to hail damage does not match the period in which there is a greater frequency of hailstorms in North Kazakhstan. As a result although hail is a recurrent phenomenon, because hail occurs relatively early in the crop season, it does not cause much losses in spring wheat crops and farmers do not perceive this peril as an significant risk that can affect their spring wheat crops. Table A1.2 presents the return periods in terms of years for the occurrence of hailstorms for the different months of the year. Map A1.4 shows the geographical distribution of the hail areas in Kazakhstan. Table A1.2. Kazakhstan: Monthly return period for the occurrence of hailstorms Recurrence Period for Hailstorms for each month throughout the year (in years) Oblast I II III IV V VI VII VIII IX X XI XII Kostanay 42 5 3 6 8 21 84 Akmola 5 8 5 13 63 NKO 7 5 11 5 7 Pavlodar 21 8 9 7 30 10 63 SKO 11 4 3 8 11 21 42 42 42 Source: Kazakh-Hydromet Hail frequency Statistics (1990-2010) Map A1.5. Kazakhstan: Average annual number of hailstorms Source: (Burt, 2007) 28. Crop pest and diseases are a serious issue for spring wheat crop production in Kazakhstan. Locust attacks are not infrequent in North Kazakhstan and the country has suffered recurrent crop and pasture damages from locusts. There are 2 main species of locusts in Kazakhstan (i) the ―Asian‖ locust which is not considered a major problem and (ii) the Italian locust ( Callitamus italicus) which is the most - 208 - common and dangerous pest found in nearly all of Kazakhstan: in 1999 the country experienced a severe outbreak of Italian locusts and as this species has a peak cycle of every 10 to 12 years, the next outbreak is expected in 2010 or 201193. The scale of the problem increased dramatically after independence (during1996–2001), when cessation of state subsidies for wheat production in the northern steppe lands led to the abandonment of up to one third of the former wheat lands. The resulting mosaic of weedy fields, pastures, and bare ground provided ideal breeding grounds for locusts. In 1999, more than seven million hectares of land were invaded by Italian locusts and 220,000 hectares of crops were destroyed causing a total damage equivalent to US$ 15 million. In 2008, again more than 200,000 hectares of crops were destroyed in SKO.94 The Government was forced to conduct a massive chemical control campaign throughout the country with assistance from Food and Agriculture Organization (FAO). 29. Spring wheat fungal leaf diseases are also a problem for spring wheat production in Kazakhstan. Despite the dry climate, cultivation of susceptible varieties results in epidemics of leaf rust on average in 1 year out of two or three years, affecting over a million hectares with yield losses of up to 15–35%.95 Monitoring by the Kazakh Research Institute of Crop Protection (Koyshibayev, 2002) suggests that from 1970 to 2002, there were 14 local leaf rust epidemics. Excess moisture and high humidity during the month of July generates the conditions for the development of rust. Most of the wheat cultivars planted in the region are susceptible to leaf rust and several resistant lines and new varieties have been tested in trials only recently 96. Stripe rust, caused by Puccinia striiformis f. sp. tritici is considered the most important disease of wheat in Central Asia and the Caucasus (CAC). Although stripe rust has been present in the region for a long time, it has only become a serious constraint to wheat production in the past 10 years. Rust attacks were observed in northern areas of Kazakhstan in 2007 and in 2009. Key risk exposures and their impact on spring wheat crop production. 30. This section describes the spring wheat crop-yield risk assessment at the rayon level in 8 Oblasts including West Kazakhstan, Aktobe, Kostanay, Akmola, North Kazakhstan, Karaganda, Pavlodar, and East Kazakhstan. The principal objectives of the rayon crop-yield risk assessment are (i) to assist decision makers in assessing the spatial distribution of crop production values and (ii) to quantify the risk of crop production and yield loss for spring wheat crops in each of the 8 selected Oblasts. The section is based on the output generated by the Crop Risk Assessment Model (CRAM), which has been specially designed in the context of spring wheat crop production in Kazakhstan. The key underlying crop production, yield, and valuation data and assumptions on which the CRAM model for the eight selected Oblast in Kazakhstan is built include the following:  Selected crops: Spring wheat in the eight selected Oblasts in Kazakhstan for which rayon-level crop area, production and yield data are available for the past 17years, 1994 to 2010.  Cultivated area: In order to remove seasonal variations from the cultivated and harvested area in each rayon, the model takes the average sown area per rayon for each spring wheat crop year for 93 IRIN Asia 2007. Kazakhstan: Locust invasion in west under control officials say. Almaty 10 July 2007. http://www.irinnews.org/report?reoportid=73115 94 Locust in Kazakhstan. Presentation made Alex Latchininsky & Ramesh Sivanpillai. University of Wyoming 95 Reynolds M.P., J. Pietragalla, and H.-J. Braun, eds. 2008. International Symposium on Wheat Yield Potential: Challenges to International Wheat Breeding. Mexico, D.F.: CIMMYT. 96 Leaf rust of spring wheat in Northern Kazakhstan and Siberia: incidence, virulence, and breeding for resistance* A. Morgounov A D, L. Rosseeva B and M. Koyshibayev CAustralian Journal of Agricultural Research 58(9) 847–853 http://dx.doi.org/10.1071/AR07086 Submitted: 8 March 2007 Accepted: 8 June 2007 Published: 28 September 2007 - 209 - the past four seasons: 2007, 2008, 2009, and 2010. The model then assumes that the cultivated area has remained constant over the past four years.  Crop yields: the crop yields are based on the Agency of Statistics reported Rayon-level average yields (total production, in metric tons, divided by sown area - acres) for Commercial Farmers (CFs) and Production Enterprises (PEs). For the purposes of eliminating the effects of the increase in yield due to technology improvements (seed genetics, crop management practices, use of agrochemicals, etc), the 17-year historical yields have been de-trended and readjusted to an expected yield based on the most recent five-year average.  Crop output prices: the spring wheat price is valued at the 3-year average market farm-gate price for September (month of harvest). 31. Assessing yield losses and the value of losses for CRAM: The risk assessment model assumes that the losses occur when the actual average yield at rayon level falls short of the expected yield for the rayon, defined as the average yield for the most recent five crop years. In any year where the actual yield is below the rayon average expected yield, the amount of yield loss is calculated as a percentage of the expected yield to derive the pure loss cost (value of loss / gross value of production x 100 percent). The average pure loss cost is then calculated as a simple average over the 17 years of yield data. In summary, the CRAM uses a historical database of 17 years of yield data, adjusted by (i) the100 percent area losses to represent more accurately the average yields sown area-basis and (ii) technological improvements in crop yields in all the eight selected oblasts in Kazakhstan in order to establish the expected value of losses and to estimate probable maximum losses for the national portfolio. National Aggregate Crop Values 32. The total values at risk (VAR) for the spring wheat portfolio amount to KZT 452 billion (approximately US$ 3 billion). Spring wheat crops cultivated by agribusiness enterprises (PEs) present higher exposure than the spring wheat crop cultivated by commercial farmers (CFs). The spring wheat cultivated by PEs with a value at risk (VaR) of KZT 310 billion accounts for 69 percent of the total portfolio‘s VAR, while the spring wheat crop cultivated by CFs with a value at risk (VaR) of G$ 141 billion accounts for only 31 percent of the portfolio‘s VAR. Several assumptions were made in order to arrive to the VAR figures for spring wheat crop production in Kazakhstan. This assumption can be summarized as follows: The assumptions made to calculate the aggregate spring wheat crop values in the 8 selected oblasts in Kazakhstan were: i) both spring wheat crops cultivated by agribusiness enterprises and commercial farmers are included, ii) the full 3-year average sown area of spring wheat crop in each rayon included with a total of 13.3 million hectares, iii) 100% coverage level on the spring wheat expected yield for each type of farmer (agribusiness enterprises or commercial farmer) and rayon, and iv) the crop production is valuated at the 2008 to 2010 farm gate prices for the month of harvest, September. 33. In the planning of any public-private crop insurance program for Kazakhstan, due consideration must be given to the spatial and temporal distribution of crop values and careful accumulation control exercised. The temporal distribution of VAR is determined by the length of the crop cycles, the predominant cropping patterns, and the crop prices that will impact directly on the exposed values. The temporal distribution of VAR for spring wheat in Northern Kazakhstan runs from May to September. Spring wheat VARs in North Kazakhstan are far from being evenly geographically distributed. Kostanay, NKO and Akmola Oblast, have a concentration of around 86 percent of the VARs for spring wheat crop production in northern Kazakhstan (KZT 390 billion) in an area of 400,000 square kilometers (approximately 630 kilometers times 630 kilometers). Although this area is huge, there is evidence of high correlations among spring wheat yields between these three oblasts. On average, the correlation coefficients among these Oblasts are above 0.84. The remaining five oblasts selected for the analysis account for only 14 percent of the VARs for spring wheat crop production in Northern - 210 - Kazakhstan (KZT 62 billion) in an area of approximately 1,330,000 square kilometers (approximately 2900 kilometers times 500 kilometers). Even taking into consideration the huge area comprised by these oblasts, moderate to high levels of yield correlations are observed between them (except for those oblasts situated in the antipodes of the area. Table A1.3 shows the distribution of spring wheat VARs per oblast and type of farmer in Northern Kazakhstan. Table A1.4 shows the correlation matrix between annual average spring wheat yields at Oblast level in North Kazakhstan. Map. A1.5 shows the distribution of spring wheat VARs per rayon for the 8 selected oblasts under analysis. Table A1.3. Kazakhstan. Spring Wheat Crops. Total Values at risk (KZT Billions) Agribusiness Enterprises Commercial Farmers Total Oblast VAR % VAR % VAR % Akmola 87 28% 23 16% 110 24% Aktobe 7 2% 5 3% 12 3% EKO 6 2% 7 5% 14 3% Karaganda 7 2% 7 5% 14 3% Kostanay 98 31% 51 36% 148 33% NKO 97 31% 36 25% 132 29% WKO 4 1% 8 5% 11 2% Pavlodar 5 1% 6 4% 10 2% Total 310 69% 142 31% 452 100% Source: Authors from CRAM Table A1.4. Kazakhstan. Spring Wheat Crops. Yield Correlation matixs Akmola Aktobe EKO Karaganda Kostanay NKO Pavlodar WKO Akmola 1.00 Aktobe 0.47 1.00 EKO 0.48 0.03 1.00 Karaganda 0.77 0.31 0.74 1.00 Kostanay 0.88 0.59 0.21 0.62 1.00 NKO 0.84 0.31 0.17 0.51 0.84 1.00 Pavlodar 0.53 0.03 0.84 0.61 0.32 0.51 1.00 WKO 0.18 0.65 -0.13 0.22 0.45 0.25 0.05 1.00 Source: Authors from Agency of Statistics - 211 - Map A1.5. North Kazakhstan: Spring Wheat Exposures in North Kazakhstan Region.. Source: Authors from Agency of Statistics. Estimation of Spring Wheat Crop Losses 34. The estimation of the crop losses for the spring wheat crop production in North Kazakhstan is performed based on an “as if” analysis over the simulated output yields generated in CRAM . That means that the CRAM, according to the assumptions made for the simulation, estimates the expected losses for the portfolio and their associated pure loss ratios based on what would have occurred for each of 5,000 yields generated by Monte Carlo Simulation under CRAM for each type of farmer and each rayon (details of the methodology followed for CRAM are provided in Appendix 1.A). The process for this estimation can be described in 3 steps. The first step consists in calculating - for each rayon and type of farmer – the percentage of yield shortfall for each of the 5,000 yield simulations generated by Monte Carlo Simulation under CRAM. Then, again for each rayon and farmer typology, if the yield generated by Monte Carlo methodology is below the expected yield calculated based on the average of the five most recent years‘ annual average yields determined for each type of farmer and rayon , then the percentage of the deviation is recorded as a loss, otherwise it is recorded as zero loss. The second step consists, for each rayon and farmer typology, in applying the percentage of the loss to the respective value at risk (VAR) to obtain the amount of losses per each of the 5,000 yields generated through Monte Carlo Methodology in CRAM. The third and last step consists in adding up, for each rayon and farmer typology, the calculated loss figures per each of the 5,000 yields generated through Monte Carlo Methodology in CRAM. 35. The main feature of this risk assessment is that the spring crop production in northern Kazakhstan is a risky endeavor. This is evidenced by the annual average expected losses for the spring wheat portfolio under analysis that, according to CRAM estimates, amount to KZT 66.5 billion per year (approximately US$ 443 million), equivalent to an annual average loss cost ratio of 14.71 percent of the total value of the expected production for major crops. The commercial farmers accounting for 31 percent of the VAR for spring wheat crop production exhibit the highest loss cost of 17.52 percent. The VAR of - 212 - the spring wheat cultivated by production enterprises– 69 percent of VAR for spring wheat production - exhibit the lowest loss costs with average annual losses of 13.43 percent. Table A1.5 below shows the average spring wheat annual expected loss at Oblast level for each type of farmer in Northern Kazakhstan. Table A1.5. North Kazakhstan. Spring Wheat Crop.. Annual Average Value of Crop Losses (KZT Billion) Region Crop Season 4-Year Total % of Values Average Losses as % Average Values at Values of of Total Planted Risk (KZT losses (KZT Values Area billions) billions) (hectares) Akmola 0.87 23.26 16% 4.41 18.98% Aktobe 0.29 5.03 4% 1.37 27.24% EKO 0.24 7.36 5% 1.41 19.13% Commercial Karaganda 0.36 6.65 5% 1.31 19.68% Farms Kostanay 1.26 50.59 36% 8.43 16.66% NKO 0.73 35.52 25% 4.78 13.47% WKO 0.32 7.75 5% 2.08 26.77% Pavlodar 0.24 5.84 4% 1.11 19.03% Subtotal Commercial Farm 4.31 141.99 31% 24.90 17.54% Akmola 2.86 86.67 28% 12.12 13.99% Aktobe 0.38 7.43 2% 1.67 22.52% EKO 0.18 6.35 2% 1.20 18.90% Agribusiness Karaganda 0.28 7.25 2% 1.08 14.94% Enterprises Kostanay 2.57 97.53 31% 13.72 14.07% NKO 2.32 96.64 31% 10.23 10.58% WKO 0.18 3.54 1% 0.87 24.45% Pavlodar 0.19 4.63 1% 0.73 15.80% Subtotal Agribusiness Enterprises 8.96 310.03 69% 41.63 13.43% Akmola 3.73 109.93 24% 16.54 15.04% Aktobe 0.67 12.45 3% 3.04 24.43% EKO 0.42 13.71 3% 2.61 19.02% Whole Karaganda 0.64 13.89 3% 2.39 17.21% Portfolio Kostanay 3.82 148.12 33% 22.15 14.96% NKO 3.06 132.16 29% 15.01 11.36% WKO 0.50 11.29 2% 2.94 26.05% Pavlodar 0.42 10.47 2% 1.84 17.60% Subtotal Whole Portfolio 13.26 452.02 100% 66.52 14.72% Source: Authors from CRAM 36. The North Kazakhstan region is heterogeneous in terms of the risk faced by spring wheat crop producers. The Oblasts situated towards the west of the region are more risky for spring wheat crop production than those oblasts situated in Central North Kazakhstan. While the oblasts situated toward the west of the region, such as WKO and Aktobe, show average loss cost ratios above 24%, the oblasts situated in Central North areas of Kazakhstan ( such as NKO, Kostanay, and Akmola) show average annual loss cost ratios between 10 percent to 15 percent of the Total VARs. The Oblasts located in the north east of Kazakhstan (EKO, Pavlodar, and parts of Karaganda Oblast) are on the mezzanine level in terms of risk for spring wheat crop production. Annual average loss costs at oblast level for spring wheat - 213 - in the oblasts situated towards east of north Kazakhstan region range from 17 percent of the total VARs to 19 percent of the total VARs. 37. Spring wheat crop production seems to be more risky when performed by commercial farmers than when it is performed by agribusiness enterprises. This is evidenced by the fact that annual average loss cost for spring wheat crop production for Commercial enterprises is higher than the annual average loss cost for spring wheat crop production calculated for Agribusiness Enterprises. While Commercial Farmers spring wheat crop production shows an average loss cost ratio of 17.54% of the total VARs, Agribusiness enterprises show lower average loss cost ratios of 13.43% of the total VARs. The reasons for the lower average loss cost ratio observed for agribusiness enterprises can be explained by the better technical package applied by these farmers to their crops. Probable Maximum Loss 38. The analysis of 17-year (1994 -2010) spring wheat rayon level average yields shows that 1998 was the worst loss year in this series with total losses in spring wheat valued at KZT 179.4 billions, which represented 51 percent of the 1998 annual spring wheat crop expected value of production97. However, although 1998 was a severe loss year in Northern Kazakhstan, even worse crop losses could occur in future. From an insurance view point, underwriters need to know with a high degree of confidence the maximum losses that they might incur (termed the Probable Maximum Loss, PML98) either 1 in 100 years, or if it is necessary to be even more conservative, 1 in 250 years. This information is an invaluable aid to structuring an insurance and reinsurance program and to determining how much capital must be reserved to cover the PML loss year. Figure A1.6 and Table A1.6 show the results of the World Bank‘s PML loss cost analysis for return periods of 1 in 2 years up to a maximum of 1 in 250 years for the spring wheat crop production in the 8 selected Oblast in the northern region in Kazakhstan simulated under CRAM, assuming a 100% Insured Yield Coverage level. The analysis shows that: (a) The losses in 1998 with 51% loss cost at 100 percent coverage level equate approximately to a 1 in 60- year return period; and (b) The 1 in a 100 year estimated PML loss cost is 54.6% at 100 percent coverage level, equivalent to a financial loss of KZT 247 billons (US$ 1.64 billion). Figure A1.6. North Kazkahstan Region: Spring Wheat Crop Portfolio Modeled PML Loss Cost (at 100% coverage level) 70.00% 59.90% 61.34% 58.11% 60.00% 54.63% 49.76% Loss Cost (% of liability) 50.00% 40.00% 34.01% 30.00% 20.00% 10.00% 0.00% 0 10 50 100 150 200 250 Recurrrence Period (ys.) 97 Valued at the average price for spring wheat during September of the most recent 3-years and assuming that the planted area for 2008 is the same that the 4-year average planted area for the period 2007 to 2010. 98 The Probable maximum Loss is defined as ―An estimate of the maximum loss that is likely t o arise on the occurrence of a single event considered to be within the realms of probability, remote coincidences and possible but unlikely catastrophes being ignored‖. - 214 - Table A1.6 North Kazakhstan: Spring Wheat Crop Portfolio modeled PML Loss Costs for different return periods Return Period (years) 1 years 10 years 50 years 100 years 150 years 200 years 250 years Expected Loss (KZT Billions) 66.31 153.74 224.93 246.92 262.68 270.75 277.25 Expected Loss (US$ Billions) 0.44 1.02 1.50 1.65 1.75 1.80 1.85 Loss Cost 14.67% 34.01% 49.76% 54.63% 58.11% 59.90% 61.34% Source: Authors form CRAM 1. The results of the spring wheat crop risk assessment for 100% coverage level at rayon level provide an idea about the potential exposure to losses of spring wheat cultivated in the 8 selected oblasts in north Kazakhstan region and assist to delineate a reinsurance strategy for this crop. The results of the Spring Wheat Portfolio PML analysis confirm that the 1 in 10 year spring wheat crop losses in Kazakhstan could be as severe as 34 percent loss cost (KZT 153.7 billon) and as high as 54.6 percent loss cost 1 in 100 years (KZT 246.9 billons). These are catastrophic losses to agriculture if the 1998 and 2010 droughts are likely to be repeated in the near future. The pattern of the PML figures for each of the analyzed return periods suggests that the retention of the total liability arising out of retaining 100% of the risk in the country is not possible; thus, the risk layering and risk financing issues must be considered seriously in delineating the risk financing strategy. 2. The PML analysis performed separately for commercial farmers (CF) and agribusiness enterprises (PE), shows that the PML for commercial farmers is higher than the PML for agribusiness enterprises.. For commercial farmers the 1 in 100 year estimated PML loss cost is 58.11 percent at 100 percent coverage level, equivalent to a financial loss of KZT 82.5 billion (US$ 550 million). Conversely, the estimated 1 in 100 year PML loss cost for agribusiness enterprises is lower at 55.95 percent at 100 percent coverage level, equivalent to a financial loss of KZT 173.5 billion (US$ 1.16 billion). Figures A1.7 a and b, show the results of the World Bank‘s PML loss cost analysis for return periods of 1 in 10 years up to a maximum of 1 in 250 years for the commercial farmers spring wheat crop portfolio and Agribusiness enterprises spring wheat crop portfolio, assuming a 100% Insured Yield Coverage level. Figure A1.6a. North Kazkahstan: Commercial Figure A1.6b. North Kazkahstan: Agribusiness Farmer Spring Wheat Portfolio Modeled PML Loss Enterprises Spring Wheat Portfolio Modeled PML Cost (at 100% coverage level) Loss Cost (at 100% coverage level) 0.7 63.58% 64.79% 0.7 61.07% 60.78% 62.30% 58.11% 58.68% 0.6 0.6 55.95% 53.76% Loss Cost (% of liability) Loss Cost (% of liability) 50.45% 0.5 0.5 38.38% 0.4 0.4 33.67% 0.3 0.3 0.2 0.2 0.1 0.1 0 0 0 0 0 10 50 100 150 200 250 0 10 50 100 150 200 250 Recurrrence Period (ys.) Recurrrence Period (ys.) Source: Authors from CRAM Source: Authors from CRAM Conclusions 3. The analysis of rayon level crop production and yields for spring wheat in the eight selected oblasts in North Kazakhstan shows that spring wheat production is exposed mainly to droughts, but - 215 - also to late frost, hail, pest attacks and leaf diseases to the crop in almost all the region. This is evidenced by the high average loss cost estimated through CRAM for a 17-years period, 1994 up to 2010, estimated at 14.72 percent of the total gross value of spring wheat production in the region and a calculated 1 in 100 year PML of 54.6 percent of the total gross value of production of spring wheat in the selected oblast in north Kazakhstan. 4. In an eventual crop insurance program for spring wheat production in north Kazakhstan special consideration should be taken to crop management issues. As can be seen from this spring wheat crop risk assessment, agribusiness enterprises are able to implement more improved crop management measures than the commercial farmers and, therefore, are able to obtain better and more stable spring wheat yields. In any eventual MPCI loss of yield crop insurance program for spring wheat production in Kazakhstan, the insurers/ reinsurers will need take all the provisions in order to identify the crop management practices implemented by the insured and make a risk selection based on the best practice in terms of spring wheat crop production and risk reduction practices. For example, zero tillage is a crop management practice that helps to save soil moisture for crops and, thus, to reduce and/or mitigate drought risks. The risk reduction caused by the application of these sorts of crop management practices by farmers should not be promoted by the insurance industry and to do so they can increase premium rates, increase deductibles or, even, not write the risk. 5. Any eventual crop insurance program for spring wheat production in north Kazakhstan, in order to avoid adverse selection of risks within the insurance portfolio, should consider a distinction, in terms of guaranteed yields and rates among the different rayons and type of farmers. The types of farmers, agribusiness and commercial farms analyzed, as well as the different rayons in the selected oblasts in North Kazakhstan show different exposures to risk. Spring wheat crops produced by commercial farmers are more risky than this crop produced by agribusiness enterprises. Spring wheat produced by commercial farmers reaches an annual average loss cost that on average for North Kazakhstan is 17.54 percent, the same crop produced in the same area but by agribusiness enterprises reaches a much lower annual average loss cost of 13.43 percent. While risky oblasts for spring wheat crop production like WKO face an average loss cost of 26.5 percent, other oblasts that are less risky such as NKO have an average loss cost of 11.36 percent. - 216 - Appendix 1.A. Crop Portfolio Risk Assessment Model – Design Features – This section presents the basic design features of the Crop Risk Assessment Model (CRAM) for spring wheat production in the selected oblasts in North Kazakhstan. The CRAM is constructed based on analysis of variation of Oblast and Rayon-level spring wheat annual average yields for a 17year time- series, from 1994 up to and including 2010. The CRAM was developed using the sown area, harvested area, production and annual average yield statistics for each of the rayons in NKO, Akmola, Kostanay, Karaganda, Pavlodar, EKO, Aktobe, and WKO in the northern Kazakhstan spring wheat belt. The original spring wheat production data used for this analysis were provided by the ARKS. Selected Crops The selected crop for the CRAM was spring wheat. According to information obtained from ARKS of Kazakhstan, the average sown area of spring wheat was 13.26 million hectares for the period 2007 – 2010. Out of the 13.26 million hectares planted with spring wheat in North Kazakhstan region, 8.96 million hectares (68 percent) are planted by agribusiness enterprises and 4.31 million hectares (32 percent) are planted by commercial farmers. Cultivated Area The CRAM assumes that the annual spring wheat planted area has remained constant at the four year average (period 2007-2010) over the 17 years sown area series. The reason for this assumption is to remove seasonal variations for each rayon from the areas. The four year average total sown area for spring wheat in the selected oblasts in North Kazakhstan amounts to 13.26 million hectares. The breakdown of this information at oblast level is shown in Table A1.7. Table A1.7. Spring Wheat . Four-year Average Sown area at Oblast level (Hectares) Commercial Agribusiness Farmers Enterprises Grand Total Oblast Sown Area Sown Area (hectares) (hectares) (hectares) Akmola 0.87 2.86 3.73 Aktobe 0.29 0.38 0.67 EKO 0.24 0.18 0.42 Karaganda 0.36 0.28 0.64 Kostanay 1.26 2.57 3.82 NKO 0.73 2.32 3.06 WKO 0.32 0.18 0.50 Pavlodar 0.24 0.19 0.42 Grand Total 4.31 8.96 13.26 Source: ARKS In order to be eligible for CRAM, two criteria have been set: minimum sown planted area per Rayon and a minimum of 17 years continuous annual average yield data. In order to ensure that there are sufficient numbers of farmers growing the crop in a selected rayon, a minimum area of 10,000 hectares has been provisionally set as a requirement for a crop in a certain rayon to be eligible for the model. The second criterion is that at least 17 continuous years of yield data must be available for each rayon in order to qualify for the CRAM.. Most of the rayons in the 8 selected oblasts in North Kazakhstan region, met the eligibility criteria. As a result, the total acreage considered for CRAM was 13.26 million hectares. Out of - 217 - the total area considered for CRAM, 8.96 million hectares (68 percent) were planted by agribusiness enterprises (PEs) and 4.31 million hectares (32 percent) were planted by commercial farmers (CFs). 6.58. The main geographic concentration of spring wheat is observed in Kostanay, Akmola, and NKO in north Kazakhstan region. The most important spring wheat rayons with between 320,000 Ha and 640,000 Ha of spring wheat are located in Kostanay, NKO and Akmola: conversely in EKO and WKO the area of spring wheat is less than 2,500 Ha in many rayons. Map A1.6 shows the distribution of spring wheat planted area (in hectares) in the 8 Oblasts of northern and central Kazakhstan at rayon level. Map A1.6. North Kazakhstan: Planted area with spring wheat at Rayon level. Source: Authors from Agency of Statistics. Rayon Crop Yield Data The CRAM uses rayon annual average yields for spring wheat crops for the period starting 1994 and up to crop year 2010 as reported by ARKS. The original rayon annual average yields from 1994 to 2010 are included in Appendix 3. The ARKS reports average yields on a sown area basis at rayon level. This is an important advantage for risk modeling purposes, since the yields calculated on a sown area basis capture, both, the variations due to partial yield loss as well as the yield variations due to total crop area losses. The ARKS reports average yields for two categories of farm typology, agribusiness enterprises and commercial farmers. This is also an important advantage since it allowed the team to perform, for each rayon, an assessment of the risk with a breakdown per each type of farm typology. Spring wheat production shows significant differences in terms of yield performance between the crops produced by commercial farmers and the crops produced by agribusiness enterprises. Spring wheat average yields for the period 1994 to 2010 are similar between agribusiness enterprises and commercial farmers. While the spring wheat average yields for crops produced by agribusiness enterprises during the - 218 - period 1994 to 2010 is 8.70 centners per hectare, the average of the annual average yields for spring wheat crops during the same period is 8.81 centners hectare. However, agribusiness enterprises‘ annual average spring wheat crop yields are less volatile than commercial farmers spring wheat annual average yields. Spring wheat crops produced by agribusiness enterprises showed an average coefficient of variation (CoV), calculated per rayon along the 17-years annual average yield series of 29 percent. Conversely, the CoV figures for spring wheat yields on crops produced by commercial farmers were higher with an average CoV of 34 percent. The commercial farmers and agribusiness enterprises spring wheat annual average yields imputed in CRAM showed an increasing trend in their productivity during the period, 1994 to 2010. Table A1.8 and the Figure A1.7.a and A1.7.b summarize the features of the crops included into the CRAM. Table A1.8. Features of zone annual average yields inputted ino CRAM. (1994 – 20101) Commercial Farmers Commercial Agribusiness Farmers Enterprises Whole Portfolio Average Yield (centners/he) 8.81 8.70 8.78 Standard Deviation (centners/he) 2.96 2.55 2.54 COV% 34% 29% 29% Minimum Yield (Centners/he) 4.12 3.80 3.99 Maximum Yield (Centners/he) 13.88 12.60 12.69 Source: Authors from GRDB Figure A1.7a.: North Kazakhstan Region: Spring Figure A1.7.b.1: North Kazakhstan Region: Spring Wheat. Historic Average Yields for Agribusiness Wheat. Historic Average Yields for Commercial Enterprises (1994- 2010) (centner/hectare) Farmers (1994- 2010) (centner/hectare) 20 20 18 18 16 16 14 14 12 12 10 10 8 8 6 6 4 4 2 2 0 y = 0.2517x + 6.4343 0 y = 0.2145x + 6.8796 Akmola Aktobe Akmola Aktobe EKO Karaganda EKO Karaganda Kostanaskay NKO Kostanaskay NKO Pavlodar WKO Pavlodar WKO Aggregate Average Aggregate Average Source: Authors based on ARKS Source: Authors based on ARKS Valuation Prices For CRAM risk modeling purposes, spring wheat has been valued at the average market average price per centner for the month of harvest for the crop years 2008, 2009, 2010. These crop prices are maintained as - 219 - a constant value for all the past 17 years. Table A1.9 shows the crop prices used for modeling under CRAM. Table A1.9. Average Springwheat market prices for the month of Septemeber (Period 2008-2010). Spring Wheat Price for Year September (KZT/Centner) 2008 2,802 2009 2,560 2010 3,998 Average 2008-2010 3,120 Source: Arka Consulting Yield Data Cleaning and trending to establish the Central Tendency The annual average yield series at zone level used to feed CRAM must be adjusted in order to reflect the current state of the art in terms of expected yields and yield variability for the selected crops for the risk assessment. This sub-section describes the methodologies followed to clean the yield data, determine the trend in yield data and, finally, to adjust the historical yields to the current expected yield at Region level. Eliminate Yield Outliers The first step was to detect and eliminate the statistical outliers from the annual average yield series for each of the selected crop and regions by applying the Chauvenet99 criteria. Each of the 17 years annual average yields records for each agribusiness enterprises and commercial farmers on each of the rayon in the 8 selected oblast in North Kazakhstan. If, by applying the Chauvenet criteria a yield outlier was detected, then the annual average crop yield was compared with the annual average crop yield performance for the same crop and year in neighboring rayon. If, as result of this comparison, it was detected that the crop yield performance in neighboring rayons diverged significantly in respect to the annual average yield for the target rayon and year, then yield, production, and harvested area figures were revisited to identify the cause of the divergence. If, any anomaly with yield, production, and harvested area figures was detected and there was no reason for the anomaly, then the procedure was to replace the outlier with the average annual average yield for contiguous rayon. Adjusting Zonal Average Yield Data for Trends The next step was to adjust the annual average yield series for maximum central tendency over the 17 years period. The crop yield central tendency is associated with crop management and technology practices; crop yield deviations from the central tendency are associated with effects of nature. The main objective of adjusting the historic annual average yield series was to isolate the effect on yields of the improvement on crop management practices and the increase in technology application to the crops along 17-year period considered for the analysis. A simplified method was adopted for determining the central tendency for each crop and each zone in the CRAM. The method aims to capture the non-linear yield tendency in the 17-years of annual average yield series at zonal level by using this yield series fitted to a lineal trend line and to an exponential trend line, and the five year moving average the 17 years annual average yield series. The yield trending method followed to determine the central tendency is summarized for spring wheat for agribusiness enterprises in Enbekshilderski Rayon in Akmola Oblast in Figure A1.8 99 In statistical theory, the Chauvenet‘s Criterion is a means of assessing whether one piece of experimental data – an outlier- from a set of observations, is likely to be spurious. - 220 - Figure A1.8: Akmola- Enbekshilderski Rayon. Spring Wheat: Historic average yields and yield trends (1994 - 2010) (centners/hectare) 18 16 14 12 10 8 6 4 2 0 Historic Actual Yield Lineal Trend Exponential Trend 5-year moving average TRENDED YIELD Source: Authors from GRDB Annual Reports 1995 - 1998 Expected Yields and adjusted crop variability. The last step was to estimate the expected yields and adjust the crop variability for spring wheat in each rayon within the 8 selected oblasts under analysis in North Kazakhstan Region. The expected yields are those used as inputs for risk modeling. The design of the CRAM is based on the spring wheat annual average yields for the period 2006 – 2010 at rayon level and their standard deviation; thus, these inputs must be representative of the current state of the art of spring wheat crop production in each of the analyzed rayons. That is, all the long terms and cyclical effects of crop management practice and of technology application on the historic annual average yields must be isolated prior to estimating these parameters for risk modeling purposes. In order to calculate the expected annual average yield for spring wheat for each rayon in the 8 oblasts under analysis, the simple average of the most recent five years historic annual average yields was calculated. This method to estimate annual average expected yields for a certain crop located in a certain zones is common in the agricultural insurance practice in countries where the constraint of scarce annual average crop yield data is a problem. The second part of this analysis was to estimate the expected annual average yield volatility of the annual average yield. The method used for this purpose was to measure the deviations between the historic actual annual average yields for each year of the series in respect to the corresponding annual average yield of the trend line. Then, these deviations were applied to the expected yield to obtain an adjusted annual average yield series. The method used to estimate the central tendency for yields and the yield deviation in respect to the central tendency is illustrated below in Table A1.10 and Figure A1.9. - 221 - Table A1.10. Akmola- Enbekshilderski Rayon. Ilustration of the Spring Wheat de-trended/actulized yield calculation Historic Historic Average Lineal Exponential 5-years annual Average annual annual Adjusted Yield Yield moving average Yield 2003- Crop Year average yield Yield Trendline Trendline Average yields 2007 yields trendline (center/he.) (center/he.) (center/he.) (center/he.) deviations (center/he.) (center/he.) (center/he.) from trend 1994 6 4.7 4.4 4.6 32% 11.1 14.7 1995 3.3 5.2 4.7 4.9 -33% 11.1 7.5 1996 6 5.6 4.9 5.3 14% 11.1 12.7 1997 3.4 6.1 5.2 5.7 -40% 11.1 6.7 1998 1.6 6.5 5.6 6.0 -73% 11.1 3.0 1999 12.4 7.0 5.9 4.1 5.6 120% 11.1 24.5 2000 7.8 7.4 6.3 5.3 6.3 23% 11.1 13.7 2001 9.2 7.8 6.7 6.2 6.9 33% 11.1 14.8 2002 9.3 8.3 7.1 6.9 7.4 26% 11.1 14.0 2003 9.4 8.7 7.5 8.1 8.1 16% 11.1 12.9 2004 5.4 9.2 7.9 9.6 8.9 -39% 11.1 6.7 2005 11.3 9.6 8.4 8.2 8.8 29% 11.1 14.4 2006 13.4 10.1 9.0 8.9 9.3 44% 11.1 16.0 2007 16.2 10.5 9.5 9.8 9.9 63% 11.1 18.2 2008 4.1 10.9 10.1 11.1 10.7 -62% 11.1 4.3 2009 15.8 11.4 10.7 10.1 10.7 47% 11.1 16.4 2010 6.2 11.8 11.4 12.2 11.8 -47% 11.1 5.9 5-year avg. 11.1 Source: Authors from ARKS Figure A1.9. Akmola- Enbekshilderski Rayon.. Adjusted Variability of Average Yields. 30.0 25.0 20.0 15.0 10.0 5.0 0.0 ACTUAL Expected Yield = 5-year average yield Source: World Bank based on ARKS data Estimation of losses for the National Paddy Portfolio The estimation of losses for the spring wheat crop portfolio was performed through a risk modeling exercise using the CRAM. Risk modeling is a fundamental step in agricultural insurance program design and ratemaking procedures. The main objective of crop risk modeling is to estimate, based on the available information, a yield probability density function that reflects the stochastic nature of yield outcomes. The model has two components: (a) the normal risk component, and (b) the catastrophic risks component. The normal risk component of CRAM is based on probability density functions that reflect the stochastic nature of yields outcomes for each crop season and zone. The model relies on two basic fundamentals: (a) - 222 - a crop yield probability density function inferred from the historic spring wheat annual average yields for each rayon and type of farmer in the analyzed portfolio, and (b) a correlation matrix of rayon-level and farmer-type level spring wheat annual average which reflects the covariant risk under the portfolio. The probability density functions were inferred from the technology adjusted annual average yields from the annual average yield series 1994 to 2010 that were fitted to a Weibull probability distribution. The outputs of the yield probability density functions obtained for each rayon and type of farmer were correlated in order to reflect the covariance on yields for risk modeling purposes. Spring wheat crop production in Kazakhstan is exposed to drought which is a very systemic risk. Variations in spring wheat crop yields are often caused by factors that typically affect a large area. The issue of a portfolio being exposed to systemic risk, since it affects the degree on which the risks can be diversified, has severe implications for the designing of crop insurance. In light of the systemic risk faced by spring wheat crop production in Kazakhstan, the CRAM considered the correlations among each rayon and type of farmer in order to simulate the potential losses for the portfolio. The Correlation matrix is presented in Appendix 1.4. The CRAM simulates 5,000 iterations of the model in order to arrive to the final spring wheat crop yield output. The final output yield for each zone and crop season is given by the multiplication of the crop yield generated by Monte Carlo simulation with a Weibull. This formula is simulated, through the Monte Carlo statistical methodology, by using @Risk software with 5,000 interactions. - 223 - Appendix1. B: Spring Wheat in North Kazakhstan. Historic Actual Spring Wheat Gross Value of Production Losses Estimation. Period 1994 - 2010 Table A1.11. North Kazakhstan. Spring Wheat. Estimation of Gross Value of Production Losses. 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Planted Area (million hectares) (1) 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 13.3 Wheat Prices (KZT/centner)(2) 3120 3120 3120 3120 3120 3120 3120 3120 3120 3120 3120 3120 3120 3120 3120 3120 3120 Calculation of the expected Total Value of Production Average Trend Yields (Centners/he) 5.6 6.7 7.4 7.8 8.2 8.5 8.7 9.0 9.1 9.3 9.5 9.6 9.7 9.9 10.0 10.1 10.2 Expected Value of Production (KZT billion) 239.3 286.6 314.3 333.9 349.1 361.6 372.1 381.2 389.3 396.5 403.0 408.9 414.4 419.4 424.1 428.5 432.7 Calculation of the actual Total Value of Production Actual Average Yield (Centner/he) 7.3 5.0 6.3 7.5 4.0 12.4 8.7 11.6 10.1 9.5 7.7 8.4 10.8 12.7 9.4 11.3 6.4 Actual Value of Production (KZT billion) 310.8 212.2 267.9 320.4 169.8 528.0 372.2 493.7 432.1 405.5 328.9 359.8 458.1 540.1 401.1 480.3 274.2 Calculation of the losses (in terms of loss of benefits) Average Yield Shortfall (Centner/he) 0.0 1.7 1.1 0.3 4.2 0.0 0.0 0.0 0.0 0.0 1.7 1.2 0.0 0.0 0.5 0.0 3.7 Percentage of the Average Yield Shortfall 0% 26% 15% 4% 51% 0% 0% 0% 0% 0% 18% 12% 0% 0% 5% 0% 37% Actual Value of Production (KZT billion) 0.0 74.4 46.4 13.5 179.4 0.0 0.0 0.0 0.0 0.0 74.0 49.1 0.0 0.0 23.0 0.0 158.5 Source: Authors Assumptions: (1) Total planted area is assumed to be equal to the average planted area for the most recent 4 years (Period 2007 -2010) (2) Wheat Prices are assumed to be equal to KZT 3120 per center, which is the average price for the month of September for the years 2008, 2009, and 2010 - 224 - Appendix 1.C: North Kazakhstan region. Spring Wheat -CRAM Results for Commercial Farmers Expected Yield Value at Risk Crop / Oblast / Rayon Sown Area (Ha) Loss cost (%) (Centner/he) (KZT) C F/ Akmola/ Akkol 27,460 8.06 710,060,825 19.98% C F/ Akmola/ Arshaly 42,200 10.76 1,457,924,943 17.26% C F/ Akmola/ Astrashanski 114,120 6.06 2,220,117,571 20.59% C F/ Akmola/ Atbasarski 49,720 9.28 1,481,036,816 19.82% C F/ Akmola/ Bulandinski 20,020 10.84 696,672,789 17.37% C F/ Akmola/ Celinogradski 32,740 7.47 785,330,857 20.67% C F/ Akmola/ Enbekshilderski 25,260 10.74 870,555,481 18.87% C F/ Akmola/ Esilski 90,640 7.61 2,214,994,634 19.94% C F/ Akmola/ Kokshetau 675 5.64 12,222,929 19.30% C F/ Akmola/ Korgalzhinski 11,760 4.71 177,796,402 19.62% C F/ Akmola/ Sandiktauski 40,620 10.49 1,367,787,848 19.14% C F/ Akmola/ Shortandinski 12,900 8.85 366,301,853 19.30% C F/ Akmola/ Shuchenski 15,380 11.52 568,800,837 16.18% C F/ Akmola/ Stepnogor 1,620 5.40 28,104,059 34.32% C F/ Akmola/ Yegindikolski 96,520 6.00 1,858,263,101 20.63% C F/ Akmola/ Zerendinski 35,760 11.81 1,355,972,962 16.49% C F/ Akmola/ Zhaksi 81,100 11.55 3,006,673,880 19.05% C F/ Akmola/ Zharkainski 151,900 7.24 3,528,167,515 17.01% C F/ Akmola/Erementauski 20,680 8.30 550,765,272 23.62% Subtotal C F/ Akmola 871,075 8.32 23,257,550,575 18.98% C F/ Aktobe/ Aktobe 10,160 6.11 199,410,046 27.27% C F/ Aktobe/ Alga 46,580 4.27 639,201,782 26.31% C F/ Aktobe/ Bugetkol 32,600 5.95 622,810,641 24.70% C F/ Aktobe/ Hromtau city 32,600 6.11 639,213,639 26.43% C F/ Aktobe/ Kargaly 23,960 6.77 521,019,876 24.81% C F/ Aktobe/ Kobda 16,420 5.28 278,234,036 30.53% C F/ Aktobe/Aitecebi 54,180 4.47 777,962,913 30.60% C F/ Aktobe/Baiganin 2,000 4.28 27,506,632 31.50% C F/ Aktobe/Martoc 41,020 6.80 895,507,598 25.49% C F/ Aktobe/Mugadjar 18,960 4.17 253,647,088 30.77% C F/ Aktobe/Oil 1,420 2.95 13,444,447 36.67% C F/ Aktobe/Temir 9,740 5.09 159,121,893 32.69% Subtotal C F/Aktobe 289,640 5.41 5,027,080,590 27.24% C F/EKO/ Ayagoz 9,120 6.14 179,680,836 18.48% C F/EKO/ Beskaragay 26,500 5.63 478,848,484 25.63% C F/EKO/ Boroduliha 47,080 8.66 1,308,349,069 22.46% C F/EKO/ Glubokoe 17,920 14.54 836,587,802 16.95% C F/EKO/ Katonkaragay 7,700 13.87 342,717,477 16.47% C F/EKO/ Kokpekti 24,100 10.22 790,962,633 16.68% C F/EKO/ Kurchum 1,440 10.87 50,230,899 15.54% C F/EKO/ Ridder city 140 17.08 7,677,625 23.16% C F/EKO/ Semey city 9,720 5.58 174,234,100 28.48% C F/EKO/ Shemonaiha 17,040 14.37 786,143,734 17.78% C F/EKO/ Tarbagatay 1,820 10.55 61,608,124 13.34% C F/EKO/ Ulan 24,080 8.82 682,121,256 19.52% C F/EKO/ Urzhar 29,580 9.21 874,867,049 17.96% C F/EKO/ Ust-Kamenogorsk city 1,000 11.97 38,425,479 17.38% C F/EKO/ Zaysan 1,920 15.45 95,235,429 12.44% C F/EKO/ Zharma 6,920 9.57 212,670,128 21.37% C F/EKO/ Zyryan 9,420 14.70 444,393,895 15.15% Subtotal C F/EKO 235,500 9.74 7,364,754,018 19.13% C F/Karaganda/ Abai city 10,800 4.96 172,124,626 21.91% C F/Karaganda/ Aktogay 3,400 6.14 67,006,404 15.84% C F/Karaganda/ Buharzhirau 48,000 5.90 909,461,948 22.38% C F/Karaganda/ Karkaraly 58,680 5.97 1,125,372,724 19.22% C F/Karaganda/ Nura 60,540 4.35 845,830,468 18.94% C F/Karaganda/ Osakarov 103,680 6.64 2,211,169,651 20.46% C F/Karaganda/ Shetski 16,960 5.27 286,641,858 17.57% - 225 - Expected Yield Value at Risk Crop / Oblast / Rayon Sown Area (Ha) Loss cost (%) (Centner/he) (KZT) C F/Karaganda/ Ulytau 33,240 5.81 619,487,391 17.53% C F/Karaganda/ Zhanaarka 26,720 4.76 408,191,093 16.74% Subtotal C F/Karaganda 362,020 5.72 6,645,286,163 19.68% C F/Kostanay/ Altynsarin 38,440 14.96 1,845,586,570 17.53% C F/Kostanay/ Amangedi 36,400 7.74 904,397,028 16.69% C F/Kostanay/ Arkalyk city 90,960 6.17 1,800,175,154 20.74% C F/Kostanay/ Auliykolski 46,960 8.24 1,242,584,188 18.34% C F/Kostanay/ Denisovski 38,400 12.02 1,481,392,499 21.64% C F/Kostanay/ Fedorovski 136,640 16.21 7,112,122,283 15.87% C F/Kostanay/ Kamisty 53,020 8.82 1,500,279,593 24.28% C F/Kostanay/ Karabalyk 46,420 14.50 2,160,228,955 17.86% C F/Kostanay/ Karasu 167,900 11.39 6,137,687,754 15.64% C F/Kostanay/ Uzunkolski 55,320 16.88 2,996,840,433 16.15% C F/Kostanay/ Zhangeldin 8,620 6.00 166,053,342 19.49% C F/Kostanay/ Zhetikara 30,800 8.45 835,523,856 21.06% C F/Kostanay/Kostanay 139,440 16.71 7,480,172,177 16.08% C F/Kostanay/Mendikara 81,180 17.29 4,505,724,488 13.81% C F/Kostanay/Nauirzym 125,800 8.81 3,557,518,105 18.36% C F/Kostanay/ Rudnyi city 10,800 4.61 159,650,542 20.68% C F/Kostanay/Sarykol 106,580 14.93 5,106,690,335 13.15% C F/Kostanay/Taranovski 41,340 12.01 1,593,807,783 19.92% Subtotal C F/Kostanay 1,255,020 12.56 50,586,435,087 16.66% C F/NKO/ Airtau 44,820 15.48 2,227,422,830 15.21% C F/NKO/ Akkayn 36,460 16.58 1,940,704,814 14.47% C F/NKO/ Akzhar 92,900 13.69 4,081,926,985 11.29% C F/NKO/ Esil 29,900 15.44 1,482,035,529 13.45% C F/NKO/ G.Musrepov 86,880 14.82 4,133,478,134 14.96% C F/NKO/ Kyzylzhar 32,520 17.06 1,780,865,841 13.40% C F/NKO/ M.Zhumabayev 107,780 15.23 5,267,536,762 10.83% C F/NKO/ Mamliut 27,600 15.98 1,415,509,852 14.44% C F/NKO/ Shalakin 42,740 14.73 2,020,976,667 12.53% C F/NKO/ Taiynsha 66,020 14.72 3,118,549,855 14.00% C F/NKO/ Timiryazev 78,220 14.89 3,738,946,933 14.79% C F/NKO/ Ualihanov 26,780 12.24 1,051,934,539 15.23% C F/NKO/ Zhambil 62,160 16.33 3,257,858,445 14.39% Subtotal C F/NKO 734,780 15.06 35,517,747,186 13.47% C F/WKO/ Akzhayk 4,460 3.13 44,853,235 32.68% C F/WKO/ Burlinski 19,500 8.43 527,901,061 29.58% C F/WKO/ Karatobinsky 9,740 4.18 130,678,597 35.16% C F/WKO/ Kaztalovski 44,600 4.36 623,961,189 26.03% C F/WKO/ Oral city 3,720 7.37 87,979,490 24.96% C F/WKO/ Shyngyrkau 10,800 7.14 247,451,624 33.27% C F/WKO/ Syrymski 22,960 6.23 459,208,975 34.79% C F/WKO/ Taskalynski 26,660 7.08 606,164,666 24.84% C F/WKO/ Terektinski 38,180 8.58 1,052,015,774 29.25% C F/WKO/ Zelenovski 91,360 9.26 2,715,886,667 24.53% C F/WKO/ Zhangalinski 4,140 3.34 44,370,753 30.75% C F/WKO/ Zhanibek 47,040 8.02 1,210,893,867 24.28% Subtotal C F/WKO 323,160 7.47 7,751,365,899 26.77% C F/Pavlodar/ Aksu city 780 4.84 12,118,392 24.40% C F/Pavlodar/ Aksuisky 3,020 5.36 51,960,912 23.73% C F/Pavlodar/ Aktogaisky 17,260 5.78 320,238,588 19.92% C F/Pavlodar/ Bayanaulski 520 4.68 7,811,856 18.70% C F/Pavlodar/ Ekibastuz city 48,740 9.08 1,420,615,032 22.05% C F/Pavlodar/ Irtyshski 108,560 7.04 2,453,282,304 16.96% C F/Pavlodar/ Kashyrski 38,280 8.74 1,073,960,712 18.55% C F/Pavlodar/ Pavlodar area 860 5.94 16,397,964 25.42% C F/Pavlodar/ Shernaktinsky area 11,660 7.48 279,965,928 18.02% C F/Pavlodar/ Uspenka area 7,840 7.70 193,781,280 23.67% C F/Pavlodar/ Zhelezninski 780 4.33 10,851,462 17.87% Subtotal C F/Pavlodar 238,300 7.64 5,840,984,430 19.03% - 226 - Expected Yield Value at Risk Crop / Oblast / Rayon Sown Area (Ha) Loss cost (%) (Centner/he) (KZT) Total Commercial Farm 4,309,495 10.26 141,991,203,948 17.54% Appendix 1.D: North Kazakhstan region. Spring Wheat - CRAM Results for Agribusiness Enterprises Sown Area Expected Yield Value at Risk Crop / Oblast / Rayon Loss cost (%) (hes) (Centner/he) (KZT) P E/ Akmola/ Akkol 88,480 8.85 2,512,784,751 16.82% P E/ Akmola/ Arshaly 94,720 9.54 2,900,532,724 11.43% P E/ Akmola/ Astrashanski 182,440 7.56 4,425,241,766 14.87% P E/ Akmola/ Atbasarski 299,360 9.22 8,857,519,376 14.94% P E/ Akmola/ Bulandinski 167,620 10.40 5,597,308,250 13.77% P E/ Akmola/ Celinogradski 184,740 6.99 4,143,495,277 13.61% P E/ Akmola/ Enbekshilderski 114,180 12.14 4,448,062,120 15.91% P E/ Akmola/ Esilski 317,600 7.94 8,093,469,852 13.89% P E/ Akmola/ Kokshetau 520 3.64 6,069,357 15.85% P E/ Akmola/ Korgalzhinski 53,980 5.80 1,005,248,410 13.60% P E/ Akmola/ Sandiktauski 210,360 11.96 8,077,478,971 11.99% P E/ Akmola/ Shortandinski 171,920 8.74 4,821,375,331 14.74% P E/ Akmola/ Shuchenski 100,740 11.19 3,618,260,650 12.26% P E/ Akmola/ Stepnogor 1,620 5.41 28,127,928 32.27% P E/ Akmola/ Yegindikolski 165,880 7.23 3,849,247,720 14.92% P E/ Akmola/ Zerendinski 124,520 14.10 5,634,119,037 12.39% P E/ Akmola/ Zhaksi 284,700 11.80 10,788,137,838 13.52% P E/ Akmola/ Zharkainski 232,920 8.62 6,446,671,872 14.91% P E/ Akmola/Erementauski 59,400 7.43 1,416,301,794 18.29% Subtotal PE/ Akmola 2,855,700 9.45 86,669,453,024 13.99% P E/ Aktobe/ Aktobe 5,100 5.41 88,528,984 19.79% P E/ Aktobe/ Alga 28,960 4.15 385,875,066 25.56% P E/ Aktobe/ Bugetkol 29,260 7.04 661,135,666 23.08% P E/ Aktobe/ Hromtau city 29,260 7.02 659,539,048 21.90% P E/ Aktobe/ Kargaly 61,280 7.90 1,554,350,473 20.15% P E/ Aktobe/ Kobda 28,920 3.57 331,688,542 28.23% P E/ Aktobe/Aitecebi 104,720 5.62 1,889,263,842 21.98% P E/ Aktobe/Baiganin 13,740 4.82 212,545,354 36.81% P E/ Aktobe/Martoc 57,900 7.43 1,380,451,244 20.31% P E/ Aktobe/Mugadjar 13,740 3.99 175,773,229 29.58% P E/ Aktobe/Oil 1,420 2.22 10,119,878 37.83% P E/ Aktobe/Temir 5,680 4.21 76,718,074 29.49% Subtotal PE/Aktobe 379,980 6.09 7,425,989,400 22.52% P E/EKO/ Ayagoz 7,533 4.51 109,178,717 20.66% P E/EKO/ Beskaragay 21,240 6.01 409,898,204 26.78% P E/EKO/ Boroduliha 27,780 11.09 989,027,196 19.54% P E/EKO/ Glubokoe 15,840 13.73 698,110,813 17.04% P E/EKO/ Katonkaragay 7,280 13.15 307,190,529 14.71% P E/EKO/ Kokpekti 24,500 7.72 607,101,213 17.77% P E/EKO/ Kurchum 250 4.61 3,700,901 23.56% P E/EKO/ Ridder city 160 15.15 7,782,956 22.56% P E/EKO/ Semey city 9,640 6.49 200,708,920 24.90% P E/EKO/ Shemonaiha 27,420 15.79 1,389,667,537 16.56% P E/EKO/ Tarbagatay 480 5.35 8,249,148 14.67% P E/EKO/ Ulan 11,540 11.85 438,878,181 27.96% P E/EKO/ Urzhar 5,760 7.35 135,899,749 16.66% P E/EKO/ Ust-Kamenogorsk city 520 11.95 19,940,013 33.16% P E/EKO/ Zaysan 350 10.95 12,300,861 31.26% P E/EKO/ Zharma 4,460 7.85 112,456,426 18.93% P E/EKO/ Zyryan 19,800 14.13 898,026,448 15.72% Subtotal C F/EKO 184,553 10.72 6,348,117,812 18.90% P E/Karaganda/ Abai city 2,100 10.87 73,292,280 18.16% P E/Karaganda/ Aktogay 80 4.91 1,261,392 20.98% P E/Karaganda/ Buharzhirau 37,680 8.31 1,005,531,899 13.83% P E/Karaganda/ Karkaraly 2,280 3.18 23,304,286 16.01% - 227 - P E/Karaganda/ Nura 130,220 8.49 3,546,878,414 15.18% P E/Karaganda/ Osakarov 78,300 7.89 1,983,059,254 14.09% P E/Karaganda/ Shetski 2,480 5.75 45,753,740 19.13% P E/Karaganda/ Ulytau 26,060 6.17 515,820,003 16.51% P E/Karaganda/ Zhanaarka 3,080 5.51 54,451,062 27.35% Subtotal C F/Karaganda 282,280 8.00 7,249,352,331 14.94% P E/Kostanay/ Altynsarin 136,860 13.33 5,856,824,100 14.53% P E/Kostanay/ Amangedi 26,940 7.23 624,892,848 15.64% P E/Kostanay/ Arkalyk city 145,180 9.16 4,267,031,367 14.65% P E/Kostanay/ Auliykolski 129,200 9.91 4,111,717,130 15.97% P E/Kostanay/ Denisovski 196,800 11.94 7,543,114,580 15.16% P E/Kostanay/ Fedorovski 172,680 16.43 9,104,630,571 13.10% P E/Kostanay/ Kamisty 241,180 9.18 7,103,774,036 15.49% P E/Kostanay/ Karabalyk 171,660 14.72 8,108,623,798 13.37% P E/Kostanay/ Karasu 369,120 10.80 12,796,043,873 14.59% P E/Kostanay/ Uzunkolski 181,120 13.99 8,131,717,338 11.73% P E/Kostanay/ Zhangeldin 900 4.79 13,843,961 25.19% P E/Kostanay/ Zhetikara 159,800 8.40 4,306,454,111 15.76% P E/Kostanay/Kostanay 120,640 12.72 4,927,730,587 14.28% P E/Kostanay/Mendikara 146,980 14.04 6,626,119,660 12.73% P E/Kostanay/Nauirzym 94,500 6.79 2,058,842,713 16.24% P E/Kostanay/ Rudnyi city 2,100 10.87 73,292,280 17.51% P E/Kostanay/Sarykol 188,740 14.61 8,852,721,341 12.79% P E/Kostanay/Taranovski 83,500 11.27 3,022,050,464 14.69% Subtotal PE/Kostanay 2,567,900 11.83 97,529,424,758 14.07% P E/NKO/ Airtau 208,540 13.15 8,803,324,609 10.37% P E/NKO/ Akkayn 119,380 14.44 5,534,283,571 9.42% P E/NKO/ Akzhar 179,620 11.00 6,341,688,989 10.37% P E/NKO/ Esil 179,220 13.76 7,917,090,376 9.65% P E/NKO/ G.Musrepov 430,400 13.16 18,187,418,497 11.42% P E/NKO/ Kyzylzhar 86,480 14.31 3,971,772,420 8.36% P E/NKO/ M.Zhumabayev 178,020 13.66 7,803,702,607 9.12% P E/NKO/ Mamliut 88,560 13.69 3,891,730,880 10.18% P E/NKO/ Shalakin 147,800 12.91 6,125,248,394 12.27% P E/NKO/ Taiynsha 326,460 13.27 13,905,412,334 10.22% P E/NKO/ Timiryazev 94,260 12.36 3,739,024,090 11.56% P E/NKO/ Ualihanov 160,700 11.35 5,854,091,295 12.14% P E/NKO/ Zhambil 123,120 11.56 4,567,170,536 11.79% Subtotal PE/NKO 2,322,560 12.96 96,641,958,599 10.58% P E/WKO/ Akzhayk 5,180 2.98 49,503,494 30.54% P E/WKO/ Burlinski 13,740 7.20 317,366,112 24.39% P E/WKO/ Karatobinsky 25,300 6.78 550,770,542 29.48% P E/WKO/ Kaztalovski 17,720 4.13 235,138,068 24.63% P E/WKO/ Oral city 6,760 7.24 157,210,504 22.27% P E/WKO/ Shyngyrkau 7,800 4.93 123,511,043 27.40% P E/WKO/ Syrymski 25,300 6.27 509,163,744 25.28% P E/WKO/ Taskalynski 4,740 4.39 66,822,563 25.28% P E/WKO/ Terektinski 14,520 8.56 398,779,106 23.63% P E/WKO/ Zelenovski 38,320 6.94 854,136,285 21.57% P E/WKO/ Zhangalinski 5,180 2.91 48,346,159 28.13% P E/WKO/ Zhanibek 12,320 5.77 228,202,743 20.14% Subtotal PE/WKO 176,880 6.23 3,538,950,365 24.45% P E/Pavlodar/ Aksu city 520 3.88 6,476,496 25.83% P E/Pavlodar/ Aksuisky 2,420 4.46 34,646,172 25.07% P E/Pavlodar/ Aktogaisky 9,040 6.90 200,226,960 15.63% P E/Pavlodar/ Bayanaulski 1,260 4.74 19,171,404 14.50% P E/Pavlodar/ Ekibastuz city 30,900 9.40 932,376,600 18.70% P E/Pavlodar/ Irtyshski 67,500 6.80 1,473,390,000 12.98% P E/Pavlodar/ Kashyrski 20,860 6.76 452,653,656 15.02% P E/Pavlodar/ Pavlodar area 4,940 6.24 98,950,176 17.77% P E/Pavlodar/ Shernaktinsky area 15,400 9.76 482,475,840 18.66% - 228 - P E/Pavlodar/ Uspenka area 30,640 9.18 902,893,392 15.75% P E/Pavlodar/ Zhelezninski 1,860 4.34 25,912,404 14.55% Subtotal PE/Pavlodar 185,340 7.78 4,629,173,100 15.80% Total Agribusiness Enterprises 8,955,193 10.79 310,032,419,388 13.43% - 229 - Appendix 1.E: North Kazakhstan region. Spring Wheat -CRAM Results for the whole portfolio Sown Area Expected Yield Value at Risk Crop / Oblast / Rayon Loss cost (%) (hes) (Centner/he) (KZT) Akmola/ Akkol 115,940 8.66 3,222,845,575 17.52% Akmola/ Arshaly 136,920 9.92 4,358,457,667 13.38% Akmola/ Astrashanski 296,560 6.98 6,645,359,337 16.78% Akmola/ Atbasarski 349,080 9.23 10,338,556,192 15.64% Akmola/ Bulandinski 187,640 10.45 6,293,981,039 14.16% Akmola/ Celinogradski 217,480 7.06 4,928,826,134 14.74% Akmola/ Enbekshilderski 139,440 11.88 5,318,617,601 16.39% Akmola/ Esilski 408,240 7.87 10,308,464,487 15.19% Akmola/ Kokshetau 1,195 4.77 18,292,287 18.15% Akmola/ Korgalzhinski 65,740 5.61 1,183,044,812 14.51% Akmola/ Sandiktauski 250,980 11.72 9,445,266,819 13.03% Akmola/ Shortandinski 184,820 8.74 5,187,677,184 15.06% Akmola/ Shuchenski 116,120 11.23 4,187,061,487 12.79% Akmola/ Stepnogor 3,240 5.41 56,231,986 33.30% Akmola/ Yegindikolski 262,400 6.78 5,707,510,821 16.78% Akmola/ Zerendinski 160,280 13.59 6,990,092,000 13.18% Akmola/ Zhaksi 365,800 11.75 13,794,811,718 14.73% Akmola/ Zharkainski 384,820 8.08 9,974,839,387 15.65% Akmola/Erementauski 80,080 7.65 1,967,067,067 19.78% Subtotal PE/ Akmola 3,726,775 9.19 109,927,003,599 15.04% Aktobe/ Aktobe 15,260 5.88 287,939,030 24.97% Aktobe/ Alga 75,540 4.23 1,025,076,848 26.03% Aktobe/ Bugetkol 61,860 6.47 1,283,946,307 23.86% Aktobe/ Hromtau city 61,860 6.54 1,298,752,687 24.13% Aktobe/ Kargaly 85,240 7.58 2,075,370,349 21.32% Aktobe/ Kobda 45,340 4.19 609,922,578 29.28% Aktobe/Aitecebi 158,900 5.23 2,667,226,755 24.49% Aktobe/Baiganin 15,740 4.75 240,051,986 36.20% Aktobe/Martoc 98,920 7.17 2,275,958,842 22.35% Aktobe/Mugadjar 32,700 4.09 429,420,317 30.29% Aktobe/Oil 2,840 2.58 23,564,325 37.17% Aktobe/Temir 15,420 4.76 235,839,966 31.65% Subtotal PE/Aktobe 669,620 5.79 12,453,069,990 24.43% EKO/ Ayagoz 16,653 5.40 288,859,554 19.30% EKO/ Beskaragay 47,740 5.80 888,746,688 26.16% EKO/ Boroduliha 74,860 9.56 2,297,376,265 21.21% EKO/ Glubokoe 33,760 14.16 1,534,698,615 16.99% EKO/ Katonkaragay 14,980 13.52 649,908,006 15.64% EKO/ Kokpekti 48,600 8.96 1,398,063,847 17.15% EKO/ Kurchum 1,690 9.94 53,931,800 16.09% EKO/ Ridder city 300 16.05 15,460,581 22.86% EKO/ Semey city 19,360 6.03 374,943,020 26.56% EKO/ Shemonaiha 44,460 15.25 2,175,811,271 17.00% EKO/ Tarbagatay 2,300 9.46 69,857,271 13.50% EKO/ Ulan 35,620 9.80 1,120,999,437 22.82% EKO/ Urzhar 35,340 8.91 1,010,766,798 17.78% EKO/ Ust-Kamenogorsk city 1,520 11.96 58,365,491 22.77% EKO/ Zaysan 2,270 14.76 107,536,289 14.59% EKO/ Zharma 11,380 8.90 325,126,553 20.52% EKO/ Zyryan 29,220 14.31 1,342,420,343 15.53% Subtotal C F/EKO 420,053 10.17 13,712,871,830 19.02% Karaganda/ Abai city 12,900 5.93 245,416,906 20.79% Karaganda/ Aktogay 3,480 6.11 68,267,796 15.94% Karaganda/ Buharzhirau 85,680 6.96 1,914,993,848 17.89% Karaganda/ Karkaraly 60,960 5.87 1,148,677,010 19.16% Karaganda/ Nura 190,760 7.17 4,392,708,882 15.90% Karaganda/ Osakarov 181,980 7.18 4,194,228,905 17.45% Karaganda/ Shetski 19,440 5.33 332,395,597 17.78% Karaganda/ Ulytau 59,300 5.96 1,135,307,394 17.07% - 230 - Sown Area Expected Yield Value at Risk Crop / Oblast / Rayon Loss cost (%) (hes) (Centner/he) (KZT) Karaganda/ Zhanaarka 29,800 4.84 462,642,155 17.99% Subtotal C F/Karaganda 644,300 6.72 13,894,638,494 17.21% Kostanay/ Altynsarin 175,300 13.69 7,702,410,670 15.25% Kostanay/ Amangedi 63,340 7.52 1,529,289,876 16.26% Kostanay/ Arkalyk city 236,140 8.00 6,067,206,521 16.46% Kostanay/ Auliykolski 176,160 9.47 5,354,301,319 16.52% Kostanay/ Denisovski 235,200 11.95 9,024,507,079 16.22% Kostanay/ Fedorovski 309,320 16.33 16,216,752,854 14.31% Kostanay/ Kamisty 294,200 9.11 8,604,053,629 17.02% Kostanay/ Karabalyk 218,080 14.67 10,268,852,754 14.32% Kostanay/ Karasu 537,020 10.98 18,933,731,627 14.93% Kostanay/ Uzunkolski 236,440 14.66 11,128,557,771 12.92% Kostanay/ Zhangeldin 9,520 5.89 179,897,302 19.93% Kostanay/ Zhetikara 190,600 8.40 5,141,977,967 16.62% Kostanay/Kostanay 260,080 14.86 12,407,902,764 15.36% Kostanay/Mendikara 228,160 15.20 11,131,844,148 13.17% Kostanay/Nauirzym 220,300 7.94 5,616,360,817 17.58% Kostanay/ Rudnyi city 12,900 5.63 232,942,823 19.68% Kostanay/Sarykol 295,320 14.73 13,959,411,675 12.92% Kostanay/Taranovski 124,840 11.52 4,615,858,247 16.50% Subtotal PE/Kostanay 3,822,920 12.07 148,115,859,845 14.96% NKO/ Airtau 253,360 13.56 11,030,747,439 11.35% NKO/ Akkayn 155,840 14.94 7,474,988,385 10.73% NKO/ Akzhar 272,520 11.92 10,423,615,974 10.73% NKO/ Esil 209,120 14.00 9,399,125,906 10.25% NKO/ G.Musrepov 517,280 13.44 22,320,896,631 12.07% NKO/ Kyzylzhar 119,000 15.06 5,752,638,260 9.92% NKO/ M.Zhumabayev 285,800 14.25 13,071,239,369 9.81% NKO/ Mamliut 116,160 14.23 5,307,240,732 11.31% NKO/ Shalakin 190,540 13.32 8,146,225,061 12.34% NKO/ Taiynsha 392,480 13.51 17,023,962,189 10.92% NKO/ Timiryazev 172,480 13.51 7,477,971,023 13.17% NKO/ Ualihanov 187,480 11.48 6,906,025,834 12.61% NKO/ Zhambil 185,280 13.16 7,825,028,982 12.87% Subtotal PE/NKO 3,057,340 13.47 132,159,705,785 11.36% WKO/ Akzhayk 9,640 3.05 94,356,730 31.55% WKO/ Burlinski 33,240 7.92 845,267,173 27.63% WKO/ Karatobinsky 35,040 6.06 681,449,139 30.57% WKO/ Kaztalovski 62,320 4.29 859,099,257 25.65% WKO/ Oral city 10,480 7.29 245,189,994 23.23% WKO/ Shyngyrkau 18,600 6.21 370,962,667 31.32% WKO/ Syrymski 48,260 6.25 968,372,719 29.79% WKO/ Taskalynski 31,400 6.68 672,987,229 24.88% WKO/ Terektinski 52,700 8.58 1,450,794,880 27.70% WKO/ Zelenovski 129,680 8.58 3,570,022,953 23.82% WKO/ Zhangalinski 9,320 3.10 92,716,913 29.39% WKO/ Zhanibek 59,360 7.55 1,439,096,610 23.62% Subtotal PE/WKO 500,040 7.03 11,290,316,264 26.05% Pavlodar/ Aksu city 1,300 4.46 18,594,888 24.90% Pavlodar/ Aksuisky 5,440 4.96 86,607,084 24.26% Pavlodar/ Aktogaisky 26,300 6.16 520,465,548 18.27% Pavlodar/ Bayanaulski 1,780 4.72 26,983,260 15.72% Pavlodar/ Ekibastuz city 79,640 9.20 2,352,991,632 20.72% Pavlodar/ Irtyshski 176,060 6.95 3,926,672,304 15.47% Pavlodar/ Kashyrski 59,140 8.04 1,526,614,368 17.51% Pavlodar/ Pavlodar area 5,800 6.20 115,348,140 18.85% Pavlodar/ Shernaktinsky area 27,060 8.78 762,441,768 18.43% Pavlodar/ Uspenka area 38,480 8.88 1,096,674,672 17.15% Pavlodar/ Zhelezninski 2,640 4.34 36,763,866 15.53% Subtotal PE/Pavlodar 423,640 7.70 10,470,157,530 17.60% Total Portfolio 13,264,688 10.62 452,023,623,336 14.72% - 231 - Appendix 1.F: North Kazakhstan region. Spring Wheat -CRAM Yield Correlation Matrix - 232 - Annex 2: Mandatory Crop Insurance Program Statutory Features This Annex contains the following legal documents relating to the Obligatory Crop Insurance Program in Kazakhstan: Annex 2.1. English translation of the Law on Comulsory Crop Insurance, Law No 5333 dated 10 March 2004 Annex 2.2. English translation of the Crop Insurance Policy Wording Annex 2.3. 2009/10 Normative Costs of Production for spring wheat in each Oblast and Zone in Kazakhstan - 233 - Annex 2.1. Law On Compulsory Crop Insurance The Law of the Republic of Kazakhstan N 533 dated March 10, 2004 This Law regulates public relations arising in the sphere of compulsory crop insurance and establishes legal, financial, and organizational frameworks for its implementation. Article 1. Main definitions used in this Law This Law shall operate the following main definitions: 1) agent: a joint-stock company founded by the resolution of the Government of the Republic of Kazakhstan, a member of national holding in the sphere of agribusiness, the only shareholder of which is the Government; 2) assessor (independent expert): a physical person or a legal entity that has a license to perform assessment; 3) total loss of crops: consequences of the impact of adverse weather events on crops when the expenses for further growing of crops and harvesting exceed the expected income from the yield; 4) partial loss of crops: consequences of the impact of adverse weather events on crops when the assessed or actual income from 1 hectare of crop production is less than the normative cost per 1 hectare of crop production which is set at the moment of concluding compulsory insurance contract; 5) adverse weather event: natural phenomenon (long-lasting - drought, frost-killing, lack of warmth, excess moisture in soil, excess moisture in air, flood, shallow, dry wind; and short-lasting - hail, excess rain, frost, strong wind, mudflow), which result in loss or damage of crop products; 6) inspection act on the fact of adverse weather event (hereinafter referred to as the ―Inspection act‖): document which confirms the cause-and-effect relationship of the partial and total loss of crops and the impact of the adverse natural event(s), the requirements to which are stipulated by the Law on the form established by the Government of the Republic of Kazakhstan; 7) expected income from yield: revenue determined by the commission in compliance with the methodology for the determining of crop loss area which is approved by the authorized body in the sphere of crop production; 8) crop production: a group of manufacturers in agricultural industry including agricultural producers, i.e. individuals and legal entities engaged in production of crop products; 9) contract of compulsory crop insurance (hereinafter referred to as compulsory insurance contact): a contract concluded by the insurer and the insured on terms and conditions specified by this Law; 10) mutual crop insurance company (hereinafter referred to as the ―mutual‘s‖): a legal entity founded in the organizational and legal form of consumer co-operative to provide mutual insurance of property interests of its members in compulsory crop insurance. 11) crop product: product obtained by means of agricultural crops cultivation (cereals, oil-bearing crops, sugar beet, cotton); 12) authorized state body in crop production: a state authority established by the Government of the Republic of Kazakhstan to carry out state regulation in the sphere of crop production development; 13) insured event: an accident the occurrence of which can result in payment of insurance indemnity as stipulated by compulsory insurance contract; 14) sum insured: a sum of money which an object of compulsory insurance has been insured for, and which represents the top amount of the insurer‘ s responsibility at the occurrence of insured event; 15) insurance premium: a sum of money which the insured must pay to the insurer for undertaking the liability to pay an insurance indemnity to the insured (beneficiary) in the amount provided for by the compulsory insurance contract; - 234 - 16) insurance indemnity: amount payable by the insurer to the insured (beneficiary) within amount of the sum insured upon the occurrence of an insured event; 17) insurer: a legal entity granted a license to implement compulsory crop insurance in the order prescribed by the legislation of the Republic of Kazakhstan, which shall pay the insurance indemnity to the insurer or any other person in favor of whom the contract was concluded (beneficiary) within the amount provided for by the contract (the sum insured); 18) insured: a person engaged in producing crop products and who concluded a compulsory insurance contract with the insurer; 19) franchise: release of the insurer from the obligation to indemnify the loss when the extent of loss does not meet the threshold; 20) normative costs: expenditures related to separate technological processes, types of work and Clauses of expenses calculated per one hectare of crop production (in KZT); Article 2. Legislation of the Republic of Kazakhstan on compulsory crop insurance 1. The legislation of the Republic of Kazakhstan on compulsory crop insurance shall be based on the Constitution of the Republic of Kazakhstan and comprise the Civil code of the Republic of Kazakhstan, the law of the Republic of Kazakhstan On insurance activity, the law of the Republic of Kazakhstan On mutual insurance, this Law and other normative and legal acts of the Republic of Kazakhstan. 2. If any international treaty as ratified by the Republic of Kazakhstan stipulates rules other than provided for by this Law, then the rules of the international treaty shall prevail. 3. This Law shall not apply to any activity related to the growing of crop products in non-agricultural lands and its storage. Article 3. The object of compulsory crop insurance Subject matter of compulsory crop insurance is the property interests of the insured related to full or partial reimbursement for the insured‘s losses at the occurrence of an insured event, the amount of which shall be determined in accordance with this Law. Article 4. The aims of compulsory crop insurance The compulsory crop insurance is aimed to: 1) ensure protection of property interests of producers of crop products against the consequences of adverse weather events by paying insurance indemnities in the cases, amounts and order set forth in this Law; 2) to provide relevant conditions to credit producers of crop products on security of the crops insured; 3) to enhance the effectiveness of the state crop production support programs. Article 4-1. Competence of the Government of the Republic of Kazakhstan The Government of the Republic of Kazakhstan shall: 1) develop state crop production programs; 2) approve standard form of compulsory crop insurance contracts and also contracts stipulating the terms and conditions of partial reimbursement of insurance indemnities; 3) approve normative costs for production of crop products which are subject to compulsory insurance per one hectare of sown areas; 4) approve the money use procedure which are allocated to support the compulsory crop insurance, and also amount of payment to the agent for its service; 5) approve criteria and description of adverse weather events which are provided by the competent environmental government agency and natural/manmade emergency authority; - 235 - 6) approve the procedure on establishing of a commission and arranging its work by a local executive body of a rayon (city of oblast status) to define the extent of sown area affected by the adverse weather event, and the form of inspection act; 7) approve the standard form of certificate from the hydro-meteorological service and/or competent natural/manmade emergency authorities which is used to confirm the occurrence of an adverse weather event; Article 5. State control and supervision in compulsory crop insurance 1. The competent government crop production authority and its territorial bodies shall control the producers‘ compliance with obligation related to the concluding of compulsory insurance contracts set forth in this Law. 2. The competent government crop production authority shall: 1) implement national strategy in crop production; 2) develop and implement national and other programmes related to crop production; 3) exercise state control for compliance with the laws of the Republic of Kazakhstan related to crop production;4) improve legal and economic conditions for crop production development; 5) study the current domestic and international situation in the crop production sector; 6) exercise control for the work of the agent, the mutual‘s, their compliance with the Law of RK on compulsory crop insurance; 7) consider cases of avoidance of concluding compulsory crop insurance contracts, and of non- compliance with the requirement stipulated by the Law of RK on Mutual insurance and with this Law by the Mutual‘s. 8) request and receive from the Insured, the Insurer, and the Agent information and documents required to fulfill its supervising functions; 9) set format and time of submission of information and documents by the Insured, the Insurer, and the Agent which are required to fulfill its supervising functions; 10) develop and approve the methodology for determining the extent of crop loss area; 11) design and approve the forms of reporting, checklists, risk assessment criteria, annual schedules of inspection as provided for by the Law of RK On State Control and Supervision in the Republic of Kazakhstan; 12) submit to the competent authority which exercises regulation and supervision of the financial market and financial organizations, a list of the mutual‘s indicating name and location; 3. Local executive body of a rayon (city of oblast status) shall: 1) arrange compulsory insurance for economic entities; 2) request and receive from the Insured, the Insurer, and the Agent information and documents required to fulfill its supervising functions on the form established by the authorized state body in crop production; 3) set optimal starting and completion date of sowing on particular area and with breakdown in the natural and climatic zones and types of crop products which are subject to compulsory crop insurance; 3-1) submit a list of crop producers which are subject to insurance in current year to the authorized state body in crop production; 4) establish commissions as provided for by Article 9 of this Law. - 236 - 4. The state control for the activities of the insurance agencies shall be exercised by the competent authority which regulates and supervises the financial market and financial organizations in compliance with the legislation of the Republic of Kazakhstan. Article 5-1. Procedure of state control in compulsory crop insurance For the purpose of state supervision, the territorial body of the competent government crop production authority shall: 1) request the agent to submit a list of insured persons/entities which have concluded compulsory crop insurance contracts, indicating all types of crop products insured on particular area 2) excluded by the Law of RK dated 30.12.2009 N 234-IV; 3) excluded by the Law of RK dated 17.07.2009 N 188-IV; 4) draw up protocols on administrative violations and impose penalty in compliance with the administrative law of the Republic of Kazakhstan. State control in the sphere of compulsory crop insurance shall be exercised in the form of inspection and other forms. The inspection shall be carried out in compliance with the Law of RK On State Control and Supervision in the Republic of Kazakhstan. Other forms of state control shall be carried out in compliance with this Law. Article 5-2. Peculiarities of the compulsory crop insurance implementation 1. The Provisions provided for in this Law shall be applied to the mutuals‘ taking into consideration the specifics established by the legislative acts of the Republic of Kazakhstan that govern the mutuals‘ activity. 2. Insurance of property interests of the mutuals‘ members shall be implemented in compliance with the rules of mutual insurance. 3. Producers of crop products being the members of the mutual‘s are not subject to compulsory insurance by the insurer. 4. A legal entity registered as insurance company before obtaining a license for the right to carry out the compulsory crop insurance is obliged to have branches and/or insurance agents in the capital and the cities of republican, oblast and rayon status. 5. It is not allowed to carry out any activity aimed to restrain or eliminate competition, to grant or gain unfounded advantages in concluding compulsory crop insurance contracts by some insurer over the others, and also to infringe rights and legal interests of the insured. Article 6. Types of insured events in compulsory crop insurance Insured events in compulsory crop insurance shall include loss or damage of crop products by adverse weather events or combination thereof, indicated in the inspection act and entailing losses. Article 7. Determining the amount of sum insured and types of normative costs 1. The amount of sum insured shall be determined by the contract concluded in compliance with specifics of compulsory insurance specified by this Law, separately for each type of crop products according to normative costs per one hectare multiplied by the area on which the insured incurs costs on producing this type of crop. 2. The sum insured shall be calculated based on the normative costs for one of the following types: 1) science-based agricultural technology; 2) simplified agricultural technology; 3) by three types of costs: fuel and lubricants, seeds, wages. - 237 - Article 8. Determining of the amount of insurance premium and payment procedure 1. The amount of insurance premium for each type of crop products shall be established by compulsory insurance contract but it cannot be less than the amount of insurance rate fixed by this Law and multiplied by the relevant sum insured. To establish the maximum amounts of insurance rates calculated in percentage to the sum insured for a type of crop products set by compulsory insurance contract, subject to the state subsidy assistance of insurance indemnity: 1) grain crops (for groups of oblasts): No Name of Oblast Insurance rate, in % Min. Max. 1 Аkmola, Аlmaty, East-Кazakhstan, Zhambyl, Kostanay, North- 1.78 3.48 Kazakhstan 2 Кaragandy, Kyzylorda, Pavlodar, South-Kazakhstan 3.17 5.83 3 Аktobe, West-Каzakhstan 5.21 9.15 2) oil-bearing plants (across the country) – min. 2.01%, max. 3.44%; 3) sugar beet (across the country) – min. 5.76%, max. 8,39%; 4) cotton (across the country) – min.0.92%, max.1.33%. 2. Insurance premiums shall be paid by the insured and charged to the cost of production of the insured type of crop products. 3. Insurance premiums of compulsory crop insurance shall be paid one-time by the insured to the insurer, or upon agreement with the insurer by installment, in the order and terms stipulated in the compulsory insurance contract. Should the insured fail to duly pay insurance premium on terms of payment by instalment established in the compulsory insurance contract, the contract shall be deemed unconcluded. Should the insured fail to duly pay the next sum of the insurance premium on terms of payment by installment established in the compulsory insurance contract, the insured shall pay a penalty to the insurer in the order and amount provided for by the civil legislation of the Republic of Kazakhstan. Should the insured event occur before the payment of the next insurance premium or the payment of an insurance premium was delayed, the insurer has the right to reckon the sum of the unpaid sum insured to the insurance indemnity. Article 9. Determining of amount of insurance indemnity and payment procedure 1. The insurance indemnity shall be paid in the amount of the loss incurred by the insured within the limits of the sum insured, applying no franchise. No franchise (contingent or absolute) shall be set by the insurer in the compulsory crop insurance; when the franchise is set it shall be deemed invalid. 2. The extent of the loss shall be determined as a positive difference between the extent of the normative costs per one hectare of crop production, which is set at the moment of concluding a compulsory insurance contract, and the revenue from one hectare of the area affected by the adverse natural event, multiplied by the exact area on which the given crop was produced and which was affected by an adverse weather event. The extent of areas affected by the adverse weather event shall be ascertained by a commission which must be established by a local executive body of rayon (city of oblast status) within five working days from the date of receipt of the insured‘s application. - 238 - The commission shall comprise the representatives of the local executive body of rayon (city of region status), the authorized state body in the areas of crop production, an agent, the insurer or the mutual‘s, and the insured. The commission shall inspect the sown area indicated by the insured by means of the methodology of determining the crop loss area, which is approved by the competent state a agency in the area of crop production and on the basis of findings shall define the extent of crop production loss, i.e. total or partial. Based on the results of inspection, on the day of inspecting the commission shall draw up an inspection act in three copies, for each case of adverse weather event separately or in combination thereof and for each type of crop products. The inspection act shall be signed by each member of the commission and submitted to the representatives of the agent, the insurer or the mutual‘s, and the insured. Should one of the commission members disagree with the taken decision and refuse to sign the act, he/she must provide his/her justification and attach it to the inspection act. The inspection act is deemed accepted when it is signed by three members of the commission. In case of partial loss of crops, the volume of products harvested from the areas affected by the adverse weather event is subject to assessment during the harvesting. In case of total loss of crops, the extent of loss shall be determined as the amount of normative costs per one hectare of crop production, which is set at the moment of conducting a compulsory crop insurance contract, multiplied by the area on which the given crop was produced and which was affected by the adverse weather event. 3. The income can be actual, i.e. calculated after the sale of total volume of the given crop affected by adverse weather event. 4. The income can be assessed, i.e. calculated after gathering the crops and prior to the sale of the total volume of the given crop affected by the adverse weather event. The assessed income shall be determined by the insurer upon the application of the insured or its representative. If case of disagreement the assessed income shall be determined by the assessor (independent expert). In case of disagreement with the results of assessment, the parties have the rights to prove their opinion. In case of total loss of crops, the assessed income shall not be assessed income shall be determined by the insurer. In case of total loss of crop the insurer shall not determine the assessed income. 5. In case of partial loss of crops, the indemnity shall be paid not earlier than one month but not later than three months after harvesting. The income shall be calculated 1) in compliance with Clause 3 of this Article, if the yield was realized; 2) in compliance with Clause 4 of this Article, if the yield was not realized; 3) in compliance with Clause 3 of this Article, if the yield was partially realized, while the income related to the non-realized part of the yield shall be calculated in compliance with Clause 4 of this Article. 4). In case of total loss of crops, the insurer must pay insurance indemnity to the insured in full within ten working days from the moment the insurer has received documents specified in Clause 2 of Article 10 hereof. Article 10. Terms of payment of the insurance indemnity 1. The insured shall put in a claim for insurance indemnity to the insurer in writing and enclose relevant documents. 2. The documents to be enclosed are as follows: - a copy of a compulsory insurance contract; - a copy of map (scheme) of fields location; - a certificate from hydro-meteorological service and/or the competent natural/manmade emergency authorities which ascertains the occurrence of the adverse weather event; - 239 - - a copy of the insured‘s application, which is to be marked as registered, to a local executive body of a rayon (city of oblast status) on the establishing a commission to determine the extent of area affected by the adverse weather event; - documents (listed in the a compulsory insurance contract) which confirm the receipt of actual income or in case of their absence, the assessor‘s (independent expert) report on assessment of income in case of partial loss of crops. The insurer may not request from the insured any other additional documents. 3. On receipt of documents the insurer shall draw up a certificate noting the full list of received documents and date of receipt. The certificate shall be made in 2 copies, one for the Claimant and the second one for the insurer. Article 11. Grounds to release the insurer from paying the insurance indemnity 1. The insurer shall have the right to refuse to pay the insurance indemnity to the insured in full or in part if the insured event resulted from: - deliberate actions of the insured which result in or contribute to the occurrence of the insured event, except for the actions taken in the event of self-defense and emergency; - actions of the insured which are deemed deliberate crimes or administrative offences according to the legal act of the Republic of Kazakhstan, which cause the insured event . 2. The insure may refuse to pay indemnity to the insured on the following grounds: - the insured provided intentionally false information about the object of insurance, insurance risk, insured event and its consequence; - the insurer deliberately did not take measures to minimize the loss resulted from the insured event; - the insured create obstacles to the insurer in investigating the circumstances of the insured event occurrence, and in determining the extent of loss; - non-notification of the insurer of the occurrence of an insured event, if it will not be proved that the insurer duly knew about or that the lack knowledge could not prevent him from the duty to indemnify; - submission of an application to a local executive body of a rayon (city of oblast status) on establishing a commission to determine the extent of areas affected by the adverse weather event, violating the terms set forth in Paragraph 6 of Clause 2 of Article 15 of this Law; - other events prescribed by the civil law of the Republic of Kazakhstan. 3. The decision to refuse to pay the insurance indemnity shall be taken by the insurer and communicated to the insured in writing together with justified reasons of such refuse, within seven calendar days from the date of submission of all documents specified in Clause 2 of Article 10 hereof. 4. The insured may appeal in court the insurer‘s refuse to pay the insurance indemnity. Article 12. Government support of compulsory crop insurance 1. Government support of compulsory crop insurance shall be performed by allotting budgetary funds as defined by the law on republican budget for appropriate fiscal year to the authorized state body in crop production with the purpose to reimburse fifty percent of insurance indemnity to insurers and the mutuals‘ on the insured events resulted from the adverse weather events and to pay for the agent‘s services. The funds allotted to pay for the agent‘s services may be transferred to the agent‘s account in a second-level bank as provided for by the contract between the competent crop production authority and the agent. 2. Аn agent shall open an account in the National bank of the Republic of Kazakhstan (hereinafter – the National Bank of Kazakhstan) to manage funds which are transferred by the competent crop production authority on the basis of the contract made between the competent crop production authority and the agent. Temporary free funds from the current account in the National bank of Kazakhstan may be placed on deposits in the National Bank of Kazakhstan and fund in state securities. Income side of current account in the National Bank of Kazakhstan shall be formed out of budgetary funds transferred by the competent crop production authority on the basis of the contract and income from - 240 - the placement of temporary free funds in deposits in the National Bank of Kazakhstan and in state securities. Expense side of the current account in the National Bank of Kazakhstan shall include amounts placed to deposit, and expenses to compensate for a part of insurance indemnity to the insurers and the mutuals‘. Account balance in the National Bank of Kazakhstan available at the end of fiscal year is not subject to return to the competent crop production authority and to budget. Article 13. Procedure to claim reimbursement for a part of insurance indemnity made 1. Demand for reimbursement of a part of the paid insurance indemnity shall be placed by the insurer/mutual‘s on the Agent in the form of written application with enclosed documents according to Clause 2 of this Article. 2. The documents which shall be enclosed to the application for reimbursement of a part of the paid insurance indemnity are the following: - a notarized copy of the contract of compulsory crop insurance; - documents confirming that the indemnity has been made (an original of a payment order, a copy of a withdrawal slip, a copy of a collection order stamped by the insurer/mutual‘s, or an original thereof); - a copy of court decision if the insurance indemnity was paid by the effective decision of the court; - a copy of document on assessed income certified by the insurer/mutual‘s in case of partial loss of crop. In case it will be revealed that the documents listed in part 1 of Clause 2 of this Article do not meet the requirements set by the Law of the Republic of Kazakhstan, those documents shall be returned to the insurer or the mutual‘s to eliminate comments. However, the part of the paid insurance indemnity shall be reimbursed to the insurer or the mutual‘s within 7 working days beginning from the date when the document have been submitted for the second time. 2-1. The agent shall reimburse a part of the paid insurance indemnity on the basis of contract on procedure and terms of partial reimbursement of insurance indemnity which is concluded with the insurer according to this Law. 3. The agent has the right to refuse the insurer/mutual‘s to reimburse a part of the paid insurance indemnity in case the latter communicated and/or submitted deliberately false information to the agent. 4. The insurer/mutual‘s shall bear responsibility for accuracy of data related to the amount of insurance indemnity. Article 14. Rights and obligations of the insurer 1. The insurer has the rights: - at the occurrence of the insured event, together with the insured and the assessor (independent expert) to carry out work to determine actual and/or assessed income from the crop products affected by the adverse weather events; - to request from the competent bodies relevant documents confirming the occurrence of the insured event; - to use the reports of an assessor (independent expert) (if any) to determine the amount of insurance indemnity as a result of the insured event occurrence; - to receive reimbursement for a part of insurance indemnity from the agent from of budgetary funds; - to make up proposals on the improvement of the compulsory crop insurance and submit them to the agent; - to determine an assessed income independently or by involving an appraiser (independent expert); - to refuse to conclude the compulsory insurance contract in case the insured failed to conclude it on due time as set forth in Clause 3 of Article 17 hereof. 2. The insurer is obliged: - if the insured fails to submit documents specified in Clause 2 of Article 10 hereof, to notify the insured of the missing documents immediately but not later that within 3 working days; - 241 - - at the occurrence of an insured event which entailed losses, to pay insurance indemnity against the kinds of crop products produced by the insured, in the amount, order and terms as prescribed by this Law and compulsory insurance contract; - to conclude a contract with the agent on procedures, terms and conditions of partial reimbursement of the insurance indemnity, in order to obtain a part of the insurance indemnity on the insured events occurred due to adverse weather event, provided that the compulsory insurance contract with the insured has been concluded in time as set forth in Clause 3 of Article 17 hereof; - to record and keep statistical data on performing compulsory crop insurance and submit it to the agent upon his request; - to agree the worked out regulations related to compulsory crop insurance with the competent authorities regulating and supervising the financial market and financial organizations; - to provide the agent with information on compulsory insurance contracts which have entered into force not later than on the fifth day of a month next to the month of contract concluding; - to familiarize the insured with the regulations on compulsory crop insurance; - to ensure confidentiality of insurance; - to pay for the services of the engaged by him assessor (independent expert) - to inform the agent about all court decisions and /or determinations related to the paid insurance indemnity and its amount within three days from the date they were given; - at the occurrence of adverse weather event to send his representative to a commission of a local executive body of a rayon (city of oblast status) created on the basis of the insured‘s application, to inspect and determine the extent of areas affected by the adverse weather event. 2-1. The insurer has no right to place demands related to the observance of agricultural technologies in crop production. 3. The insurer has other rights and obligations provided for by the legal acts of the Republic of Kazakhstan and the compulsory insurance contract. Article 15. Rights and obligations of the insured 1. The insured has the rights: - to be paid the insurance indemnity within the time limit stipulated by the compulsory insurance contract; - to demand for explanations from the insurer on conditions of the compulsory insurance, his rights and obligations under the compulsory insurance contract; - to familiarize with the results of assessment of losses incurred; - to dispute the insurer‘s decision to refuse to pay insurance indemnity or reduce its amount in compliance with the procedure provided for by the laws of the Republic of Kazakhstan and the compulsory insurance contract; - to apply to the court in case of disagreement with the results of assessment of actual or assessed income on each kind of crop products produced in the sown areas affected by the adverse weather event; - to choose one of the normative costs indicated in Article 7 of this Law to calculate the sum insured for the compulsory insurance contract. 2. The insured is obliged: - to make a compulsory insurance contract with the insurer within the time limit prescribed in Clause 3 of Article 17 hereof; - to pay insurance premiums in the amount, order and term stipulated by the compulsory insurance contract; - immediately but not later than five working days after he has known about the occurrence of adverse weather event, which can entail the occurrence of an insured event, to notify of it the insurer by any possible way (verbally, in writing). Verbal notification shall be later (within seventy two hours) confirmed in writing. The failure to do it due to serious reasons shall be proved by relevant document. - to provide relevant conditions for the representatives of the agent, insurer and assessor (independent expert) to observe the area on which the adverse weather event occurred, and not to prevent them to observe the state of the crops before the completion of harvesting; - 242 - - to submit an application to the local authority of the rayon (city of oblast status) on the establishment of a commission to determine the extent of area affected by the adverse weather event: - in case of short-lasting event: within three working days from the date it has occurred; - in case of short-lasting event: within ten working days after revealing its impact on crops, if the certificate from the hydro meteorological service and/or the competent natural/manmade emergency authorities confirming the insured event occurrence is available ; - to assure property interests on each separate field of sowing; - to take measures to reduce losses resulted from the event insured; - submit to the insurer documents listed in Clause 2 of Article 10 of this Law, which are required for obtaining a insurance indemnity. 3. The insured has other rights and obligations provided for by the legal acts of the Republic of Kazakhstan and the compulsory insurance contract. Article 16. Rights and obligations of the agents 1. The agent has the rights: - to work out proposals related to the improvement of the compulsory crop insurance; - to sum up the experience of the implementation of the compulsory crop insurance; - to request for relevant information including confidential insurance data from the stakeholders of the compulsory crop insurance with the purpose to exercise his/her rights and obligations. 2. the agent is obliged: - to inform the insurer and the insured of the approved normative costs for the kind of crop production; - at the occurrence of the adverse weather event, being a member of the commission established upon the insured‘s application to the local executive body of the rayon (city of oblast status), to carry out work on determining the extent of the crop area affected by the adverse weather events; - reimburse a part of the insurance indemnity to the insurer within seven working days from the moment the documents have been received according to Clause 2 of Article 13 of this Law, in the amount stipulated hereby; - conclude contract with the insured/the mutual‘s on the procedure and conditions of partial reimbursement of insurance indemnities on the basis of a standard form of reimbursement of part of indemnities on insured events occurred as a result of the adverse weather event. 3. The agent has other obligations assigned to him by the legislation of the Republic of Kazakhstan. 4. The agent shall appropriate funds to the insurers/ the mutuals‘ within the time limit stipulated by Clause 2 of this Article, only after the insurers/ the mutual‘s has executed their liabilities related to insurance indemnity provided that the insurer/the mutual‘s complies with the terms and conditions of the contract concluded with the agent. 5. The performance of the agent shall be supervised by the authorized state bodies in compliance with the legislation of the Republic of Kazakhstan. Article 16-1. Rights and obligations of the hydro meteorological service and/or the competent natural/manmade emergency authorities 1. The hydro meteorological service and/or the competent natural/manmade emergency authorities have the right when preparing a certificate confirming the occurrence of an adverse weather event, in case of dispute to carry out on-site observation. 2. The hydro meteorological service and/or the competent natural/manmade emergency authorities are obliged to issue a certificate which confirms or does not confirm the occurrence of an adverse weather event: 1) to the insured - free of charge, in two copies, within three calendar days in case of short-lasting event, and within five calendar days in case of long-lasting event; - 243 - 2) to other stakeholders - upon official written requests in compliance with the established procedure. Article 17. Contract of compulsory insurance 1. The conclusion of a contract of compulsory insurance is obligatory for crop producers and insurers. 2. A contract of compulsory insurance shall be made in writing and in two copies for the period not less than five months but not longer than twelve months. Compulsory insurance contract shall be concluded on the basis of the Insured‘s application. The requirements to its content are established by the normative and legal acts of the authorized body which regulates and supervises the financial market and financial organizations. 3. A contract of compulsory insurance for each kind of crop production shall be concluded not later than within fifteen calendar days after finishing sowing as defined by the local executive body of rayon (city of Oblast status) in compliance with Clause 3 of Article 5 hereof. 4. A contract of compulsory insurance shall terminate in the cases of: 1) expiry of the period of its effect; 2) making payment of insurance indemnity on all insured events occurred during the validity period of that contract. 5. A contract of compulsory insurance may be cancelled preterm in cases stipulated by the Civil code of the Republic of Kazakhstan. 6. A beneficiary under the contract of compulsory crop insurance shall be the insured or any other person in favor of whom the contract has been made. Article 18. Settlement of disputes Any disputes arising from the relationships of the contract of compulsory crop insurance shall be considered in order fixed by the legislation of the Republic of Kazakhstan. Article 19. Responsibility for violations of the legislation of the Republic of Kazakhstan оn compulsory crop insurance Persons who are guilty of violation of the legislation of the Republic of Kazakhstan on compulsory crop insurance shall bear responsibility in compliance with the laws of the Republic of Kazakhstan. Article 20. Procedure for entry into force of this Law This Law shall enter into force from April 1, 2004. President of the Republic of Kazakhstan - 244 - Annex 2.2. Kazakhstan Compulsory Crop Insurance Policy (English) Approved by the Enactment of the Government of the Republic of Kazakhstan of 31 October 2006 N 1036 Standard Form of Compulsory Crop Insurance Contract _______ series ________ N __ "__" _______ 200_г. _____________________________________, hereinafter referred to as the ―Insurer‖, acting on the basis of ___________________________________________________, in the person of ____________________________________________________________, on the one part, and ________________________, hereinafter referred to as the ―Insured‖, acting on the basis of ___________ on the other part, collectively referred to as the ―Parties‖, and individually referred to as the ―Party‖, on the basis if the Civil Code and the Law of the Republic of Kazakhstan of 10 March 2004 ‗On Compulsory Crop Insurance‖ (hereinafter referred to as the ―Law‖) have concluded this Contract of Compulsory insurance (hereinafter referred to as the ―Contract‖) on the following basis: 1. Terminology and Definitions 1. The following terminology and definitions are used herein: 1) inspection act on the fact of adverse weather event (hereinafter referred to as the ―Inspection act‖): document which confirms the cause-and-effect relationship of the partial and total loss of crops and the impact of the adverse natural event(s), the requirements to which are stipulated by the Law on the form established by the Government of the Republic of Kazakhstan; 2) beneficiary is the Insured or other person in behalf of whom the Contract was concluded; 3) adverse weather event: natural phenomenon (long-lasting - drought, frost-killing, lack of warmth, excess moisture in soil, excess moisture in air, flood, shallow, dry wind; and short-lasting - hail, excess rain, frost, strong wind, mudflow), which result in loss or damage of crop products; 4) object of compulsory crop insurance is the Insured‘s property interests related to partial or full indemnity of the Insured‘s losses upon the occurrence of an insured event, the amount of which is determined in accordance with the Law. 5) assessor (independent expert): a physical person or a legal entity that has a license to perform assessment; 6) total loss of crops: consequences of the impact of adverse weather events on crops when the expenses for further growing of crops and harvesting exceed the expected income from the yield; 7) partial loss of crops: consequences of the impact of adverse weather events on crops when the assessed or actual income from 1 hectare of crop production is less than the normative cost per 1 hectare of crop production which is set at the moment of concluding compulsory insurance contract; 8) expected income from yield: revenue determined by the commission in compliance with the methodology for the determining of crop loss area which is approved by the authorized body in the sphere of crop production; 9) crop product: product obtained by means of agricultural crops cultivation (cereals, oil-bearing crops, sugar beet, cotton); 10) crop production: a group of manufacturers in agricultural industry including agricultural producers, i.e. individuals and legal entities engaged in production of crop products; 11) insurer: a legal entity granted a license to implement compulsory crop insurance in the order prescribed by the legislation of the Republic of Kazakhstan, which shall pay the insurance - 245 - indemnity to the insurer or any other person in favor of whom the contract was concluded (beneficiary) within the amount provided for by the contract (the sum insured); 12) insured: a person engaged in producing crop products and who concluded a compulsory insurance contract with the insurer; 13) sum insured: a sum of money which an object of compulsory insurance has been insured for, and which represents the top amount of the insurer‘ s responsibility at the occurrence of insured event; 14) insurance premium: a sum of money which the insured must pay to the insurer for undertaking the liability to pay an insurance indemnity to the insured (beneficiary) in the amount provided for by the compulsory insurance contract; 15) insured event: an accident the occurrence of which can result in payment of insurance indemnity as stipulated by compulsory insurance contract; Note. Clause 1 as revised by the Enactment of the Government of RK dated 08.10.2010 No 1042 2. Subject of Contract 2. Under this Contract the Insurer shall provide insurance coverage for the Insured‘s property interests related to partial or full indemnity of the Insured‘s losses upon the occurrence of an insured event, the amount of which shall be determined in compliance with the law. 3. The effect of this Contract shall not be applied to any activities related to growth of crop products on non-agricultural land and crop products storage. 4. The Insurer shall not provide insurance coverage for the Insured‘s losses which occurred due to causes others than adverse weather events. 3. Amount of Sum Insured and Insurance Premium 5. The amount of sum insured shall be established for each type of crop products according to normative costs per one hectare multiplied by the area on which the insured incurs costs on producing this type of crop. 6. The sum insured under this Contract shall be determined in the amount of __________ (________) KZT in accordance with the Annex hereto, соnstituting an integral part hereof. 7. The indemnity of loss and/or damage in excess of the sum insured shall be born by the Insured. 8. The insurance premium under this Contract shall be __________ (________) KZT and shall be determined within the time limit and in order set forth in Annex hereto which is considered to be its integral part. 9. Should the insured fail to duly pay insurance premium on terms one-time payment as provided for in the compulsory insurance contract, the contract shall be deemed unconcluded. 10. Should the insured fail to duly pay the next sum of the insurance premium on terms of payment by installment as provided for in this compulsory insurance contract, the insured shall pay a penalty to the insurer in the order and amount provided for by the civil legislation of the Republic of Kazakhstan. 11. Should the insured event occur before the payment of the next insurance premium or the payment of an insurance premium was delayed, the insurer has the right to reckon the sum of the unpaid sum insured to the insurance indemnity. 4. Insured event 12. Insured events in compulsory crop insurance shall include loss or damage of crop products by adverse weather events or combination thereof, indicated in the inspection act and entailing losses. Note. Clause 12 with amendments made by the Enactment of the Government of RK dated 08.10.2010 No 1042. 13. The proof of insured event occurrence and loss shall rest with the Insured. 5. Rights and Obligations of the Parties - 246 - 14. The insurer has the rights: 1) at the occurrence of the insured event, together with the insured and the assessor (independent expert) to carry out work to determine actual and/or assessed income from the crop products affected by the adverse weather events; 2) to request from the competent bodies relevant documents confirming the occurrence of the insured event; 3) to use the reports of an assessor (independent expert) (if any) to determine the amount of insurance indemnity as a result of the insured event occurrence; 4) to determine an assessed income independently or by involving an appraiser (independent expert); 15. The Insurer is obliged: 1) to familiarize the Insured with the rules of compulsory crop insurance; 2) to ensure confidentiality of insurance; 3) at the occurrence of adverse weather event to send his representative to a commission of a local executive body of a rayon (city of oblast status) created on the basis of the insured‘s application, to inspect and determine the extent of areas affected by the adverse weather event; 4) at the occurrence of an insured event which entailed losses, to pay insurance indemnity against the kinds of crop products produced by the insured, in the amount, order and terms as prescribed by the Law and this compulsory insurance contract; 5) to pay for the services of the engaged by him assessor (independent expert); 16. The insurer has other rights and obligations provided for by the legal acts of the Republic of Kazakhstan and this compulsory insurance contract; 17. The insurer has no right to place demands related to the observance of agricultural technologies in crop production. 18. The insured has the right: 1) to be paid the insurance indemnity within the time limit specified herein; 2) to demand for explanations from the insurer on compulsory insurance conditions and his rights and obligations set forth herein; 3) to familiarize with the results of assessment of losses incurred; 4) to dispute the insurer‘s decision to refuse to pay insurance indemnity or reduce its amount in compliance with the procedure provided for by the laws of the Republic of Kazakhstan and this Contract; 5) to apply to the court in case of disagreement with the results of assessment of actual or assessed income on each kind of crop products produced in the sown areas affected by the adverse weather event; 6) to chose one of normative costs to calculate sum insured when concluding this Contract. 19. The is obliged: 1) to assure property interests on each separate field of sowing; 2) to pay insurance premiums in an amount, manner and terms specified herein; 3) to take measures to reduce losses resulted from the event insured; 4) immediately but not later than within five working days after he has known about the occurrence of adverse weather event, which can entail the occurrence of an insured event, to notify of it the insurer by any possible way (verbally, in writing). Verbal notification shall be later (within seventy two hours) confirmed in writing. The failure to do it due to serious reasons shall be proved by documentary evidence; 5) to submit an application to the local authority of the rayon (city of oblast status) on the establishment of a commission to determine the extent of area affected by the adverse weather event: - in case of short-lasting event: within three working days from the date its occurrence; - in case of long-lasting event: within ten working days after revealing its impact on crops, if the certificate from the hydro meteorological service and/or the competent natural/manmade emergency authorities confirming the insured event occurrence is available; 6) to notify the insurer of all Contracts concluded or being under conclusion in respect of the object of insurance; 7) to provide relevant conditions for the representatives of the insurer and assessor (independent expert) to observe the area on which the adverse weather event occurred, and not to prevent them from observation of the crops state until the completion of harvesting; 8) submit to the insurer all documents which are required for obtaining the insurance indemnity. 20. The insured has other rights and obligations provided for by the legal acts of the Republic of - 247 - Kazakhstan and this Contract. . 6. Determining the amount of insurance indemnity and payment procedure 21. The amount of insurance indemnity amount shall be determined in compliance with the Law and this Contract: 1) in case of total loss of crops, the extent of loss shall be determined as the amount of normative costs per one hectare of crop production, which is set at the moment of conducting a compulsory crop insurance contract, multiplied by the area on which the given crop was produced and which was affected by the adverse weather event; in case of full damage EL = (NC * S ha); S – area affected by adverse weather event; AL – extent of loss; NC – normative costs; 2) in case of partial crop damage, the volume of production which was collected from the area effected by the adverse weather event, is subject to obligatory appraisal during the harvesting. The revenue shall be calculated in the following way: after realization of the whole volume of the kind of crop production on the basis of actual income; prior to realization of the whole volume of the kind of crop production on the basis of estimated income; in case the yield was realized partially and there is a part of yield unrealized, then in compliance with Paragraph 2 & 3, Sub-clause 2, Clause 21 thereof; in case of partial loss EL = (NC –income from 1 ha of the area which was affected by the adverse weather event) * S ha (area which was affected by adverse weather event). Note. Clause 21 amended by the Enactment of the Government of the Republic of Kazakhstan dated 28.06.2008 N 642. 22. .The insured shall put in a claim for insurance indemnity to the insurer in writing and enclose the following documents: - a copy of a compulsory insurance contract; - a copy of map (scheme) of fields location; - a certificate from hydro-meteorological service and/or the competent natural/manmade emergency authorities which ascertains the occurrence of the adverse weather event; - a copy of the insured‘s application, which is to be marked as registered, to a local executive body of a rayon (city of oblast status) on the establishing a commission to determine the extent of area affected by the adverse weather event; in case of partial loss the insurer shall also submit documents which confirm obtaining of income, and namely: - copies of crop production purchase-and-sale contracts; - copies of invoices and shipping documents; - income estimation certificate issued by the insurer or an independent expert. Note. Clause 22 with amendments by the Enactment of the Government of RK dated 30.10.2009 No 1728 (for procedure for entry into force, please see Clause 2) 23. On receipt of documents the insurer shall draw up a certificate noting the full list of received documents and date of receipt. The certificate shall be made in 2 copies, one for the Claimant and the second one for the insurer. Note. Clause 23 with amendments by the Enactment of the Government of RK dated 08.10.2010 No 1042. 23-1. Should the insured fail to submit the documents as provided for by Clause 22 of this Contract, the insurer shall immediately, but not later that within three working days notify the insurer in writing of the missing documents. Note. Section 6 is amended by adding Clause 12 by the Enactment of the Government of RK dated 08.10.2010 No 1042. - 248 - 24. In case of total crop loss, no estimated income shall be determined by the insurer. 25. In case of total crop loss, the insurance indemnity shall be paid within 10 working days from the date of signing the inspection act, on the basis of the inspection act and the documents stipulated in Clause 22 hereof. 26. The insurance indemnity shall be paid if event insured occurred during the period validity if this Contract. 26-1. The beneficiary under the Contract is __________________. The details of the beneficiary shall be indicated by the insureв in application for insurance indemnity. Note. The standard contract form is amended by adding Clause 26-1 according to Enactment of the Government of RK dated 30.10.2009 No 1728 (for procedure for entry into force, please see Clause 2) 27. If under this Contract the sum insured is paid in full, the Contract shall be deemed terminated. 28. The decision to refuse payment of the insurance indemnity shall be taken by the insurer and communicated to the insured in writing together with justified reasons of such refuse, within seven calendar days from the date of submission of all documents specified in Clause 22 of this Contract. 7. Release of the Insurer from paying insurance indemnity 29. The insurer shall have the right to refuse to pay the insurance indemnity to the insured in full or in part if the insured event resulted from: 1) deliberate actions of the insured which result in or contribute to the occurrence of the insured event, except for the actions taken in the event of self-defense and emergency; 2) actions of the insured which are deemed deliberate crimes or administrative offences according to the legal act of the Republic of Kazakhstan, which cause the insured event. 30. The insurer may refuse to pay indemnity to the insured on the following grounds: 1) the insured provided intentionally false information about the object of insurance, insurance risk, insured event and its consequence; 2) the insurer deliberately did not take measures to minimize the loss resulted from the insured event; 3) the insured create obstacles to the insurer in investigating the circumstances of the insured event occurrence, and in determining the extent of loss; 4) non-notification of the insurer of the occurrence of an insured event, if it will not be proved that the insurer duly knew about or that the lack knowledge could not prevent him from the duty to indemnify; 5) submission of an application to a local executive body of a rayon (city of oblast status) on establishing a commission to determine the extent of areas affected by the adverse weather event, violating the terms set forth in Sub-clause 5 of Clause 19 of this Contract. 6) other cases prescribed by the civil law of the Republic of Kazakhstan. 8. Force-Majeure 31. The Parties shall be released from any liability for partial or non-fulfillment of the obligations under this Contract if such improper fulfillment was caused by force-majeure including natural disaster, hostilities, nuclear exposure, strikes, civil disorders and also prohibitive measures envisaged by the legal acts of the government agencies of the Republic of Kazakhstan, if these circumstances directly influenced the fulfillment by the Parties of their obligations under this Contract. 32. In case of force majeure which makes impossible the fulfillment for the Party of its obligations under the Contract, then this Party shall inform the other Party of it and provide relevant evidences not later than within 5 working days from the date of its occurrence. 33. The circumstances indicated in Clause 31 shall be certified by the competent government authorities and organizations. 34. In case of undue notification, the Party shall be ruled ineligible to plead to any of the above circumstances as a reason for exemption from responsibility for non-fulfillment or improper fulfillment of liabilities under this Contract. 35. If force-majeure lasts for more than one month, then any of the Parties is entitled to solely terminate this Contract - 249 - 9. Validity Period and Place of the Insurance Contract 36. This Contract is deemed valid and binding for all Parties from the moment of payment of the insurance premium by the Insured and it is in force till ___________ 20__. Note. Clause 36 with amendments by the Enactment of the Government of RK dated 30.10.2009 No 1728 (for procedure for entry into force, please see Clause 2). 37. In case the insurance premium is paid by installment as provided for by this Contract, it shall be deemed valid from the date of payment of the first installment unless otherwise specified by the legal acts. 38. The insurance coverage period shall be equal to the Contract validity period. 39. Under the Contract, the place of the Contract is the territory of the Republic of Kazakhstan. 10. Grounds for Terminating Insurance Contract 40. This Contract shall be terminated in the following cases: 1) expiry of the Contract validity period; 2) insurance indemnity has been paid out for all insured events which occurred during the Contracts validity period. 41. The Contract may be terminated pre-term in cases stipulated by the Civil Law of the Republic of Kazakhstan. 11. Liabilities of the Parties 42. In case of non-fulfillment or improper fulfillment of obligations undertaken by the Parties under this Contract, the Parties shall bear responsibility as required by the effective law of the Republic of Kazakhstan. 12. Procedure for Introduction of Amendments in the Contract 43. Introduction of amendments in the conditions hereto shall be valid on written consent of the Parties of this Contract. 13. Dispute Settlement 44. Any disputes that may arise with compulsory crop insurance shall be regulated in accordance with the legislation of the Republic of Kazakhstan. 14. Additional Conditions 45. Annexes, changes and amendments to this Contract hereto shall be deemed integral part hereof and valid only if made in writing and signed by both Parties. 46. Neither of the Parties shall be entitled to transfer its rights and obligations under the Contract to third party without the written approval of the other Party. 47. Other interrelations not stipulated by the conditions of this Contract shall be regulated by the civil law of the Republic of Kazakhstan. 48. The Contract is made in duplicate having equal legal force for each of the Parties. 15. Bank Details of the Parties "Insurer" "Insured" Name: _____________ Name: ____________ Address: ____________________ Address: ___________________ TRN (Taxpayer Registration No) ____ TRN ______________________ - 250 - IIC (Individual Identity Code) ____ IIC ______________________ BIC (Bank Identifier Code) _________ BIC ______________________ - 251 - Annex to Standard form of Compulsory Crop Insurance Contract Sum Insured and Insurance Premium Calculation For Compulsory Crop Insurance Contract of ______________ 2010 concluded with _______________________ (farmer‘s name) _________ rayon _______________ oblast _____________________________ natural climatic zone Field No Total Free area Actual sown Kind of Normative Sum Insurance Insurance as area (ha, non- area (ha) crop costs per 1 ha insured rate (%) premium indicated (ha) sown product (KZT) total By Ref.No KZT in the area, kinds fields including map fallow land) Agrotechnology Reference Number: Payment of insurance premium on one-time basis shall be made not later than ____ banking days after signing this Contract. Payment of insurance premium by installment shall be made on the following terms: 1st installment ___ % of insurance premium at the amount of _______ KZT within _ days after signing this Contract; 2nd installment ___ % of insurance premium at the amount of _______ KZT until ___________ 2011. Insurer _________________ Insured _______________ (signature, stamp) (signature, stamp) - 252 - Annex 2.3. 2009/2010 Normative Costs of Production for Spring Wheat by Oblast and by Zone (KZT/Hectare) Science- Simplified Costs based on 3 based Agro- Agro- types of expenses technology Technology (salary, fuels & lubricants and Oblast Agro-ecological Zone seeds) Akmola 2nd Steppe Zone 8,175 6,239 3,457 Akmola 3rd Dry Steppe 8,511 5,787 3,426 Aktobe II Steppe Zone 9,489 6,678 3,789 Aktobe III Dry Steppe Zone 8,606 6,033 3,739 Aktobe IV Semi-Desertic Zone 5,758 3,318 1,923 Aktobe V Desert Zone 5,719 2,978 1,920 Almaty Irrigated V Desert Zone 10,960 7,183 3,822 Almaty IV Priedmont Desert Steppe Zone 10,964 7,183 3,829 Almaty IX Central Asian Rock Zone 10,788 7,059 3,737 Almaty Rainfed VI Priedmont Desert Steppe 8,396 5,559 3,551 Almaty IX Central Asian Rock Zone 7,939 5,529 3,580 EKO II Steppe Zone 8,050 5,295 3,307 EKO III Dry Steppe Zone 8,545 5,201 3,412 EKO IV Semi-desertic Zone 8,298 5,078 3,357 EKO V Desert Zone 7,678 5,078 3,235 EKO VI Priedmont Desert Steppe Zone 7,678 5,078 3,319 EKO IX Central Asian Rock Zone 7,664 5,078 3,404 EKO X South Siberian Rock Zone 7,490 5,078 3,482 Karanganda Oblast III Dry Steppe Zone 7,252 5,306 2,648 Karanganda Oblast IV Semi-desertic Zone 7,405 5,553 2,521 Karanganda Oblast V Desert Zone 7,295 5,408 2,355 Kostanay Oblast II Steppe Zone 9,489 6,678 3,789 Kostanay Oblast III Dry Steppe Zone 8,606 6,033 3,739 Kostanay Oblast IV Semi-desertic Zone 8,580 5,666 3,824 Kostanay Oblast V Desert Zone 8,375 5,443 3,682 Kyzylorda Oblast V Desert Zone (Irrigated) 10,302 6,851 3,071 NKO I Forest Steppe Zone 10,511 7,259 3,686 NKO II Steppe Zone 9,053 6,333 3,663 Pavlodar Oblast II Steppe Zone 8,762 5,739 3,642 Pavlodar Oblast III Dry Steppe Zone 8,502 5,890 3,015 SKO Irrigated VII Subtropic Desert Zone 11,225 7,381 3,939 VIII Subtropic Priedmont Desert SKO Zone 11,053 7,258 3,852 SKO IX Central Asian Rock Zone 11,022 7,212 3,784 SKO Rainfed V Desert Zone 8,867 5,741 3,506 SKO VI Priedmont Desert Steppe Zone 8,801 5,697 3,504 - 253 - VIII Subtropic Priedmont Desert SKO Zone 8,757 5,697 3,476 SKO IX Central Asian Rock Zone 8,620 5,574 3,397 WKO III Dry Steppe Zone 5,887 3,606 2,159 WKO IV Semi-desertic Zone 5,767 3,839 2,112 WKO V Desert Zone 5,502 3,136 2,074 Zhambyl Oblast Irrigated V Desert Zone 14,118 8,540 4,748 Zhambyl Oblast VI Priedmont Desert Steppe Zone 13,577 8,011 5,039 Zhambyl Oblast IX Central Asian Rock Zone 12,330 7,764 4,857 Zhambyl Oblast Rainfed VI Priedmont Desert Steppe 8,131 5,511 3,539 Source: GORK, Regulation of 25 March 2009, No 410 - 254 - Annex 3: Kazakhstan: Mandatory Crop Insurance Results The mandatory crop insurance scheme in Kazakhstan has been implemented since the crop year 2005. Evolution of Insured Area, Sum Insured and Premiums Owing to its mandatory feature the crop insurance program is issuing a significant number of insurance policies each year. According to the Fund for Financial support in Agriculture (FFSA), the mandatory crop insurance scheme has been insuring an average of 23,420 insurance policies per year, with the peak number of policies issued under the program during the crop year 2007 with 34,000 thousand policies issued. The average area per insurance policy is 560 hectares. Figure A3.1. Number of Crop Insurance Policies issued from 2005 to 2010 Number of Insurance policies (2005-2010) 40.0 1,000 Avg. Area per Insured Policy Insurance Policies (in'000) 35.0 800 30.0 25.0 600 20.0 (hes) 15.0 400 10.0 200 5.0 0.0 0 2005 2006 2007 2008 2009 2010 Insured Policies (in ' 000) Avg. Area per Insurance Policy Source: FFSA 2011 The level of adoption of crop insurance in wheat has increased since the inception of the program in 2005. The trend during the period 2005-2010 was to increase the area insured in the program. The total insured area under the mandatory crop insurance program in Kazakhstan has increased from an insured area of 10.5 million hectares in 2005, to 12.7 million hectares in 2010. Figure A3.2. Evolution of the Total Insured Area from 2005 to 2010 (Million Hectares) Evolution of the Total Insured Area (2005-2010) 16.0 Planted Area (hectares) 14.0 12.0 10.0 8.0 15.0 14.5 6.0 12.1 12.7 10.5 4.0 9.1 2.0 0.0 2005 2006 2007 2008 2009 2010 Insured Area Source: FFSA, 2011 The crop insurance portfolio is concentrated on spring crops. During the period 2005-2010, the spring crops accounted, on average, for 98 percent of the total insured area during the period. There are some minor areas of winter crops (e.g. winter sown cereals) in the south of Kazakhstan. Out of the spring - 255 - crops, spring wheat, accounting for 86 percent of the total crop area, is the main insured crop in Kazakhstan. Figure A3.3. Distribution of Crop Insurance Portfolio between Winter and Spring crops 2005-10 (% of total insured area) Distribution of the crop insurance portfolio between winter and spring crops (2005-2010) 2% 98% Spring Crop Winter Crop Source: FFSA, 2011 Despite the mandatory nature of the crop insurance scheme, the levels of adoption of this risk transfer financial tool is not 100%. Over the period 2005-2010, and average of 73 percent of the total planted area of the compulsorily insurable crop in Kazakhstan has been insured. The program experienced issues in 2005 during its first year of implementation. Because of this reason the level of adoption of crop insurance was reduced in 2006 to only 59 percent of total insurable area in the second year of implementation. During the crop years 2008 and 2009, the adoption of crop insurance reached its highest level of uptake in 2008, the total insured area accounted for 84 percent of the total planted area in the country. Figure A3.4.Evolution of Insured Area as a percentage of Total Planted Area 2005-2010 Evolution of the Total Planted and the Insured Area (2005-2010) 100% 22% 16% 18% Planted Area (hectares) 30% 32% 80% 41% 60% 40% 78% 84% 82% 70% 68% 59% 20% 0% 2005 2006 2007 2008 2009 2010 Insured Area Not Insured Area Source: FFSA, 2011 The insured area expressed as a percentage of the total planted area of insurable crops is not homogeneous across the different Oblasts in the country. In the northernmost Oblasts of NKO, Kostanay and Akmola, where average rainfall is relatively high and where the major spring wheat farming areas are located, a very high percentage or more than 75% of the total crop area has been insured over the past 6 years. Conversely, in the southern areas of Kazakhstan where average farm size is much smaller and where crop production takes place under irrigation, the ratio of insured area over total crop - 256 - area is much lower. For instance, in Almaty – where crop production mostly take place under irrigation- the ratio between insured area and total crop area is lower than in any northern Oblast. Map A3.1. Average Insured Area as a percent of Total cultivated Area per Oblast (2005-2010) Source: FFSA, 2011 The total sum insured of the compulsory crop insurance program has increased over the past 6 years. The trend along the period 2005-2010 was to increase the total sum insured in the program. The total sum insured under the mandatory crop insurance program in Kazakhstan increased from KZT 34.4 billion in 2005, to KZT 47.3 billion in 2010. The fall in the total sum insured in the crop year 2006 was due to the reduced crop insurance uptake during this year. The reduction on the uptake of insurance was due to problems the program experienced during the prior 2005 crop season. Figure A3.5. Evolution of Total Sum Insured on the Compulsory Crop Insurance Scheme 2005-10 (KZT Billion) Evolution of the Total Sum Insured (2005-2010) 60.0 Sum Insured (billion KZT 50.0 40.0 30.0 52.9 46.6 47.3 20.0 34.4 34.8 26.7 10.0 0.0 2005 2006 2007 2008 2009 2010 Sum Insured (billion KZT) Source: FFSA, 2011 - 257 - The sum insured is not homogeneously distributed across the country. The major concentration of insured liability (total sum insured) is observed in the main crop production areas (Akmola, Kostanay, and NKO). These three oblasts account for 77 percent of the total sum insured under the mandatory crop insurance scheme. Within the areas comprised by these Oblasts, the central and north rayons of Kostanay, the rayons situated in the southwest of NKO, and the rayons situated in the western area of Akmola Oblast are those which show the highest risk exposures in Kazakhstan. Map A3.2. Distribution of Average Total Annual Sum Insured by Oblast and by Rayon 2005-10 (KZT Million) Source: Authors‘ analysis of FFSA results 2011 The average sum insured per hectare is very low in Kazakhstan. The average sum insured is equivalent to the normative cost which is determined for each crop and rayon on each year by MoA (See Annex 2.3 for full details of 2009-10 normative costs per hectare for each insured crop in each Oblast). The normative cost determined by MoA is usually very low. An analysis for the average normative cost for wheat in the period 2005-2010 shows that the average normative cost for this period was KZT 2,380 per hectare, being the average maximum the normative cost for wheat the one established by MoA for the crop season 2010, KZT 3728 per hectare and the minimum the one established by MoA for wheat crop season 2007, KZT 2871/hectare. - 258 - Figure A3.6. Evolution of the Average Sum Insured per Hectare 2005-2010 (KZT/Ha) Evolution of the Average Sum Insured per hectare (2005-2010) 4,000.0 Average Sum Insured ( KZT/hectare) 3,728 3,500.0 3,532 3,288 3,226 3,000.0 2,935 2,871 2,500.0 2,000.0 1,500.0 1,000.0 500.0 0.0 2005 2006 2007 2008 2009 2010 Source: FFSA, 2011 The equivalent wheat covered yields are extremely low. As a result of the extremely low average sum insured per hectare, the equivalent wheat yield covered under this insurance program is also very low. The average equivalent wheat yield covered under the mandatory crop insurance scheme during the period 2005-2010 was only 1.6 centrums per hectare (160 Kg/Ha). Owing to the uptrend in the price of international agricultural commodities, the average wheat yield equivalent to the average sum insured shows a decreasing trend during the same period. The wheat prices in Kazakhstan have increased 3.5 times during the period 2005 to 2010, from KZT 1,149 per centrum in 2005 to KZT 3,988 per Centrum in 2010. During the same period the average normative costs of production/average sum insured per hectare has only increased marginally from KZT 3,288 per hectare in 2005 to KZT 3,728 per hectare in 2010,an increase equivalent to 14 percent only over 6 years (Figure A3.6). As a result of the increase in the price of wheat and the relative low increase in the insured normative costs of production, the average wheat yield equivalent to the average sum insured has been reduced significantly from an average level in 2005 of 2.86 centrums per hectare, to less than one centrum per hectare in 2010 (Figure A3.7). Figure A3.7. Evolution of Average wheat prices used to value crop salvage (KZT/centrum) and equivalent Insured Spring Wheat Crop Yields (Centrum’s per hectare). Evolution of wheat yield equivalent coverage and wheat average farm gate prices (2005-2010) 4,500.0 3.50 Wheat yield equivalent (centrum/he) Wheat Prices ( KZT/Centrum) 4,000.0 3,988 3.00 3,500.0 2.86 3,083 2.50 3,000.0 2.30 2,500.0 2,560 2.00 2,221 2,000.0 1.50 1.29 1.38 1,500.0 1.05 1,149 1,277 0.93 1.00 1,000.0 500.0 0.50 0.0 0.00 2005 2006 2007 2008 2009 2010 Price (KZT/centrum) Wheat Yield Equivalent (Centrum/he) Source: FSSA (2011), JS "KazAgroMarketing" - 259 - Average Premium rates have been declining since the inception of the program in 2005. Two periods can be distinguished in terms of premium rates application. The first period comprises the first three years of the implementation of the mandatory crop insurance program (2005, 2006, and 2007). During this period the average premium rates for the program were maintained relatively high between 2.60% and 2.90%. The second period comprises the last three years of the program (2008, 2009, and 2010). During the last three years of application of the program the average premium rate fell from an average of 2.68 percent for the three-year period 2005-2007, to only 2.24 percent on average between 2008 and 2010. The reason is because GoK decided to modify the rates for the mandatory crop insurance program and to permit companies to charge flexible rates between an agreed maximum and minimum. Competition especially among the Mutual Insurance companies which have been permitted to underwrite this program since 2008 means that average rates have been forced down to the minimum levels set by government. (Figure A3.8). Figure A3.8. Evolution of Average Premium Rates 2005 to 2010 (rate as % of sum insured) Evolution of the Average Premium rate (2005-2010) 3.00% Average Premium Rate ( % TSI) 2.90% 2.87% 2.80% 2.70% 2.60% 2.61% 2.57% y = -0.001x + 0.0282 2.50% R² = 0.5008 2.40% 2.34% 2.30% 2.27% 2.20% 2.10% 2.11% 2.00% 2005 2006 2007 2008 2009 2010 Source: Authors from FSSA (2011) The average premium rates charged by the companies are not homogenous throughout the country. The average premium rate at national level for the period 2008-2010 was 2.24 percent. However, different Oblasts throughout the country show different premium rates according to the risk they face. For instance, the Oblasts situated in the main crop production areas in the country (Kostanay, NKO and Akmola ) show lower average premium rates – 1.66 percent in average - than those Oblast situated in more risky areas from the agricultural production standpoint of view such as in WKO – 7.42 percent- or Pavlodar Oblast – 3.24 percent. The same situation is observed at rayon level within an Oblast. Those rayons situated in low risk areas from the agricultural production standpoint of view, show lower average premium rates than those rayons situated in high risk agricultural production areas in the same Oblast. (Map A3.3.). - 260 - Map A3.3. Average Crop Insurance premium rates charged by Rayon and by Oblast 2008-2010 Source: Authors based on FSSA 2011 Analysis of Claims on Compulsory Crop Insurance Scheme The program shows acceptable levels of claims frequency. Over the six year period on average claims have been declared on 5.9 percent of the total policies issued under the program each year. Out of the 5.9 percent insurance policies denounced, 79 percent of these declared losses resulted in a claims payment by insurers. For each of the insurance policies denounced to the insurance company an average of 1.3 field- based inspections were carried out. Each inspection represents an additional cost for the MoA/FFSA and the Insurers. (Figure 3.9). Figure A3.9. Claims Frequency by year (percent of policies declaring a loss and receiving a claims payment) Claim Frequency (2005-2010) 14.0% 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 0.0% 2005 2006 2007 2008 2009 2010 Source: Authors‘ analysis of FFSA results - 261 - The value of paid claims has increased significantly in the past three years which also coincide which bad drought years. The average value of paid claims paid is KZT 1.37 billion per year: the lowest claims were incurred in 2006 and the highest claims valued at nearly KZT 2.8 billion were paid in 2010. The claims data for 2010 should be interpreted carefully. The expected return period for a drought with the severity that the one occurred in 2010 is 1 every 50 or more years. The annual average loss cost (annual claims divided by annual total liability and expressed as a percentage) for the mandatory crop insurance program in Kazakhstan is 3.21 percent over the past 6 years. The lowest loss cost value observed during the period 2005-2010 is 1.79 percent in 2006. The highest loss cost value observed during the period 2005-2010 was observed as consequence of the drought of 2010 when total losses reached 5.93% of the liabilities written under the crop insurance program during that year. (Figure A3.10) Figure A3.10. Evolution of Paid Claims (KZT Billion) and Annual Loss Cost (2005-2010) Evolution of claims and average loss cost for the Program (2005-2010) 7.00% 3.00 Average Loss Cost ( Claims/ TSI) Total Claims (KZT billions) 6.00% 5.93% 2.50 5.00% 2.00 4.00% 3.67% 1.50 3.00% 3.10% 2.77% 1.00 2.00% 1.79% 2.01% 1.00% 0.50 0.00% 0.00 2005 2006 2007 2008 2009 2010 Source: Authors‘ analysis of FFSA results The main cause of loss under the mandatory crop insurance scheme is drought. Drought events have accounted for 91.5 percent of the total damaged area in the period 2006-2010. Including the losses due to drought in combination with other perils such as hail or freeze, the importance of drought peril in the total losses of mandatory crop insurance scheme rises to 97 percent. The second most important cause of loss over the past six years has been hail storm accounting for 2.5 percent of the total damaged area. Finally, losses due to excess moisture and freeze have been insignificant to date. (Figure A3.11). - 262 - Figure A3.11. Main Causes of Loss on Compulsory Crop Insurance Program (% of damaged area) Kazakhstan Mandatory Crop Insurance : Main Causes of Losses Drought & Freeze 1.5% Drought & Hailstorms Drought 4.0% 91.5% Excess of Moisture 0.3% Freeze 0.2% Hailstorms 2.5% Source: Authors‘ analysis of FFSA results The claims pattern is very different in the different geographic regions of the country and according to the differential drought and other climatic risk exposures faced in each region. The average loss cost for the period 2005-2010 is 3.21 percent. Different areas of the country show different loss costs according to the risk they face. For instance, the Oblasts situated in the higher rainfall main crop production areas in the country (Kostanay, NKO and Akmola Oblast) show very low average 6-year loss costs varying from 0.5 percent to 2 percent of the total liabilities. Conversely the Oblasts situated in the much drier western part of the country show very high loss cost ratios. For instance, WKO show an average loss cost ratio of 39.6 percent of the total liability. Aktobe, which is situated between the main crop production area and WKO, also shows a very high average loss cost ratio of 22.2% between 2005 and 2010 The central and eastern Oblasts of Kazakhstan (Karaganda, EKO and Almaty) show loss cost values between 4.6 percent and 6.1 percent. The southernmost Oblast in Kazakhstan, where crop production is mostly irrigated, show loss cost ratios lower than 1 percent of the total liabilities. (Map A3.4). The claims pattern also varies considerably within Oblasts at the Rayon level which clearly indicates a need to introduce a system or Rayon-level actuarial rating. Loss cost values for the period 2005-2010 also vary very locally. Even within a same Oblast it is possible to find very dissimilar loss costs for different rayons. For instance, in Akmola Oblast the rayons situated in the eastern areas of the Oblast, which is dryer than the rest of the Oblast, show higher loss cost ratios than the north-western rayons of Akmola Oblast which enjoy higher average rainfall and are therefore less risky. (Map A3.4). - 263 - Map A3.4. Spring Wheat: 6-Year Average Loss Costs per Oblast and Rayon Source: Authors‘ analysis of FFSA results Analysis of Mandatory Crop Insurance Program Results 2005 to 2010 Overall, the mandatory crop insurance program has been unprofitable over the past 6 years. . The average gross loss ratio for the program for the period 2005 to 2010 is 140 percent prior to the operation of the government 50% claims compensation scheme. This means that the gross claims paid under the program have been 40 percent higher than the premiums collected from the program. This loss ratio indicates that the program has not been financially sustainable along its history. Only in two years (2006 and 2007) out of the six years of the series have the premiums been higher than the claims. After application of the Government‘s 50% claims subsidy payments, the average loss ratio for the insurance companies over the past 6 years has been 75% on completion of the 2010 season. (Table A3.1 and Figure A3.12). Over the past three years the crop insurance results have deteriorated. Over the past three years the program has incurred major drought losses with loss ratios of 156% in 2008, 131% in 2009 and then finally 261% in 2010. This indicates that the scheme is under-rated and that local scheme management needs to review where the major claims are occurring and to take necessary actions to reverse this trend. The loss ratio for year 2010 should be considered carefully. The recurrence period for a drought like the one occurred in 2010 is 1 year out of about 50 years. Having said it would not be fair for the calculation of the expected loss ratios to load in full the losses due to 2010 drought. - 264 - Table A3.1. Summary of Crop Insurance Results (Premiums, Claims and Loss Ratios) 2005 to 2010 Item 2005 2006 2007 2008 2009 2010 Total No. of Policies (000) 19.0 13.6 25.4 34.0 32.2 16.8 141.0 Total Insured Area (Million Ha) 10.5 9.1 12.1 14.5 15.0 12.7 73.8 Sum Insured (Million KZT) 34,372 26,650 34,796 46,645 52,903 47,266 242,631 Premiums (Million KZT) 899 685 997 1,093 1,114 1,074 5,862 Average Premium Rate % 2.61% 2.57% 2.87% 2.34% 2.11% 2.27% 2.42% Claim payments (Million KZT) 1,065 478 701 1,710 1,465 2,805 8,223 Loss Ratio (%) 119% 70% 70% 156% 131% 261% 140% Loss cost (%) 3.1% 1.8% 2.0% 3.7% 2.8% 5.9% 3.4% FFSA Compensation (Million KZT) 520 236 350 819 693 1,225 3,843 FFSA Compensation % of total claims 49% 49% 50% 48% 47% 44% 47% Loss Ratio % (Net of FFSA support) 61% 35% 35% 81% 69% 147% 75% Source: FFSA 2011 Figure A3.12. Crop Insurance Program Summary of premium, claims and loss ratio (2005-10) Evolution of premium, claims and loss ratio (2005-2010) 300% 3.00 Total Claims (KZT billions) 250% 2.50 200% 2.00 150% 1.50 100% 1.00 50% 0.50 0% 0.00 2005 2006 2007 2008 2009 2010 Source: Authors‘ analysis of FFSA data The analysis of the crop insurance program results clearly shows that some Oblasts and Rayons with good performance over time are cross subsidizing the rates and claims of poor performing Oblasts and Rayons. In general the low risk Oblasts and Rayons are subsidizing the high risk Oblasts and Rayons. For instance, the 6-year average loss ratio for NKO at end of 2010 was only 24 percent, which indicates that the program is making considerable profits in this Oblast. On account of the very low loss ratio inn NKO, there is scope to reduce the average premium rates offered to farmers in this Oblast. However, this reduction in the price for insurance to NKO farmers‘ will not happen because, the profit obtained by the system from NKO is used to compensate or cross-subsidise the major recurrent losses in some of the other Oblasts. (See loss ratios in Map A3.5.) Over the past 6 years the compulsory crop insurance program has performed very poorly in WKO and Aktobe which are located in the drier western regions of Kazakhstan and the results in these two Oblasts are negatively affecting the overall financial viability of the program. Over the past 6 years, 2005 to 2010 WKO and Aktobe have accounted for a very small <5% of total scheme liability (Total Sum Insured) and less than 13% of total earned premium: however, these two Oblasts have incurred total - 265 - claims valued at KZT 31.1 billion or nearly 41% of the total claims on the scheme. This major imbalance means that at end 2006 the long term loss ratio in WKO stands at 507% and in Aktobe 381% which is not sustainable and which is prejudging the financial viability of the whole compulsory crop insurance scheme. (Map A3.5. and Appendix A3.2) The analysis also shows that there is also considerable variation in risk exposures and loss rati o’s within each Oblast and this implies a need to introduce a system of Rayon by Rayon actuarially determined crop insurance premium rates. For example in Kostanay Oblast which has an average 6 year-loss ratio of only 73%, there is a wide range in premium: claims experience at the individual rayon level varying from 0% to 5% loss ratio in the north (higher average rainfall) to as high as 320% - 640% in one rayon in the south west. In WKO which has experienced very high losses over time (21% of total claims), the loss ratios vary between 160%-320% and >1280% loss ratio. This evidence suggests the need to introduce a system of zonal rating at a more disaggregated level than the Oblast. Map A3.5. Kazakhstan Crop Insurance Program: Average Loss Ratios (2005-2010) by Oblast and by Rayon Source: Authors based on FSSA 2011 AS IF Analysis The mandatory crop insurance scheme was tested through an As if analysis. . The As if analysis is a standard practice of the insurance and reinsurance industry for property treaty underwriting. The As If consists of simulating what would be the results of any particular program by assuming changes in the terms and conditions of the treaty. For the case of Kazakhstan, the analysis consisted of two steps. The first step consisted of performing an analysis in order to see what would be the results of the program under the assumptions that the total sum insured in the program follows the same geographical distribution per rayon and Oblast as for the period 2009-2010. The second step of this AS IF consisted in estimating the rates that would be necessary to ask in order to reach a target loss ratio for the program of 60 percent in each Rayon and in each Oblast. - 266 - The aggregate loss ratio for the scheme, assuming a requested premium rate in average for the years 2009 and 2010, is expected to be worse than the loss ratio based on the actual historic conditions observed for each of the years since the inspection of the program . The reason for such deterioration in the expected loss ratio for the scheme is because of the erosion of the annual average premium rates for the program. The average annual rates for the program have passed from a high of 2.87% in 2007 to a low average of 2.27% in 2010. This represents a reduction on the rates of 10% in respect to the previous year. As a result of the impact in the reduction of the rates, the expected loss ratio (considering other factors as ceteris paribus) would have risen from 146% to 163%. The graph in Figure A3.13 presents this analysis. Figure A3.13. Comparison between the Historic As If Claims and Loss Ratio 2005 to 2010 Comparisson between the historic AS IF and force, claims and loss ratio (2005-2010) 300.00% Total Claims (KZT billions) 250.00% 200.00% 150.00% 100.00% 50.00% 0.00% 2006 2007 2008 2009 2010 Source: Author‘s analysis of FFSA data Crop Insurance premium rates at Rayon and Oblast level under the mandatory crop insurance scheme should be adjusted on an actuarial basis and where necessary increased to achieve a target loss ratio of no more than about 60% and in order to ensure the program is financially sustainable in future. The second step of the As If analysis was to simulate how much the current Oblast premium rates (average 2009-2010) should be increased in order to reach an average loss ratio of 60 percent for the scheme. In order to reach such ratio of 60%, the rates have to be increased in average 148.8 percent from the last two years average rate, 2.09%, to 5.20% which is the rate that is required to achieve a 60% loss ratio in each Rayon and Oblast. This increase on the rates will need to be very significant in those rayons situated in WKO (average premium rate increase to achieve 60% target loss ratio from 7.8% to about 46%), in Akmola (average premium rate of 5.8% would need to be increased to about 37%, and in Pavlodar (actual average rate 3.6% will need to increase to 13.1%) and in EKO from 3.4% to 8.6%. In Zhambyl Oblast southern Kazakhstan which has underwritten a very small crop insurance portfolio over the past 6 years, the results have also been poor and the average premium rates would need to increase from 3.5% to 21% to achieve the 60% target loss ratio. (See Map A3.6. for As If Rate increases per Rayon and Oblast and actual average rates per Oblast are shown in Appendix A3.2.) Map A3.6. As If Crop Insurance Average Premium Rates per Oblast and per Rayon to achieve a target loss ratio of 60%. - 267 - Source: Authors‘ analysis of FFSA data - 268 - - 269 - - 270 - Annex 3, Appendix 1. Kazakhstan Summary of Crop Insurance Results by Year 2005 to 2010 Total Avg Sum Avg Claim No. of Insured Sum Insured Insured Premiums Loss Ratio Loss cost Year Premium payments Policies Area ('000 TH) (KZT ('000 KZT) (%) (%) Rate ('000 KZT) (Hectares) /Hectare) 2005 19,008 10,454,252 34,372,458 3,288 898,607 2.61% 1,064,870 119% 3.10% 2006 13,619 9,078,612 26,650,054 2,935 684,722 2.57% 477,670 70% 1.79% 2007 25,446 12,119,581 34,796,059 2,871 997,392 2.87% 700,538 70% 2.01% 2008 33,957 14,460,541 46,644,737 3,226 1,093,232 2.34% 1,709,623 156% 3.67% 2009 32,165 14,979,524 52,902,505 3,532 1,114,366 2.11% 1,465,129 131% 2.77% 2010 16,766 12,678,405 47,265,625 3,728 1,073,639 2.27% 2,804,945 261% 5.93% Total 140,961 73,770,915 242,631,438 3,289 5,861,958 2.42% 8,222,776 140% 3.39% Source: FFSA 2011 Annex3, Appendix 2. Annual Crop Insurance Results after application of FFSA 50% Claims Subsidy Item 2005 2006 2007 2008 2009 2010 Total No. of Policies (000) 19.0 13.6 25.4 34.0 32.2 16.8 141.0 Total Insured Area (Million Ha) 10.5 9.1 12.1 14.5 15.0 12.7 73.8 Sum Insured (Million KZT) 34,372 26,650 34,796 46,645 52,903 47,266 242,631 Premiums (Million KZT) 899 685 997 1,093 1,114 1,074 5,862 Average Premium Rate % 2.61% 2.57% 2.87% 2.34% 2.11% 2.27% 2.42% Claim payments (Million KZT) 1,065 478 701 1,710 1,465 2,805 8,223 Loss Ratio (%) 119% 70% 70% 156% 131% 261% 140% Loss cost (%) 3.1% 1.8% 2.0% 3.7% 2.8% 5.9% 3.4% FFSA Compensation (Million KZT) 520 236 350 819 693 1,225 3,843 FFSA Compensation % of total claims 49% 49% 50% 48% 47% 44% 47% Loss Ratio % (Net of FFSA support) 61% 35% 35% 81% 69% 147% 75% Source: FFSA 2011 - 271 - Annex 3, Appendix 3. Kazakhstan Summary of Crop Insurance Results by Oblast (2005-2010) Total % of Sum Avg Sum Avg Claim % of No. of % of Premiums % of Loss Loss Oblast Insured Insured Insured Insured Premium payments Claims Policies TSI ('000 KZT) Premium Ratio cost Area (Ha) Area ('000 KZT) (KZT/Ha) Rate ('000 KZT) Payments Akmola 11,505 19,007,668 25.8% 60,701,280 25% 3,194 1,176,023 20% 1.9% 1,229,528 15% 105% 2.0% Aktobe 3,824 2,283,860 3.1% 7,306,712 3% 3,199 425,673 7% 5.8% 1,620,556 20% 381% 22.2% Almaty 23,362 1,281,919 1.7% 5,772,595 2% 4,503 201,019 3% 3.5% 264,704 3% 132% 4.6% EKO 17,137 3,508,694 4.8% 12,393,819 5% 3,532 419,989 7% 3.4% 740,631 9% 176% 6.0% Karaganda 5,708 2,788,646 3.8% 6,700,677 3% 2,403 314,790 5% 4.7% 315,953 4% 100% 4.7% Kostanay 19,657 21,618,796 29.3% 74,659,628 31% 3,453 1,300,112 22% 1.7% 942,639 11% 73% 1.3% Kyzylorda 1,160 279,814 0.4% 1,268,097 1% 4,532 65,819 1% 5.2% 1,165 0% 2% 0.1% NKO 14,475 16,310,981 22.1% 53,470,496 22% 3,278 1,081,629 18% 2.0% 254,290 3% 24% 0.5% Pavlodar 4,318 3,395,155 4.6% 9,997,225 4% 2,945 355,498 6% 3.6% 708,106 9% 199% 7.1% SKO 31,904 583,280 0.8% 3,146,062 1% 5,394 83,916 1% 2.7% 25,277 0% 30% 0.8% WKO 2,944 2,033,252 2.8% 4,276,128 2% 2,103 333,614 6% 7.8% 1,691,907 21% 507% 39.57% Zhambyl 4,967 678,850 0.9% 2,938,718 1% 4,329 103,875 2% 3.5% 428,021 5% 412% 14.56% Total 2005-10 140,961 73,770,915 100.0% 242,631,438 100% 42,865 5,861,958 100% 2.42% 8,222,776 100% 140% 3.39% Source: Authors‘ analysis of FFSA data - 272 - Annex 3, Appendix 4. Actual Crop Insurance Results by Oblast and by Year (2005 to 2010) Avge. Sum No. of Total Insured Sum Insured Premiums ('000 Avg Premium Claim payments Loss Ratio Oblast Insured per Ha Loss cost % Policies Area (Ha) ('000 KZT) KZT) Rate % ('000 KZT) % (KZT/Ha) Akmol a 1,837 3,520,477 10,187,199 2,894 213,201 2.09% 150,315 71% 1.48% Akmol a 1,054 1,940,482 5,489,787 2,829 121,432 2.21% 121,479 100% 2.21% Akmol a 1,642 2,697,201 7,732,267 2,867 182,150 2.36% 108,898 60% 1.41% Akmol a 2,177 3,597,384 11,257,865 3,129 198,912 1.77% 241,464 121% 2.14% Akmol a 2,618 3,746,433 12,825,612 3,423 199,655 1.56% 90,510 45% 0.71% Akmol a 2,177 3,505,691 13,208,551 3,768 260,673 1.97% 516,861 198% 3.91% Akmol a 11,505 19,007,668 60,701,280 3,194 1,176,023 1.94% 1,229,528 105% 2.03% Aktobe 638 245,789 511,186 2,080 44,063 8.62% 135,699 308% 26.55% Aktobe 13 40,806 70,984 1,740 3,622 5.10% 21,685 599% 30.55% Aktobe 472 266,415 496,681 1,864 38,283 7.71% 70,471 184% 14.19% Aktobe 924 573,538 1,921,700 3,351 105,358 5.48% 40,411 38% 2.10% Aktobe 967 660,656 2,395,023 3,625 126,689 5.29% 383,453 303% 16.01% Aktobe 810 496,656 1,911,138 3,848 107,659 5.63% 968,837 900% 50.69% Aktobe 3,824 2,283,860 7,306,712 3,199 425,673 5.83% 1,620,556 381% 22.18% Al ma ty 1,997 21,149 490,620 23,198 13,933 2.84% 19,443 140% 3.96% Al ma ty 2,958 155,046 634,430 4,092 19,444 3.06% 3,735 19% 0.59% Al ma ty 4,436 305,304 1,031,655 3,379 37,736 3.66% 136,577 362% 13.24% Al ma ty 4,473 260,708 1,019,560 3,911 36,775 3.61% 76,197 207% 7.47% Al ma ty 5,240 309,652 1,343,940 4,340 47,857 3.56% 19,287 40% 1.44% Al ma ty 4,258 230,060 1,252,391 5,444 45,273 3.61% 9,465 21% 0.76% Al ma ty 23,362 1,281,919 5,772,595 4,503 201,019 3.48% 264,704 132% 4.59% EKO 5,190 370,794 2,221,205 5,990 72,046 3.24% 132,848 184% 5.98% EKO 2,566 597,731 1,762,944 2,949 59,592 3.38% 41,314 69% 2.34% EKO 2,988 628,958 1,824,594 2,901 62,788 3.44% 15,813 25% 0.87% EKO 2,521 712,145 2,389,920 3,356 78,751 3.30% 479,832 609% 20.08% EKO 2,811 736,307 2,574,412 3,496 82,597 3.21% 53,286 65% 2.07% EKO 1,061 462,759 1,620,744 3,502 64,215 3.96% 17,538 27% 1.08% EKO 17,137 3,508,694 12,393,819 3,532 419,989 3.39% 740,631 176% 5.98% Ka ra ga nda 952 372,604 895,612 2,404 47,259 5.28% 91,652 194% 10.23% Ka ra ga nda 561 323,740 681,909 2,106 40,169 5.89% 21,432 53% 3.14% Ka ra ga nda 1,108 528,433 1,128,747 2,136 54,484 4.83% 32,025 59% 2.84% Ka ra ga nda 1,086 550,958 1,301,119 2,362 59,530 4.58% 75,412 127% 5.80% Ka ra ga nda 1,238 629,007 1,648,690 2,621 71,751 4.35% 54,879 76% 3.33% Ka ra ga nda 763 383,904 1,044,600 2,721 41,596 3.98% 40,552 97% 3.88% Ka ra ga nda 5,708 2,788,646 6,700,677 2,403 314,790 4.70% 315,953 100% 4.72% Kos ta na y 3,404 3,098,000 11,244,498 3,630 240,641 2.14% 54,986 23% 0.49% Kos ta na y 3,039 3,051,785 9,378,746 3,073 208,994 2.23% 212,868 102% 2.27% Kos ta na y 3,267 3,546,038 10,638,960 3,000 219,939 2.07% 32,609 15% 0.31% Kos ta na y 3,318 3,910,402 13,112,304 3,353 210,599 1.61% 30,450 14% 0.23% Kos ta na y 3,557 4,174,791 15,768,815 3,777 193,390 1.23% 142,779 74% 0.91% Kos ta na y 3,072 3,837,780 14,516,305 3,782 226,549 1.56% 468,947 207% 3.23% Kos ta na y 19,657 21,618,796 74,659,628 3,453 1,300,112 1.74% 942,639 73% 1.26% Source: FFSA 2011 - 273 - Annex 3, Appendix 4. Continued Actual Crop Insurance Results by Oblast and by Year (2005 to 2010) Avge. Sum No. of Total Insured Sum Insured Premiums ('000 Avg Premium Claim payments Loss Ratio Oblast Insured per Ha Loss cost % Policies Area (Ha) ('000 KZT) KZT) Rate % ('000 KZT) % (KZT/Ha) Kyzyl orda 102 26,979 134,997 5,004 6,787 5.03% 1,165 17% 0.86% Kyzyl orda 63 18,476 88,699 4,801 4,755 5.36% 0 0% 0.00% Kyzyl orda 255 60,683 240,788 3,968 14,082 5.85% 0 0% 0.00% Kyzyl orda 274 63,342 273,753 4,322 16,014 5.85% 0 0% 0.00% Kyzyl orda 244 48,739 231,923 4,758 11,653 5.02% 0 0% 0.00% Kyzyl orda 222 61,595 297,936 4,837 12,527 4.20% 0 0% 0.00% Kyzyl orda 1,160 279,814 1,268,097 4,532 65,819 5.19% 1,165 2% 0.09% NKO 2,304 1,917,364 5,982,959 3,120 140,407 2.35% 32,447 23% 0.54% NKO 2,404 2,345,438 6,896,639 2,940 160,743 2.33% 15,189 9% 0.22% NKO 2,752 2,781,224 8,075,440 2,904 213,984 2.65% 10,497 5% 0.13% NKO 2,812 3,245,547 10,339,148 3,186 183,909 1.78% 105,021 57% 1.02% NKO 2,449 3,192,024 11,376,077 3,564 183,531 1.61% 4,876 3% 0.04% NKO 1,754 2,829,384 10,800,233 3,817 199,056 1.84% 86,260 43% 0.80% NKO 14,475 16,310,981 53,470,496 3,278 1,081,629 2.02% 254,290 24% 0.48% Pa vl oda r 791 489,024 1,550,662 3,171 51,250 3.31% 78,236 153% 5.05% Pa vl oda r 682 501,193 1,294,177 2,582 49,362 3.81% 34,974 71% 2.70% Pa vl oda r 732 595,553 1,493,307 2,507 72,775 4.87% 56,169 77% 3.76% Pa vl oda r 813 708,795 2,184,021 3,081 74,231 3.40% 442,721 596% 20.27% Pa vl oda r 863 703,882 2,218,821 3,152 65,806 2.97% 5,010 8% 0.23% Pa vl oda r 437 396,708 1,256,237 3,167 42,073 3.35% 90,996 216% 7.24% Pa vl oda r 4,318 3,395,155 9,997,225 2,945 355,498 3.56% 708,106 199% 7.08% SKO 369 4,662 100,147 21,482 3,446 3.44% 942 27% 0.94% SKO 0 0 0 0 0 0.00% 0 0% 0.00% SKO 6,510 202,489 927,747 4,582 24,895 2.68% 14,695 59% 1.58% SKO 13,636 225,331 1,227,412 5,447 32,679 2.66% 9,460 29% 0.77% SKO 10,312 118,619 747,776 6,304 18,953 2.53% 0 0% 0.00% SKO 1,077 32,179 142,980 4,443 3,943 2.76% 179 5% 0.13% SKO 31,904 583,280 3,146,062 5,394 83,916 2.67% 25,277 30% 0.80% WKO 700 330,435 653,718 1,978 52,887 8.09% 359,651 680% 55.02% WKO 78 37,198 79,518 2,138 7,101 8.93% 682 10% 0.86% WKO 510 354,824 640,764 1,806 56,494 8.82% 39,288 70% 6.13% WKO 600 452,102 930,454 2,058 71,197 7.65% 20,051 28% 2.15% WKO 640 506,230 1,136,179 2,244 90,174 7.94% 672,888 746% 59.22% WKO 416 352,463 835,496 2,370 55,762 6.67% 599,347 1075% 71.74% WKO 2,944 2,033,252 4,276,128 2,103 333,614 7.80% 1,691,907 507% 39.57% Zha mbyl 724 56,975 399,655 7,015 12,688 3.17% 7,485 59% 1.87% Zha mbyl 201 66,717 272,221 4,080 9,508 3.49% 4,311 45% 1.58% Zha mbyl 774 152,459 565,111 3,707 19,781 3.50% 183,495 928% 32.47% Zha mbyl 1,323 160,289 687,481 4,289 25,276 3.68% 188,605 746% 27.43% Zha mbyl 1,226 153,184 635,235 4,147 22,309 3.51% 38,161 171% 6.01% Zha mbyl 719 89,226 379,015 4,248 14,313 3.78% 5,964 42% 1.57% Zha mbyl 4,967 678,850 2,938,718 4,329 103,875 3.53% 428,021 412% 14.56% Source: FFSA 2011 - 274 - Annex 4: Individual Grower MPCI Opportunities for Kazakhstan This Annex presents the Rating methodology used to derive the individual grower Spring Wheat Multiple Peril Crop Insurance (MPCI) insured yields and premium rates for different levels of insured yield coverage. The model used termed the Crop Risk Assessment Model, CRAM, is constructed based on analysis of variation of spring wheat annual average yields for a 17year time-series, from crop year 1994 up to and including the crop year 2010, at the rayon level. The CRAM was developed for spring wheat using the sown area, harvested area, production and annual average yield statistics for each and every Rayon in the following 8 Oblasts: NKO, Akmola, Kostanay, Karaganda, Pavlodar, EKO, Aktobe, and WKO in North Kazakhstan region. The original series used for this analysis were provided by the ARKS. Planted Area According to information obtained from ARKS, spring wheat crops sown area amount to 13.26 million hectares on average for the period 2007 – 2010. Out of the 13.26 million hectares planted with spring wheat in North Kazakhstan region, 8.96 hectares (68 percent) are planted by Agribusiness Enterprises (Production Enterprises) and 4.31 million hectares (32 percent) are planted by Commercial Farmers. The CRAM assumes that the annual spring wheat planted area has remained constant at the four year average (period 2007-2010) over the 17 years sown area series. The reason for this assumption is to remove seasonal variations for each rayon from the areas. In order to be eligible for CRAM, two criteria have been set: minimum planted area per region and a minimum of 17 years continuous annual average yield data. In order to ensure that there are sufficient numbers of farmers growing the crop in a selected region, a minimum area of 10,000 hectares has been provisionally settled as a requirement for a crop in a certain region to be eligible for the model. The second criterion, at least 17 continuous years of yield data available for each rayon to qualify for the CRAM, has been settled to have continuous series in order to establish possible yield correlations among different production zones. Rayon Crop Yield Data The CRAM uses rayon annual average yields for spring wheat crops for the period starting 1994 and up to crop year 2010 as reported by ARKS. The original rayon annual average yields from 1994 to 2010 are included in Appendix 4.3. The ARKS reports average yields on sown area basis at rayon level. This is an important advantage for risk modeling purposes, since the yields on sown capture, both, the variations due to yield performance as well as the yield variations due to full crop losses. The ARKS reports average yields for two categories of farm typology, agribusiness enterprises and commercial farmers. This fact allows CRAM to perform the risk assessment at rayon level with a breakdown per each type of farm typology. Valuation Prices - 275 - For CRAM risk modeling purposes, spring wheat has been valued at the average market average price per centner for the period 2008-2010 for the month of September, KZT 3120 per centner. The crop price is maintained as a constant value for all the past 17 years. Yield Data Cleaning and trending to establish the Central Tendency The annual average yield series at zone level used to feed CRAM must be adjusted in order to reflect the current state of the art in terms of expected yields and yield variability for the selected crops for the risk assessment. This sub-section describes the methodologies followed to clean the yield data, determine the trend in yield data and, finally, to adjust the historical yields to the current expected yield at Region level. Eliminate Yield Outliers The first step was to detect and eliminate the statistical outliers from the annual average yield series for each of the selected crop and regions by applying the Chauvenet100 criteria to each of the 17 years annual average yield records for each agribusiness enterprises and commercial farmers on each of the rayon. If, by applying the Chauvenet criteria a yield outlier was detected, then the annual average crop yield was compared with the annual average crop yield performance for the same crop and year in the neighboring rayon. If, as result of this comparison, it was detected that the crop yield performance in the neighboring rayons diverged significantly in respect to the annual average yield for the target rayon and year, then yield, production, and harvested area figures were revisited to identify the cause of the divergence. Adjusting Zonal Average Yield Data for Trends The crop yield central tendency is associated with crop management and technology practices; crop yield deviations from the central tendency are associated with effects of nature . The main objective of adjusting the historic annual average yield series was to isolate the effect on yields of the improvement on crop management practices and the increase in technology application to the crops along 17-year period considered for the analysis. A simplified method was adopted for determining the central tendency for each crop and each zone in the CRAM. The method aims to capture the non-linear yield tendency in the 17-years of annual average yield series at zonal level by using this yield series fitted to a lineal trend line and to an exponential trend line, and the five year moving average the 17 years annual average yield series. Expected Yields and adjusted crop variability. The design of the CRAM is based on the spring wheat annual average yields for the period 2006 – 2010 at rayon level and their standard deviation; thus, these inputs must be representative of the current state of the art of spring wheat crop production in each of the analyzed rayons. That is, all the long terms and cyclical effects of crop management practice and of technology application on the historic annual average yields must be isolated prior to estimating these parameters for risk modeling purposes. 100 In statistical theory, the Chauvenet‘s Criterion is a means of assessing whether one piece of experimental data – an outlier- from a set of observations, is likely to be spurious. - 276 - In order to calculate the expected annual average yield for spring wheat for each rayon in the 8 oblasts under analysis, the simple average of the most recent five years historic annual average yields was calculated. This method to estimate annual average expected yields for a certain crop located in a certain rayon is common in the agricultural insurance practice in countries where the constraint of scarce annual average crop yield data is a problem. The second part of this analysis was to estimate the expected annual average yield volatility of the annual average yield. The method used for this purpose was to measure the deviations between the historic actual annual average yields for each year of the series in respect to the corresponding annual average yield of the trend line. Then, these deviations were applied to the expected yield to obtain an adjusted annual average yield series. Modeling Expected Yields The estimation of losses for the spring wheat crop portfolio was performed through a risk modeling exercise using the CRAM. Risk modeling is a fundamental step in agricultural insurance program design and ratemaking procedures. The main objective of crop risk modeling is to estimate, based on the available information, a yield probability density function that reflects the stochastic nature of yield outcomes. The yield model relies on two basic fundamentals: (a) a crop yield probability density function inferred from the historic spring wheat annual average yields for each rayon and type of farmer in the analyzed portfolio, and (b) a correlation matrix of rayon-level and farmer-type level spring wheat annual average yields which reflects the covariant risk under the portfolio. The probability density functions were inferred from the technology adjusted annual average yields from the annual average yield series 1994 to 2010 that were fitted to a Weibull probability distribution. For the purpose of assessing risks for an individual grower multiple peril crop insurance (MPCI) scheme is necessary to the Probability functions fitted based on rayon level actualized annual average yields into spring wheat yield probability function that reflect the yield variability at individual farm level. Individual crop yield variability is always bigger than aggregate rayon level crop yield variability. The challenge when there is not individual farm level crop yield records for a significant number of crop years is to infer the probability function for individual farm yields. International best practice uses several ways to infer individual yields depending on the availability and quality of the data. One of these ways – which is mostly used in the international reinsurance market- is to assume, based on the farmer size, increase of 15 percent on the crop yield variability for individual farmers in respect of the crop yield variability at rayon level. In order to infer the crop yield probability functions for spring wheat at individual farm level, the crop yield probability functions fitted based on rayon level actualized annual average yields were tuned in such way that maintain the same average yield, but increase the coefficient of variation for the inferred individual farm yields has 15 basis points more than the rayon-level coefficient of variation of yields.. For instance, commercial farmers spring wheat crop rayon-level yields in Bulandinsky rayon in Akmola Oblast show an average of 10.8 centner per hectare and have a standard deviation of 4.1 centners per hectare, which is equal to 38 percent coefficient of variation in yields. In order to infer individual commercial farmer spring wheat crop yield from commercial farmer rayon level yields in Bulandinsky rayon in Akmola two assumptions are made. The first assumption is that the individual farmer expected average yield is equal to the rayon – level expected average yield. For the case of commercial farmer spring wheat production in Bulandinsky rayon in Akmola Oblast it is expected in 10.8 centners per hectare; thus, individual farmers spring wheat yields in Bulandinsky rayon are also expected in 10.8 centners per hectare.. The second assumption made to infer individual commercial farmer spring wheat - 277 - crop yield from commercial farmer rayon level yields is that the coefficient of variation of yields for an individual farmer will has 15 basis points higher than the coefficient of variation of spring wheat yields at rayon level. For the case of commercial farmer rayon level spring wheat production in Bulandinsky rayon in Akmola Oblast the coefficient of variation of spring wheat yields at individual farm level is 37.57 percent; thus the coefficient of variation of spring wheat yields for individual farmers are calculated in 52.57 percent. (See Figure A4.1 for details). Figure A4..1: Akmola- Bulandinsky: Spring Wheat. Comparisson of Rayon-Level and simulated individual farm level yields Spring Wheat. Akmola-Bulandinski Rayon. 4.36 Comparison Rayon-level and Farm-level simulated yields 17.81 0.1 0.09 0.08 0.07 0.06 frequency 0.05 0.04 0.03 0.02 0.01 0 10 15 20 25 30 35 0 5 Yield (Centner/he.) Spring Wheat Yield Rayon Level Spring Wheat Yield Individual Farm Level Source: Authors from CRAM The outputs of the yield probability density functions obtained for each rayon and type of farmer were correlated in order to reflect the covariance on yields for risk modeling purposes. Spring wheat crop production in Kazakhstan is exposed to drought which is a very systemic risk. Variations in spring wheat crop yields are often caused by factors that typically affect a large area. The issue of a portfolio being exposed to systemic risk, since it affects the degree on which the risks can be diversified, has severe implications for the design and rating of a crop insurance program. In light of the systemic risk faced by spring wheat crop production in Kazakhstan, the CRAM considered the correlations among each rayon and type of farmer in order to simulate the potential losses for the portfolio. CRAM Simulation Based on the stochastic distributions defined for each rayon and type of farmer and based on the spring wheat rayon level correlation matrix, the CRAM, by using @ Risk Software, applies Monte Carlo methodology and generates 5,000 iterations of yields based on the defined stochastic function for each rayon and type of farmer which are correlated each other. CRAM Output As a result of CRAM a database comprised of 221 combinations of rayon and farm type level and 5000 yield iterations for each of the combination rayon and farmers‘ type is generated. This - 278 - database is the one used for all subsequent risk assessment and risk pricing purposes for individual grower spring wheat MPCI. Crop Portfolio Risk Assessment Model – Rating Exercise Insured Yield Coverage Levels Individual Farmer multiple peril crop insurance (MPCI) was device for spring wheat production in the 8 selected oblasts in north Kazakhstan offering yield coverage levels of between 40% maximum and 10% minimum of the 5-years average yield for the period 2006-2010. MPCI rating methodology MPCI pure loss cost rating methodology The Technical rates rating methodology is based on standard MPCI rating procedures. The loss cost formula is given by: For the MPCI Program, the loss cost formula is given by: Where, Y= year, 2006-2010; U = rayon-farm typology, 1 to 221; C = crop, spring wheat Coverage Level is between a minimum of 10% and maximum of 40% of average yield. Using Contiguous Counties to Smooth Rates The MPCI insurance pure loss costs obtained for each rayon-type of farmer are smoothed by utilizing information from contiguous rayons (Skees, 1997). The smoothed pure loss cost for each rayon-type of farmer is calculated as a weighted average of the pure loss cost for that rayon- type of farmer and the pure loss cost for each contiguous rayon for the same type of farmer. The formula to calculate the smoothed pure loss cost is: The weights are calculated as follows: Subject to: - 279 - Where, is the weight assigned to the target rayon-type of farmer and is the average hectares planted in the target rayon over the most recent 4-year period; and Where, is the weight assigned to the ith contiguous rayon, and is the average area planted over the most recent 4-year period for each contiguous rayons. All weights sum to one. Loading the smoothed pure loss cost rates to derive technical rates a) Loading for idiosyncratic risks (Hail, wind, fire, etc.) Losses due to idiosyncratic risks (e.g. crop hail) are not captured in the yield data used as underlying for MPCI insurance pure loss costs. Therefore, in order to reflect the impact of idiosyncratic risk (e.g. crop hail) on rates, a loading factor was included. For the purpose of the idiosyncratic risk calculation a total loss at spring wheat at field level with recurrence period of one in 50 years, which is equal to an annual rate of 2 percent of the total sum insured, was assumed. This rate of 2 percent is applied to the level of coverage and added to the smoothed pure loss cost rate in order to get a pure loss cost which includes the potential damage due to idiosyncratic risks. b) Security Loadings A security loading was added to the model in order to consider those infrequent, but potentially catastrophic losses that were not captured within the 17-years of actual spring wheat crop yield series used as the basis for risk modeling. With that purpose, the method of risk exposure was introduced in the rate calculation. The method of risk exposure consists in calculate the loading factor based on the 1 in 100 year probable maximum loss (PML) for the whole portfolio under the assumption that a loss approximately equal to the PML will take place within a certain number of years ahead. For the purposes of the rating calculations for individual grower MPCI in Kazakhstan it was assumed that a spring wheat portfolio loss similar to the spring wheat PML will take place within the next 10 years. Security loadings are Coverage specific. The PML for the portfolio varies according to the selected level of coverage. Table A4.1 presents the security loadings used to load individual farmer spring wheat MPCI for the different levels of coverage analysed in the study Table A4.1. Catastrophe Loadings applied to MPCI pure loss cost rates by Coverage level - 280 - Coverage Level 10% 20% 30% 40% 50% PML 1-100% 5.43% 16.80% 28.70% 39.17% 46.98% Return Period 10-years 10-years 10-years 10-years 10-years Loading Factor 0.543% 1.68% 2.87% 3.92% 4.70% Source: Authors Finally, the calculation of the technical rates consists on the integration of the smoothed loss cost calculation, and the loadings due to idiosyncratic risks and catastrophic risks. The formula summarizing the calculation of the technical rates is presented below: Loading Technical to derive Commercial Premium Rates The technical rates calculated by the model are then loaded to cover various cost components in order to derive final commercial premium rates which are paid by growers. The general formulae for developing the final premium rates include: The study did not include a detailed analysis of the Kazakhstan insurance companies’ actual administrative cost structures - acquisition costs, administrative cost, and insurers and reinsurers profit margins expectations. For these reasons the current study, based on the international experience in multiple period crop insurance products, assumes a target loss ratio of 60 percent which was estimated to adequate to cover all of the above expenses and reasonable profit expectations. Therefore, the conversion of Technical Premium Rates into indicative Commercial Premium Rates is given by the following formulae: It is noted 60 percent is a reasonable target loss ratio for an MPCI cover. We understand that, if the scheme reaches economies of scale, the administrative expenses could be substantially reduced; thus, the target loss ratio could be increased. The illustrative Commercial Premium Rates for Individual grower MPCI are shown by Oblast and by Rayon and by insured yield coverage level from 10% to 50% of 5-year average yield in Table A4.2 for Commercial Farmers and then in Table A4.3 for Agricultural Enterprises (Production Enterprises) and then finally in the average Rayon-level rates for both types of farmer combined are shown in Table A4.4. - 281 - Table A4.2 Estimated Commercial Premium Rates for target loss ratios equal to 60% and Guaranteed Yields for individual grower MPCI for Spring Wheat produced by Commercial Farmers in North Kazakhstan Region Coverage Level 10% 20% 30% 40% 50 % Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Akmola Akkol 0.81 2.82% 1.61 7.89% 2.42 13.79% 3.22 19.95% 4.03 26.03% Akmola Arshaly 1.08 1.83% 2.15 5.69% 3.23 10.45% 4.31 15.65% 5.38 20.98% Akmola Astrashanski 0.61 3.68% 1.21 9.61% 1.82 16.22% 2.42 22.93% 3.03 29.40% Akmola Atbasarski 0.93 3.00% 1.86 8.34% 2.78 14.49% 3.71 20.87% 4.64 27.13% Akmola Bulandinski 1.08 2.19% 2.17 6.41% 3.25 11.48% 4.34 16.94% 5.42 22.45% Akmola Celinogradski 0.75 2.87% 1.49 8.02% 2.24 14.00% 2.99 20.23% 3.74 26.36% Akmola Enbekshilderski 1.07 2.25% 2.15 6.62% 3.22 11.82% 4.29 17.36% 5.37 22.91% Akmola Esilski 0.76 2.35% 1.52 6.96% 2.28 12.48% 3.05 18.38% 3.81 24.28% Akmola Korgalzhinski 0.47 2.49% 0.94 7.24% 1.41 12.88% 1.88 18.86% 2.35 24.81% Akmola Sandiktauski 1.05 2.24% 2.10 6.69% 3.15 12.05% 4.20 17.77% 5.24 23.52% Akmola Shortandinski 0.88 2.62% 1.77 7.42% 2.65 13.06% 3.54 19.00% 4.42 24.90% Akmola Shuchenski 1.15 1.59% 2.30 5.07% 3.46 9.44% 4.61 14.30% 5.76 19.34% Akmola Yegindikolski 0.60 2.83% 1.20 7.96% 1.80 13.94% 2.40 20.18% 3.00 26.32% Akmola Zerendinski 1.18 1.63% 2.36 5.27% 3.54 9.85% 4.73 14.90% 5.91 20.12% Akmola Zhaksi 1.15 2.18% 2.31 6.61% 3.46 11.98% 4.62 17.75% 5.77 23.57% Akmola Zharkainski 0.72 1.77% 1.45 5.61% 2.17 10.38% 2.89 15.62% 3.62 21.00% Akmola Erementauski 0.83 3.72% 1.66 9.64% 2.49 16.19% 3.32 22.82% 4.15 29.20% EKO Beskaragay 0.56 4.44% 1.13 11.08% 1.69 18.27% 2.25 25.43% 2.81 32.21% EKO Boroduliha 0.87 3.12% 1.73 8.53% 2.60 14.69% 3.46 21.04% 4.33 27.21% EKO Glubokoe 1.45 1.76% 2.91 5.50% 4.36 10.12% 5.82 15.18% 7.27 20.39% EKO Kokpekti 1.02 1.51% 2.04 4.86% 3.07 9.01% 4.09 13.55% 5.11 18.19% EKO Semey city 0.56 6.04% 1.12 13.70% 1.68 21.56% 2.23 29.15% 2.79 36.19% EKO Shemonaiha 1.44 1.91% 2.87 5.93% 4.31 10.86% 5.75 16.22% 7.19 21.68% EKO Ulan 0.88 2.28% 1.76 6.78% 2.65 12.15% 3.53 17.87% 4.41 23.58% EKO Urzhar 0.92 1.74% 1.84 5.53% 2.76 10.21% 3.69 15.27% 4.61 20.40% EKO Zyryan 1.47 1.39% 2.94 4.55% 4.41 8.56% 5.88 13.03% 7.35 17.69% Karaganda Buharzhirau 0.59 3.14% 1.18 8.55% 1.77 14.70% 2.36 21.05% 2.95 27.23% Karaganda Karkaraly 0.60 2.17% 1.19 6.41% 1.79 11.46% 2.39 16.84% 2.99 22.20% Karaganda Nura 0.44 2.11% 0.87 6.45% 1.31 11.71% 1.74 17.36% 2.18 23.06% Karaganda Osakarov 0.66 2.44% 1.33 7.11% 1.99 12.62% 2.66 18.44% 3.32 24.19% Karaganda Shetski 0.53 1.70% 1.05 5.47% 1.58 10.16% 2.11 15.29% 2.63 20.52% Karaganda Ulytau 0.58 1.79% 1.16 5.67% 1.74 10.47% 2.32 15.71% 2.90 21.08% Karaganda Zhanaarka 0.48 1.61% 0.95 5.24% 1.43 9.81% 1.90 14.84% 2.38 20.02% Kostanay Altynsarin 1.50 1.82% 2.99 5.76% 4.49 10.63% 5.98 15.96% 7.48 21.40% Kostanay Amangedi 0.77 1.76% 1.55 5.51% 2.32 10.14% 3.10 15.21% 3.87 20.40% Kostanay Arkalyk city 0.62 2.77% 1.23 7.89% 1.85 13.87% 2.47 20.12% 3.08 26.27% Kostanay Auliykolski 0.82 2.02% 1.65 6.18% 2.47 11.27% 3.30 16.79% 4.12 22.40% Kostanay Denisovski 1.20 3.03% 2.40 8.33% 3.61 14.44% 4.81 20.79% 6.01 27.02% Kostanay Fedorovski 1.62 1.30% 3.24 4.41% 4.86 8.43% 6.49 13.03% 8.11 17.93% Kostanay Kamisty 0.88 4.81% 1.76 11.28% 2.64 18.10% 3.53 24.84% 4.41 31.22% Kostanay Karabalyk 1.45 1.72% 2.90 5.52% 4.35 10.29% 5.80 15.55% 7.25 20.98% Kostanay Karasu 1.14 1.57% 2.28 5.12% 3.42 9.59% 4.56 14.56% 5.69 19.72% Kostanay Uzunkolski 1.69 1.54% 3.38 5.08% 5.06 9.55% 6.75 14.52% 8.44 19.69% Kostanay Zhetikara 0.85 2.85% 1.69 7.90% 2.54 13.77% 3.38 19.93% 4.23 26.01% Kostanay Kostanay 1.67 1.48% 3.34 4.86% 5.01 9.17% 6.68 13.99% 8.36 19.04% - 282 - Coverage Level 10% 20% 30% 40% 50 % Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Kostanay Mendikara 1.73 1.21% 3.46 4.12% 5.19 7.87% 6.92 12.13% 8.65 16.68% Kostanay Nauirzym 0.88 2.10% 1.76 6.34% 2.64 11.47% 3.52 17.00% 4.40 22.59% Kostanay Rudnyi city 0.46 3.53% 0.92 8.86% 1.38 14.74% 1.84 20.71% 2.30 26.50% Kostanay Sarykol 1.49 1.19% 2.99 4.02% 4.48 7.66% 5.97 11.79% 7.46 16.20% Kostanay Taranovski 1.20 2.42% 2.40 7.04% 3.60 12.55% 4.80 18.43% 6.01 24.33% NKO Airtau 1.55 1.43% 3.10 4.73% 4.64 8.93% 6.19 13.62% 7.74 18.52% NKO Akkayn 1.66 1.31% 3.32 4.35% 4.97 8.20% 6.63 12.50% 8.29 17.00% NKO Akzhar 1.37 1.08% 2.74 3.55% 4.11 6.66% 5.48 10.17% 6.84 13.95% NKO Esil 1.54 1.17% 3.09 3.95% 4.63 7.51% 6.18 11.55% 7.72 15.83% NKO G.Musrepov 1.48 1.43% 2.96 4.69% 4.45 8.82% 5.93 13.43% 7.41 18.27% NKO Kyzylzhar 1.71 0.98% 3.41 3.13% 5.12 5.58% 6.82 8.02% 8.53 10.26% NKO M.Zhumabayev 1.52 1.02% 3.05 3.36% 4.57 6.28% 6.09 9.57% 7.61 13.11% NKO Mamliut 1.60 1.23% 3.20 4.17% 4.79 7.91% 6.39 12.11% 7.99 16.51% NKO Shalakin 1.47 1.13% 2.95 3.80% 4.42 7.23% 5.89 11.14% 7.37 15.32% NKO Taiynsha 1.47 1.28% 2.94 4.28% 4.41 8.11% 5.89 12.42% 7.36 16.97% NKO Timiryazev 1.49 1.30% 2.98 4.40% 4.47 8.37% 5.96 12.86% 7.45 17.61% NKO Ualihanov 1.22 1.57% 2.45 4.99% 3.67 9.25% 4.89 13.95% 6.12 18.81% NKO Zhambil 1.63 1.24% 3.27 4.19% 4.90 7.97% 6.53 12.25% 8.16 16.77% Pavlodar Aktogaisky 0.58 2.90% 1.16 7.93% 1.73 13.66% 2.31 19.56% 2.89 25.28% Pavlodar Ekibastuz city 0.91 2.71% 1.82 7.39% 2.72 12.67% 3.63 18.02% 4.54 23.12% Pavlodar Irtyshski 0.70 2.09% 1.41 6.30% 2.11 11.36% 2.82 16.78% 3.52 22.23% Pavlodar Kashyrski 0.87 2.84% 1.75 7.89% 2.62 13.70% 3.50 19.72% 4.37 25.61% Pavlodar Shernaktinsky 0.75 2.93% 1.50 7.80% 2.24 13.31% 2.99 18.96% 3.74 24.42% Pavlodar Uspenka area 0.77 5.99% 1.54 13.24% 2.31 20.58% 3.08 27.61% 3.85 34.07% Pavlodar Zhelezninski 0.43 2.35% 0.87 6.90% 1.30 12.33% 1.73 18.09% 2.17 23.83% - 283 - Table A4.3 Estimated Commercial Premium Rates for target loss ratios equal to 60% and Guaranteed Yields for individual grower MPCI for Spring Wheat produced by Agribusiness Enterprises in North Kazakhstan Region Coverage Level 10% 20% 30% 40% 50 % Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Akmola Akkol 0.88 1.75% 1.77 5.58% 2.65 10.34% 3.54 15.57% 4.42 20.92% Akmola Arshaly 0.95 1.12% 1.91 3.65% 2.86 6.79% 3.82 10.30% 4.77 14.03% Akmola Astrashanski 0.76 1.31% 1.51 4.44% 2.27 8.46% 3.02 13.00% 3.78 17.78% Akmola Atbasarski 0.92 1.30% 1.84 4.40% 2.77 8.40% 3.69 12.93% 4.61 17.73% Akmola Bulandinski 1.04 1.23% 2.08 4.14% 3.12 7.89% 4.16 12.15% 5.20 16.70% Akmola Celinogradski 0.70 1.18% 1.40 4.02% 2.10 7.67% 2.79 11.83% 3.49 16.28% Akmola Enbekshilderski 1.21 1.58% 2.43 5.11% 3.64 9.52% 4.85 14.40% 6.07 19.45% Akmola Esilski 0.79 1.23% 1.59 4.17% 2.38 7.94% 3.18 12.22% 3.97 16.77% Akmola Korgalzhinski 0.58 1.19% 1.16 4.07% 1.74 7.77% 2.32 11.99% 2.90 16.49% Akmola Sandiktauski 1.20 1.06% 2.39 3.59% 3.59 6.82% 4.78 10.51% 5.98 14.50% Akmola Shortandinski 0.87 1.34% 1.75 4.48% 2.62 8.49% 3.49 13.00% 4.37 17.76% Akmola Shuchenski 1.12 1.11% 2.24 3.73% 3.36 7.08% 4.48 10.90% 5.59 15.02% Akmola Yegindikolski 0.72 1.31% 1.45 4.43% 2.17 8.46% 2.89 13.01% 3.61 17.83% Akmola Zerendinski 1.41 1.10% 2.82 3.71% 4.23 7.08% 5.64 10.91% 7.05 15.04% Akmola Zhaksi 1.18 1.18% 2.36 3.99% 3.54 7.62% 4.72 11.74% 5.90 16.14% Akmola Zharkainski 0.86 1.30% 1.72 4.42% 2.59 8.45% 3.45 13.01% 4.31 17.85% Akmola Erementauski 0.74 1.93% 1.49 5.95% 2.23 10.86% 2.97 16.18% 3.71 21.60% EKO Beskaragay 0.60 5.21% 1.20 12.01% 1.80 19.06% 2.40 25.90% 3.01 32.28% EKO Boroduliha 1.11 2.61% 2.22 6.91% 3.33 11.89% 4.44 17.14% 5.55 22.37% EKO Glubokoe 1.37 1.95% 2.75 5.67% 4.12 10.23% 5.49 15.23% 6.86 20.38% EKO Kokpekti 0.77 2.36% 1.54 6.25% 2.32 10.78% 3.09 15.60% 3.86 20.45% EKO Shemonaiha 1.58 1.77% 3.16 5.19% 4.74 9.39% 6.32 14.00% 7.89 18.78% EKO Ulan 1.18 7.15% 2.37 14.68% 3.55 22.10% 4.74 29.17% 5.92 35.71% EKO Zyryan 1.41 1.81% 2.83 5.22% 4.24 9.38% 5.65 13.95% 7.06 18.69% Karaganda Buharzhirau 0.83 1.18% 1.66 4.01% 2.49 7.66% 3.33 11.80% 4.16 16.21% Karaganda Nura 0.85 1.37% 1.70 4.60% 2.55 8.73% 3.39 13.37% 4.24 18.24% Karaganda Osakarov 0.79 1.27% 1.58 4.24% 2.37 8.00% 3.16 12.23% 3.94 16.68% Karaganda Ulytau 0.62 1.54% 1.23 5.07% 1.85 9.53% 2.47 14.48% 3.08 19.62% Kostanay Altynsarin 1.33 1.28% 2.67 4.34% 4.00 8.27% 5.33 12.73% 6.67 17.44% Kostanay Amangedi 0.72 1.41% 1.45 4.72% 2.17 8.95% 2.89 13.70% 3.61 18.68% Kostanay Arkalyk city 0.92 1.30% 1.83 4.39% 2.75 8.37% 3.66 12.87% 4.58 17.63% Kostanay Auliykolski 0.99 1.48% 1.98 4.90% 2.97 9.24% 3.97 14.06% 4.96 19.08% Kostanay Denisovski 1.19 1.36% 2.39 4.56% 3.58 8.67% 4.78 13.30% 5.97 18.17% Kostanay Fedorovski 1.64 1.14% 3.29 3.86% 4.93 7.37% 6.57 11.38% 8.21 15.68% Kostanay Kamisty 0.92 1.32% 1.84 4.45% 2.75 8.43% 3.67 12.89% 4.59 17.54% Kostanay Karabalyk 1.47 1.15% 2.94 3.92% 4.41 7.49% 5.89 11.57% 7.36 15.95% Kostanay Karasu 1.08 1.35% 2.16 4.49% 3.24 8.49% 4.32 12.98% 5.40 17.69% Kostanay Uzunkolski 1.40 1.06% 2.80 3.58% 4.20 6.80% 5.59 10.48% 6.99 14.44% Kostanay Zhetikara 0.84 1.33% 1.68 4.48% 2.52 8.51% 3.36 13.02% 4.20 17.73% Kostanay Kostanay 1.27 1.25% 2.54 4.21% 3.82 8.03% 5.09 12.35% 6.36 16.93% Kostanay Mendikara 1.40 1.09% 2.81 3.69% 4.21 7.05% 5.62 10.89% 7.02 15.03% Kostanay Nauirzym 0.68 1.46% 1.36 4.87% 2.04 9.22% 2.71 14.07% 3.39 19.14% Kostanay Sarykol 1.46 1.15% 2.92 3.90% 4.38 7.42% 5.84 11.41% 7.31 15.68% Kostanay Taranovski 1.13 1.29% 2.25 4.35% 3.38 8.28% 4.51 12.73% 5.64 17.44% NKO Airtau 1.32 0.98% 2.63 3.28% 3.95 6.16% 5.26 9.43% 6.58 13.00% NKO Akkayn 1.44 0.95% 2.89 3.14% 4.33 5.84% 5.78 8.89% 7.22 12.21% - 284 - Coverage Level 10% 20% 30% 40% 50 % Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) NKO Akzhar 1.10 1.01% 2.20 3.34% 3.30 6.23% 4.40 9.49% 5.50 13.01% NKO Esil 1.38 0.98% 2.75 3.27% 4.13 6.17% 5.50 9.50% 6.88 13.16% NKO G.Musrepov 1.32 1.07% 2.63 3.57% 3.95 6.74% 5.27 10.33% 6.58 14.20% NKO Kyzylzhar 1.43 0.93% 2.86 3.03% 4.29 5.57% 5.72 8.38% 7.15 11.42% NKO M.Zhumabayev 1.37 0.94% 2.73 3.07% 4.10 5.68% 5.46 8.58% 6.83 11.72% NKO Mamliut 1.37 0.98% 2.74 3.27% 4.11 6.13% 5.48 9.37% 6.84 12.89% NKO Shalakin 1.29 1.09% 2.58 3.69% 3.87 7.03% 5.16 10.85% 6.46 14.96% NKO Taiynsha 1.33 1.00% 2.65 3.31% 3.98 6.20% 5.31 9.46% 6.63 12.99% NKO Timiryazev 1.24 1.04% 2.47 3.53% 3.71 6.71% 4.94 10.34% 6.18 14.26% NKO Ualihanov 1.13 1.13% 2.27 3.75% 3.40 7.04% 4.54 10.77% 5.67 14.75% NKO Zhambil 1.16 1.07% 2.31 3.62% 3.47 6.89% 4.62 10.64% 5.78 14.68% Pavlodar Ekibastuz city 0.52 1.50% 1.05 4.68% 1.57 8.49% 2.09 12.50% 2.62 16.43% Pavlodar Irtyshski 0.67 1.18% 1.35 3.97% 2.02 7.54% 2.69 11.57% 3.37 15.82% Pavlodar Kashyrski 0.82 1.68% 1.63 5.49% 2.45 10.30% 3.26 15.64% 4.08 21.16% Pavlodar Shernaktinsky 0.76 1.93% 1.51 5.69% 2.27 10.04% 3.02 14.54% 3.78 18.86% Pavlodar Uspenka area 0.67 1.71% 1.35 5.18% 2.02 9.27% 2.69 13.53% 3.37 17.68% Pavlodar Zhelezninski 0.90 1.50% 1.80 4.98% 2.70 9.45% 3.60 14.49% 4.49 19.78% - 285 - Table A4.4 Estimated Commercial Premium Rates for target loss ratios equal to 60% and Guaranteed Yields for individual grower MPCI for Spring Wheat produced by Combined Commercial Farmers and Agribusiness Enterprises in North Kazakhstan Region Coverage Level 10% 20% 30% 40% 50 % Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Akmola Akkol 0.87 1.98% 1.73 6.09% 2.60 11.10% 3.46 16.53% 4.33 22.05% Akmola Arshaly 0.99 1.36% 1.98 4.33% 2.97 8.01% 3.97 12.09% 4.96 16.36% Akmola Astrashanski 0.70 2.10% 1.40 6.17% 2.09 11.05% 2.79 16.32% 3.49 21.66% Akmola Atbasarski 0.92 1.54% 1.85 4.96% 2.77 9.27% 3.69 14.07% 4.61 19.08% Akmola Bulandinski 1.04 1.33% 2.09 4.39% 3.13 8.29% 4.18 12.68% 5.22 17.33% Akmola Celinogradski 0.71 1.45% 1.41 4.66% 2.12 8.68% 2.82 13.17% 3.53 17.89% Akmola Enbekshilderski 1.19 1.69% 2.38 5.35% 3.56 9.90% 4.75 14.88% 5.94 20.02% Akmola Esilski 0.79 1.47% 1.57 4.77% 2.36 8.92% 3.15 13.55% 3.93 18.39% Akmola Korgalzhinski 0.56 1.39% 1.12 4.54% 1.68 8.54% 2.24 13.02% 2.80 17.74% Akmola Sandiktauski 1.17 1.23% 2.34 4.04% 3.52 7.58% 4.69 11.56% 5.86 15.81% Akmola Shortandinski 0.87 1.43% 1.75 4.69% 2.62 8.81% 3.50 13.43% 4.37 18.26% Akmola Shuchenski 1.12 1.17% 2.25 3.91% 3.37 7.40% 4.49 11.37% 5.62 15.61% Akmola Yegindikolski 0.68 1.80% 1.36 5.58% 2.03 10.24% 2.71 15.35% 3.39 20.59% Akmola Zerendinski 1.36 1.20% 2.72 4.02% 4.08 7.61% 5.43 11.69% 6.79 16.03% Akmola Zhaksi 1.17 1.83% 2.34 5.63% 3.50 10.29% 4.67 15.36% 5.84 20.55% Akmola Zharkainski 0.81 1.47% 1.62 4.84% 2.42 9.13% 3.23 13.94% 4.04 18.96% Akmola Erementauski 0.77 2.43% 1.53 6.98% 2.30 12.35% 3.06 18.04% 3.83 23.72% EKO Beskaragay 0.58 4.80% 1.16 11.51% 1.74 18.63% 2.32 25.65% 2.90 32.24% EKO Boroduliha 0.96 2.90% 1.91 7.83% 2.87 13.49% 3.82 19.36% 4.78 25.13% EKO Glubokoe 1.42 1.84% 2.83 5.57% 4.25 10.17% 5.66 15.20% 7.08 20.39% EKO Kokpekti 0.90 1.88% 1.79 5.46% 2.69 9.78% 3.58 14.44% 4.48 19.17% EKO Semey city 0.56 6.04% 1.12 13.70% 1.68 21.56% 2.23 29.15% 2.79 36.19% EKO Shemonaiha 1.52 1.82% 3.05 5.46% 4.57 9.92% 6.10 14.80% 7.62 19.83% EKO Ulan 0.98 4.19% 1.96 9.87% 2.94 16.05% 3.92 22.29% 4.90 28.33% EKO Urzhar 0.92 1.74% 1.84 5.53% 2.76 10.21% 3.69 15.27% 4.61 20.40% EKO Zyryan 1.43 1.67% 2.86 5.00% 4.29 9.11% 5.72 13.64% 7.16 18.36% Karaganda Buharzhirau 0.70 2.11% 1.39 6.16% 2.09 11.00% 2.79 16.19% 3.48 21.44% Karaganda Karkaraly 0.60 2.17% 1.19 6.41% 1.79 11.46% 2.39 16.84% 2.99 22.20% Karaganda Nura 0.72 1.51% 1.43 4.95% 2.15 9.30% 2.87 14.14% 3.59 19.17% Karaganda Osakarov 0.72 1.89% 1.44 5.75% 2.15 10.44% 2.87 15.50% 3.59 20.64% Karaganda Shetski 0.53 1.70% 1.05 5.47% 1.58 10.16% 2.11 15.29% 2.63 20.52% Karaganda Ulytau 0.60 1.68% 1.19 5.40% 1.79 10.04% 2.39 15.15% 2.98 20.41% Karaganda Zhanaarka 0.48 1.61% 0.95 5.24% 1.43 9.81% 1.90 14.84% 2.38 20.02% Kostanay Altynsarin 1.37 1.41% 2.74 4.68% 4.11 8.84% 5.48 13.50% 6.84 18.39% Kostanay Amangedi 0.75 1.61% 1.50 5.19% 2.26 9.66% 3.01 14.59% 3.76 19.69% Kostanay Arkalyk city 0.80 1.73% 1.60 5.43% 2.40 10.00% 3.20 15.02% 4.00 20.19% Kostanay Auliykolski 0.95 1.61% 1.89 5.20% 2.84 9.71% 3.79 14.70% 4.73 19.85% Kostanay Denisovski 1.20 1.63% 2.39 5.18% 3.59 9.62% 4.78 14.53% 5.98 19.62% Kostanay Fedorovski 1.63 1.21% 3.27 4.10% 4.90 7.84% 6.53 12.10% 8.17 16.66% Kostanay Kamisty 0.91 1.93% 1.82 5.64% 2.73 10.12% 3.64 14.97% 4.56 19.92% Kostanay Karabalyk 1.47 1.27% 2.93 4.26% 4.40 8.08% 5.87 12.41% 7.33 17.01% Kostanay Karasu 1.10 1.42% 2.20 4.69% 3.30 8.85% 4.39 13.49% 5.49 18.35% Kostanay Uzunkolski 1.47 1.19% 2.93 3.98% 4.40 7.54% 5.87 11.57% 7.33 15.85% Kostanay Zhetikara 0.84 1.58% 1.68 5.04% 2.52 9.36% 3.36 14.14% 4.20 19.08% Kostanay Kostanay 1.49 1.39% 2.97 4.61% 4.46 8.71% 5.94 13.34% 7.43 18.20% Kostanay Mendikara 1.52 1.14% 3.04 3.86% 4.56 7.38% 6.08 11.39% 7.60 15.70% - 286 - Coverage Level 10% 20% 30% 40% 50 % Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Kostanay Nauirzym 0.79 1.87% 1.59 5.80% 2.38 10.64% 3.18 15.93% 3.97 21.32% Kostanay Rudnyi city 0.46 3.53% 0.92 8.86% 1.38 14.74% 1.84 20.71% 2.30 26.50% Kostanay Sarykol 1.47 1.17% 2.95 3.94% 4.42 7.50% 5.89 11.55% 7.36 15.87% Kostanay Taranovski 1.15 1.68% 2.30 5.27% 3.46 9.76% 4.61 14.70% 5.76 19.82% NKO Airtau 1.36 1.07% 2.71 3.57% 4.07 6.72% 5.43 10.28% 6.78 14.11% NKO Akkayn 1.49 1.04% 2.99 3.45% 4.48 6.45% 5.98 9.83% 7.47 13.45% NKO Akzhar 1.19 1.04% 2.38 3.42% 3.57 6.40% 4.77 9.76% 5.96 13.38% NKO Esil 1.40 1.01% 2.80 3.38% 4.20 6.38% 5.60 9.82% 7.00 13.58% NKO G.Musrepov 1.34 1.14% 2.69 3.78% 4.03 7.13% 5.38 10.91% 6.72 14.95% NKO Kyzylzhar 1.51 0.95% 3.01 3.06% 4.52 5.57% 6.02 8.27% 7.53 11.06% NKO M.Zhumabayev 1.42 0.97% 2.85 3.19% 4.27 5.92% 5.70 8.98% 7.12 12.28% NKO Mamliut 1.42 1.05% 2.85 3.51% 4.27 6.60% 5.69 10.10% 7.12 13.85% NKO Shalakin 1.33 1.10% 2.66 3.72% 4.00 7.08% 5.33 10.92% 6.66 15.05% NKO Taiynsha 1.35 1.05% 2.70 3.49% 4.05 6.55% 5.41 10.00% 6.76 13.72% NKO Timiryazev 1.35 1.17% 2.70 3.96% 4.05 7.54% 5.40 11.60% 6.75 15.94% NKO Ualihanov 1.15 1.20% 2.30 3.94% 3.44 7.38% 4.59 11.25% 5.74 15.37% NKO Zhambil 1.32 1.14% 2.63 3.85% 3.95 7.34% 5.26 11.31% 6.58 15.55% Pavlodar Aktogaisky 0.58 2.90% 1.16 7.93% 1.73 13.66% 2.31 19.56% 2.89 25.28% Pavlodar Ekibastuz city 0.74 2.32% 1.47 6.54% 2.21 11.35% 2.95 16.28% 3.68 21.01% Pavlodar Irtyshski 0.69 1.75% 1.38 5.43% 2.08 9.94% 2.77 14.84% 3.46 19.84% Pavlodar Kashyrski 0.85 2.45% 1.71 7.08% 2.56 12.55% 3.41 18.35% 4.27 24.11% Pavlodar Shernaktinsky 0.75 2.13% 1.51 6.11% 2.26 10.70% 3.02 15.43% 3.77 19.99% Pavlodar Uspenka area 0.72 3.69% 1.43 8.92% 2.15 14.52% 2.86 20.06% 3.58 25.28% Pavlodar Zhelezninski 0.61 1.86% 1.23 5.81% 1.84 10.69% 2.46 16.04% 3.07 21.53% - 287 - - 288 - Annex 5: International Experience with Crop Insurance Pools This Annex provides information on the Spanish and Turkish Agricultural Insurance Pool Programs which are examples of national Public-Private Partnerships where the insurance industry has voluntarily elected to form a specialist agricultural co-insurance pool company rather than to underwrite agricultural insurance on an individual company basis. The structure of these two programs and the roles played by goveremnt may be of interest to key stakeholders in Kazakhstan who are currently studying future institutional options for the compulsory crop insurance program. Spain: The Combined Agrarian Insurance Program (Seguro Agrario Combinado) The agricultural insurance system in Spain is structured around an established public-private partnership, PPP, known as AGROSEGURO (Agrupación Española de Entidades Aseguradores de los Seguros Agrarios Combinados), formed in 1980 to provide farmers with insurance for crop, livestock, aquaculture, and most recently, forestry. The organizational structure of the Agroseguro program is shown in Figure A5.1. There are three main groups involved in the implementation of this national agricultural insurance program: (1) government including the Autonomous State Governments, the Ministries of Agriculture and Finance and ENESA, the National Agricultural Insurance Agency; (2) the insurance and reinsurance sectors, comprised of the Pool Coinsurers and Agroseguro, the managing agricultural underwriter company owned by the pool coinsurers, Consorcio de Compensacion, the national catastrophe reinsurer and international reinsurers and (3) the producer associations and individual farmers located throughout Spain and its territories. Figure A5.1. Organisational Structure of the Spanish Agricultural Insurance Program Source: ENESA 2007 - 289 - Key Parties Involved Are  Administrator: ENESA (The National Agricultural Insurance Agency or La Entidad Estatal de Seguros Agrarios) coordinates the system and manages resources for subsidizing insurance premiums.  Pool Co-insurers: In 2010, 28 private and mutual Spanish and international insurance companies and the state catastrophe reinsurer, the Consorcio de Compensacion de Seguros, CCS participated in the Agricultural Insurance Pool on a voluntary basis. Mapfre Insurance Company and CCS are the largest shareholders in the Pool while some companies maintain shares of less than 1 percent. The co-insurance pool clearly illustrates the principle of a large number of companies pooling risks and each company bearing a small share of risk.  Managing Underwriter: AGROSEGURO, which is owned by the 36 shareholders/co- insurers has been appointed by the co-insurers to underwrite, adjust, and settle claims on their collective behalves. The key functions of the Pool Coinsurers include: (i) to provide their financial capacity to underwrite the scheme and (ii) to promote and market the AGROSEGURO standard agricultural insurance policies through their networks of sales agents for which they receive an agreed commission.  International Commercial Reinsurers: Providers of (i) stop loss reinsurance to pool reinsurers on their viable line retentions and (ii) multi-year catastrophe stop loss to CCS. The public sector entities are ENESA, which coordinates the system and manages resources for subsidizing insurance premiums; and CCS, which, together with private reinsurers, provides reinsurance for the agricultural insurance market. Local governments are involved only to the extent that they are allowed to augment premium subsidies offered at the national level. On the private side, insurance contracts are sold by AGROSEGURO. Farmers, insurers, and institutional representatives are all part of a general commission hosted by ENESA that functions as the managing board of the Spanish agricultural insurance system. Forms of Government Support to Agricultural Insurance The Spanish Federal and Regional autonomous governments support the AGROSEGURO program in a number of ways including:  Insurance Legislation.  Product design and rating (data and information provision, statistical studies and actuarial research)  Subsidies on Agriculture Insurance Premiums Paid by Farmers/Herders.  Co-insurance and Reinsurance through the Insurance Compensation Consortium or Consorcio de Compensación de Seguro (CCS). In 2010 the premium subsidies on Agroseguro amounted to US$ 539 million (€ 406) million or 62% of the total earned premiums of US$ 864 million (€ 650 million) in 2010. In Spain, government’s rationale for subsidizing agricultural insurance is that it wishes to encourage all farmers to buy insurance against natural and climatic calamities and to replace free ad hoc government disaster relief mechanisms by a much more rigorous ex-ante agricultural insurance program. To reinforce the point, Spanish producers are not eligible for - 290 - disaster payments for perils for which insurance is offered. For non-covered perils, ad hoc disaster payments are available, but only if the producer has already purchased agricultural insurance for covered perils. AGROSEGURO In 2010 AGROSEGURO employed a full-time professional staff of 255 comprising management, technicians, administrators and clerical support staff. AGROSEGURO maintains its own network of 348 trained and specialized crop loss adjusters located throughout Spain. In addition, the company retains a network of 121 livestock loss adjusters. Its operating costs in 2010 amounted to about US$ 31 million or 3.55% of total earned premium. Agroseguro acts as the specialist insurance company or managing underwriter acting on behalf of the28 pool coinsurers. The company is responsible for the design and rating of agricultural insurance products in conjunction with ENESA throughout Spain, and for then issuing standard policies to the Pool Insurers who then market these standard policies on their own paper. Agricultural insurance is voluntary is Spain. AGROSEGURO cannot reject any application for insurance by a farmer on the grounds of the risk exposure: however, if a farmer does not adopt the recommended husbandry or production practices for that crop or livestock commodity the company many refuse the risk on these grounds. Agroseguro is responsible for underwriting the scheme and for adjusting claims on behalf of the Pool Coinsurers. The operational procedures are outlined in Figure A5.2. Figure A5.2. Agroseguro Operating Systems and Procedures EXPERIENCIA DE ESPAÑA EN EL DESARROLLO DE LOS SEGUROS AGRICOLAS Agroseguro Spain: Operating Procedures Flow Chart Payment of Premium (Managing Underwriter) AGROSEGURO Pool Insurance Intermediaries Policy holders Companies Farmers Policies Communication of Losses Loss Assessment and Payment of Claims Source: ENESA 2007 Agroseguro underwrites a large and diversified crop and livestock portfolio and also insures aquaculture and forestry. Crop Insurance includes individual grower multiple-peril crop insurance MPCI. However, only a very small fraction of Agroseguro‘s overall liability is MPCI - 291 - and the company underwrites crop hail and named peril policies, as well as a large livestock portfolio, neither of which accumulate with the MPCI drought and flood policies. Policies are available for crops, livestock, and aquaculture activities, with these risks being pooled across the country by AGROSEGURO. Unlike the United States and Canada, farmer associations are more actively involved in implementation and development of agricultural insurance. Government has reserves to cover extreme losses, and as a final resort, the government treasury is used to cover losses that may occur beyond these reserves. After more than 30 years operations, Agroseguro is today a very large mature agricultural insurance program underwriting practically every class of crop and livestock business throughout Spain. Total premiums for agriculture insurance policies purchased reached US$900 million (€678 million) in 2008 before falling back slightly to US$ 864 million (€ 651 million) in 2010. In 2010 the Total sum Insured of the program amounted to US$ 14.7 billion (€11.1 billion) Direct premium subsidies (from both the central government and the autonomous regions) were equal to 57% percent to the total premium volume in 2010, with farmers paying the remaining 38% of commercial premiums. In 2010 AGROSEGURO underwrote about 485,000 crop and livestock insurance policies with an earned premium volume of US$ 864 million (€651 million). In 2010, AGROSEGURO adjusted losses on and compensated a total of 1,108,177 claims divided into: 94,717 crop insurance claims, 124,460 livestock insurance claims and 889,000 claims relating to individual animal carcass removal and destruction. In 2010 the total cost of claims amounted to US$ 736 million (€554 million) equivalent to a loss ratio of 85%.101 Viable lines include the less volatile and lower-risk crop programs, which are insured by the pool of 28 coinsurers. Experimental lines are insured directly by CCS and include the more volatile crops and peril combinations, including the systemic perils of drought and flood, which can lead to catastrophe losses. Drought is only offered as an experimental cover for selected crops and programs (e.g., the Integral Winter Cereal Program; and yield shortfall policies for wine grapes); flood has been only included as an insured peril since 1999. In order to minimize anti- selection, flood is a compulsory peril on all crop insurance lines; Drought and flood are only offered on experimental lines and are reinsured by CCS. In 2010 AGROSEGURO underwrote about 200 viable and experimental crops, livestock, and marine aquaculture lines, and forestry insurance covering a wide range of crop types including cereals, oilseeds, horticultural crops, leaf and fibers, tree fruits and vines, and livestock types. The company offers comprehensive range of single-peril hail, named-peril, and multi-peril crop insurance policies and a remote sensing pasture-grazing NDVI index cover. AGROSEGURO Reinsurance CCS the national catastrophe reinsurer provides commercially priced stop loss reinsurance protection to Agroseguro on the Viable Lines and the Experimental Lines. The viable lines stop loss reinsurance protection provides layered reinsurance cover and the pool insurers are required to retain (coinsure) between 5% and 10% of the losses on these layers. However, the law then permits pool coinsurers to protect their stop loss retentions by purchasing reinsurance from international reinsurers. In addition, CCS has traditionally purchased a multi-year catastrophe reinsurance protection on its retention. This is a multi-year stop loss retrocession treaty placed with international reinsurers that in the 2000‘s reinsured CCS losses of 40 percent in excess of 145 percent of GNPI on the Agroseguro program. (It is not known whether CCS currently purchases this stop loss retrocession protection in 2011). The relationships between Agroseguro, and Consorcio and international reinsurers are illustrated in Figure A5.3. 101 Agroseguro 2010 Annual Report. www.agroseguro.es - 292 - Figure A5.3. Agroseguro: Financial Flows and Reinsurance Structure EXPERIENCIA DE ESPAÑA EN EL DESARROLLO DE LOS SEGUROS AGRICOLAS Flujos económicos Regional International Goverments Reinsurers National Reinsurer (Consorcio de Compensación de Seguros) ENESA Farmers Payment of Premium AGROSEGURO. Payment of Claims Source: ENESA 2007 Turkey: The TARSIM Agricultural Insurance Pool History of Agricultural Insurance Turkey has a lengthy history of agricultural crop insurance daring back to 1957 while livestock insurance was introduced in 1960. Until 2006 the agricultural crop and livestock insurance in Turkey was underwritten by 15 private commercial insurance companies that competed against each other to underwrite a series of products, including, most notably, crop hail (plus pilot frost), livestock, poultry, and greenhouse insurance. In 2005, the market leaders were Guven and Basak with 37% and 34% shares, respectively, of total agricultural insurance market premiums of Turkish Lira 49 million (USD 36 million). During this period, a market tariff system was applied by international reinsurers and with the agreement of local insurers in the crop hail business; each company placed its own reinsurances with the national reinsurer (Milli Re) and/or international reinsurers on a proportional and/or non-proportional basis. At this time there was no government financial or other support to agricultural insurance in Turkey. In 2005, with the agreement of government and the private commercial insurers, legislation was enacted under the Agricultural Insurance Law No 5363, dated June 14, 2005, to create an Agricultural Insurance Pool under the administration of a new managing underwriter, TARSIM, and to define the role and functions of federal government support in the form of financial subsidies and excess of loss reinsurance protection. - 293 - Figure A5.4. shows the institutional structure for the TARSIM Pool. It is a public-private partnership involving the government, the private insurance companies, and supporting organizations (insurance association, Ministry of Agriculture, etc). A management committee comprised of representation from each of these organizations is responsible for policy decisions regarding the operations of the Pool, for determination of crops, risks, and regions to be supported, and for determination of subsidy levels. The TARSIM Agricultural Insurance Pool functions as a conventional coinsurance pool, and its shareholders and coinsurers today include 22 insurance companies licensed to transact agricultural Insurance102. The coinsurers issue TARSIM‘s approved and standard insurance contracts (policies) on their own paper; the companies receive an agreed commission for bringing business to the Pool, and all risks and premium are 100% ceded to the Insurance Pool. TARSIM is responsible for product design and setting standard rates, for premium collection, for loss assessment and claims settlement, and for reinsurance arrangements. Figure A5.4: TARSIM Pool Public-Private Partnership for Agricultural Insurance in Turkey Turkey Public Private Partnership for Agricultural Insurance Organizations Insurance Government Private Sector and Agriculture Ministry of Association of the Union of the Turkish Undersecretariat Insurance Insurance and Agriculture and Chamber of of the Treasury Companies Reinsurance Rural Affairs Agriculture Companies of Turkey Agricultural Insurance Pool TARSIM TSV (field loss assessment services) Source: TARSIM (2006) Agricultural Insurance Products Available TARSIM offers a wide range of specialist agricultural crop, livestock and greenhouse and aquaculture insurance products. The company does not, however, underwrite Multi-Peril Crop Insurance (MPCI) covers. Rather, it offers a named-peril hail policy plus additional perils for all crops. For fruit and vegetables and ornamentals additional cover may be purchased against frost damage. The company also underwrites a material damage policy for loss of greenhouse structures and the crops grown under cover. The company insures dairy cattle against a wide range of perils including diseases, but excluding notifiable diseases, and a similar comprehensive cover is offered for poultry. The company also underwrites a marine aquaculture policy against a wide range of perils including pollution, diseases, and algae bloom. 102 Bora, B. 2010 Subsidised agricultural insurance in Turkey. B. Bora, General Manager of Tarsim Turkey - 294 - Delivery Channels The main delivery channel for agricultural insurance services is the insurance agents’ network (for both crop and livestock). Financial institutions and input suppliers are the second important delivery channel for crop insurance. Another significant delivery channel for both crop and livestock insurance is producers‘ associations and cooperatives. Insurance brokers are marginal. There are no specific institutions for delivering agricultural insurance to small and marginal farmers in Turkey. Voluntary vs. Compulsory Insurance Agricultural insurance is voluntary for all farmers. Agricultural Reinsurance TARSIM is responsible for deciding on its risk retention and reinsurance strategy . The company is permitted to retrocede business back to the insurers and/or to reinsure through MilliRe and international reinsurers. It is also understood that, in cases where sufficient reinsurance capacity is not available through commercial reinsurers, government may accept to share in the risk financing.103 Types of Public Support for Agricultural Insurance in Turkey Public support to agricultural insurance is important in Turkey. The government provides a wide range of support under the new TARSIM Pool arrangement including:  Agricultural insurance legislation enacted in 2005 to create the national Pool Scheme and to define the roles of public and private sectors;  Agricultural insurance premium subsidies, which are fixed at 50% of the premium cost for both crops and livestock and which are paid by government directly to the Pool (TARSIM);  Subsidies on TARSIM‘s administration and operating expenses and on loss adjustment expenses;  Government support to the reinsurance program; and  Agricultural insurance premiums sales tax exemptions. The costs of government premium subsidies and other forms of financial support to the Program are detailed in Table A5.1 for 2007, which was the first full year of operation of the new TARSIM Pool agricultural insurance scheme. Table A5.1.: TARSIM Subsidies Paid by Federal Government, 2007 A&O Premium Expenses LAE Subsidies Subsidies Subsidies Total Percent of Crop 50% 6% 5% 61% Premium Livestock 50% 6% 5% 61% 103 Bora 2005, Public-Private Partnerships for Risk Management in Agriculture: Turkish Experience. - 295 - Crop 20.0 2.4 2.0 24.4 Amount Livestock 6.1 0.7 0.6 7.4 (USD mil) Total 26.1 3.1 2.6 31.8 Source: World Bank Survey (2008). Tarsim's results for the period 2007 to 2009 are summarized in Table A5.2. Table A5.2. Tarsim’s results Item 2007 2008 2009 No of Policies 113,413 134,881 158,661 Premium Income (in TRY 000) 33,147 50,885 66,244 Sum Insured (In TRY 000) 764,340 1,150,088 1,500,169 No. of Insured Cattle 28,703 37,193 58,028 Insured area (Dunams)* 1,804,304 2,293,434 2,938,557 Number of Risk Inspections and loss notifications 31,083 44,048 57,265 Source: Bora 2010 * 1 dunam = 0.1 Hectares TRY 1.00 = Euro 0.464 at 31/12/2009 - 296 - Annex 6: Named Peril Crop Insurance Opportunities in Kazakhstan This Annex presents further information on the key features of named -peril damage based crop insurance and indemnity products . This section is based on Iturrioz 2009 104. Named peril (damage based) as the name suggests provides indemnity against those adverse events that are explicitly listed in the policy. This subclass has a number of distinctive features: - The sum insured is agreed at the inception of the contract and may be based on production costs, or on the expected crop revenue; - The loss is determined as a percentage of the damage incurred by the insured party as established by a loss adjuster as soon after the damage occurs; - The indemnity is calculated as the product of the percentage of the damage and the sum insured; - Deductibles105 and franchises106 are normally applied to reduce the incidence of false claims and to encourage improvements in risk management. An example of indemnity under a named peril contract is illustrated in figure A6.1 below. Figure A6.1: Example of Indemnity of a named peril insurance contract Insured Unit Insurance contract Conditions: Insured Peril: Hail Sum Insured: US$ 10,000 Indemnity Limit: Full Value Deductible: Damage = 0% Option A) 5% of the total sum insured Option B) 10% of the loss Loss Adjustment: - 50% of the insured unit with 0% damage. - 50% of the insured unit with 40% damage. Consequently, Total Damage = 50%*0% +%50*40% = 20% Damage = 40% Indemnity Calculation: Indemnity = Damage (%) * Total Sum Insured – Deductible Option A) 20% * US$ 10,000 – US$ 10,000*5% = US$ 1,500 Option B) 20% * US$ 10,000 – US$ 10,000*20%*10% = US$ 1,800 Source: Iturrioz, R (2009) Named peril is a popular type of insurance and accounts for a significant portion of the agricultural premium worldwide. From the perspective of the insured parties it appeals where firms are located in areas frequently subjected to one of the perils covered; from the insurer‘s point of view it is suitable to situations where the damages caused by the named perils are both measurable and have sudden impact. 104 Iturrioz R, 2009: Agricultural Insurance: A Primer. The World Bank, Washington DC. 105 A deductible is an amount or a percentage of the loss that is deducted from the indemnity and represents the first portion of the claim that the insured bears. The purpose of a deductible is to reduce moral hazard by encouraging the insured to prevent losses. Deductibles can be either a percentage of the sum insured or a percentage of the loss and can be applied to each and every loss or to the total losses over a specified period, normally the currency of the contract. 106 A franchise is a loss threshold that the insured has to reach in order to be able to receive the indemnity. Once the threshold is reached the amount of any subsequent loss is paid in full. The purpose of a franchise is to reduce claim frequency. - 297 - Named peril agricultural insurance products account for a considerable proportion of agricultural insurance worldwide. Named peril insurance contracts are used extensively to protect against hail damage and are used in horticulture and floriculture in addition to crops and fruit but are also used in livestock, bloodstock aquaculture, forestry and greenhouses insurance. Hail Risk modeling for hail insurance in Kazakhstan. This section presents the basic design features of the hail crop risk assessment model for spring wheat production in 7 selected rayons (including Altynsarinski and Auliyekolski Rayons in Kostanay Oblast; Aktogaiski and Zhelezinski Rayons in Pavlodar Oblast; Bulandinski and Enbekshilderski Rayons in Akmolinsk Oblast; and Tole-bi Rayon in South Kazakhstan) and the rating tool designed for hail insurance. The hail crop risk assessment model is constructed based on analysis of the frequency and severity of occurrence of hailstorms in the selected weather stations in the selected rayons . The weather stations that were used for this assessment are. Schuchinsk, Egindykol, and Stepnogorsk in Akmola Oblast; Diyevskaya, Kostanay, Kushmurun in Kastanay Oblast; Aktogay, Zholboldy, and Mikhailovka in Pavlodar Oblast, and Kazygurt and Shimkent in SKO. The original series on frequency of occurrence of hail by month used for this analysis were provided by KHM. The data on hail severity was based on secondary qualitative information obtained from the farmers on the field visits. Modeling hail damage on spring wheat The hail crop risk assessment model used for Kazakhstan has four modules, namely, exposure module, hazard module, vulnerability model, am loss financial model. The first module, the exposure module, is aimed for estimating the total sum insured on each of the locations under assessments. The second module, the loss hazard module, is aimed for determining the frequency and severity of the occurrence of hailstorms in the selected locations. The third module, the vulnerability module, associates the frequency and severity of hail with a certain level of damage to the crop. The fourth module, the loss financial module, calculates (based on the outputs of the hazard module, the vulnerability module, and the exposure module) the amount of a financial gross loss due to the occurrence of hail. The financial gross loss due to the occurrence of hail was tested against the eventual insurance policy terms and conditions to arrive to the technical rate. The exposure module contains the information on risk accumulation per each weather station under analysis and for the months on which the crops are on the field. Spring wheat in NKO is standing on the field from mid May to mid September. In south Kazakhstan, spring wheat is on the field from February to May and, occasionally, begging of June. For the purposes of estimating the spring wheat crop exposure the first and last day of the period at risk were considered at 50 percent of the total exposure. Table A6.1 summarizes the spring wheat risk exposures in northern and southern oblast in Kazakhstan and the assumptions made for the hail crop risk assessment - 298 - Table A6.1 Spring wheat crop risk exposures in North and South Kazakhstan. Region of the Jan Feb Mar Apr May Jun Jul Ago Sep Oct Nov Dic Country 50% 100% 100% 100% 50% North TSI TSI TSI TSI TSI 50% 100% 100% 100% 50% South TSI TSI TSI TSI TSI Source: Authors from Arka Consulting The hazard module of the hail crop risk assessment is compounded by a frequency sub-module and a hazard sub-module. The frequency sub-module is aimed for assess the probability of occurrence of hail in the selected locations, the hail severity sub-module is aimed for - once the occurrence of hail is determined - to assess the intensity of the hail event. The frequency of occurrence of hail is calculated based on the information on the number of hailstorms observed for each month of the year in each of the selected weather stations during the period from 1990 up to and including 2000 according with the information provided from Kazhydromet. The frequency of occurrence of hail is modeled by using a binomial distribution. The severity of hail in the selected locations is estimated based on information obtained from the farmers‘ focal meetings. All the farmers we met in the selected locations agree to mention that, although hail is frequent event in the area, it is rarely severe. Based on this information, in order to reflex hail severity, the study has assumed an histogram frequency distribution with the following assigned probabilities of severity if the hail event occurs: Low; medium, severe, and very severe each of them with probabilities of occurrence of 0.75, 0.15, 0.75, and 0.025, respectively. Table A6.2 presents the probabilities of occurrence of hail for each of the selected weather stations in Kazakhstan. Table A6.2. Probability of Occurrence of Hail for selected weather stations in Kazakhstan. Weather Probability of Occurrence (from Kaz-Hydromet data) Oblast Station I II III IV V VI VII VIII IX X XI XII Akmola Schuchinsk 0.00 0.00 0.00 0.05 0.29 0.24 0.19 0.14 0.05 0.00 0.00 0.00 Akmola Egindykol 0.00 0.00 0.00 0.00 0.24 0.29 0.14 0.24 0.05 0.00 0.00 0.00 Akmola Stepnogorsk 0.00 0.00 0.00 0.00 0.14 0.38 0.19 0.05 0.05 0.00 0.00 0.00 Kostanay Diyevskaya 0.00 0.00 0.00 0.05 0.14 0.38 0.10 0.05 0.05 0.05 0.00 0.00 Kostanay Kostanay 0.00 0.00 0.00 0.00 0.24 0.10 0.19 0.05 0.05 0.00 0.00 0.00 Kostanay Kushmurun 0.00 0.00 0.00 0.00 0.24 0.19 0.19 0.05 0.00 0.00 0.00 0.00 Kostanay Mikhailovka 0.00 0.00 0.00 0.00 0.14 0.19 0.10 0.19 0.14 0.00 0.00 0.00 NKO Balkashino 0.00 0.00 0.00 0.00 0.10 0.10 0.29 0.14 0.00 0.00 0.00 0.00 Pavlodar Aktogay 0.00 0.00 0.00 0.10 0.19 0.19 0.14 0.00 0.19 0.00 0.05 0.00 Pavlodar Zholboldy 0.00 0.00 0.00 0.05 0.14 0.05 0.14 0.00 0.05 0.00 0.00 0.00 Pavlodar Mikhailovka 0.00 0.00 0.00 0.00 0.05 0.10 0.14 0.10 0.05 0.00 0.00 0.00 SKO Kazygurt 0.00 0.00 0.14 0.38 0.43 0.24 0.14 0.05 0.05 0.05 0.05 0.00 SKO Shimkent 0.00 0.00 0.05 0.10 0.33 0.00 0.05 0.05 0.00 0.00 0.00 0.00 Source: Authors from KHM. The model simulates 10,000 hail events for each selected weather station. Based on the hail frequency and severity inputs and the corresponding stochastic distributions defined for weather - 299 - station, the hail risk model, by using @ Risk Software, applies Monte Carlo methodology and generates 10,000 hail events based on the defined stochastic function for each weather station. The crop vulnerability module of the hail crop risk assessment is aimed for, given a certain hail severity on a given phenological stage of the spring wheat crop, to estimate a potential crop loss. Hail damage is produced by the lost on leaf area and by the lost of grains on the last stages of the crop. The basis of the vulnerability module is a vulnerability table with two entrances. The first entrance is the phenological stage of the crop and the second entrance is the severity of the hail storm. Hail damage on wheat crops vary according to the phenological stage of the crop. Hail damage is low on the vegetative stage of the crops and becomes worse from stem elongation to physiological maturity. Indeed, intense hailstorms cause more damage crops than the damage caused by the mild ones. For the case of spring wheat hail risk assessment in Kazakhstan we used a wheat vulnerability tables that are based on the hail vulnerability curves that were developed by Munich Re and are of widespread use in the market. Table A6.3 presents the vulnerability curve for spring wheat crop production for each month on which the crop is exposed to hail losses. Table A6.3. Spring wheat vulnerability to hail losses in North and South Kazakhstan. Month Low Mild Severe Very Severe North South Kazakhstan Kazakhstan Min Max Min Max Min Max Min Max July April 0.00 0.08 0.08 0.16 0.16 0.32 0.32 0.64 August May 0.00 0.12 0.12 0.24 0.24 0.48 0.48 0.96 September June 0.00 0.15 0.15 0.30 0.30 0.60 0.60 1.00 Source: Authors adapted from Munich Re Hail Vulnerability Tables The financial loss module integrates the results from the exposure module, the hazard model, and the vulnerability model in order to estimate a financial gross loss for the hail event. Under the financial loss module, the 10,000 events are tested against the crop vulnerability curves in order to determine the proportion of the loss. In case there is a loss, this proportion is applied to the exposure module in order to determine the size of the loss. The model deducts the losses from the remaining sum insured in the crop. That means that if the crop was insured for KZT 10,000 and during the first month of the coverage it was a loss of 20 percent, the crop remains insured for KZT 8,000. Where, TU = rayon-farm typology, 1 to 221; To = beginning of month T-1= previous month Crop Portfolio Risk Assessment Model – Rating Exercise Hail pure loss cost rating methodology The Technical rates rating methodology is based on standard hail rating procedures. The loss cost formula is given by: - 300 - Loading pure loss cost to derive Commercial Premium Rates The technical rates calculated by the model are then loaded to cover various cost components in order to derive final commercial premium rates which are paid by growers. The general formulae for developing the final premium rates include: The study did not include the analysis we have still not performed a detailed analysis of the potential interested insurance companies cost structure - acquisition cost, administrative cost, insurers and reinsurers profit margins expectations). For these reasons the current study, based on the international experience in multiple period crop insurance products, assumes a target loss ratio of 80 percent. Therefore, the derivation of pure premium rates into indicative commercial premium rates is given by the following formulae: - 301 - Annex 7: Area-Yield Index Insurance Opportunities in Kazakhstan This annex provides an analysis of the key design and rating issues and methodology for an Area-Yield Index program for paddy production in Kazakhstan and draws where relevant on international experience. Outline proposals are presented for a prototype Area-Yield Index program for spring and autumn paddy in Kazakhstan, but it is stressed that further design work will be required if GRK decides to go forward with this initial idea. The section embraces all the steps involved on the design of an AYII product and concludes with the notional cost for potential AYII for spring wheat in Kazakhstan. It starts with the description of the features of this insurance product and the review of the international experience in AYII. Next, the section deals with aspects related with the potential basis risk of area yield index agricultural insurance products. Once the issues related with basis risk are addressed, the section explains the aspects related with the coverage design. Next to this, through an applied example for the selected crops and region, the section describes the rating methodology for an AYII product. Following to the explanations regarding the ratting process, the section describes the pricing issues. Finally, the outputs of these processes, that is the market reference rates, and a simulation of a possible loss scenario are presented. Features of Area Yield Index Insurance AYII represents an alternative approach to MPCI insurance which aims to overcome many of the drawbacks of traditional MPCI crop insurance. The key feature of this product is that it does not indemnify crop yield losses at the individual field or grower level. Rather, an Area-Yield- Index product makes indemnity payments to growers according to yield loss or shortfall against an average area yield (the index) in a defined geographical area (e.g., Rayon or spring wheat production zone). An area-yield index policy establishes an Insured Yield which is expressed as a percentage (termed the ―Coverage Level‖) of the historical average yield for each crop in the defined geographical region which forms the Insured Unit. Farmers whose fields are located within the Insured Unit (IU) may purchase optional coverage levels which typically vary between a minimum of 50 percent and a maximum of 80 percent of historical average yield. The actual average yield for the insured crop is established by sample field measurement (usually involving crop cutting) in the Insured Unit and an indemnity is paid by the amount that the actual average yield falls short of the Insured Yield Coverage level purchased by each grower. The key advantages of the Area-Yield approach are that moral hazard and anti-selection are minimized, and the costs of administering such a policy are much reduced and this offers the potential to market this product at lower premium costs to growers. The main disadvantage of an AYII policy is that an individual grower may incur severe losses due to localized perils e.g. hail, or flooding by a nearby river, but because these localized losses do not impact on the county or departmental average yield, the grower does not receive an indemnity. (See Box A7.1. for further details). - 302 - Box A7.1. Area-Yield Index Insurance: Advantages and Disadvantages ADVANTAGES DISADVANTAGES Adverse selection and Moral Hazard are Basis risk issues minimized The occurrence of basis risk depends on the extent to The indemnity is based on average area yields and not which individual farmer‘s yield outcomes are positively on individual farmers‘ yields. Individual farmers correlated with the area-yield index. cannot therefore influence the yield outcome. Yield Data Availability for Insurance Not suitable for localized perils Time-series Regional-level or Zonal-level area-yield Area-Yield insurance will not work in areas with high data is available at Guyana Rice Development Board. losses due to localized perils e.g. hail, or localized frost pockets. Comprehensive Multi-peril Insurance suited to the Requires Homogeneous Agro climatic Risk regions insurance of systemic risk and Cropping Systems. The policy acts as an All Risk Yield shortfall Area-yield insurance works best in a homogeneous guarantee policy and is best suited to situations where climatic zone and where cropping systems for the severe systemic risk (e.g. drought) impacts equally insured crop are uniform (e.g. same varieties, planting over the defined area Insured Unit (e.g. Paddy dates, management practices). Production Zone). Lower Underwriting and Delivery Costs Accuracy of Historical Area Yield Data There is no need to conduct pre-inspections on Methods of yield measurement and reporting may not individual farms or to collect individual grower yield be accurate racing doubts about the historical area- data. yields. Lower Loss Adjusting Costs Problems of Accurate Measurement of Area Yields There is no requirement for individual grower infield Sampling error and enumerator bias can be a major area loss assessment which is very time consuming problem in determination of average area yields. and costly. Affordability of Product Time delays in settling claims The combination of reduced exposure to yield loss Farmers often have to wait for at least 3 to 6 months and reduced administrative costs offers the potential post harvest for the official results of the area yields to for cheaper premiums than for individual farmer be published and for indemnities to be paid if MPCI. applicable. Source: Authors . International Experience with Area Yield Index Insurance Area-Yield Index crop insurance has been implemented in many countries. In the late 1970‘s, India introduced area-yield index crop insurance and the USA and Canada introduced this product in the early 1990‘s. Recently, other countries like Morocco, Mexico, Sudan, and Brazil have developed area based crop insurance products. In India area-yield crop insurance has operated for over 20 years and is currently the world’s largest crop insurance program insuring about 20 million farmers. The Agricultural Insurance Company of India (AICI) is responsible for implementing area-yield crop insurance under the National Agricultural Insurance Scheme (NAIS). The program is targeted at small and marginal farmers (with less than 2 hectares), and whom are highly dependent on access to seasonal crop credit. Crop Insurance is compulsory for borrowing farmers and voluntary for non-borrowing farmers. The Insured Unit is normally the block or panchayet which comprises a group of nearby villages and which may include up to 27,000 acres or more of a single crop and several thousands of small and marginal farmers. Farmers may select coverage levels of 60 percent, 80 percent or maximum of 90 percent of the 5-year average area-yield. The program is administered through the rural agricultural bank branch network in each state and department and block (group of villages). Actual area-yields are established through sample crop-cutting. This is a major and - 303 - costly exercise and suffers from delays in processing the results. Indemnity payments are therefore often delayed for 6 months or more. In USA, AYII is marketed under the name Group Risk Plan (GRP). Under the GRP the pay outs of the coverage rather than be based on the individual farmer's yield loss experience; they are based on the actual value of an area-yield index in a certain area, namely the insured unit, which in US is defined by the county level ( 2500 km2 average insured unit). The indemnities on the GRP proceeds when the actual yield for the insured crop at the county on which the insured is situated, as determined by the National Agricultural Statistics Service (NASS), falls below the guaranteed Yield chosen by the farmer. Under the GRP, farmers can choose among different coverage levels (Insured Yield options): 90 percent, 85 percent, 80 percent, 75 percent or 70 percent of the expected county yield. The sum insured value for each crop is based on a percentage of the expected market price. The grower may elect an insured value of between a minimum of 90% and a maximum of 150% of the expected market price. The justification for permitting growers to insure at up to 150% of the expected market price is that this affords adequate protection for growers whose own yields are higher than the county average. Final payments are not determined until six months after the crop harvest when NASS released the actual county yields for each county. Payments are then made within 30 days. GRP insurance policies are easier to administrate and less costly than the traditional individual grower MPCI policy. However, individual crop losses may not be covered if the county yield does not suffer a similar level of loss. This type of insurance is most appropriate for farmers whose crop production and yields (and losses) typically follow the county pattern. The Issue of Basis Risk Basis Risk can be defined as the potential mismatch in terms of yield performance between the individual field and the geographical area defined as the IU for the AYII. The feature that the indemnity payments of an AYII are based on a yield loss or shortfall against an average area yield (the index) in a defined geographical area, makes room to the existence of basis risk on these kinds of insurance products. Because of this reason, two undesirable situations may occur: (a) growers who not suffered any yield shortfall below the coverage level receive indemnities from the insurance because the insured unit where they are situated has suffered a yield shortfall in respect to the guaranteed yield, and (b) growers that have actual yields below the coverage level, do not received any indemnity from the insurance because the actual yield for the IU on which are situated is above the coverage level. Both situations are serious drawbacks for the sustainability of an AYII product. The issue of the basis risk must be seriously addressed on the design of AYII. Basis Risk can be mitigated but it cannot be eliminated from an AYII portfolio. The issue of the basis risk is related to how correlated are the yields at growers field level and the yields in the geographical area selected as IU for the coverage. The choice of the guaranteed yields for the coverage and the selection of IU are key topics that need to be addressed on the design of area- yield insurance products to mitigate the basis risk. The experience with AYII products demonstrates that as higher the coverage level is settled, the bigger the basis risk problem is; likewise, as bigger the geographical area selected as IU is, the bigger the basis risk problem is. Basis risk is a serious issue for area-yield index products that have coverage levels settled close to the expected yields. Small yield shortfalls in respect to the expected yields are more in relation with idiosyncratic risks, like crop management and crop husbandry practices, than with weather events. At high coverage levels, the correlations between the yield performance at the individual grower field level and the yield performance at the geographical area selected as IU are not strong enough. The correct definition of the insured unit is also a key factor for the mitigation of basis risk issues in an AYII coverage. Area zone boundaries for an AYII must be selected so as to group together the largest possible number of farms with similar climate and soils (Skees, 1997). - 304 - In other words, as bigger the geographical area selected as IU is, the lower the probability to group together the largest possible number of farms with similar climate and soils is. The issue of Basis risk was examined under the current study for spring wheat gown in Kazakhstan. In order to assess basis risk under an AYII program, it is necessary to compare the crop yields obtained by individual farmers and to check the differences between farmers‘ yields and actual average yield in the same year at Rayon level. For the Rayons which were selected for detailed analysis under this study, ARKA Consulting kindly collected some individual farmer- level spring wheat crop yield data for the three year period 2008 to 2010. This individual farm- level spring wheat yield data for a sample of about 10 farmers has been analysed for Aktogay and Zhelezninski Rayons in Pavlodar Oblast and the results are summarized in Table A7.1 at the end of this Annex. For Aktogay the average expected yield is 5.8 centners/Ha and assuming a 50% AYII insured yield coverage level of 2.9 centners/Ha, Table 7.1 shows that in 2008 when the average rayon spring wheat yield was 1.3 centners per hectare all farmers would have received an indemnity of 2.9 – 1.3 = 1.6 centners per hectare. However, the actual yields for this small sample of farms was considerably higher than the Rayon yield in 2008 and in this case the AYII Cover would have paid out bigger losses than were actually incurred by all these farmers. The 2008 farmer yields were highly variable (COV 42%) and the degree of basis risk is evidenced by the major differences (divergence) between individual farmer yields and the Rayon Yield Index. In 2009 which was a well above average yield year the sampled farmers achieved a very high average yield of 9.8 centners per Ha with low variation (COV 8%) against the Rayon average or expected yield of 5.8 Centners/Ha. In 2009 the AYII policy would not have paid out given the very good yields achieved by most farmers in Aktoby Rayon and would have therefore performed as intended. In 2010 however, which was again a severe drought year, the policy would not have paid out any indemnities as the Rayon average yield of 3.2 centners/Ha would have been above the 50% Insured Yield coverage level of 2.9 centners/Ha. In this case nearly 50% of the sampled farmer achieved average yields of less than the 50% coverage level, but they would not have received an indemnity because other farmers achieved higher yields thereby raising the Rayon average yield to 3.2 centners/Ha. Basis risk would therefore have been a problem in 2010 with 50% coverage level. For Zhelezninski Rayon which has an expected yield of spring wheat of only 6.1 centners/Ha the same assumed 50% AYII coverage level would amount to 3.05 centners/Ha. In 2008 which was a below average yield year with a Rayon average yield of 4.4 centners/Ha there would have been no indemnities because the this yield would have exceeded the 50% insured coverage level of 3.05 centners/Ha. In this case the actual average yields achieved by individual farmers very closely approximated the rayon average yield and no farmer achieved less than 4 centners/Ha. In other words, the policy responded as intended and basis risk would have been low. 2009 and 2010 were extremely good yield years with the sampled farmers achieving average yields of 17.8 centners and 10.5 centners/Ha respectively or well above the average yield and again there would have been no payouts. The above analysis indicates that under any future possible AYII program for spring wheat the planners should analyse on a Rayon by Rayon case the individual grower spring wheat yields and compare these against the Rayon average yields in order to assess Basis Risk. As a guideline the AYII product will operate best where there is a low degree of yield variability between individual farmers in a the same year and it will not perform so well where individual farmer yields are highly variable. - 305 - Insured Yield Coverage Levels Area-Yield Insurance policies usually offer Insured Yield Coverage levels of between 90% maximum and 50% minimum of annual average yield for the spring wheat production zone. The methodology for cleaning and detrending the actual rayon 17-year spring-wheat yields and then in simulating Rayon-level yields with 5000 iterations is outlined in Appendix 7.1. at the end of Annex 7. The average yields for the AYII Cover are presented separately for Commercial Farmers, Agricultural Enterprises and then the aggregate or combined yields for both types of farmer are presented in Table A7.2. The corresponding Rayon-level 10% up to a maximum of 80% insured yield coverage levels are shown in Tables A7.3, A7.4. and A7.5 respectively for Commercial Farmers, Agricultural Enterprises and finally for the combined yields for both sets of farmers.. Estimation of Technical Rates and Indicative Commercial Premium Rates for Area-Yield Index Insurance The rating methodology for the Rayon-level AYII spring wheat product is set out in Appendix 7.1. and closely approximates the procedure used for individual grower MPCI, the main differences being (i) there is no need to model the higher yield variability expected at the individual farmer level and (ii) the average calculated loss costs do not need adjusting for idiosyncratic risk (e.g. crop hail) under the AYII cover. The technical and indicative commercial premium rates for AYII cover have been estimated separately for Commercial Farmers and Agricultural Enterprises (production Enterprises and then finally for the Combined aggregate cover for both types of farmer at Rayon level. The technical rates can be provided to interested parties on request and the indicative commercial premium rates for an assumed 70% target loss ratio are presented in Tables A7.2, A7.3 and A7.4 respectively for the Commercial Framers, Production Enterprises and finally for both types of farmer and for insured yield coverage levels of 10% up to a maximum of 80% of the Rayon average expected yields for each type of farmer and in aggregate. Estimation of the Probable Maximum Loss (PML) for the selected area and crops The CRAM is programmed to calculate the Probable Maximum Losses which might be expected under an Area Yield Index AYII program for spring wheat in Kazakhstan. The calculated PML‘s for the combined Commercial Farmer and Production Enterprise yields are presented in Figure A7.1. and Table A7.a. below. For a 50% coverage level the 1-in-a-100 year PML would be about PML would be about 24% of TSI or a loss of KZT 51 billion (US$ 343 million) assuming 100% scheme uptake. This would be equivalent to a loss ratio of about 370%. For the highest 80% coverage level the 1-in-a-100 year PML would be about 45% of TSI or a loss of KZT 152.7 billion (US$ 1.02 billion) assuming 100% scheme uptake. This would be equivalent to a loss ratio of about 259%. - 306 - Figure A7.1. Spring Wheat AYII scheme PML estimates by coverage level from 10% to 80% 60.00% 50.00% "10% Coverage" 40.00% "20% Coverage" Loss Cost % "30% Coverage" 30.00% "40% Coverage" 20.00% "50% Coverage" "60% Coverage" 10.00% "70% Coverage" 0.00% "80% Coverage" 1 50 100 150 200 250 Return Period (years) Table A7.a. Estimated 1 in 100 Year PML for AYII Wheat Scheme (100% figures) Coverage level Item 10% 20% 30% 40% 50% 60% 70% 80% PML (% of TSI) 1.11% 4.53% 10.04% 17.40% 24.29% 31.84% 39.15% 44.99% PML (KZT Billion ) 0.5 3.8 12.8 29.5 51.5 81.0 116.3 152.7 PML (US$ Million) 3 25 85 197 343 540 775 1,018 1 in 100 year PML 466% 461% 440% 414% 372% 334% 297% 259% Loss Ratio % Source: Authors Appendix 7.1. Risk Modeling and Rating for Area Yield Index Insurance (AYII) Crop Portfolio Risk Assessment Model – Design Features – This section presents the basic design features of the Crop Risk Assessment Model (CRAM) for spring wheat production in the selected oblasts in North Kazakhstan and the rating tool designed for AYII The CRAM is constructed based on analysis of variation of spring wheat annual average yields for a 17year time-series, from crop year 1994 up to and including the crop year 2010, at the rayon level. The CRAM was developed using the sown area, harvested area, production and annual average yield statistics for each of the rayons in North Kazakhstan, Akmola, Kostanay, Karaganda, Pavlodar, East Kazakhstan, Aktobe, and WKO in North Kazakhstan region. The original series used for this analysis were provided by ARKS. Planted Area - 307 - According to information obtained from ARKS, spring wheat crops sown area amount to 13.26 million hectares on average for the period 2007 – 2010. Out of the 13.26 million hectares planted with spring wheat in North Kazakhstan region, 8.96 hectares (68 percent) are planted by agribusiness enterprises and 4.31 million hectares (32 percent) are planted by commercial farmers. The CRAM assumes that the annual spring wheat planted area has remained constant at the four year average (period 2007-2010) over the 17 years sown area series. The reason for this assumption is to remove seasonal variations for each rayon from the areas. In order to be eligible for CRAM, two criteria have been set: minimum planted area per region and a minimum of 17 years continuous annual average yield data. In order to ensure that there are sufficient numbers of farmers growing the crop in a selected region, a minimum area of 10,000 hectares has been provisionally settled as a requirement for a crop in a certain region to be eligible for the model. The second criterion, at least 17 continuous years of yield data available for each rayon to qualify for the CRAM, has been settled to have continuous series in order to establish possible yield correlations among different production zones. Rayon Crop Yield Data The CRAM uses rayon annual average yields for spring wheat crops for the period starting 1994 and up to crop year 2010 as reported by ARKS. The original rayon annual average yields from 1994 to 2010 are available on request from ARKS. The ARKS reports average yields on sown area basis at rayon level. This is an important advantage for risk modeling purposes, since the yields on sown capture, both, the variations due to yield performance as well as the yield variations due to full crop losses. The ARKS reports average yields for two categories of farm typology, agribusiness enterprises and commercial farmers. This fact allows CRAM to perform the risk assessment at rayon level with a breakdown per each type of farm typology. Valuation Prices For CRAM risk modeling purposes, spring wheat has been valued at the average market average price per centner for the period 2008-2010 for the month of September, KZT 3120 per centner. The crop price is maintained as a constant value for all the past 17 years. Yield Data Cleaning and trending to establish the Central Tendency The annual average yield series at zone level used to feed CRAM must be adjusted in order to reflect the current state of the art in terms of expected yields and yield variability for the selected crops for the risk assessment. This sub-section describes the methodologies followed to clean the yield data, determine the trend in yield data and, finally, to adjust the historical yields to the current expected yield at Region level. Eliminate Yield Outliers - 308 - The first step was to detect and eliminate the statistical outliers from the annual average yield series for each of the selected crop and regions by applying the Chauvenet107 criteria. Each of the 17 years annual average yields records for each agribusiness enterprises and commercial farmers on each of the rayon If, by applying the Chauvenet criteria a yield outlier was detected, then the annual average crop yield was compared with the annual average crop yield performance for the same crop and year in neighboring rayon. If, as result of this comparison, it was detected that the crop yield performance in neighboring rayons diverged significantly in respect to the annual average yield for the target rayon and year, then yield, production, and harvested area figures were revisited to identify the cause of the divergence. Adjusting Zonal Average Yield Data for Trends The crop yield central tendency is associated with crop management and technology practices; crop yield deviations from the central tendency are associated with effects of nature. The main objective of adjusting the historic annual average yield series was to isolate the effect on yields of the improvement on crop management practices and the increase in technology application to the crops along 17-year period considered for the analysis. A simplified method was adopted for determining the central tendency for each crop and each zone in the CRAM. The method aims to capture the non-linear yield tendency in the 17-years of annual average yield series at zonal level by using this yield series fitted to a lineal trend line and to an exponential trend line, and the five year moving average the 17 years annual average yield series. Expected Yields and adjusted crop variability. The design of the CRAM is based on the spring wheat annual average yields for the period 2006 – 2010 at rayon level and their standard deviation; thus, these inputs must be representative of the current state of the art of spring wheat crop production in each of the analyzed rayons. That is, all the long terms and cyclical effects of crop management practice and of technology application on the historic annual average yields must be isolated prior to estimating these parameters for risk modeling purposes. In order to calculate the expected annual average yield for spring wheat for each rayon in the 8 oblasts under analysis, the simple average of the most recent five years historic annual average yields was calculated. This method to estimate annual average expected yields for a certain crop located in a certain rayon is common in the agricultural insurance practice in countries where the constraint of scarce annual average crop yield data is a problem. The second part of this analysis was to estimate the expected annual average yield volatility of the annual average yield. The method used for this purpose was to measure the deviations between the historic actual annual average yields for each year of the series in respect to the corresponding annual average yield of the trend line. Then, these deviations were applied to the expected yield to obtain an adjusted annual average yield series. Modeling Expected Yields 107 In statistical theory, the Chauvenet‘s Criterion is a means of assessing whether one piece of experimental data – an outlier- from a set of observations, is likely to be spurious. - 309 - The estimation of losses for the spring wheat crop portfolio was performed through a risk modeling exercise using the CRAM. Risk modeling is a fundamental step in agricultural insurance program design and ratemaking procedures. The main objective of crop risk modeling is to estimate, based on the available information, a yield probability density function that reflects the stochastic nature of yield outcomes. The yield model relies on two basic fundamentals: (a) a crop yield probability density function inferred from the historic spring wheat annual average yields for each rayon and type of farmer in the analyzed portfolio, and (b) a correlation matrix of rayon-level and farmer-type level spring wheat annual average which reflects the covariant risk under the portfolio. The probability density functions were inferred from the technology adjusted annual average yields from the annual average yield series 1994 to 2010 that were fitted to a Weibull probability distribution. For the purpose of assessing risks for an individual grower AYII scheme is necessary to the Probability functions fitted based on rayon level actualized annual average yields into spring wheat yield probability function that reflect the yield variability at individual farm level. Individual crop yield variability is always bigger than aggregate rayon level crop yield variability. Figure A7.2: Akmola- Bulandinsky: Spring Wheat. Rayon-Level yield probability density function Spring Wheat. Akmola-Bulandinski Rayon. 4.36 17.81 0.1 0.09 0.08 0.07 0.06 frequency 0.05 0.04 0.03 0.02 0.01 0 0 5 10 15 20 25 30 Yield (Centner/he.) 35 Spring Wheat Yield Rayon Level Source: Authors from CRAM The outputs of yield probability density functions obtained for each rayon and type of farmer were correlated in order to reflect the covariance on yields for risk modeling purposes. Spring wheat crop production in Kazakhstan is exposed to drought which is a very systemic risk. Variations in spring wheat crop yields are often caused by factors that typically affect a large area. The issue of a portfolio being exposed to systemic risk, since it affects the degree on which the risks can be diversified, has severe implications for the designing of crop insurance. In light of the systemic risk faced by spring wheat crop production in Kazakhstan, the CRAM considered the correlations among each rayon and type of farmer in order to simulate the potential losses for the portfolio. The Correlation matrix is presented in Annex 1. - 310 - CRAM Simulation Based on the stochastic distributions defined for each rayon and type of farmer and based on the spring wheat rayon level correlation matrix, the CRAM, by using @ Risk Software, applies Monte Carlo methodology and generates 5,000 iterations of yields based on the defined stochastic function for each rayon and type of farmer which are correlated each other. CRAM Output As a result of CRAM a database compounded by 221 combinations of rayon and farm type level and 5000 yield iterations for each of the combination rayon and farmers‘ type is generated. This database is the one used as underlying for risk assessment and risk pricing purposes. Crop Portfolio Risk Assessment Model – Rating Exercise Insured Yield Coverage Levels AYII was devised for spring crop production in the 8 selected oblast in north Kazakhstan offering yield coverage levels of between 80% maximum and 10% minimum of the 5-years average yield for the period 2006-2010. AYII rating methodology AYII pure loss cost rating methodology The Technical rates rating methodology is based on standard AYII rating procedures. The loss cost formula is given by: For the AYII Program, the loss cost formula is given by: Where, Y= year, 2006-2010; U = rayon-farm typology, 1 to 221; C = crop, spring wheat Coverage Level is between a minimum of 10% and maximum of 40% of average yield. Using Contiguous Counties to Smooth Rates The AYII insurance pure loss costs obtained for each rayon-type of farmer are smoothed by utilizing information from contiguous rayons (Skees, 1997). The smoothed pure loss cost for each rayon-type of farmer is calculated as a weighted average of the pure loss cost for that rayon-type of farmer and the pure loss cost for each contiguous rayon for the same type of farmer. The formula to calculate the smoothed pure loss cost is: - 311 - The weights are calculated as follows: Subject to: Where, is the weight assigned to the target rayon-type of farmer and is the average hectares planted in the target rayon over the most recent 4-year period; and Where, is the weight assigned to the ith contiguous rayon, and is the average area planted over the most recent 4-year period for each contiguous rayons. All weights sum to one. Loading the smoothed pure loss cost rates to derive technical rates Likewise for MPCI calculation, a security loading was added to the area yield index insurance rating model in order to consider those catastrophic losses that were not captured within the 17- years spring wheat crop yield series used as basis for risk modeling. With that purpose, the method of risk exposure was introduced in the rate calculation. The method of risk exposure consists in calculate the loading factor based on the 1 in 100 year probable maximum loss (PML) for the whole portfolio under the assumption that a loss approximately equal to the PML will take place within a certain number of years ahead. For the purposes of the rating calculations for AYII in Kazakhstan it was assumed that a spring wheat portfolio loss similar to the spring wheat PML will take place within the next 10 years. Security loadings are Coverage specific. The PML for the portfolio varies according to the selected level of coverage. Table A7.b presents the security loadings used to load the Rayon-level AYII spring wheat average loss costs for the different levels of coverage analysed in the study. Table A7.b. Security Loadings applied to AYII average loss cost rates to derive AYII Technical Rates Coverage Level 10% 20% 30% 40% 50% 60% 70% 80% PML 1-100% 1.11% 4.53% 10.04% 17.40% 24.29% 31.84% 39.15% 44.99% Return Period 10-years 10-years 10-years 10-years 10-years 10-years 10-years 10-years Loading Factor 0.111% 0.453% 1/00% 1.74% 2.43% 3.18% 3.91% 4.45% Last, the calculation of the technical rates consist on the integration of the smoothed loss cost calculation, and the loadings due to idiosyncratic risks and catastrophic risks. The formula summarizing the calculation of the technical rates is presented below: - 312 - Loading Technical to derive Commercial Premium Rates The technical rates calculated by the model are then loaded to cover various cost components in order to derive final commercial premium rates which are paid by growers. The general formulae for developing the final premium rates include: The study did not include the analysis we have still not performed a detailed analysis of the potential interested insurance companies cost structure - acquisition cost, administrative cost, insurers and reinsurers profit margins expectations). For these reasons the current study, based on the international experience in multiple period crop insurance products, assumes a target loss ratio of 70 percent. Therefore, the derivation of pure premium rates into indicative commercial premium rates is given by the following formulae: It is noted 70 percent is a reasonable target loss ratio for an MPCI cover. We understand that, if the scheme reaches economies of scale, the administrative expenses could be substantially reduced; thus, the target loss ratio could be increased - 313 - Annex Table A7.1. Comparison of infidel farmer spring wheat yields with Rayon Average Yields, Pavlodar District Aktogay Year 2008 2009 2010 Rayon Average Yield 1.3 8.5 3.2 Aktogay: Individual Yields (25km away WS) vs Rayon 12.0 Farm Kairat 8.7 2.1 Yields Sagat 2.2 10.5 1.2 10.0 Yields (centner/he) Rayon Average Yield Elita 2.9 10.5 4.0 8.0 Farm Kairat Pobeda 8.5 1.7 Sagat Kazakhstan 9.8 1.2 6.0 Elita Aslan 9.8 2.0 Vesna 2.3 10.5 4.0 4.0 Pobeda Loza 4.7 9.8 5.4 Kazakhstan 2.0 Urozhay 9.7 3.0 Aslan Group Average 3.0 9.8 2.7 0.0 Vesna Group Standard Deviation 1.3 0.7 1.5 Loza 2008 2009 2010 Coeficient of Variation 42% 8% 53% Zhelezninski Year 2008 2009 2010 Rayon Average Yield 4.4 17.5 7.9 Zhelezninski: Individual Yields (25km away WS) vs 35.0 Rayon Yields Akbiday 5.8 30.0 14.0 Rodnik 4.1 18.0 11.0 30.0 Rayon Average Yield Salun 4.0 18.0 11.0 Akbiday Yields (centner/he) Angelina 4.5 20.0 12.0 25.0 Ageenko 4.3 19.0 10.0 Rodnik Nurgaliyeva 4.1 15.0 9.0 20.0 Salun Kirilov 4.2 18.0 10.0 Angelina Svintitsky 4.1 18.0 9.0 15.0 Ageenko Tupitsin 4.1 18.0 9.0 Nurgaliyeva Toktushakov 4.6 18.0 11.0 10.0 Kirilov Druzik 4.5 15.0 12.0 5.0 Svintitsky Trusova 4.3 10.0 10.0 Tupitsin Andreenko 4.2 15.0 9.0 Group Average 4.4 17.8 10.5 0.0 Toktushakov Group Standard Deviation 0.5 4.5 1.5 2008 2009 2010 Druzik Coeficient of Variation 11% 25% 14% Source: Author's analysis of Individual grower spring wheat yield data provided by Arka Consulting August 2011 - 314 - Annex 7. Table C o m m e r- C o m m e r- c ia l P ro d u c t io n A ll c ia l P ro d u c t io n A ll C ro p / D is t ric t / S u b d is t ric t F a rm e rs E n t e rp ris e s F a rm e rs C ro p / D is t ric t / S u b d is t ric t F a rm e rs E n t e rp ris e s F a rm e rs S u b t o t a l C F / Ka ra g a n d in s k a y 5 .7 8 .1 6 .8 Akm o la / Akko l 8.1 8.8 8.7 Ko s ta na s ka y/ Altyns a rin 15.0 13.3 13.7 Akm o la / Ars ha ly 10.8 9.5 9.9 Ko s ta na s ka y/ Am a nge di 7.7 7.2 7.5 Akm o la / As tra s ha ns ki 6.1 7.6 7.0 Ko s ta na s ka y/ Arka lyk c ity 6.2 9.2 8.0 Akm o la / Atba s a rs ki 9.3 9.2 9.2 Ko s ta na s ka y/ Auliyko ls ki 8.2 9.9 9.5 Akm o la / B ula ndins ki 10.8 10.4 10.4 Ko s ta na s ka y/ De nis o vs ki 12.0 11.9 12.0 Akm o la / C e lino gra ds ki 7.5 7.0 7.1 Ko s ta na s ka y/ F e do ro vs ki 16.2 16.4 16.3 Akm o la / Enbe ks hilde rs ki 10.7 12.1 11.9 Ko s ta na s ka y/ Ka m is ty 8.8 9.2 9.1 Akm o la / Es ils ki 7.6 7.9 7.9 Ko s ta na s ka y/ Ka ra ba lyk 14.5 14.7 14.7 Akm o la / Ko ks he ta u 5.6 3.6 0.0 Ko s ta na s ka y/ Ka ra s u 11.4 10.8 11.0 Akm o la / Ko rga lzhins ki 4.7 5.8 5.6 Ko s ta na s ka y/ Uzunko ls ki 16.9 14.0 14.7 Akm o la / S a ndikta us ki 10.5 12.0 11.7 Ko s ta na s ka y/ Zha nge ldin 6.0 4.8 0.0 Akm o la / S ho rta ndins ki 8.8 8.7 8.7 Ko s ta na s ka y/ Zhe tika ra 8.5 8.4 8.4 Akm o la / S huc he ns ki 11.5 11.2 11.2 Ko s ta na s ka y/Ko s ta na s ka y 16.7 12.7 14.9 Akm o la / S te pno go r 5.4 5.4 0.0 Ko s ta na s ka y/M e ndika ra 17.3 14.0 15.2 Akm o la / Ye gindiko ls ki 6.0 7.2 6.8 Ko s ta na s ka y/Na uirzym 8.8 6.8 7.9 Akm o la / Ze re ndins ki 11.8 14.1 13.6 Ko s ta na s ka y/ R udnyi c ity 4.6 10.9 4.6 Akm o la / Zha ks i 11.5 11.8 11.7 Ko s ta na s ka y/S a ryko l 14.9 14.6 14.7 Akm o la / Zha rka ins ki 7.2 8.6 8.1 Ko s ta na s ka y/Ta ra no vs ki 12.0 11.3 11.5 Akm o la /Ere m e nta us ki 8.3 7.4 7.7 S u b t o t a l P E / Ko s t a n a s k a y 12 .6 11.8 12 .1 S u b t o t a l P E / A k m o la 8 .3 9 .5 9 .2 NKO/ Airta u 15.5 13.2 13.6 EKO/ Aya go z 6.1 4.5 0.0 NKO/ Akka yn 16.6 14.4 14.9 EKO/ B e s ka ra ga y 5.6 6.0 5.8 NKO/ Akzha r 13.7 11.0 11.9 EKO/ B o ro duliha 8.7 11.1 9.6 NKO/ Es il 15.4 13.8 14.0 EKO/ Glubo ko e 14.5 13.7 14.2 NKO/ G.M us re po v 14.8 13.2 13.4 EKO/ Ka to nka ra ga y 13.9 13.1 0.0 NKO/ Kyzylzha r 17.1 14.3 15.1 EKO/ Ko kpe kti 10.2 7.7 9.0 NKO/ M .Zhum a ba ye v 15.2 13.7 14.2 EKO/ Kurc hum 10.9 4.6 0.0 NKO/ M a m liut 16.0 13.7 14.2 EKO/ R idde r c ity 17.1 15.2 0.0 NKO/ S ha la kin 14.7 12.9 13.3 EKO/ S e m e y c ity 5.6 6.5 5.6 NKO/ Ta iyns ha 14.7 13.3 13.5 EKO/ S he m o na iha 14.4 15.8 15.2 NKO/ Tim irya ze v 14.9 12.4 13.5 EKO/ Ta rba ga ta y 10.5 5.4 0.0 NKO/ Ua liha no v 12.2 11.3 11.5 EKO/ Ula n 8.8 11.8 9.8 NKO/ Zha m bil 16.3 11.6 13.2 EKO/ Urzha r 9.2 7.4 9.2 S u b t o t a l P E / N KO 15 .1 13 .0 13 .5 EKO/ Us t-Ka m e no go rs k c ity 12.0 11.9 0.0 P a vlo da r/ Aks u c ity 4.8 4.3 0.0 EKO/ Za ys a n 15.5 10.9 0.0 P a vlo da r/ Aks uis ky 5.4 4.3 0.0 EKO/ Zha rm a 9.6 7.9 0.0 P a vlo da r/ Akto ga is ky 5.8 5.4 5.8 EKO/ Zyrya n 14.7 14.1 14.3 P a vlo da r/ B a ya na uls ki 4.7 5.7 0.0 S u b t o t a l C F / E KO 9 .7 11.4 10 .4 P a vlo da r/ Ekiba s tuz c ity 9.1 5.2 7.4 Ka ra ga ndins ka y/ Aba i c ity 5.0 10.9 0.0 P a vlo da r/ Irtys hs ki 7.0 6.7 6.9 Ka ra ga ndins ka y/ Akto ga y 6.1 4.9 0.0 P a vlo da r/ Ka s hyrs ki 8.7 8.2 8.5 Ka ra ga ndins ka y/ B uha rzhira u 5.9 8.3 7.0 P a vlo da r/ P a vlo da r a re a 5.9 5.3 0.0 Ka ra ga ndins ka y/ Ka rka ra ly 6.0 3.2 6.0 P a vlo da r/ S he rna ktins ky a re a 7.5 7.6 7.5 Ka ra ga ndins ka y/ Nura 4.4 8.5 7.2 P a vlo da r/ Us pe nka a re a 7.7 6.7 7.2 Ka ra ga ndins ka y/ Os a ka ro v 6.6 7.9 7.2 P a vlo da r/ Zhe le znins ki 4.3 9.0 6.1 Ka ra ga ndins ka y/ S he ts ki 5.3 5.7 5.3 S u b t o t a l P E / P a v lo d a r 6 .7 7 .5 7 .0 Ka ra ga ndins ka y/ Ulyta u 5.8 6.2 6.0 A ll S e a s o n s a g g re g a t e 10 .9 11.1 11.0 Ka ra ga ndins ka y/ Zha na a rka 4.8 5.5 4.8 S u b t o t a l C F / Ka ra g a n d in s k a y 5 .7 8 .1 6 .8 - 315 - Table A7.2 Estimated Commercial Premium Rates for target loss ratios equal to 70% and Guaranteed Yields Area Yield Index Insurance for Spring Wheat produced by Commercial Farmers in North Kazakhstan Region Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Akmola Akkol 0.81 0.47% 1.61 1.95% 2.42 4.37% 3.22 7.67% 4.03 11.43% 4.83 15.80% 5.64 20.62% 6.44 25.58% Akmola Arshaly 1.08 0.32% 2.15 1.30% 3.23 3.03% 4.31 5.59% 5.38 8.64% 6.46 12.41% 7.53 16.82% 8.61 21.62% Akmola Astrashanski 0.61 0.52% 1.21 2.05% 1.82 4.47% 2.42 7.83% 3.03 11.75% 3.64 16.32% 4.24 21.30% 4.85 26.47% Akmola Atbasarski 0.93 0.42% 1.86 1.84% 2.78 4.16% 3.71 7.36% 4.64 11.10% 5.57 15.49% 6.50 20.37% 7.42 25.45% Akmola Bulandinski 1.08 0.30% 2.17 1.30% 3.25 3.05% 4.34 5.64% 5.42 8.78% 6.50 12.64% 7.59 17.10% 8.67 21.89% Akmola Celinogradski 0.75 0.53% 1.49 2.19% 2.24 4.72% 2.99 8.19% 3.74 12.20% 4.48 16.71% 5.23 21.64% 5.98 26.74% Akmola Enbekshilderski 1.07 0.40% 2.15 1.63% 3.22 3.85% 4.29 6.95% 5.37 10.52% 6.44 14.68% 7.52 19.26% 8.59 24.09% Akmola Esilski 0.76 0.49% 1.52 2.00% 2.28 4.43% 3.05 7.68% 3.81 11.39% 4.57 15.74% 5.33 20.60% 6.09 25.65% Akmola Korgalzhinski 0.47 0.45% 0.94 1.85% 1.41 4.13% 1.88 7.32% 2.35 11.05% 2.83 15.43% 3.30 20.20% 3.77 25.21% Akmola Sandiktauski 1.05 0.45% 2.10 1.66% 3.15 3.88% 4.20 7.06% 5.24 10.75% 6.29 14.94% 7.34 19.62% 8.39 24.54% Akmola Shortandinski 0.88 0.43% 1.77 1.72% 2.65 3.87% 3.54 6.84% 4.42 10.38% 5.31 14.61% 6.19 19.41% 7.08 24.47% Akmola Shuchenski 1.15 0.27% 2.30 1.13% 3.46 2.64% 4.61 4.92% 5.76 7.71% 6.91 11.27% 8.06 15.48% 9.22 20.17% Akmola Yegindikolski 0.60 0.57% 1.20 2.11% 1.80 4.61% 2.40 8.03% 3.00 11.95% 3.60 16.46% 4.20 21.44% 4.80 26.60% Akmola Zerendinski 1.18 0.25% 2.36 1.13% 3.54 2.72% 4.73 5.19% 5.91 8.21% 7.09 11.92% 8.27 16.20% 9.45 20.82% Akmola Zhaksi 1.15 0.38% 2.31 1.71% 3.46 3.85% 4.62 6.80% 5.77 10.34% 6.93 14.58% 8.08 19.36% 9.24 24.36% Akmola Zharkainski 0.72 0.29% 1.45 1.29% 2.17 3.01% 2.89 5.50% 3.62 8.57% 4.34 12.37% 5.07 16.77% 5.79 21.53% Akmola Erementauski 0.83 1.10% 1.66 3.53% 2.49 6.88% 3.32 11.00% 4.15 15.46% 4.98 20.29% 5.81 25.43% 6.64 30.56% EKO Beskaragay 0.56 1.37% 1.13 4.23% 1.69 8.00% 2.25 12.55% 2.81 17.54% 3.38 22.90% 3.94 28.44% 4.50 33.87% EKO Boroduliha 0.87 0.83% 1.73 2.89% 2.60 5.95% 3.46 9.81% 4.33 14.00% 5.19 18.71% 6.06 23.75% 6.93 28.78% EKO Glubokoe 1.45 0.32% 2.91 1.35% 4.36 3.15% 5.82 5.59% 7.27 8.56% 8.73 12.26% 10.18 16.58% 11.63 21.30% EKO Kokpekti 1.02 0.21% 2.04 1.02% 3.07 2.50% 4.09 4.64% 5.11 7.24% 6.13 10.54% 7.16 14.38% 8.18 18.57% EKO Semey city 0.56 2.61% 1.12 6.51% 1.68 11.14% 2.23 16.32% 2.79 21.58% 3.35 27.15% 3.91 32.76% 4.47 38.15% EKO Shemonaiha 1.44 0.33% 2.87 1.41% 4.31 3.39% 5.75 6.17% 7.19 9.46% 8.62 13.44% 10.06 17.92% 11.50 22.68% EKO Ulan 0.88 0.43% 1.76 1.79% 2.65 4.16% 3.53 7.29% 4.41 10.87% 5.29 15.12% 6.18 19.80% 7.06 24.65% EKO Urzhar 0.92 0.36% 1.84 1.42% 2.76 3.27% 3.69 5.75% 4.61 8.63% 5.53 12.24% 6.45 16.47% 7.37 20.99% EKO Zyryan 1.47 0.21% 2.94 0.94% 4.41 2.26% 5.88 4.24% 7.35 6.73% 8.82 9.95% 10.29 13.79% 11.76 18.13% Karaganda Buharzhirau 0.59 0.83% 1.18 2.75% 1.77 5.72% 2.36 9.57% 2.95 13.89% 3.54 18.65% 4.13 23.74% 4.72 28.88% - 316 - Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Karaganda Karkaraly 0.60 0.47% 1.19 1.75% 1.79 3.85% 2.39 6.74% 2.99 10.16% 3.58 14.13% 4.18 18.57% 4.78 23.21% Karaganda Nura 0.44 0.38% 0.87 1.60% 1.31 3.64% 1.74 6.57% 2.18 10.11% 2.61 14.35% 3.05 19.07% 3.48 24.01% Karaganda Osakarov 0.66 0.48% 1.33 1.88% 1.99 4.24% 2.66 7.50% 3.32 11.24% 3.99 15.53% 4.65 20.21% 5.32 25.13% Karaganda Shetski 0.53 0.30% 1.05 1.22% 1.58 2.92% 2.11 5.48% 2.63 8.56% 3.16 12.28% 3.69 16.59% 4.21 21.22% Karaganda Ulytau 0.58 0.33% 1.16 1.35% 1.74 3.13% 2.32 5.78% 2.90 8.92% 3.48 12.80% 4.06 17.22% 4.64 22.01% Karaganda Zhanaarka 0.48 0.24% 0.95 1.05% 1.43 2.67% 1.90 5.13% 2.38 8.03% 2.86 11.68% 3.33 15.91% 3.81 20.48% Kostanay Altynsarin 1.50 0.35% 2.99 1.38% 4.49 3.23% 5.98 5.96% 7.48 9.14% 8.97 12.98% 10.47 17.40% 11.97 22.14% Kostanay Amangedi 0.77 0.32% 1.55 1.33% 2.32 3.05% 3.10 5.52% 3.87 8.51% 4.64 12.16% 5.42 16.38% 6.19 20.97% Kostanay Arkalyk city 0.62 0.67% 1.23 2.46% 1.85 5.10% 2.47 8.58% 3.08 12.50% 3.70 17.02% 4.32 21.88% 4.93 26.86% Kostanay Auliykolski 0.82 0.43% 1.65 1.64% 2.47 3.64% 3.30 6.55% 4.12 9.96% 4.95 14.06% 5.77 18.62% 6.59 23.45% Kostanay Denisovski 1.20 0.86% 2.40 2.76% 3.61 5.59% 4.81 9.26% 6.01 13.35% 7.21 18.06% 8.41 23.06% 9.61 28.23% Kostanay Fedorovski 1.62 0.19% 3.24 0.89% 4.86 2.17% 6.49 4.16% 8.11 6.74% 9.73 10.12% 11.35 14.23% 12.97 18.94% Kostanay Kamisty 0.88 1.85% 1.76 4.88% 2.64 8.53% 3.53 12.84% 4.41 17.37% 5.29 22.24% 6.17 27.26% 7.05 32.25% Kostanay Karabalyk 1.45 0.31% 2.90 1.31% 4.35 3.10% 5.80 5.63% 7.25 8.72% 8.70 12.68% 10.15 17.27% 11.60 22.22% Kostanay Karasu 1.14 0.26% 2.28 1.15% 3.42 2.68% 4.56 4.93% 5.69 7.69% 6.83 11.15% 7.97 15.24% 9.11 19.77% Kostanay Uzunkolski 1.69 0.27% 3.38 1.11% 5.06 2.67% 6.75 4.98% 8.44 7.81% 10.13 11.41% 11.81 15.58% 13.50 20.18% Kostanay Zhetikara 0.85 0.76% 1.69 2.61% 2.54 5.30% 3.38 8.78% 4.23 12.79% 5.07 17.33% 5.92 22.20% 6.76 27.28% Kostanay Kostanay 1.67 0.22% 3.34 1.05% 5.01 2.55% 6.68 4.75% 8.36 7.47% 10.03 11.03% 11.70 15.20% 13.37 19.85% Kostanay Mendikara 1.73 0.17% 3.46 0.81% 5.19 1.95% 6.92 3.78% 8.65 6.07% 10.37 9.07% 12.10 12.73% 13.83 17.02% Kostanay Nauirzym 0.88 0.45% 1.76 1.72% 2.64 3.74% 3.52 6.63% 4.40 10.09% 5.29 14.16% 6.17 18.70% 7.05 23.52% Kostanay Rudnyi city 0.46 1.08% 0.92 3.17% 1.38 6.05% 1.84 9.66% 2.30 13.57% 2.76 17.94% 3.22 22.58% 3.68 27.32% Kostanay Sarykol 1.49 0.18% 2.99 0.77% 4.48 1.89% 5.97 3.66% 7.46 5.79% 8.96 8.59% 10.45 12.09% 11.94 16.21% Kostanay Taranovski 1.20 0.50% 2.40 1.84% 3.60 4.04% 4.80 7.19% 6.01 10.81% 7.21 15.18% 8.41 20.07% 9.61 25.24% NKO Airtau 1.55 0.24% 3.10 1.03% 4.64 2.40% 6.19 4.56% 7.74 7.24% 9.29 10.60% 10.84 14.62% 12.39 19.08% NKO Akkayn 1.66 0.19% 3.32 0.84% 4.97 2.12% 6.63 4.05% 8.29 6.46% 9.95 9.53% 11.61 13.28% 13.27 17.71% NKO Akzhar 1.37 0.18% 2.74 0.75% 4.11 1.74% 5.48 3.22% 6.84 4.93% 8.21 7.27% 9.58 10.21% 10.95 13.82% NKO Esil 1.54 0.17% 3.09 0.77% 4.63 1.88% 6.18 3.62% 7.72 5.75% 9.26 8.53% 10.81 12.10% 12.35 16.35% NKO G.Musrepov 1.48 0.23% 2.96 1.01% 4.45 2.40% 5.93 4.48% 7.41 7.03% 8.89 10.27% 10.38 14.20% 11.86 18.64% NKO Kyzylzhar 1.71 0.16% 3.41 0.68% 5.12 1.56% 6.82 2.79% 8.53 4.09% 10.24 5.89% 11.94 9.43% 13.65 14.38% NKO M.Zhumabayev 1.52 0.16% 3.05 0.69% 4.57 1.60% 6.09 2.92% 7.61 4.45% 9.14 6.61% 10.66 9.41% 12.18 12.99% NKO Mamliut 1.60 0.17% 3.20 0.79% 4.79 2.01% 6.39 3.88% 7.99 6.16% 9.59 9.16% 11.18 13.03% 12.78 17.56% - 317 - Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) NKO Shalakin 1.47 0.17% 2.95 0.74% 4.42 1.80% 5.89 3.38% 7.37 5.36% 8.84 8.02% 10.31 11.40% 11.78 15.36% NKO Taiynsha 1.47 0.20% 2.94 0.88% 4.41 2.14% 5.89 4.03% 7.36 6.34% 8.83 9.42% 10.30 13.12% 11.77 17.36% NKO Timiryazev 1.49 0.19% 2.98 0.84% 4.47 2.12% 5.96 4.08% 7.45 6.58% 8.93 9.82% 10.42 13.75% 11.91 18.24% NKO Ualihanov 1.22 0.27% 2.45 1.15% 3.67 2.63% 4.89 4.86% 6.12 7.54% 7.34 10.86% 8.57 14.79% 9.79 19.17% NKO Zhambil 1.63 0.19% 3.27 0.82% 4.90 2.00% 6.53 3.85% 8.16 6.18% 9.80 9.27% 11.43 13.13% 13.06 17.60% Pavlodar Aktogaisky 0.58 0.68% 1.16 2.38% 1.73 4.94% 2.31 8.33% 2.89 12.11% 3.47 16.41% 4.05 20.99% 4.62 25.63% Pavlodar Ekibastuz city 0.91 0.65% 1.82 2.28% 2.72 4.73% 3.63 7.90% 4.54 11.37% 5.45 15.21% 6.36 19.28% 7.26 23.36% Pavlodar Irtyshski 0.70 0.36% 1.41 1.49% 2.11 3.39% 2.82 6.06% 3.52 9.19% 4.22 13.01% 4.93 17.36% 5.63 21.96% Pavlodar Kashyrski 0.87 0.54% 1.75 2.06% 2.62 4.43% 3.50 7.59% 4.37 11.18% 5.24 15.37% 6.12 19.88% 6.99 24.51% Pavlodar Shernaktinsky 0.75 0.56% 1.50 1.97% 2.24 4.04% 2.99 6.82% 3.74 10.03% 4.49 13.80% 5.24 17.91% 5.98 22.10% Pavlodar Uspenka area 0.77 1.59% 1.54 4.37% 2.31 8.00% 3.08 12.24% 3.85 16.68% 4.62 21.48% 5.39 26.39% 6.16 31.18% Pavlodar Zhelezninski 0.43 0.40% 0.87 1.64% 1.30 3.72% 1.73 6.56% 2.17 9.88% 2.60 13.90% 3.03 18.42% 3.47 23.13% - 318 - Table A7.3 Estimated Commercial Premium Rates for target loss ratios equal to 70% and Guaranteed Yields Area Yield Index Insurance for Spring Wheat produced by Agribusiness Enterprises in North Kazakhstan Region Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Akmola Akkol 0.88 0.28% 1.77 1.25% 2.65 2.97% 3.54 5.45% 4.42 8.43% 5.31 12.18% 6.19 16.52% 7.08 21.23% Akmola Arshaly 0.95 0.18% 1.91 0.78% 2.86 1.82% 3.82 3.32% 4.77 5.11% 5.72 7.46% 6.68 10.36% 7.63 13.94% Akmola Astrashanski 0.76 0.19% 1.51 0.90% 2.27 2.11% 3.02 4.10% 3.78 6.57% 4.53 9.86% 5.29 13.90% 6.05 18.41% Akmola Atbasarski 0.92 0.19% 1.84 0.89% 2.77 2.16% 3.69 4.19% 4.61 6.71% 5.53 9.96% 6.45 13.94% 7.37 18.41% Akmola Bulandinski 1.04 0.19% 2.08 0.82% 3.12 1.94% 4.16 3.71% 5.20 5.93% 6.24 8.88% 7.28 12.60% 8.32 16.92% Akmola Celinogradski 0.70 0.17% 1.40 0.76% 2.10 1.86% 2.79 3.57% 3.49 5.73% 4.19 8.67% 4.89 12.28% 5.59 16.55% Akmola Enbekshilderski 1.21 0.23% 2.43 1.13% 3.64 2.77% 4.85 5.07% 6.07 7.80% 7.28 11.26% 8.50 15.37% 9.71 19.92% Akmola Esilski 0.79 0.18% 1.59 0.81% 2.38 1.99% 3.18 3.81% 3.97 6.10% 4.76 9.12% 5.56 12.87% 6.35 17.20% Akmola Korgalzhinski 0.58 0.17% 1.16 0.76% 1.74 1.90% 2.32 3.67% 2.90 5.91% 3.48 8.87% 4.06 12.55% 4.64 16.81% Akmola Sandiktauski 1.20 0.16% 2.39 0.70% 3.59 1.64% 4.78 3.12% 5.98 4.94% 7.18 7.43% 8.37 10.62% 9.57 14.56% Akmola Shortandinski 0.87 0.19% 1.75 0.88% 2.62 2.18% 3.49 4.16% 4.37 6.57% 5.24 9.80% 6.12 13.69% 6.99 18.10% Akmola Shuchenski 1.12 0.18% 2.24 0.75% 3.36 1.73% 4.48 3.28% 5.59 5.18% 6.71 7.73% 7.83 11.03% 8.95 15.00% Akmola Yegindikolski 0.72 0.18% 1.45 0.89% 2.17 2.18% 2.89 4.20% 3.61 6.72% 4.34 9.97% 5.06 13.94% 5.78 18.43% Akmola Zerendinski 1.41 0.17% 2.82 0.71% 4.23 1.68% 5.64 3.19% 7.05 5.13% 8.46 7.76% 9.87 11.09% 11.28 15.09% Akmola Zhaksi 1.18 0.16% 2.36 0.75% 3.54 1.88% 4.72 3.67% 5.90 5.88% 7.08 8.81% 8.26 12.44% 9.44 16.63% Akmola Zharkainski 0.86 0.18% 1.72 0.84% 2.59 2.07% 3.45 4.06% 4.31 6.52% 5.17 9.78% 6.04 13.75% 6.90 18.31% Akmola Erementauski 0.74 0.30% 1.49 1.47% 2.23 3.54% 2.97 6.38% 3.71 9.71% 4.46 13.63% 5.20 18.10% 5.94 22.89% EKO Beskaragay 0.60 1.95% 1.20 5.33% 1.80 9.43% 2.40 13.97% 3.01 18.66% 3.61 23.64% 4.21 28.61% 4.81 33.47% EKO Boroduliha 1.11 0.84% 2.22 2.37% 3.33 4.57% 4.44 7.41% 5.55 10.56% 6.65 14.27% 7.76 18.52% 8.87 23.03% EKO Glubokoe 1.37 0.55% 2.75 1.64% 4.12 3.39% 5.49 5.87% 6.86 8.81% 8.24 12.51% 9.61 16.80% 10.98 21.52% EKO Kokpekti 0.77 0.80% 1.54 2.10% 2.32 4.02% 3.09 6.53% 3.86 9.42% 4.63 12.89% 5.40 16.81% 6.18 21.05% EKO Semey city 1.58 0.47% 3.16 1.46% 4.74 3.07% 6.32 5.33% 7.89 8.05% 9.47 11.42% 11.05 15.36% 12.63 19.71% EKO Shemonaiha 1.18 3.62% 2.37 7.68% 3.55 12.29% 4.74 17.20% 5.92 22.12% 7.11 27.24% 8.29 32.36% 9.48 37.32% EKO Ulan 1.41 0.51% 2.83 1.48% 4.24 3.03% 5.65 5.19% 7.06 7.81% 8.48 11.06% 9.89 14.91% 11.30 19.27% EKO Urzhar 0.83 0.17% 1.66 0.72% 2.49 1.80% 3.33 3.54% 4.16 5.79% 4.99 8.72% 5.82 12.34% 6.65 16.55% EKO Zyryan 0.85 0.19% 1.70 0.88% 2.55 2.26% 3.39 4.38% 4.24 7.00% 5.09 10.29% 5.94 14.21% 6.79 18.73% Karaganda Buharzhirau 0.79 0.18% 1.58 0.82% 2.37 2.04% 3.16 3.92% 3.94 6.29% 4.73 9.29% 5.52 12.89% 6.31 17.06% Karaganda Karkaraly 0.62 0.25% 1.23 1.07% 1.85 2.66% 2.47 5.04% 3.08 7.95% 3.70 11.58% 4.32 15.79% 4.93 20.47% Karaganda Nura 1.33 0.18% 2.67 0.85% 4.00 2.08% 5.33 4.04% 6.67 6.50% 8.00 9.73% 9.33 13.58% 10.67 17.96% - 319 - Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Karaganda Osakarov 0.72 0.21% 1.45 0.99% 2.17 2.44% 2.89 4.59% 3.61 7.28% 4.34 10.69% 5.06 14.80% 5.78 19.39% Karaganda Shetski 0.92 0.18% 1.83 0.82% 2.75 2.06% 3.66 4.02% 4.58 6.49% 5.49 9.68% 6.41 13.59% 7.32 18.08% Karaganda Ulytau 0.99 0.23% 1.98 1.05% 2.97 2.55% 3.97 4.84% 4.96 7.67% 5.95 11.23% 6.94 15.40% 7.93 20.03% Karaganda Zhanaarka 1.19 0.18% 2.39 0.90% 3.58 2.24% 4.78 4.27% 5.97 6.78% 7.16 10.11% 8.36 14.12% 9.55 18.67% Kostanay Altynsarin 1.64 0.16% 3.29 0.73% 4.93 1.83% 6.57 3.50% 8.21 5.54% 9.86 8.37% 11.50 11.89% 13.14 16.02% Kostanay Amangedi 0.92 0.19% 1.84 0.89% 2.75 2.19% 3.67 4.16% 4.59 6.64% 5.51 9.87% 6.42 13.71% 7.34 18.02% Kostanay Arkalyk city 1.47 0.17% 2.94 0.76% 4.41 1.88% 5.89 3.56% 7.36 5.67% 8.83 8.52% 10.30 12.07% 11.77 16.25% Kostanay Auliykolski 1.08 0.21% 2.16 0.95% 3.24 2.27% 4.32 4.28% 5.40 6.81% 6.48 10.02% 7.56 13.85% 8.64 18.21% Kostanay Denisovski 1.40 0.16% 2.80 0.69% 4.20 1.69% 5.59 3.20% 6.99 5.00% 8.39 7.45% 9.79 10.54% 11.19 14.31% Kostanay Fedorovski 0.84 0.20% 1.68 0.92% 2.52 2.18% 3.36 4.15% 4.20 6.57% 5.04 9.82% 5.88 13.72% 6.72 18.12% Kostanay Kamisty 1.27 0.19% 2.54 0.81% 3.82 2.03% 5.09 3.95% 6.36 6.36% 7.63 9.49% 8.91 13.30% 10.18 17.64% Kostanay Karabalyk 1.40 0.16% 2.81 0.69% 4.21 1.68% 5.62 3.28% 7.02 5.25% 8.43 7.91% 9.83 11.32% 11.24 15.45% Kostanay Karasu 0.68 0.23% 1.36 1.03% 2.04 2.57% 2.71 4.84% 3.39 7.65% 4.07 11.21% 4.75 15.46% 5.43 20.15% Kostanay Uzunkolski 1.46 0.18% 2.92 0.77% 4.38 1.88% 5.84 3.49% 7.31 5.51% 8.77 8.27% 10.23 11.72% 11.69 15.76% Kostanay Zhetikara 1.13 0.18% 2.25 0.84% 3.38 2.11% 4.51 4.08% 5.64 6.59% 6.76 9.83% 7.89 13.78% 9.02 18.23% Kostanay Kostanay 1.32 0.16% 2.63 0.66% 3.95 1.51% 5.26 2.76% 6.58 4.23% 7.89 6.32% 9.21 9.07% 10.52 12.52% Kostanay Mendikara 1.44 0.16% 2.89 0.65% 4.33 1.46% 5.78 2.62% 7.22 3.92% 8.67 5.72% 10.11 8.15% 11.55 11.33% Kostanay Nauirzym 1.10 0.16% 2.20 0.70% 3.30 1.61% 4.40 2.92% 5.50 4.43% 6.60 6.52% 7.70 9.23% 8.80 12.62% Kostanay Rudnyi city 1.38 0.16% 2.75 0.67% 4.13 1.52% 5.50 2.72% 6.88 4.06% 8.26 5.88% 9.63 8.39% 11.01 11.58% Kostanay Sarykol 1.32 0.17% 2.63 0.72% 3.95 1.68% 5.27 3.13% 6.58 4.93% 7.90 7.36% 9.21 10.43% 10.53 14.12% Kostanay Taranovski 1.43 0.16% 2.86 0.65% 4.29 1.45% 5.72 2.53% 7.15 3.62% 8.58 4.94% 10.02 6.66% 11.45 9.21% NKO Airtau 1.37 0.16% 2.73 0.65% 4.10 1.46% 5.46 2.62% 6.83 3.89% 8.19 5.63% 9.56 7.97% 10.92 11.02% NKO Akkayn 1.37 0.16% 2.74 0.67% 4.11 1.53% 5.48 2.77% 6.84 4.17% 8.21 6.13% 9.58 8.78% 10.95 12.18% NKO Akzhar 1.29 0.18% 2.58 0.73% 3.87 1.73% 5.16 3.26% 6.46 5.10% 7.75 7.65% 9.04 10.95% 10.33 14.91% NKO Esil 1.33 0.16% 2.65 0.68% 3.98 1.56% 5.31 2.83% 6.63 4.30% 7.96 6.34% 9.29 9.02% 10.62 12.37% NKO G.Musrepov 1.24 0.16% 2.47 0.67% 3.71 1.63% 4.94 3.04% 6.18 4.80% 7.41 7.26% 8.65 10.39% 9.89 14.15% NKO Kyzylzhar 1.13 0.18% 2.27 0.77% 3.40 1.81% 4.54 3.36% 5.67 5.21% 6.81 7.69% 7.94 10.90% 9.08 14.75% NKO M.Zhumabayev 1.16 0.16% 2.31 0.71% 3.47 1.69% 4.62 3.20% 5.78 5.01% 6.93 7.50% 8.09 10.66% 9.24 14.45% NKO Mamliut 0.52 0.24% 1.05 1.09% 1.57 2.62% 2.09 4.78% 2.62 7.26% 3.14 10.18% 3.66 13.46% 4.19 16.87% NKO Shalakin 0.67 0.18% 1.35 0.81% 2.02 1.97% 2.69 3.78% 3.37 5.89% 4.04 8.69% 4.72 12.15% 5.39 16.10% NKO Taiynsha 0.82 0.25% 1.63 1.17% 2.45 2.96% 3.26 5.68% 4.08 8.90% 4.89 12.94% 5.71 17.66% 6.52 22.75% - 320 - Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) NKO Timiryazev 0.76 0.28% 1.51 1.21% 2.27 2.77% 3.02 4.99% 3.78 7.46% 4.53 10.35% 5.29 13.48% 6.04 16.65% NKO Ualihanov 0.67 0.26% 1.35 1.16% 2.02 2.64% 2.69 4.72% 3.37 7.04% 4.04 9.80% 4.71 12.83% 5.39 15.92% NKO Zhambil 0.90 0.22% 1.80 1.04% 2.70 2.63% 3.60 5.09% 4.49 8.01% 5.39 11.76% 6.29 16.23% 7.19 21.16% Pavlodar Aktogaisky 0.88 0.28% 1.77 1.25% 2.65 2.97% 3.54 5.45% 4.42 8.43% 5.31 12.18% 6.19 16.52% 7.08 21.23% Pavlodar Ekibastuz city 0.95 0.18% 1.91 0.78% 2.86 1.82% 3.82 3.32% 4.77 5.11% 5.72 7.46% 6.68 10.36% 7.63 13.94% Pavlodar Irtyshski 0.76 0.19% 1.51 0.90% 2.27 2.11% 3.02 4.10% 3.78 6.57% 4.53 9.86% 5.29 13.90% 6.05 18.41% Pavlodar Kashyrski 0.92 0.19% 1.84 0.89% 2.77 2.16% 3.69 4.19% 4.61 6.71% 5.53 9.96% 6.45 13.94% 7.37 18.41% Pavlodar Shernaktinsky 1.04 0.19% 2.08 0.82% 3.12 1.94% 4.16 3.71% 5.20 5.93% 6.24 8.88% 7.28 12.60% 8.32 16.92% Pavlodar Uspenka area 0.70 0.17% 1.40 0.76% 2.10 1.86% 2.79 3.57% 3.49 5.73% 4.19 8.67% 4.89 12.28% 5.59 16.55% Pavlodar Zhelezninski 1.21 0.23% 2.43 1.13% 3.64 2.77% 4.85 5.07% 6.07 7.80% 7.28 11.26% 8.50 15.37% 9.71 19.92% - 321 - Table A7.4 Estimated Commercial Premium Rates for target loss ratios equal to 70% and Guaranteed Yields Area Yield Index Insurance for Spring Wheat produced by Commercial Farmers and Agribusiness Enterprises in North Kazakhstan Region Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Akmola Akkol 0.87 0.32% 1.73 1.40% 2.60 3.28% 3.46 5.94% 4.33 9.09% 5.20 12.98% 6.06 17.42% 6.93 22.19% Akmola Arshaly 0.99 0.22% 1.98 0.95% 2.97 2.22% 3.97 4.08% 4.96 6.29% 5.95 9.11% 6.94 12.52% 7.93 16.51% Akmola Astrashanski 0.70 0.30% 1.40 1.28% 2.09 2.90% 2.79 5.34% 3.49 8.30% 4.19 12.02% 4.89 16.38% 5.58 21.10% Akmola Atbasarski 0.92 0.22% 1.85 1.03% 2.77 2.45% 3.69 4.65% 4.61 7.34% 5.54 10.75% 6.46 14.86% 7.38 19.42% Akmola Bulandinski 1.04 0.20% 2.09 0.87% 3.13 2.06% 4.18 3.93% 5.22 6.24% 6.27 9.29% 7.31 13.10% 8.36 17.47% Akmola Celinogradski 0.71 0.22% 1.41 0.99% 2.12 2.32% 2.82 4.30% 3.53 6.76% 4.24 9.95% 4.94 13.77% 5.65 18.18% Akmola Enbekshilderski 1.19 0.26% 2.38 1.21% 3.56 2.95% 4.75 5.38% 5.94 8.24% 7.13 11.82% 8.32 16.01% 9.51 20.61% Akmola Esilski 0.79 0.25% 1.57 1.07% 2.36 2.51% 3.15 4.64% 3.93 7.24% 4.72 10.54% 5.51 14.53% 6.29 19.01% Akmola Korgalzhinski 0.56 0.21% 1.12 0.92% 1.68 2.23% 2.24 4.22% 2.80 6.68% 3.36 9.85% 3.92 13.70% 4.48 18.07% Akmola Sandiktauski 1.17 0.21% 2.34 0.84% 3.52 1.97% 4.69 3.69% 5.86 5.78% 7.03 8.51% 8.21 11.93% 9.38 16.01% Akmola Shortandinski 0.87 0.21% 1.75 0.94% 2.62 2.30% 3.50 4.35% 4.37 6.84% 5.25 10.14% 6.12 14.09% 7.00 18.55% Akmola Shuchenski 1.12 0.19% 2.25 0.80% 3.37 1.86% 4.49 3.50% 5.62 5.52% 6.74 8.21% 7.86 11.64% 8.99 15.70% Akmola Yegindikolski 0.68 0.30% 1.36 1.28% 2.03 2.97% 2.71 5.45% 3.39 8.42% 4.07 12.08% 4.74 16.38% 5.42 21.09% Akmola Zerendinski 1.36 0.19% 2.72 0.79% 4.08 1.88% 5.43 3.58% 6.79 5.72% 8.15 8.56% 9.51 12.08% 10.87 16.20% Akmola Zhaksi 1.17 0.21% 2.34 0.96% 3.50 2.31% 4.67 4.35% 5.84 6.85% 7.01 10.70% 8.17 13.95% 9.34 18.69% Akmola Zharkainski 0.81 0.22% 1.62 1.00% 2.42 2.40% 3.23 4.57% 4.04 7.24% 4.85 10.70% 5.65 14.82% 6.46 19.45% Akmola Erementauski 0.77 0.52% 1.53 2.05% 2.30 4.48% 3.06 7.67% 3.83 11.32% 4.59 15.50% 5.36 20.16% 6.12 25.04% EKO Beskaragay 0.58 1.64% 1.16 4.74% 1.74 8.66% 2.32 13.20% 2.90 18.06% 3.48 23.24% 4.06 28.51% 4.64 33.69% EKO Boroduliha 0.96 0.83% 1.91 2.67% 2.87 5.36% 3.82 8.78% 4.78 12.52% 5.74 16.80% 6.69 21.50% 7.65 26.31% EKO Glubokoe 1.42 0.43% 2.83 1.48% 4.25 3.26% 5.66 5.72% 7.08 8.67% 8.50 12.37% 9.91 16.68% 11.33 21.40% EKO Kokpekti 0.90 0.47% 1.79 1.49% 2.69 3.16% 3.58 5.46% 4.48 8.18% 5.38 11.56% 6.27 15.44% 7.17 19.64% EKO Semey city 0.56 2.61% 1.12 6.51% 1.68 11.14% 2.23 16.32% 2.79 21.58% 3.35 27.15% 3.91 32.76% 4.47 38.15% EKO Shemonaiha 1.52 0.42% 3.05 1.44% 4.57 3.19% 6.10 5.63% 7.62 8.56% 9.15 12.15% 10.67 16.28% 12.20 20.78% EKO Ulan 0.98 1.68% 1.96 4.10% 2.94 7.34% 3.92 11.17% 4.90 15.27% 5.88 19.87% 6.86 24.72% 7.84 29.61% EKO Urzhar 0.92 0.36% 1.84 1.42% 2.76 3.27% 3.69 5.75% 4.61 8.63% 5.53 12.24% 6.45 16.47% 7.37 20.99% EKO Zyryan 1.43 0.41% 2.86 1.30% 4.29 2.77% 5.72 4.88% 7.16 7.45% 8.59 10.69% 10.02 14.54% 11.45 18.89% Karaganda Buharzhirau 0.70 0.48% 1.39 1.69% 2.09 3.66% 2.79 6.40% 3.48 9.63% 4.18 13.43% 4.87 17.75% 5.57 22.41% Karaganda Karkaraly 0.60 0.47% 1.19 1.75% 1.79 3.85% 2.39 6.74% 2.99 10.16% 3.58 14.13% 4.18 18.57% 4.78 23.21% Karaganda Nura 0.72 0.23% 1.43 1.02% 2.15 2.53% 2.87 4.80% 3.59 7.60% 4.30 11.07% 5.02 15.15% 5.74 19.75% - 322 - Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) Karaganda Osakarov 0.72 0.34% 1.44 1.38% 2.15 3.20% 2.87 5.81% 3.59 8.90% 4.31 12.58% 5.03 16.75% 5.74 21.31% Karaganda Shetski 0.53 0.30% 1.05 1.22% 1.58 2.92% 2.11 5.48% 2.63 8.56% 3.16 12.28% 3.69 16.59% 4.21 21.22% Karaganda Ulytau 0.60 0.30% 1.19 1.22% 1.79 2.92% 2.39 5.44% 2.98 8.48% 3.58 12.25% 4.17 16.57% 4.77 21.31% Karaganda Zhanaarka 0.48 0.24% 0.95 1.05% 1.43 2.67% 1.90 5.13% 2.38 8.03% 2.86 11.68% 3.33 15.91% 3.81 20.48% Kostanay Altynsarin 1.37 0.22% 2.74 0.98% 4.11 2.35% 5.48 4.50% 6.84 7.14% 8.21 10.51% 9.58 14.49% 10.95 18.96% Kostanay Amangedi 0.75 0.27% 1.50 1.19% 2.26 2.80% 3.01 5.14% 3.76 8.01% 4.51 11.56% 5.27 15.73% 6.02 20.32% Kostanay Arkalyk city 0.80 0.33% 1.60 1.31% 2.40 2.96% 3.20 5.38% 4.00 8.27% 4.80 11.86% 5.60 16.05% 6.40 20.69% Kostanay Auliykolski 0.95 0.28% 1.89 1.19% 2.84 2.80% 3.79 5.24% 4.73 8.20% 5.68 11.89% 6.63 16.15% 7.57 20.82% Kostanay Denisovski 1.20 0.29% 2.39 1.20% 3.59 2.79% 4.78 5.09% 5.98 7.86% 7.17 11.42% 8.37 15.59% 9.56 20.24% Kostanay Fedorovski 1.63 0.18% 3.27 0.80% 4.90 1.98% 6.53 3.79% 8.17 6.07% 9.80 9.14% 11.43 12.92% 13.07 17.30% Kostanay Kamisty 0.91 0.48% 1.82 1.58% 2.73 3.29% 3.64 5.67% 4.56 8.51% 5.47 12.02% 6.38 16.07% 7.29 20.50% Kostanay Karabalyk 1.47 0.20% 2.93 0.88% 4.40 2.14% 5.87 4.00% 7.33 6.31% 8.80 9.39% 10.27 13.17% 11.74 17.50% Kostanay Karasu 1.10 0.23% 2.20 1.01% 3.30 2.40% 4.39 4.49% 5.49 7.10% 6.59 10.39% 7.69 14.30% 8.79 18.72% Kostanay Uzunkolski 1.47 0.19% 2.93 0.81% 4.40 1.95% 5.87 3.68% 7.33 5.76% 8.80 8.52% 10.26 11.90% 11.73 15.89% Kostanay Zhetikara 0.84 0.29% 1.68 1.20% 2.52 2.69% 3.36 4.90% 4.20 7.58% 5.04 11.04% 5.88 15.10% 6.72 19.61% Kostanay Kostanay 1.49 0.21% 2.97 0.96% 4.46 2.34% 5.94 4.44% 7.43 7.03% 8.92 10.42% 10.40 14.45% 11.89 18.97% Kostanay Mendikara 1.52 0.17% 3.04 0.74% 4.56 1.79% 6.08 3.48% 7.60 5.58% 9.12 8.38% 10.64 11.89% 12.16 16.09% Kostanay Nauirzym 0.79 0.37% 1.59 1.46% 2.38 3.31% 3.18 5.98% 3.97 9.19% 4.77 13.08% 5.56 17.51% 6.35 22.28% Kostanay Rudnyi city 0.46 1.08% 0.92 3.17% 1.38 6.05% 1.84 9.66% 2.30 13.57% 2.76 17.94% 3.22 22.58% 3.68 27.32% Kostanay Sarykol 1.47 0.18% 2.95 0.77% 4.42 1.88% 5.89 3.55% 7.36 5.62% 8.84 8.39% 10.31 11.85% 11.78 15.92% Kostanay Taranovski 1.15 0.30% 2.30 1.19% 3.46 2.78% 4.61 5.15% 5.76 8.05% 6.91 11.68% 8.06 15.95% 9.21 20.65% NKO Airtau 1.36 0.18% 2.71 0.73% 4.07 1.69% 5.43 3.12% 6.78 4.84% 8.14 7.18% 9.49 10.19% 10.85 13.85% NKO Akkayn 1.49 0.17% 2.99 0.70% 4.48 1.63% 5.98 2.99% 7.47 4.58% 8.97 6.71% 10.46 9.49% 11.95 12.98% NKO Akzhar 1.19 0.17% 2.38 0.72% 3.57 1.66% 4.77 3.04% 5.96 4.63% 7.15 6.81% 8.34 9.61% 9.53 13.09% NKO Esil 1.40 0.16% 2.80 0.68% 4.20 1.58% 5.60 2.86% 7.00 4.33% 8.40 6.30% 9.80 8.97% 11.20 12.33% NKO G.Musrepov 1.34 0.18% 2.69 0.77% 4.03 1.82% 5.38 3.38% 6.72 5.32% 8.07 7.90% 9.41 11.13% 10.75 14.95% NKO Kyzylzhar 1.51 0.16% 3.01 0.66% 4.52 1.48% 6.02 2.61% 7.53 3.76% 9.04 5.23% 10.54 7.52% 12.05 10.81% NKO M.Zhumabayev 1.42 0.16% 2.85 0.67% 4.27 1.51% 5.70 2.74% 7.12 4.12% 8.55 6.02% 9.97 8.55% 11.40 11.81% NKO Mamliut 1.42 0.16% 2.85 0.70% 4.27 1.66% 5.69 3.07% 7.12 4.70% 8.54 6.93% 9.96 9.92% 11.39 13.62% NKO Shalakin 1.33 0.18% 2.66 0.74% 4.00 1.75% 5.33 3.29% 6.66 5.17% 7.99 7.74% 9.32 11.06% 10.66 15.02% NKO Taiynsha 1.35 0.17% 2.70 0.72% 4.05 1.67% 5.41 3.05% 6.76 4.67% 8.11 6.90% 9.46 9.77% 10.81 13.29% - 323 - Coverage Level 10% 20% 30% 40% 50 % 60% 70% 80% Oblast Rayon Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Insured Coml. Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate Yield Rate (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) (ctr/he) (% TSI) NKO Timiryazev 1.35 0.18% 2.70 0.76% 4.05 1.87% 5.40 3.56% 6.75 5.69% 8.10 8.54% 9.45 12.07% 10.81 16.19% NKO Ualihanov 1.15 0.19% 2.30 0.83% 3.44 1.93% 4.59 3.59% 5.74 5.56% 6.89 8.18% 8.03 11.49% 9.18 15.42% NKO Zhambil 1.32 0.17% 2.63 0.75% 3.95 1.82% 5.26 3.47% 6.58 5.50% 7.89 8.24% 9.21 11.69% 10.53 15.76% Pavlodar Aktogaisky 0.58 0.68% 1.16 2.38% 1.73 4.94% 2.31 8.33% 2.89 12.11% 3.47 16.41% 4.05 20.99% 4.62 25.63% Pavlodar Ekibastuz city 0.74 0.52% 1.47 1.91% 2.21 4.06% 2.95 6.91% 3.68 10.07% 4.42 13.62% 5.16 17.44% 5.89 21.31% Pavlodar Irtyshski 0.69 0.29% 1.38 1.24% 2.08 2.86% 2.77 5.21% 3.46 7.96% 4.15 11.40% 4.85 15.42% 5.54 19.77% Pavlodar Kashyrski 0.85 0.44% 1.71 1.76% 2.56 3.94% 3.41 6.94% 4.27 10.41% 5.12 14.55% 5.97 19.13% 6.83 23.92% Pavlodar Shernaktinsky 0.75 0.34% 1.51 1.36% 2.26 3.03% 3.02 5.36% 3.77 7.98% 4.52 11.05% 5.28 14.38% 6.03 17.75% Pavlodar Uspenka area 0.72 0.88% 1.43 2.65% 2.15 5.13% 2.86 8.21% 3.58 11.52% 4.29 15.22% 5.01 19.12% 5.72 23.00% Pavlodar Zhelezninski 0.61 0.30% 1.23 1.30% 1.84 3.10% 2.46 5.73% 3.07 8.82% 3.68 12.69% 4.30 17.18% 4.91 22.01% - 324 - Annex 8: Weather Index Insurance Opportunities in Kazakhstan 1. This annex provides an overview of general features Weather Index Insurance (WII). In the following Appendixes some contract prototypes for addressing drought risk in spring wheat production in the North of Kazakhstan are presented. The contract structures illustrated are indicative and further research and design work would be required to develop the actual marketable products. Features of Weather Index Insurance108 2. The essential feature of weather index-based insurance (WII) is that the insurance contract responds to an objective parameter (e.g., measurement of rainfall or temperature) at a defined weather station during an agreed-upon time period. The parameters of the contract are set so as to correlate, as accurately as possible, with the loss of a specific crop type suffered by the policyholder. All policyholders within a defined area receive payouts based on the same contract and measurement at the same station, eliminating the need for field loss assessment. 3. WII is best suited to weather hazards that are well-correlated over a widespread area and where there is a close correlation between weather and crop yield. The strongest relationships typically involve a single crop, a marked rainy season, and no irrigation. To date, most WII efforts have focused on the risk of rainfall deficit (drought). All these conditions apply to the case of spring wheat production in the North of Kazakhstan. 4. WII is less useful where more complex conditions exist. Localized risks, such as hail, or where microclimates exist (for example, in mountainous areas) are not suitable for WII. Similarly, the scope for WII is limited where crop production is impacted by many or complex causes of loss (as may be the case in humid subtropics) or where pest and disease are major influences on yields. For a given environment, other insurance products may be more appropriate (such as AYII or named-peril crop insurance). 5. Introducing WII to an area requires willing stakeholders: insurers, national weather services, and linkages for distribution and support, including financial service providers, agri- chain participants, and government, which provides the regulatory environment. WII is best introduced using market-based principles and business practices, but often with an important developmental and social agenda. As such, private and public sector partnerships are common. There are many opportunities for technical and organizational innovation in WII. 6. The strengths and weaknesses of WII are well documented in the literature.109 Box A8.1 provides a summary of the main advantages and disadvantages of WII. 108 Sections 2 to 17 have been adapted from Weather Index-based Insurance in Agricultural Development: A Technical Guide, IFAD, Rome, 2011 109 Among others see IFAD, WFP (2010); World Bank (2005); and USAID (2006). - 325 - Box A8.1: Advantages and Disadvantages of Weather Index Insurance Advantages  Transparency: index insurance contracts allow the policyholder direct access to the information on which the payouts will be calculated. Trust is strengthened by transparency.  No on-farm loss adjustment: This is a primary advantage of index insurance, as on-farm loss adjustment is quite complex and costly, and it may not be credible in many low-income countries.  Lack of adverse selection: Adverse selection occurs when potential insured parties have hidden information about their risk exposure that is not available to the insurer, who then becomes more likely to erroneously assess the risk of the insured. Traditional insurance encourages high-risk producers to insure, whereas risk (and premium) is calculated on the average producer. Index insurance requires that all insured farmers within the defined area have the same insurance payout conditions, regardless of their specific risk exposure. Hence, insurers and clients benefit from reduced adverse selection.  Lack of moral hazard: Moral hazard occurs when individuals engage in hidden activities that increase their exposure to risk as a result of purchasing insurance, or attempt to influence the claims outcome. These hidden activities can leave the insurer exposed to higher levels of risk than had been anticipated when premium rates were established. With WII, there is no benefit in individual producers trying to influence claims. All producers in the defined area are treated equally.  Addresses correlated risks: Index products work best where there are correlated risks. Perils such as drought are challenging to insure under traditional products.  Low operational and transaction costs: Index insurance requires limited individual underwriting (client assessment). It can be distributed, and claims can be settled, at relatively lower cost. Education in the product remains important, both prior to product launch and ongoing.  Rapid payout: Measurement of weather station data, with no field loss adjustment, allows rapid payouts to be made. Disadvantages  Basis Risk: Basis risk in WII is a key constraint. Basis risk is the difference between the loss experienced by the farmer and the payout triggered. It could result in a farmer experiencing yield loss, but not receiving a payout or also in a payout being triggered without any loss being experienced. Index insurance works best where losses are homogenous in the defined area, and highly correlated with the indexed peril. Basis risk can arise from: o Spatial basis risk: Local variations in the peril occurrence (e.g., rainfall) within the area surrounding a weather station. o Temporal basis risk: Inter-annual variations in seasonal crop phases, meaning that the insurance phases are not aligned in time with the intended crop growth stage. o Product basis risk: Crop losses can be caused by many factors. Where there is not a clear-cut relationship between loss and the indexed weather peril, basis risk can be high. WII is most likely to work for rainfed crops, and at severe levels of the event, when losses may be more widespread and homogenous. - 326 -  Limited perils: WII normally only covers one, and sometimes two, weather perils. Although reducing cost compared to MPCI, the product may not provide broad enough coverage to satisfy risk management needs.  Replication: The triggers, limits, and increments of a specific product need to be adjusted to reflect the weather parameters of each weather station. Different product designs are required for different crop types (or at least generic crop types). WII requires considerable technical work to implement and be sustained.  Technical capacity and expertise is required, particularly during the initial design phase for new products, in agro-meteorology and in operationalizing the products.  Lack of weather data: WII depends on the availability and quality of weather data, which can drastically vary from country to country. In developing countries, the shortage of historical and real-time weather data is often a major hurdle. Levels of implementation of WII and International Experience 7. WII can be introduced at different levels. There are different implementation models that can be used to benefit the desired target group and classification is usually carried out according to who the actual policyholder is. At the micro level, the policyholders (the insurer‘s customer) are farmers, households, or small business owners who purchase insurance to protect themselves from potential losses caused by adverse weather events. Micro policies can also be distributed to farmers by organizations such as financial service providers, farmer associations, input suppliers, processors, or NGOs.110 In addition to having wider outreach to the target group than most insurers, these intermediaries also have vested social or commercial interests in protecting themselves and their smallholder clients against the weather risk. For example, insuring the farmers can help financial service providers, input suppliers, and other intermediaries manage their risks of default by farmers. This in turn can help to unlock development opportunities for poor smallholders, such as access to credit, or higher quality inputs. At the meso level, the above mentioned organizations can act as the policyholder. At this level, WII can be structured through a policy issued to the organization, but with payout rules which could either directly or indirectly benefit farmers– for example to alleviate mass loan defaults in a microfinance institution (MFI). Finally, index insurance can also be sold at the macro level, to aid governments and relief agencies in development and disaster management. 8. The majority of WII experience has been with micro-level applications and rainfall deficit (drought). To date, many initiatives have been piloted, but only in India has a market- based scale-up of WII taken place. Table 8A.1 provides a synoptic summary of the countries in which WII has been piloted. 110 Intermediaries may be subject to regulatory approval, and the potential regulatory implication of such a scheme should be carefully assessed. - 327 - Table A8.1. International Experience with Weather Index Insurance at Different Levels of Aggregation Micro level Weather-indexed insurance for smallholder farmers: Examples: India, Nicaragua, Malawi, Ukraine, Thailand, Ethiopia, Kenya, Ghana, The Philippines, China. Over 30 projects in about 25 countries. Scale-up only in India Meso level Weather-indexed portfolio hedge for rural financial institutions that lend to poor farmers Examples: Peru, Ghana, Vietnam (under development) Programs are too new to assess scale-up and sustainability Macro level Weather insurance or weather-indexed contingent credit line for governments or international organizations Examples: Ethiopia, Malawi, Mexico (both AYII and WII), Caribbean States (CCRIF) risk pool for hurricanes & earthquake) Mexico has achieved major scale-up across most states in the past decade. CCRIF is insuring 16 Caribbean states Source: Dick, W. (2009) Availability of Weather Data and Infrastructure 9. WII relies on historical and current weather data. Historical data is used as the basis for data analysis in product design and product pricing. Current data—as measured by local weather stations—provides the information needed in the operational phase. Unfortunately, both historical and current data are not always plentiful. The completeness of the historical dataset is highly variable for different areas, particularly for daily data, which is needed for index design. Likewise, the density of weather stations forming the national network (which could include WMO reporting stations, national core stations, and localized rain gauges) varies considerably from country to country. During operation, network density will likely need to be enhanced through installation of automatic weather stations, with dataloggers and daily reporting to headquarters via telecommunication networks. Where there is no previous station data, methodology to generate artificial datasets for new station locations can sometimes be used. 10. The National Meteorological Service (NMS) supervises data and related information on past and present weather recording. NMS can normally provide a summary of data availability and history, of current and past networks, and the terms under which access to the data may be permitted. Obtaining this information may involve fees, as many NMS are required to generate income. The active involvement of NMS in any ongoing project is critical not only for access to data and knowledge on weather patterns and risks, but also for capacity building (especially in relation to agro-meteorology). Agricultural research stations may also have analysed meteorological data and be well-informed. 11. The data used to construct the underlying weather indices should adhere to quality requirements, including: - 328 -  Reliable and trustworthy on-going daily collection and reporting procedures  Periodic checks and quality control  Independent source of data for verification (e.g., surrounding weather stations, WMO Global Telecommunication Network). Areas without access to weather data that satisfy the above criteria or areas with poor spatial coverage may not able to benefit from weather risk management products, and another pilot zone should be considered. 12. The following list provides general criteria for weather data requirements for WII applications:111  At least 20 years of historic weather data  Limited missing values and out-of-range values (preferably less than 3 per cent missing data from the entire historical data set)  Availability of a nearby station for fall-back verification purposes  Consistency of observation techniques: manual versus automated  Limited changes of instrumentation / orientation / configuration  Integrity of weather data recording procedure  Little potential for measurement tampering 13. Beyond the quality of data, it is critical to define the boundaries of the area(s) covered by the weather station(s) so that WII contracts can be written for specific areas tied to a specific station. A general rule of thumb is to consider a specific WII contract marketable within a 20 km radius of the weather station; but the applicable area may be smaller or larger and case- specific evaluation must be carried out. In general terms, the more the terrain varies, the more the acceptable distance from a station decreases. Availability of Agricultural Data and Information 14. Official statistics on crop loss or damage, information on food or cash crop loss, and risk mapping (e.g., food security agency mapping, such as Fewsnet) are all contributors to an understanding the spatial distribution of weather risk. Assessment of weather risk involves a mix of data, wide information sources, expert opinion on agronomic practices, and agro- meteorology. 15. Agricultural information is the second part of the WII contract design equation. The most relevant information to be collected is data on productivity (yield), but a clear description of the agricultural production practices carried out in the areas of interest is also necessary. Unfortunately, the availability of quality yield data series of adequate length and at the appropriate level of disaggregation is not common. However, lack of quality yield data does not pose as large a problem as lack of good weather data, since it is still possible to find alternative 111 Adapted from ISMEA, but originally presented by Brian Tobben at the Annual Meeting of the International Task Force on Commodity Risk Management, jointly sponsored by the FAO and the World Bank at the FAO, Rome, 5 and 6 May, 2004. - 329 - approaches to estimating yield variability. One possibility is to simulate synthetic yield data series through plant-growth models. 16. Whether data on productivity is collected by official sources or generated by a crop model, it is still necessary to collect specific information from potential WII end users on local cultivating practices and procedures. Information such as crop varieties adopted, planting periods, modalities used (e.g., dry planting), management practices and related costs, risk profiles, historical recollection of the impact of the peril, and the most sensitive phases in crop life are essential for designing a meaningful WII contract. WII Contract Design 17. The objective of contract design is to define a structure that effectively captures the relationship between the weather variable and the potential crop loss and to select the index that is most effective in providing payouts when losses are experienced, reducing basis risk as much as possible. The set of possible index combinations is unlimited, and numerous structures have been developed in the relatively short history of WII. One of the most commonly adopted structures is that of a continuous payout triggered and limited by a cumulative measure of the weather variable (e.g., rainfall) for each of the different crop growth stages Box A8.2: Payout Parameters in a WII Contract (Example) Using the drought coverage case represented in Figure A8.1 as an example, the parameters that characterize an incremental payout structure can be defined as follows:  Trigger: Threshold above or below which payouts are due. Payments are due when the calculated value of the index is below the trigger level (300mm).  Exit: Threshold above or below which no additional incremental payout will be applied. The maximum payout is paid if the calculated value of the index is equal to or below the exit threshold (100mm).  Tick: Incremental payout value per unit deviation increase from the trigger. With a maximum payout (the insured sum) of $200, a trigger of 300mm, and an exit of 100 mm, the monetary value of each deficit mm of rainfall below the trigger is: $200 /(300 mm-100 mm) or $1 per mm. - 330 - Figure A8.1: Payout Structure of a WII Drought Contract 250 Exit: 100mm 200 150 Payout ($) 100 50 Trigger: 300mm 0 1 51 101 151 201 251 301 351 401 Cumulated rainfall (mm) - 331 - Annex 8 - Appendix A Map 5.2: Selected Weather Stations and Wheat Map 5.3: Selected Weather Stations and Wheat Production Rayons in Kostanay Oblast Production Rayons in Akmola Oblast Map 5.4: Selected Weather Stations and Wheat Production Rayons in Pavlodar Oblast Source: Authors‘ elaborations on Google Earth - 332 - Annex 8 - Appendix B. List of KHM Meteorological Stations in selected Oblasts Type of observations Meteorological № Soil humidity Rayon Year of Type of Actinometry Evaporation Name of installat manageme settlement ions nt Manual/A utomatic NKO 1 Zhambylsky Blagoveschenka 1936 m + + 2 Zhumabayeva Bulayevo 1930 m + + 3 Vozvyshenka 1931 m + + 4 Aiyrtausky Saumalkol 1914 m + + 5 Ualikanovsky Kishkenekol 1954 m + 6 Taiynshinsky Chkalovo 1955 m + + 7 Taiynsha 1925 m + + 8 Petropavlovsk Petropavlovsk 1890 m + city 9 Musrepova Rusaevka 1935 m + + 10 Timirayzevsky Timirayzevo 1955 m + + 11 Shal Akyn Sergeevka 1968 m + + 12 Esilsky Yavlenka 1902 m + + 13 Mamlutsky Mamlutka 2009 m + 14 Akkainsky Smirnovo 2009 m + Kostanay oblast 1 Amangeldinsky Amangeldy 1935 m + 2 Arkalyksky Arkalyk 1958 m + 3 Ekidyn 1959 m + 4 Denisovsky Arshalinsky 1956 m + + 5 Kamystinsky Bestau 1960 m + 6 Zhitikarinsky Dzhetigara 1935 m + + 7 Auliekolsky Dievskaya 1957 m + + 8 Kushmurun 1940 m + + 9 Naurzumsky Dokuchaevka 1934 m + 10 Karasusky Zheleznodorozhny 1960 m + + 11 Karasu 1937 m + + 12 Karabalyksky Komsomolets 1930 m + + 13 Kostanaysky Kostanay 1962 m + + 14 Rudnay 1960 m + + 15 Mendykarinsky Mikhailovka 1958 m + + 16 Uzunkolsky Presnogorkovka 1955 m + + 17 Taranovsky Tobol 1950 m + + 18 Dzhangeldinsky Torgay 1974 m + 19 Sarykolsky Uritsky 1924 m + + Pavlodar oblast 1 Aktogaysky Zholboldy 1956 m + + 2 Aktogay 1960 m + + - 333 - 3 Zhelezinsky Mikhailovka 1907 m + + 4 Irtyshsky Golubovka 1949 m + + 5 Ertis 1935 m + + 6 Kashirsky Fedorovskaya 1956 m + + 7 Pavlodarsky Krasnoarmeika 1963 m + + 8 Uspensky Lozovaya 1954 m + + 9 Uspenka 1958 m + 10 Sharbaktinsky Sharbakty 1932 m + + 11 Shalday 1957 m + 12 Bayanaulsky Bayanaul 1925 m + + 13 Maisky Koktobe 1970 m + 14 Ekibastuz town Ekibastuz 1949 m + 15 Aksusky Akzhol 1961 m + 16 Pavlodar city Pavlodar 1891 m + Akmola oblast 1 Ereimentausky Ereimentau 1954 m + 2 Esilsky Esil 1941 m + 3 Zhaksinsky Zhaksy 1970 m + + 4 Kima 1980 m + 5 Astrakhansky Zhaltyr 1940 m + + 6 Kokshetau town Kokshetau 1895 M/А + 7 Stepnogorsk town Stepnogorsk 1935 m + 8 Atbassarsky Atbassar 1886 m + + 9 Sandyktausky Balkashino 1934 m + + 10 Egindykolsky Egindykol 1973 m + + 11 Zharkainsky Tasty-Taldy 1934 m + 12 Zerendinsky Zerenda 2004 m + + 13 Akkolsky Akkol 1909 m + 14 Astana city Astana 1870 m + 15 Arshalinsky Arshaly 1935 m + 16 Borovoe SFM 1980 M/А + 17 Burabaisky Burabay 2006 M/А + + 18 Schuchinsk 1898 m + 19 Korgalzhinsky Korgalzhin 1955 m + 20 Ereimentausky Novomarovka 2009 m + 21 Shortandinsky Shortandy 2007 m + Karaganda oblast 1 Osakarovsky Rodnikovsky 1978 m + 2 Osakarovka 2007 m + + 3 Nurinsky Kievka 2005 m + + 4 Barshino 2007 m + 5 Kertindy 1956 m + 6 Karaganda 1969 m + + Bukhar-Zhirausky Agricultural experimental station 7 Korneevka 1959 m + + 8 Karkaralinsky Karkaraly 1875 M/А + 9 Besoba 1938 m + 10 Ulytausky Dzhetykonyr 1947 m + 11 Kulzhambay 1972 m + 12 Ulytau 2006 m + 13 Zhaarkinsky Zhanaarka 1937 m + - 334 - 14 Karazhal 2008 m + 15 Kyzylzhar 1937 m + 16 Shetsky Kyzyltau 1947 m + 17 Agadyr 1948 m + 18 Aksu-Ayuly 1938 m + 19 Zharyk 1938 m + 20 Aktogaysky Aktogay 1931 m + + 21 Balkhash 1959 m + + 22 Bektauata 1938 m + 23 Saryagash 1961 m + 24 Sayak 1973 m + 25 Karaganda city Karaganda 1932 m + 26 Zheskazgansky Zheskazgan 1932 m + + SKO 1 Arysky Arys 1925 m + 2 Tulkubassky Т. Ryskulbekov aul 1951 m + + 3 Turkestansky Akchisay 1935 m + 4 Turkestan 1882 m + 5 Kazygurtsky Kazygurt 1936 m + + 6 Otrarsky Kyzylkum 1946 m + 7 Zhetysaisky Zhetysay 1957 m + + 8 Tolebiysky Tasaryk 1906 m + + 9 Shuuldak 1936 m + 10 Suzaksky Tasty 1948 m + + 11 Sholakkurgan 1935 m + 12 Shardarinsky Shardara 1934 m + + 13 Baidykbeksky Shayn 1936 m + 14 Shimkent city Shymkent 1938 m + + EKO 1 Aygozsky Aygoz 18896 m + 2 Aktogay 1961 m + 3 Barshatas 1937 m + 4 Beskaragaisky Semiyarka 1893 m + 5 Borodulikhinsky Dmitrievka 1960 m + + 6 Glubokovsky Leninogorsk 1928 m + 7 Zharminsky Shar 1931 m + 8 Shalabay 1934 m + 9 Zhalgystobe 1931 m + 10 Zaisanskyй Zaisan 1876 m + 11 Zyraynovsky Zyraynovsk 1887 m + + 12 Seleznevka 1967 m + + + 13 Katon-karagaysky Katonkaragay 1898 m + 14 Ulken naryn 1937 m + + 15 Kokpektinsky Kokpekty 1889 m + 16 Samarka 1929 m + + 17 Kurchumsky Boran 2005 M/А + 18 Zapovednik 1982 m + + Markakol 19 Terekty 1970 m + 20 Kurshim 1936 m + 21 Tarbagataisky Akzhar 1961 m + 22 Aksuat 1940 m + 23 Tugyl 1907 m + + - 335 - 24 Urzharsky Urzhar 1933 m + + 25 Bakty 1894 m + 26 Shemonaihinsky Shemonaikha 1934 m + + 27 Ust-kamenogorsk Ust-Kamenogorsk 1877 M/А + 28 Semipalatinsk Kainar 1950 m + 29 Semipalatinsk 1944 m + + 30 Abaisky Karaul 1937 m + Aktobe oblast 1 Shalraksky Aykkum 1949 m + + 2 Shalkar 1922 m + + 3 Baskuduk 1959 m + 4 Aitekebiisky Komsomolskoe 1970 m + + 5 Karabutak 1937 m + 6 Alginsky Ilyinsky 1958 m + + 7 Irgizsky Irgiz 1856 m + 8 Nura 1953 m + 9 Baiganinsky Karaulkeldy 1937 m + 10 Kargalinsky Kos-Istek 1956 m + + 11 Martuksky Martuk 1926 m + + 12 Rodnikovka 1925 m + + 13 Mulgalzharskaya 1926 m + 14 Mugalzharsky Emba 1904 m + + 15 Zhagabulak 2007 m + 16 Khobdinsky Novoalekseevka 1937 m + 17 Krasnoyrskoe 2005 m + 18 Khromtausky Novorossiyskoe 1925 m + + 19 Temirsky Temir 1894 m + 20 Uilsky Uil 1886 m + 21 Aktobe city Aktobe 1898 m + WKO 1 Burlinsky Aksay 1937 m + 2 Syrymsky Dzhambeity 1897 m + 3 Dzhangalinsky Dzhangala 1979 m + 4 Zhanakazan 2006 m + 5 Kaztalovsky Zhalpaktal 1924 m + 6 Kaztalovka 1956 m + 7 Taskalinsky Kamenka 1954 m + + 8 Karatobinsky Karatobe 1958 m + 9 Akzhaisky Taipak 1891 m + 10 Chapaevo 1930 m + 11 Uralsk city Uralsk 1899 m + + + 12 Bokei-orda Urda 1915 m + 13 Chingirlausky Chingirlau 1928 m + + 14 Zelenovsky Yanvartsevo 1955 m + 15 Dzhanibeksky Dzhanibek 1950 m + + - 336 - Annex 8 - Appendix C. Hydrotermal Indexes and Trend Analysis A simple index adopted for agricultural monitoring in Kazhakstan is the hydrothermal ratio (HTR) defined as HTR   R , where R is cumulated rainfall expressed in millimeters and  T / 10 T is the sum of temperatures in degrees for the period with average daily temperatures above 10°C. It is based on the concept that active vegetation of most crops starts from the date of stable transfer of average per diem temperature above 10°С, which on most part of farming Kazakhstan comes to the end of April – beginning of May. A very severe drought would correspond to HTR=0.3, less severe drought is usually in the range 0.31 < HTR < 0.6, a medium drought is in the range 0.61 < HTR < 0.8, and weak drought is when 0.81 < HTR < to 1.0 A slightly more complex example is humidity factor K, developed by Kelchevskay, Brinken, Sapozhnikova and Chirkova. The K index is essentially a ratio between water available for evapotranspiration (snow contribution is accounted for 0.5 of precipitation levels) and cumulated average temperature. The K index may be indicative of agricultural drought situations when the daily thermal cycle (i.e. solar radiation) is not too strong and the local balance between rainfall and evapotranspiration occurs over time scales of about a month or longer. The actual formulation 0.5  R114   R58 of the K index K  . Map A1.1 portrays a zoning of the country on the 0.118 T58 basis of the K index according to which the territory of Kazakhstan can be divided into 6 zones (Map AD.1). Main crop farming area is set in poor watered (K=0,8-1,0), poor arid (K=0,6-0,8) and mid arid zone (K=0,4-0,6). - 337 - Map AD.1 Kazakhstan climate classification according to Humidity factor K К >1,0 well watered soil K = 0,4-0,6 – mid arid zone К = 0,8 – 1,0 – poor watered zone К = 0,2 – 0,4 strongly arid zone К = 0,6 – 0,8 – weakly arid zone К =<0,2 – dry zone Source: KHM Table AD.1. Trend analysis Annual Rainfall Annual Temperature Trend P(t) Trend P(t) Aktogay 2.8 18.5% 0.05 2.6% Balkashino 0.6 84.4% 0.07 0.7% Diyevskaya -2.2 23.7% 0.09 0.1% Kostanay -1.1 59.2% 0.09 0.2% Kushmurun 0.0 99.1% 0.07 0.7% Mikhailovka 0.7 77.9% 0.05 3.6% Schuchinsk 2.4 31.7% 0.06 2.4% Stepnogorsk -1.1 51.4% 0.07 0.7% Tasaryk -0.8 87.3% 0.04 0.2% Zholboldy 0.2 87.4% 0.06 9.8% Source: Authors‘ analysis on KHM data. - 338 - Annex 8 - Appendix D. Yield data - 339 - Annex 8 - Appendix E WII Pricing Methodology 1. For the purpose of defining plausible commercial premium rates for the WII contract structures developed in this feasibility study, the Return-on-Risk (RoR) approach has been adopted.112 As for any insurance contract, pricing of WII is a highly subjective process and there is no correct approach: any industry player will make use of specific and customized pricing methodologies. Hence, the purpose of this analysis is not to provide objective market quotes, but to factor in the various elements that may provide a more realistic estimate of the cost of insuring against drought with a WII contract. 2. In order to get to the definition of a final premium, the RoR approach analyses the Expected Loss (the arithmetic average payout of the contract from the historical cleaned dataset for the station), the Probable Maximum Loss (PML, representing the cost the client has to pay to buy the right of receiving extreme indemnities ), and Administration and Business Expenses. 3. In the RoR approach the premium is calculated as follows: Premium = AEL + * [PML (1-in-100) – AEL] + Admin. & Bus. Expenses, or also Premium = Technical Premium + Admin. & Bus. Expenses where:  AEL is the Adjusted Expected Loss, i.e. the Expected Loss adjusted where necessary by a data uncertainty factor and a missing data factor; computed as follows: o Data Uncertainty Factor = F (β) * s /sqrt(N*(1-j)); with F (β) as the selected level of confidence; and s/Sqrt(N*(1-j)), the standard error of the mean corrected for the % of missing values ―j‖  PML (1-in-100) is the Probable Maximum Loss or the highest value of payouts expected 1 in a 100 years under the defined contract.  Admin. & Bus. Expenses (ABE) a percentage of the Technical Premium. 4. The rationale behind the introduction of PML (sometimes called Value at Risk, VaR) is that the insurer is exposed to the risk of extreme payouts. Therefore he decides that the client has to pay a fraction of this payout to buy the right of receiving such extreme payouts. 112 A more comprehensive illustration of the use of the RoR approach for pricing WII contracts is provided in the web based training course on WII available at www.agrisktraining.org. Literature references for the RoR are World Bank (2005), ISMEA (2006) and Henderson et al. (2002). - 340 - 5. s the target Return-on-Risk (RoR), assuming the insurer is required to reserve capital against its portfolio at the PML (1-in-100) year level and where risk is defined in terms of the payouts in excess of the AEL.. 6. Missing data in the weather stations records need to be incorporated into the pricing of the weather insurance contracts. As there is increased uncertainty in data that is received from a station with missing data, the resulting insurance contract will be priced higher due to the higher level of uncertainty regarding the underlying risk. A simple way of incorporating the ‗data quality risk‘ in the pricing is to add a data uncertainty factor to the AEL In this perspective, Kazakhstan is a rather fortunate environment as the rate of missing data is usually very low, maximum close to 1% for the daily weather data analyzed during this feasibility study, so that the final gross premium is hardly affected by data uncertainty. 7. The following are the values of the pricing parameters that have been adopted in this feasibility study: PML = Maximum historical loss β = 90% and, therefore, F(β) = 1.28 N = 25 j=0   - 341 - Annex 8 - Appendix F WII Prototype Contracts Figure A8.2: Historical Performance of Drought Index Contract for Spring Wheat in Auliekolskiy Rayon 15.000 10.000 5.000 KZT/Hectare 0 -5.000 -10.000 HTR Index Payout Revenue Loss Rayon: Auliekolskiy Weather Station: Kushmurun Indicative Yield Level Coverage: 65% Index: HTR, from 1 March to 31 July Trigger: 0.475 Index Units Exit: 0.100 Index Units Contract Maximum Payout: KZT 19,000 Reference Gross Premium: 13.7% Basis Risk Index: 18% - 342 - Figure A8.?: Historical Performance of Index Contract for Spring Wheat in Enbekshilderskiy Rayon (WS Stepnogorsk) 15.000 10.000 5.000 KZT / Hectare 0 -5.000 -10.000 -15.000 K Index Payout Revenue Loss Rayon: Enbekshilderskiy Weather Station: Stepnogorsk Indicative Yield Level Coverage: 65% Index: K: Prec: 1 Nov-31Aug; Temp: 1 May-31 Aug Trigger: 0.700 Index Units Exit: 0.200 Index Units Contract Maximum Payout: KZT 23,000 Reference Gross Premium: 15.2% Basis Risk Index: 50% - 343 - Figure A8.?: Historical Performance of WII Contract for Spring Wheat in Enbekshilderskiy Rayon (WS Schuchinsk) 15.000 10.000 5.000 KZT/Hectare 0 -5.000 -10.000 -15.000 K Index Payout Revenue Loss Rayon: Enbekshilderskiy Weather Station: Schuchinsk Indicative Yield Level Coverage: 65% Index: K: Prec: 1 Nov-31Aug; Temp: 1 May-31 Aug Trigger: 0.700 Index Units Exit: 0.200 Index Units Contract Maximum Payout: KZT 23,000 Reference Gross Premium: 15.0% Basis Risk Index: 86% - 344 - Figure A8.2: Historical Performance of Drought Index Contract for Spring Wheat in Altinsarin Rayon 25.000 20.000 15.000 10.000 KZT/Hectare 5.000 0 -5.000 -10.000 -15.000 Cumulated Precipitation Payout Revenue Loss Rayon: Altinsarin Weather Station: Kostanay Indicative Yield Level Coverage: 65% Index: Cumulative precipitation, 1 Mar-20 Jul Trigger: 95 mm Exit: 30 mm Contract Maximum Payout: KZT 24,000 Reference Gross Premium: 17% Basis Risk Index: 41% - 345 - Figure A8.3: Historical Performance of Drought Index Contract for Spring Wheat in Aktogayskiy Rayon (WS Aktogai) 15.000 10.000 5.000 KZT/Hectare 0 -5.000 -10.000 Yield reduction due to locust attack -15.000 Cumulated Rain Payout Revenue Loss Rayon: Aktogayskiy Weather Station: Aktogai Indicative Yield Level Coverage: 65% Index: Cumulative precipitation, 1 Mar - 31 Aug Trigger: 150 mm Exit: 50 mm Contract Maximum Payout: KZT 13,000 Reference Gross Premium: 20.9% Basis Risk Index: 90% - 346 - Figure A8.3: Historical Performance of Drought Index Contract for Spring Wheat in Aktogayskiy Rayon (WS Zholboldy) 15.000 10.000 5.000 KZT/Hectare 0 -5.000 -10.000 Yield reduction due to locust attack -15.000 Cumulated Rain Payout Revenue Loss Rayon: Aktogayskiy Weather Station: Zholboldy Indicative Yield Level Coverage: 65% Index: Cumulative precipitation, 1 Mar - 31 Aug Trigger: 160 mm Exit: 50 mm Contract Maximum Payout: KZT 13,000 Reference Gross Premium: 24.2% Basis Risk Index: 88% - 347 - Figure A8.3: Historical Performance of Drought Index Contract for Spring Wheat in Bulandinsky Rayon (80% yield reference level) 15.000 10.000 KZT/Hectare 5.000 0 -5.000 Yield reduction -10.000 due to excesssive rain and pest attack K Index Payout Revenue Loss Rayon: Bulandinsky Weather Station: Balkashino Indicative Yield Level Coverage: 80% Index: K: Prec: 1 Nov-31Aug; Temp: 1 May-31 Aug Trigger: 0.700 Index Units Exit: 0.300 Index Units Contract Maximum Payout: KZT 23,000 Reference Gross Premium: 9.9% Basis Risk Index: 78% - 348 - Figure A8.3: Historical Performance of Drought Index Contract for Spring Wheat in Bulandinsky Rayon (65% yield reference level) 5.000 4.000 3.000 2.000 KZT/Hectare 1.000 0 -1.000 -2.000 -3.000 -4.000 -5.000 K Index Payout Revenue Loss Rayon: Bulandinsky Weather Station: Balkashino Indicative Yield Level Coverage: 65% Index: K: Prec: 1 Nov-31Aug; Temp: 1 May-31 Aug Trigger: 0.530 Index Units Exit: 0.300 Index Units Contract Maximum Payout: KZT 19,000 Reference Gross Premium: 3.1% Basis Risk Index: 67% - 349 -