2021 The World Bank Group 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbankgroup.org Some rights reserved This work is a product of the staff of The World Bank Group with external contributions. The findings, interpre- tations, and conclusions expressed in this work do not necessarily reflect the views of the World Bank Group, its Board of Executive Directors, or the governments they represent. The World Bank Group does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of the World Bank Group, all of which are specifically reserved. RIGHTS AND PERMISSIONS This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// creative- commons. org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, dis- tribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Translations - If you create a translation of this work, please add the following disclaimer along with the attribu- tion: This translation was not created by the World Bank Group and should not be considered an official World Bank Group translation. The World Bank Group shall not be liable for any content or error in this translation. Adaptations - If you create an adaptation of this work, please add the following disclaimer along with the attri- bution: This is an adaptation of an original work by the World Bank Group. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by the World Bank Group. Third-party content - The World Bank Group does not necessarily own each component of the content con- tained within the work. The World Bank Group therefore does not warrant that the use of any thirdparty-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Photos: Shutterstock & Unsplash. ACKNOWLEDGEMENTS This note was prepared by the World Bank Agriculture and Food Practice for East and Southern Africa under the guidance of Holger Kray, Practice Manager. The report was prepared by Hazem Hanbal and John Nash (Consultant) with inputs from Paavo Eliste. Sergiy Zoray and Manievel Sene of the World Bank as well as Herbert Matsikwa and Allan Mulando of the World Food Program provided technical advice as peer reviewers. Mara Warwick (Country Director - Tanzania, Malawi, Zambia and Zimbabwe), Mukami Kariuki (Country Manager, Zimbabwe) , Sahr Kpundeh (Country Manager, Zambia), Preeti Arora (Operations Manager), and M. Yaa Oppong (Sector Leader, Sustainable Development - Tanzania, Malawi, Zambia and Zimbabwe ) provided strategic guidance and the resources to conduct the analytical work. TABLE OF CONTENTS Acknowledgements .....iv Table of Contents .....v INTRODUCTION .....vi PART 1: SUMMARY OF EVIDENCE AND LESSONS LEARNED ON INPUT SUBSIDY PROGRAMS (ISPS) IN ZAMBIA AND ELSEWHERE .....1 A. A brief history of Input Subsidy Programs in Africa .....2 B. Input subsidy programs – What are the unintended effects? .....6 C. What do we know about the aggregate costs and benefits of input subsidy programs? .....9 PART 2: RECOMMENDATIONS FOR REDESIGNING AND STRENGTHENING ZAMBIA’S FISP .....11 PART 3: RECOMMENDATIONS ON COMPLEMENTARY MEASURES TO ENHANCE GRAIN PRODUCTION AND MARKETING .....18 Annex A1: Turkey’s Sectoral Transformation through Adoption of Better Instruments of Support .....21 Annex A2: Smarter support to agriculture: Decoupled direct payments .....25 References .....27 iv v INTRODUCTION This note provides recommendations for re- The note is a key deliverable under the Advisory designing Zambia’s Farmer Input Support Program Services work carried out by the Bank on strategies (FISP) in the context of international experience for food security in Zambia and Zimbabwe. with other similar programs, especially those in This note also feeds into the ongoing technical Africa. The analysis and recommendations are dialogue between the World Bank Agriculture and based on past analytical work on input subsidies Food Global Practice team and the Ministry of carried out by the World Bank,1 and the extensive Agriculture of Zambia. literature on input subsidy programs from research in Zambia and elsewhere including the extensive This note has three sections following the analytical work from the Indaba Agricultural Policy summary: (i) a brief review of the evidence Research Institute (IAPRI) in collaboration with generated on input subsidy programs; (ii) specific Michigan State University (MSU). recommendations for Zambia’s (Farmers Input Support Program) FISP, and (iii) recommendations The objective of this note is to provide on complementary investments. recommendations that can help in improving the current programs providing subsidized inputs to Annexed to the note are other countries farmers, through enhancing the efficiency of the experience in the area of input support to program, targeting the right beneficiaries, value farmers, and also descriptions of different support perceived by beneficiary farmers, and probably instruments that proved to be successful in bringing in savings. helping farmers elsewhere. Summary of main findings and recommendations Research on the Zambia FISP and other similar • Input subsidy programs can generate impacts input subsidy programs finds: in terms of expanding food production, though at a smaller scale than what would • In 2019, the budget for the FISP (not included be expected due to (i) crowding out of in the budget of the Ministry of Agriculture) commercial fertilizer use, (ii) program was larger than the rest of the total public implementation inefficiencies, and (iii) low expenditure on agriculture, and in 2020 soil fertility, which reduces yield response to climbed to almost 2.5 times the other budget increased fertilizer use. for the sector – almost 3 times the budget of the Ministry of Agriculture itself. This • Evidence shows that most beneficiaries of commitment severely limits the Government’s subsidies (in Zambia and other countries) capacity to invest in other poverty reduction are larger farmers. In Zambia, this is an and agricultural development programs. This inevitable consequence of the FISP program’s is similar to the situation in other countries eligibility rules, which are perhaps based on which use input subsidies as major agricultural the belief that large farmers use fertilizer INTRODUCTION policy instruments. more effectively. In reality, evidence from 1 The note draws heavily on chapter 3 in Goyal and Nash (2017), which is based on a background paper by Jayne, et. al. (2016). vi vii ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM INTRODUCTION farm survey data indicates that maize yield • Corruption infrastructure is currently in place for commercial basis), capping the size of the response on small farms may actually be successful use of e-vouchers, and areas subsidy, and setting a maximum for the higher. Thus, poor targeting is both inequitable • Across many countries and over decades where this is not the case; (2) identify what number of years of eligibility. Use the savings and an inefficient use of funds. Many input of experience, detailed analysis has found is needed to prepare the latter areas; (3) to fund other more appropriate types of public subsidy programs also face challenges due to very few examples where benefits to subsidy develop a plan to carry out this preparation, expenditure in the agricultural sector. This will the capture of elites and politicization in the programs exceed their costs, even when the including piloting strategies to attract input also require good stakeholders’ engagement allocation of subsidy benefits. These problems unintended ancillary costs are excluded from suppliers, install necessary infrastructure, and a clear communication strategy around mean that subsidies are a poor tool for social the calculations. However, these programs and build capacity; (4) ascertain whether these decisions. protection, although this is sometimes viewed continue to be popular among politicians farmers are currently getting access to as an ostensible objective. since they provide a very immediate and fertilizers and other inputs appropriate for • Put in place rules to ensure transparent and tangible example of what the official can do their location and soil types under both the competitive procurement procedures for • Unintended impacts from most input for constituents. DIS and e-voucher systems, and where this purchases of inputs to supply the DIS and subsidy programs have been substantial, is not the case suggest changes to resolve reduce opportunities for corruption and although they can be mitigated by better • Across many countries and over decades this problem; and (5) suggest eligibility rules diversion. Procurement rules should include design (e.g., using vouchers instead of direct of experience, detailed analysis has found for the DIS and e-voucher that would ensure quality assurance and provide for selection supply) and execution. These unintended very few examples where benefits to subsidy that small farmers who are not capable of based on both price and quality. consequences include: programs exceed their costs, even when the benefiting from input subsidies due to their unintended ancillary costs are excluded from scale of operation, lack of labor or other • Review (and revise as necessary) regulations • Creation of negative incentives for the calculations. However, these programs factors are transferred from these programs associated with inputs to ensure that farmers developing private sector input supply continue to be popular among politicians to other more suitable programs of support are getting the product and quality they are networks since they provide a very immediate and (e.g., ‘kitchen gardening’ or cash safety nets). expecting (e.g., through truth in labeling • Crowding out of other public spending tangible example of what the official can do requirements), but their options are not on programs and investments that are for constituents. • Begin to put in place a system of nation-wide restricted by requirements for lengthy and demonstrably more productive; and soil mapping that would better inform the expensive testing for new varieties and products. type and amount of fertilizers that need to be made available to farmers under the subsidy • Where vouchers are not used, include a more programs. This would need to be organized appropriate mix of inputs in the packets, such Specific recommendations for the re-design of the Zambia FISP in the short term are: by and coordinated with the research and as lime and fertilizer types better suited to the extension services. different varieties of soils in Zambia. • In areas where it was obvious that the • As a matter of priority, an independent • Take measures to ensure proper targeting of • Establish robust monitoring and evaluation e-voucher scheme has been tried and have expert evaluation of the experience with input subsidies (including equal participation systems that allow for greater information and worked well, the program can be re-launched, the piloting and subsequent rollout of by female farmers) – whether administered analysis of program implementation, costs and while it can be piloted in other parts of the the e-voucher program and the current through the DIS or e-vouchers. This will impacts to aid policy and decision makers. country where it wasn’t tested before. Allow functioning of the disbursement system probably require using selection criteria beneficiaries to use the vouchers on a wide needs to be commissioned. The evaluation related directly to individual beneficiaries, • Develop an exit strategy or criteria for range of inputs at dealers of their choice. should cover a representative sample of farm rather than using co-op membership as a scaling down the program as conditions Ensure that funds are always released to households of different sizes, types, and major criterion, a system that has not worked eventually change, including a strategy for agro-dealers in a timely manner, which is locations. The over-arching objective would well with respect to targeting (para. 29). It will communicating the rationale to beneficiaries not currently the case. A World Bank report be to ensure that the e-voucher system can also likely require limiting FISP beneficiaries and stakeholders, and also description in (forthcoming 2021) is a synthesis of the eventually successfully replace the physical (e.g., excluding larger farmers who will replacing policies that may be implemented experiences in other countries with e-voucher distribution of inputs nationwide. Terms of purchase unsubsidized modern inputs on a in the area of support to beneficiaries. schemes, which includes lessons that may be reference would include (1) identify areas useful for Zambia’s future e-voucher program. where the commercial and electronic viii ix ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM The expected supply response to the FISP re- transport grain across districts within Zambia. design will amplify the importance of taking complementary measures to enhance the • Undertake reforms to the FRA, as discussed functioning of grain markets in general, which is in detail in a companion report on that topic. tightly linked to the FISP. These measures include: • Prioritize and expedite operationalization of the • Establish and abide by transparent rules Zambia Grain Information System (ZAGRIS). for trade restrictions on grain, including ensuring automatic access to export permits • Prioritize enactment of an appropriate except under well-defined and extraordinary regulatory framework for a warehouse circumstances. Export restrictions are highly receipts system. detrimental to market development and are very rarely invoked in countries that aspire • Foster development of local trade associations. to be significant exporters. The conditions under which restrictions are imposed should • Make additional investments in sustainable be defined such that they rarely occur and in land and water management, extension, the medium term, the government should research, education, and irrigation to achieve foreclose the use of export permits and maximum productivity gains from input use restrictions altogether, except in cases of serious emergencies. • Increase investments to reduce marketing costs–in particular rural road rehabilitation • Immediately reduce, and in the medium term and construction – especially in areas of high eliminate, levies and paperwork required to agricultural potential. PART 1 SUMMARY OF EVIDENCE AND LESSONS LEARNED ON INPUT SUBSIDY PROGRAMS (ISPS) IN ZAMBIA AND ELSEWHERE x 1 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM PART 1 | SUMMARY OF EVIDENCE AND LESSONS LEARNED ON INPUT SUBSIDY PROGRAMS (ISPS) IN ZAMBIA AND ELSEWHERE in demonstration plots or experiment stations also occurs and fertilizer is resold. Research in A. A brief history of Input Subsidy Programs in Africa using ideal technologies. So, it turns out that several countries has shown a significant impact low fertilizer use may not be due to ignorance, of commercial crowding out or diversion of 1. Input subsidy programs have a long history in countries pledged to increase spending on irrationality, or the poverty of farmers. Rather, fertilizer to the market. Within the region, it has Africa and other developing countries. The logic agriculture, and an optimistic view that a new under existing conditions, using more fertilizer been estimated that for every 1 ton of fertilizer behind these seemed persuasive: Both fertilizer2 generation of ISPs could be better designed and may not be an economically rational practice distributed through the subsidy program, total use and productivity in Africa were (and still are) ‘smarter’. The case was also bolstered by the early given its commercial price. fertilizer use is raised by only 490kg in Malawi, very low by global standards, and it seemed – and as it subsequently turned out, somewhat 536kg in Zambia and 239 kg in Kenya (Jayne reasonable to assume that there was a causal link. erroneous – reports of the “Malawi miracle” due 5. This conclusion is supported by a large et. al 2013). Crowding has been much lower in If so, raising fertilizer use would raise productivity. to its subsidy program. The new programs were number of studies over many African countries areas where commercial input markets are weak The low level of use was attributed, inter alia, to quite large scale in many countries. By 2014, 10 that have estimated the “value cost ratio” (VCR), or where households have been poor. In recent farmers’ ignorance of the benefits of fertilizer use, countries in Africa (representing over 60 percent that is, the value of additional production relative years it appears Zambia’s FISP has tried to target underdevelopment of input sales networks, and of the continent’s population) had programs that to the cost (at commercial prices) of applying areas where the private sector input suppliers are the lack of affordability for many farmers. Heavily absorbed a significant share of the agricultural additional fertilizer. In almost all cases, the not located to reduce crowding out effects. subsidizing fertilizers for a few years would make public expenditure budget. Malawi’s program cost estimated VCR is too low to justify increasing them affordable, thereby both demonstrating was about half of the budget, with Ghana and fertilizer application rates. For Zambia, for example, 8. More recently, some studies have estimated their value and generating sufficient income to Zambia not far behind at 40 percent. the VCR was estimated to range from 0.3- 1.2 that crowding out of commercial fertilizer sales eventually let farmers save up and continue to (depending on soil pH level), whereas a value of may have been substantially under-estimated due buy their own from private sector suppliers once Evidence shows that ISPs have generally 2 is generally considered a minimum to induce to fertilizer that has been diverted from subsidy the subsidy ended. This would lead to a ‘virtuous not been very successful in meeting their widespread adoption of a new practice (Burke, program channels into what can be mistaken for cycle’ of growing input use and productivity. objectives 2012). The implication is that there will never be a commercial sales (Mason and Jayne, 2013; Jayne “virtuous cycle” after the subsidies are ended. This et al., 2013). Both in Malawi and Zambia, comparing 2. In Africa, however, most ISPs were phased out 4. By now, these programs (especially those in does not mean that such a cycle is impossible. It the official subsidized fertilizer distribution volumes in the 1980s and 1990s. This occurred because Africa and Asia) have been operating for quite a does mean, however, that starting the dynamic and the estimated volume of subsidized fertilizer of considerable evidence that they (i) largely long period, and they have been the subjects of processes to trigger it will require a much more received by farmers according to nationally failed to live up to expectations and achieve their a substantial body of research. In summary, one holistic approach, including in particular a focus representative survey data suggests diversion of 25- objectives, (ii) had very high fiscal costs, and (iii) led important finding is that these programs are unlikely on soil management. 35% of subsidized fertilizer is common. Diversion to a number of adverse unintended consequences ever to produce the kind of envisaged ‘virtuous of program fertilizer has important income including corruption and suppression of private cycle’ that would allow them to be phased out 6. Some of the other more detailed lessons based distributional effects, with program implementers sector market development. Zambia and Malawi or at least reduced in scale over time. Rather, the on global experience and research include: Input receiving a major portion of the program benefits continued to implement ISPs in this period, empirical evidence has generally undermined a key subsidy programs have enabled countries to rather than farmers (Jayne et al., 2015). although both were relatively modest in scale. assumption underlying the programs – that raising increase domestic food supply in a relatively fertilizer use will in practice significantly increase short span of time – but by less than expected 9. Second, inefficiencies in program 3. Around 2005, the landscape changed quickly, productivity. In fact, experience has shown that and much less than enough to offset the cost of implementation impact production; farmers with at least 10 Africa countries beginning or increasing fertilizer use by smallholders will only the subsidy. This failure has occurred for a range do not get what they expect, and inputs arrive renewing input subsidies. This change was marginally increase productivity in areas that are of reasons: too late. A detailed study of the Zambia FISP in prompted by several factors, including a general drought-prone and have acidic soils deficient 2010 found program inefficiencies had significant relaxation of fiscal constraints in many countries in organic matter, which are characteristics of 7. First, the increase in total fertilizer use is less impacts. The 2007/08 input distribution was (due to some extent to debt forgiveness under the many parts of Africa. Furthermore, yield response than would be expected due to displacement characterized by large differences in farmer heavily indebted poor countries (HIPC) initiative, to fertilizer in farmers’ fields has generally been of demand for commercially supplied fertilizer. expectations and actual receipts. When the inputs the Maputo Declaration in 2003 in which African much less than was expected based on the results Some farmers who receive fertilizer subsidies arrived, however, only 44% of farmers actually would have purchased fertilizer even in the received the 4x4 fertilizer allowance; 55% of absence of the subsidy – thus it only partially farmers received less, and 2% received more. contributes to increased fertilizer use and crop These widespread differences in expectations and 2 All programs included fertilizer in the subsidy scheme. Many also included other inputs, and the rationale was the same, but production. In some cases, diversion of the subsidy receipts point to serious negative consequences they were a smaller part of the program. 2 3 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM PART 1 | SUMMARY OF EVIDENCE AND LESSONS LEARNED ON INPUT SUBSIDY PROGRAMS (ISPS) IN ZAMBIA AND ELSEWHERE for a farmer's ability to plan and make wise percent of the subsidized fertilizer was common 13. Input subsidy programs have often not 14. There is also considerable evidence that in production decisions including whether or not (Jayne, et. al., 2013; Mason and Jayne, 2013). reached their targeted beneficiaries. Input some countries, benefits go disproportionately to buy inputs outside the program. subsidy programs often state their targeted to those who are politically well connected or 12. Additionally, as mentioned above, low soil beneficiaries are small, resource-poor farmers. otherwise ‘elite’. In Zambia, for example, results 10. In the same Zambia FISP survey, although fertility is a limiting factor for fertilizer efficiency; Actual beneficiary characteristics usually differ, based on multiple nationally-representative surveys most farmers said the quality of FISP inputs was farmers with poor quality soil do not achieve however, with most fertilizer subsidy programs (both panel and cross-sectional) consistently good, the 2007/08 season was marked by serious much yield benefit from fertilizer subsidies. Soil benefiting larger, wealthier farmers than originally suggest that from the late 1990s through 2010, problems with late delivery. Across the sample, quality has a significant impact on crop response anticipated. In studies of Kenya, Malawi and Zambia smallholder households in constituencies won less than 4% of beneficiaries said they received to fertilizer application. Nitrogen application on input subsidy programs, subsidy beneficiaries by the ruling party (the Movement for Multi- their inputs by the end of October and 69% said carbon deficient soils (low soil organic matter) has were wealthier than non-targeted beneficiaries. Party Democracy (MMD) at that time) in the last they did not get their inputs until after the start been shown to be inefficient and economically (Mason et al, 2013). In Zambia, households with presidential election received significantly more of the rains. Timeliness of planting is extremely unprofitable. (Marenya and Barret 2009). A larger land holdings ended up benefitting more (23 kg) subsidized fertilizer than those in areas important with maize and there is no doubt detailed study on maize yield response to fertilizer from subsidies. The top quintile of farms by size lost by the ruling party; moreover, the quantity this record had a significant negative impact on application in Zambia showed limited maize yield of holding captured 41 percent of the subsidy, of subsidized fertilizer received was increasing yields. In the 2007/08 farm season, for instance, response to basal fertilizer application on highly while the smallest quintile received only 6 percent in the ruling party’s margin of victory (Mason, the World Bank study found that maize yields per acidic soils – which represent the majority of farms (see figure). The majority of fertilizer subsidies Jayne, and Mufya-Mukuka, 2013; Mason, Jayne, hectare were around 5% lower for households in Zambia (51% of farms classified as highly acidic were used by farmers cultivating more than 2 and Van de Walle, 2013). For Malawi and Nigeria, that used subsidized inputs against private inputs with pH <4.3 in the study). (Burke, 2012 cited in ha of land, while the actual majority of Zambian there is some evidence that communities with specifically because of late delivery and delivery in Jayne et. al, 2016). Crop productivity will not farmers (around 73%) cultivate less than 2 ha resident elected leaders or communities that are quantities that were different from what the farmer increase significantly with fertilizers, if soil fertility of land. (Mason et al 2013). In many countries geographically closer to the hometown of those was expecting. Similarly, a study in 2015 found that issues are not addressed through agronomic discrepancies between actual and planned leaders (e.g., MPs in Malawi and state governors late delivery reduced the technical efficiency of practices that improve soil organic matter. beneficiaries are due to political patronage, elite in Nigeria) receive significantly more subsidized maize yields by 4.2 percent (Namonje-Kapenbwa, Zambia’s conservation agriculture initiatives are capture within communities responsible for fertilizer on average, other factors constant Fisher et. al., 2015). a step in this direction but there is some concern targeting, or simple reselling of fertilizer vouchers and Kadiwa, 2014; Ricker-Gilbert and Jayne, 2011). that that shorter fallow periods and increased from small to larger farmers. In the Zambia FISP, Overall, there is mounting empirical evidence of 11. A public expenditure tracking study in Zambia expansion of land under production may lead to the rules virtually guarantee that larger farmers will the politicization of ISPs in SSA, but the nature of related to the 2010 FISP assessment3 found only lower soil fertility. disproportionately capture benefits, but to some the politicization varies across countries as well minor discrepancies in the quantities of inputs extent this program design may be due to the as within countries over time (Chinsinga and allocated to each district and the amounts perception that larger farmers will more effectively Poulton, 2014; Westberg, 2015). district authorities said they received. The analysis use the fertilizer. In reality, evidence from farm was not designed as an audit of tender winner survey data indicates that maize yield response 15. Empirical evidence from several SSA countries contracts, but the survey team found little evidence on small farms may actually be higher (Burke, et. suggests that social capital factors also leads to of systematic leakages at this level. However, as al., 2012). Thus, poor targeting is both inequitable “elite capture” of ISP benefits. In Tanzania, for the inputs moved from district warehouses to and an inefficient use of funds. There may also example, Pan and Christiaensen (2012) found that cooperatives and individual farmers, considerable be a perception that since large farmers produce 60% of the households receiving input vouchers evidence of leakage was reported. For 69 of 84 marketable surpluses while many small farmers contained a village official as a member. They cooperatives where corresponding data were produce mostly for self-consumption, more also found that households with elected officials available, the study found a 20% discrepancy in maize production by the latter does not increase and voucher committee members were 1.7 and 4 the inputs DACOs said they released and what the the quantity on the market. In fact, a ton of extra times more likely to receive input vouchers than cooperatives said they received. This is consistent maize produced and consumed on farm reduces households without such members. Similarly, with the findings of other studies, which found that household’s demand for maize purchases in evidence from Zambia and Malawi suggests that that in Zambia (and Malawi), diversion of 25- 35 the market, and so contributes just as much to households with links to traditional authorities are overall availability as an extra ton produced and more likely to receive input subsidies (Mason and sold by the larger farmer. Smale, 2013; Kilic, et. al., 2015) 3 Ministry of Agriculture could not account for 1,500 tons of fertilizer allocated as backstopping packs. 4 5 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM PART 1 | SUMMARY OF EVIDENCE AND LESSONS LEARNED ON INPUT SUBSIDY PROGRAMS (ISPS) IN ZAMBIA AND ELSEWHERE Share of subsidized fertilizer acquired in ZFISP by landholding quintile 19. The costs of input subsidy programs grow the rest of the spending on the agricultural sector, considerably over time and risk displacing other and in 2020 reached almost 2.5 times the rest development priorities. In many countries input of the sectoral budget and almost 3 times the subsidy programs have been very costly and can budget of the Ministry of Agriculture (Table 1). Quintile 2, 12% capture a large portion of the public budget at the Meanwhile, research and development – which has Quintile 1 (smallest), 6% expense of other priorities in agriculture such as consistently been shown to have very high rates Quintile 3, 16% infrastructure, irrigation, extension, and research. of return in many countries –receives 1 percent or Program costs are also often underestimated less of the total sectoral budget – 0.9%, 0.6%, and in budgets and grow over time. In Zambia even 1.0 percent in 2017, 2018, and 2019, respectively. Quintile 5 (largest), 41% though only the e-voucher part of the FISP budget is Other sector priorities – including investments in Quintile 4, 25% included in the Ministry of Agriculture’s budget, with irrigation development, agriculture infrastructure, the conventional FISP in the budget of the Finance land development, livestock development, and Ministry, high costs have important implications for agriculture technology – are each ranked higher Source: Jayne, et. al., 2016. other budget priorities in agriculture. In 2019, FISP than the FISP in the FNDP but have received spending more than doubled from 2018, exceeding significantly fewer resources. 16. In the 2010 assessment of Zambia FISP, alternatives, are left behind. regional targeting was shown to lack strategic Table 1: Key Budget Figures Kwacha focus. Thus far, the allocation of subsidized inputs 17. Input subsidy programs do appear to reduce across Zambia has been based primarily on the domestic grain prices, but the effect is very Budget FRA (a)/(c) FISP (b)/(d) Ministry of Total agricultural National budget year (a) (b) Agriculture (c) sector (d) (e) proportional distribution of farmers in each district. small. Input subsidy programs can play a role Because the most populous areas usually also in reducing imports and lowering grain prices, 2016 750,000,000 0.31 1,003,550,810 0.29 2,382,266,379 3,429,775,710 53,135,825,364 have the best developed private supply networks, but evidence indicates this has been relatively 2017 942,500,000 0.17 2,856,399,170 0.53 5,435,167,917 5,435,167,917 64,510,303,308 therefore, a large amount of the total subsidy has modest. Research on Malawi and Zambia indicate 2018 1,051,200,000 0.21 2,675,000,000 0.42 5,047,679,280 6,343,915,247 71,662,385,976 gone to areas where there is already reasonable retail maize prices are only between 1% to 4% 2019 672,000,000 0.16 5,609,000,000 1.05 4,160,914,514 5,321,190,088 86,807,894,727 private sector capacity while less populous lower as a result of the subsidy programs. (Arndt 2020 660,000,000 0.20 9,848,840,201 2.47 3,355,518,388 3,972,249,042 106,007,612,236 areas where farmers have fewer private sector et. al, 2014). Table 2: Key Budget Figures US$ B. Input subsidy programs – What are the unintended effects? Budget FRA FISP Ministry of Total Total agricultural year Agriculture agricultural sector national 18. Impact on Private Sector Development. before receiving FISP inputs compared with only sector budget Input subsidy programs can have disproportionate 43% who continued to use private supplies with 2016 72739264 97330063 231045736 332639146 5153414416 impacts on private sector development FISP support. In the past contracts for Compound depending on the channels used for procurement D have been awarded on a single-source basis to 2017 98846356 299569918 570022854 570022854 6765632229 and delivery of inputs. When the original input the parastatal firm, Nitrogen Chemicals of Zambia 2018 100569242 170772543 482915980 606928031 6856004399 subsidy program was launched, one of the two (NCZ). With Urea, most contracts since inception 2019 52045819 110597365 322259231 412121571 6723196382 main objectives of the FISP was to enhance the have been awarded to two private firms, Omnia 2020 36100096 608144510 183537174 217270562 5798310528 capacity of private companies to supply inputs Small Scale Limited and Nyiombo Investments to smallholder farmers. Despite the importance Limited, which also shared the contracts for Converted to US$ at annual average exchange rate for the year as indicated below. attached to private sector development, little Compound D with NCZ in the beginning. Private has been done by the Government of Zambia to networks are now stronger in many locations 2016 2017 2018 2019 2020 monitor the impact on private firms and people than when the program was launched and there 10.3108 9.5350 10.4525 12.9117 18.2825 interviewed for this study reported mixed effects is potential to better utilize the private sector on the private sector. For example, 50% of FISP within the subsidy program at retail and wholesale beneficiaries said they had bought private inputs procurement level. 6 7 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM PART 1 | SUMMARY OF EVIDENCE AND LESSONS LEARNED ON INPUT SUBSIDY PROGRAMS (ISPS) IN ZAMBIA AND ELSEWHERE 20. In addition, actual expenditures have improve soil quality. By incentivizing maize C. What do we know about the aggregate costs and benefits of sometimes exceeded planned by significant monocropping within seasons and over time, and margins. For example, the original budget for the by disincentivizing fallowing, Zambia’s FISP may input subsidy programs? 2007/8 FISP was US$ 39 million. Analysis of the be undermining the effectiveness of inorganic expenditure record shows that total core spending fertilizer distributed through the program. Thus, 23. Cost-benefit analysis has shown a few cases presents financial benefit cost analysis based amounted to US$ 43.62 million or 12% more than while ISPs aim to increase soil fertility, there may where benefits exceed costs, but in most cases on total cost of the subsidy programs. If the expected. When the extra-budgetary costs of staff be unintended negative consequences of the the converse is true. Research on returns to economic cost of the program is measured time, DAC meetings, and extras paid by farmers programs on the use of inputs or management input subsidy programs shows some evidence based on a lower incremental economic costs for cooperative membership and other fees are practices that are complementary to inorganic of positive returns, however it is limited. Analysis to government, benefit cost ratios are higher, with taken into account, the total cost of the 2007/08 fertilizer use. of Zambia’s FSP program in 2010 found positive the five year average ratio at 0.886 for Kenya, FISP is estimated to be at least US$ 47.43 million returns in outlying areas as measured by the value 0.765 for Malawi, 0.902 for Zambia. Another or 23% more than reported by the government. 22. Many ISPs have a ‘one size fits all’ design, with of incremental maize production when compared study of the Malawi program using data from a While similar exercises have not been undertaken potential adverse impacts. Traditional ISPs are to the costs of delivering relief maize in case of 2013 household survey estimated an economic for more recent budgets, it seems likely that this designed to support the production of one or a few emergencies. In contrast, a multi-country analysis benefit-cost ratio in the range of 0.29- 0.62, issue has persisted. crops, usually staples. In the case of Zambia, this of Kenya, Malawi and Zambia programs between depending on assumptions (Jacoby, 2016). And is maize. This design discourages diversification 2005 and 2010 show the ratio of financial benefits these do not account for the costs that the ISPs 21. ISPs may discourage good soil management and essentially locks farmers into production of to cost is consistently less than 1, and on average impose in crowding out other agricultural public practices. As noted above, soil quality (acidity and a low-value crop. The universal recommendation a bit over 0.5 – indicating that costs are close spending on activities and investments that have carbon content) is a major constraint limiting the in Zambia to apply a 4x4 maize input pack is to twice the value of benefits. The table below higher returns (see below). ability of farmers to raise productivity, particularly inconsistent with important differences in climate, from fertilizer use. But there is some reason to market access, and individual farmer objectives. In Table 3: Benefit–cost estimates for government fertilizer subsidy programs in Kenya, Malawi and Zambia think that subsidized fertilizers may increase the sampled areas, for example, 6% of beneficiaries the use of practices that are detrimental to soil said they relied more heavily on cassava for their Five year quality. ISPs that are designed to subsidize inputs staple food than maize and would likely have 2005/06 2006/07 2007/08 2008/09 2009/10 total for a specific crop obviously increase farmers’ benefited more from other kinds of support Financial Benefit-cost ratio of incremental benefits Kenya 0.800 0.792 0.378 0.631 0.602 0.524 incentives to grow that crop (Levine, 2014) also than the original FSP was designed to provide. In finds that ZFISP discourages inter-cropping and recent years, other crop seeds have been added (value of incremental maize output) to total government Malawi 0.524 0.622 1.110 0.340 0.435 0.555 crop rotation and encourages continually planting to the FISP package (Kuteya, et. al., 2016), but it program costs and incremental farmer costs Zambia 0.758 0.592 0.370 0.422 0.637 0.523 maize on the same plot. In addition, results from is not yet clear to what extent this has actually Source: Jayne, Mather, Mason, Ricker-Gilbert, 2013. Zambia suggest that the ZFISP discourages encouraged diversification. These programs also fallowing (Mason, Jayne, and Mofya-Mukuka, 2013; generally give farmers little choice of what kinds of Levine, 2014). High soil acidity and low soil organic inputs to use. In Zambia, the packages rely heavily 24. Larger studies on returns on public investment India indicates that while fertilizer subsidies matter levels on many Zambian smallholders’ on Compound D and urea, which is unsuitable in the agriculture sector show subsidy programs generated higher returns when first established in maize fields reduce fertilizer use efficiency but in large parts of the country where the soils are rank relatively low compared to other potential the 1960s and 1970s, returns declined in the 1980s intercropping, crop rotation and fallowing can acidic (Kuteya, et.al., 2016). investments. Long term analysis on returns to and 1990s. When compared to other priorities public expenditures is limited but evidence from fertilizer subsidies also ranked relatively low. 8 9 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM Table 4: Returns in Agriculture Growth and Poverty Reduction to Investments and Subsidies, India, 1960-2000 1960s Rank 1970s Rank 1980s Rank 1990s Rank Returns Returns Returns Returns Returns in Agricultural GDP (Rs produced per Rs spent) Road investment 8.79 1 3.80 3 3.03 5 3.17 2 Educational investment 5.97 2 7.88 1 3.88 3 1.53 3 Irrigation investment 2.65 5 2.10 5 3.61 4 1.41 4 Irrigation subsidies 2.24 7 1.22 7 2.28 6 na 8 Fertilizer subsidies 2.41 6 3.03 4 0.88 8 0.53 7 Power subsidies 1.18 8 0.95 8 1.66 7 0.58 6 Credit subsidies 3.86 3 1.68 6 5.20 2 0.89 5 Agricultural R&D 3.12 4 5.90 2 6.95 1 6.93 1 Returns in Rural Poverty Reduction (decrease in number of poor per million Rs spent) Road investment 1272 1 1346 1 295 3 335 1 Educational investment 411 2 469 2 447 1 109 3 Irrigation investment 182 5 125 5 197 5 67 4 Irrigation subsidies 149 7 68 7 113 6 na 8 Fertilizer subsidies 166 6 181 4 48 8 24 7 Power subsidies 79 8 52 8 83 7 27 6 Credit subsidies 257 3 93 6 259 4 42 5 Agricultural R&D 207 4 326 3 345 2 323 2 Source: Fan et al., 2007. PART 2 RECOMMENDATIONS FOR REDESIGNING AND STRENGTHENING ZAMBIA’S FISP 10 11 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM PART 2 | RECOMMENDATIONS FOR REDESIGNING AND STRENGTHENING ZAMBIA’S FISP 25. Based on the findings of the recommended likely to resolve all of the problems mentioned Box 1: What is the E-voucher System and how does it differ from traditional systems? independent evaluation, the transition to an above that are created by subsidy schemes e-voucher scheme to replace the DIS scheme but can mitigate some of them. A World Bank should continue with the goal of covering as report (forthcoming 2021) is a synopsis of the much of the country as technically feasible and experiences with e-voucher schemes, and some Use of e-Vouchers for agricultural input subsidy choices in both the inputs they use and the dealers deliveries has grown substantially in numerous Sub- from which they buy them. as quickly as possible. Analysis of the history of these have lessons that may be useful to Saharan African (SSA) countries since the start of of e-voucher systems in other countries and Zambia. Evaluations of ‘lessons learned’ from the the present decade. Nearly a dozen countries have Broadly speaking, traditional input subsidy schemes the current e-voucher pilot in Zambia suggests pilot in Zambia and of subsequent developments experimented with e-voucher mechanisms in some mainly relied on two kinds of mechanisms. One was a that e-vouchers can successfully be used to in the program4 found that despite the roll- way, and a wide variety of experiences and lessons reduction in purchase price of some specific product for learned have been gained regarding design, operations all buyers. The second was the provision to the farmer distribute FISP inputs to farmers, particularly out problems (particularly the late delivery of and scale-up approaches, platform technology of a fixed physical quantity (generally per hectare) of an in high potential agricultural regions. The use vouchers) the e-voucher scheme had significant choices, agricultural input and services subsidy support input or package of inputs, with the selection of type and of e-vouchers to administer subsidies are not benefits relative to the previous system: coverage, and subsidy payment systems. quantity of the inputs made by the government agency that administered the program. Individual recipients It should be recognized that e-vouchers are a of the subsidized inputs had no choice other than to A. It saved money on implementation costs. The savings were estimated to be 14.5 percent in direct mechanism for delivering input subsidies, but not take it or leave it. Under a voucher scheme, however, input subsidy policy per se. In most countries, farmers receive a voucher which entitles them to free or costs, and 27.4 percent including savings from undistributed or unloaded cards (Kuteya, et. al., e-voucher programs were introduced in order to subsidized (with co-payment) inputs when redeemed 2016). These savings to the national treasury can be re-allocated to other spending purposes, at least partially mitigate shortcomings of existing at an authorized dealer. In the past, some vouchers either within the agriculture sector, or to different areas of support to communities. subsidy programs, with the main objective to were paper, but recently schemes generally deliver the improve efficiency, ensure transparency, and reduce vouchers electronically thus the term “e-voucher”. The fraud in the distribution of subsidized inputs, actual technology for delivering the e-vouchers varies, B. It encouraged diversification, although the results were modest in the first year. About 15 percent thereby easing the graduation of farmers from with some countries using SMS (text messages) and of farmers used vouchers to purchase inputs other than maize seed and fertilizer, implying they getting subsidized inputs to purchasing inputs others some variant of a specialized debit card. E-voucher were producing something other than maize (Kuteya, et. al., 2016; World Bank, forthcoming at commercial prices. Voucher schemes can also programs also generally include capacity building of 2021). The area planted to non-maize crops increased by 18 percent, and an index of crop have advantages in encouraging development of a input dealers to ensure that they are able to participate competitive input market and giving farmers more efficiently in the redemption of the e-vouchers. diversity increased by 11 percent (Mason, et. al., 2018). Unfortunately, the MIS system used to monitor implementation did not have the capability of distinguishing different types of inputs, so there is no evidence on this issue for subsequent years. It is recommended that the Ministry of Agriculture takes action on this matter, as this needs to be corrected. 26. There has been some progress made in (Kuteya, et. al., 2016; World Bank, forthcoming C. It encouraged private sector development of input suppliers. The number of agro dealers transitioning to e-vouchers over the last 4 years, 2021) suggest some ways to do this. increased from 1,000 in the pilot phase to 5,000 in 2018 (World Bank, forthcoming 2021). The but it has been inconsistent (table below). There number of dealers trained under the training component of the program grew from 422 in the were some challenges in the roll-out of the pilot, 27. The Government’s current plan is to move 2015/16 season to over 1,360 in 2017/18. During the pilot, over 200 new point of sale devices and when the program went nationwide in 2017, from an 80/20 split of the program between were installed, most of them in district level dealerships (World Bank, forthcoming 2021). there were problems caused by poor connectivity the traditional DIS and e-vouchers in 2020 to and perhaps by lack of capacity of agro-dealers in 60/40 in 2021, 20/80 in 2022 and 100 percent D. While causality is complex, it appears to have raised farmer income, as e-voucher recipients gross some places. Obstructionism by parties with vested e-voucher by 2023. Whatever the eventual value of production was increased by 31 percent (Mason, et. al., 2018). interests in maintaining the old system also appears decision on the split each year, it needs to be to have played a role. After 2017 a decision was based on evidence from an objective evaluation E. It is understood that the launch of the e-voucher system will bring benefits to the majority of made to revert to the DIS in some districts, and of readiness in each region, as described in the farmers and those engaged in the agriculture sector, given that nearly 90 percent of Zambian the system continues to rely on the conventional summary of findings and recommendations in farmers are small scale. modality for a significant part of the FISP (table). But the first section of this paper. If such an evaluation the currently perceived technical and administrative indicates that some areas are still not ready for a bottlenecks of the e-voucher can be addressed in full transition to e-vouchers because of a lack of the short run, and the evaluation of the pilot and input dealers, a plan may be needed to prepare lessons from similar schemes in other countries them, as discussed below. 4 These evaluations include Kuteya, et. al., ; Mason, et. al.; and World Bank, forthcoming 2021. 12 13 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM PART 2 | RECOMMENDATIONS FOR REDESIGNING AND STRENGTHENING ZAMBIA’S FISP 30. More specifically on the gender dimension, their own production. Furthermore, it would be Table 5: FISP budget (Kwacha) the implementation of the e-voucher system a mistake to assume that all ‘small’ farmers are 2016 2017 2018 2019 2020 will allow better targeting of women farmers. non-commercial; quite small plots of high-value With gender being one of the selection criteria of crops (e.g., vegetables or fruits) can be viable for FISP Conventional 1,155,220,645 0 441,000,000* 2,969,500,000* 7,879,200,000* beneficiaries eligible for the e-voucher, women commercial production. Hence farm size should FISP e-voucher 748,330,165 2,856,399,170 1,344,000,000 2,692,300,000 1,969,800 farmers will have access to the required inputs not be used as the primary criterion to exclude for their farming. Additionally, given that women farmers from the program. This notwithstanding, Notes (i)* May not include contractual obligations to transporters and input suppliers which was done by the Ministry of Finance; (ii) 2019 figures based farmers usually engage in cultivating horticulture there may well be some owners of small plots on split in original budget, applied to the adjusted actual budget, since figures on the split in adjusted budget are not available; (iii) 2020 figures are crops, the e-voucher system will help them in rural areas who are not farmers and will not based on actual 2020 total FISP budget, with the 80/20 split, as planned in the government’s strategy. Actual split in 2020 budget is not known. acquire their specific needs, compared to the be able to use the subsidy in any case. It would traditional DIS. be reasonable to exclude these households from 28. The full benefits from using e-vouchers inputs on crops and production technologies input subsidies, and where appropriate to include in enhancing farmer productivity and private of their choosing. In addition, introduction of 31. Who should benefit from e-vouchers? The them in some other programs of support, such sector development in input supply are achieved several PIN numbers on e-vouchers, which is subsidy needs to be given to smallholders who as the ‘kitchen gardening’ program or a general only when they offer flexibility in choosing a technically easy task, would permit farmers can translate inputs into higher production and safety net cash transfer program. Development inputs and sellers of these inputs (agro-dealers). to have more bargaining power against local incomes. It implies the exclusion of large farms, of eligibility criteria at the lower end of the farm The rules of the vouchers should allow them agro-dealers, while inducing competition and for example, those with cultivated area above size spectrum deserves further consideration, to be used to purchase a wide range of inputs providing farmers’ access to a larger variety of 5 ha. This will not result in significant budget based on evidence regarding characteristics of from dealers of their choosing, and to use the affordable inputs over time. savings because there are only about 43,000 such very small farms. farmers, or 3.4 per cent of total. But it will insert Box 2: How did the e-voucher system perform discipline into the public support, confirming the 33. In addition, and consistent with the commitment to assist smallholders, given that underlying objective of input subsidies (whether large farms can afford to buy inputs commercially. administered by vouchers or direct supply) Reallocating a greater proportion of FISP subsidies the programs should have some limits on the In general, they have demonstrated advantages over the policy choices (e.g., allowing farmers no or few options traditional programs. But in some countries, there have as to which inputs to buy with their vouchers). But to relatively small farms would more effectively eligibility period for individual farmers. The major been ‘teething pains’. Problems have included poor over time, mobile networks have spread, dealers have reduce rural poverty. As noted above, the argument justifying subsidies is that farmers do not connectivity in remote areas and high setup costs. And become more numerous, the fixed costs of setting up evidence shows that small farmers are as efficient use modern inputs because they are not aware of in some countries, not all of the potential benefits have the systems have already been incurred so costs per unit as larger farmers in their use of fertilizer. So, this the benefits. So, the subsidies will allow them to been realized, either because of circumstances (e.g., have fallen, and experience has allowed other problems sparse dealer networks, limited choices for buyers) or to be resolved. reallocation – in addition to more effectively try these production technologies at low cost and alleviating poverty – could reduce households’ demonstrate to them these benefits, while also net maize purchasing dependence on the market increasing profitability. They will then be convinced for food and conserve more maize supplies for of the benefits of purchasing the inputs and can 29. In addition to the advantages mentioned can be targeted directly to beneficiaries, not urban consumers without adversely affecting total afford to do so. By this logic, once the farmers have above, another major gain from e-vouchers is cooperatives, based on the electronic database national production. received the subsidized inputs for several years, the ability to establish a clear list of intended with information on: (i) name of the head of the the subsidies no longer serve this purpose. They beneficiaries and monitor whether benefits are household; (ii) National Registration Card; (iii) 32. The question of eligibility of very small farmers may still serve to support incomes of beneficiaries, transferred to them, improving efficiency and number of household’s members; (iv) gender of is important, yet complex. Some proposals but this is more efficiently accomplished as a accountability. Under the FISP, the key selection the head of household; and (v) total cultivated would exclude such farmers on the basis that direct transfer through a safety-net program or criterion is a membership in cooperatives. There area. Once such a database is established, they do not use fertilizer efficiently or that they area-based, decoupled support payment (see have been many ‘local solutions’ to this open- and it can be done within several months, the are non-viable or that their production does not the Annex). It makes sense to limit eligibility for ended criterion, resulting in small benefit to Government will know whom to select and contribute to the supply since they consume each farmer to a period of a few years, which also smallholders and the capture of the 90 per cent can be sure that vouchers are able to reach the it on farm. The evidence described in para. 14 allows more farmers to benefit over time from a subsidized inputs for farms above 5 ha. E-vouchers intended beneficiaries. above, however, indicates that small grain famers given budget for the program. Existing (and past) do use fertilizers efficiently and do (indirectly) versions of the subsidy programs have recognized increase market supply, even if they consume this, and the rules have included some limits on 14 15 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM PART 2 | RECOMMENDATIONS FOR REDESIGNING AND STRENGTHENING ZAMBIA’S FISP the period of eligibility, but these ‘graduation’ an inefficient way to give farmers handouts. The modifications in light of the lessons learned. The e-voucher system have in the past been plagued rules have not been enforced consistently. This change in the instruments used should be clearly operational details of a voucher program including by delays in delivery. With the e-voucher, systems needs to change; otherwise, the subsides are just communicated to beneficiaries. procedures for beneficiary targeting would need need to be developed to ensure that funds are careful consideration to achieve maximum impact. always released to agro-dealers in a timely manner, Box 3: Zambia’s experience with e-vouchers, and what value they can bring Rather than operate a parallel system that does which has sometimes not been the case. not involve local retailers; however, such an arrangement would be a direct way to support 36. In addition to accelerating the transition to genuine capacity development as the original FISP e-vouchers, other measures would improve the In Zambia, there were also teething problems in the purchased other farm inputs. (Mason et al.) suggest set out to do. Both the conventional FISP and the overall impact of the FISP: E-FISP rollout, some due to legitimate technical that the 2015/16 e-FISP pilot spurred greater crop concerns (poor connectivity in some regions, platforms diversification and possibly an increase in the percent unable to identify what kinds of inputs farmers bought), of households using fertilizer relative to the conventional some apparently due to attempts to undermine the FISP. And the program in general was widely used: of the A. First, institute a clear and workable system for monitoring and evaluation. The Government system by vested interests, and some due to poor 804,260 targeted delivery (2017/18), 716,161 accessed may wish to introduce a routine system for rigorous auditing, accounting, and results-based performance by the Government. One example of the inputs. The program also supported growth in the number monitoring and evaluation of the FISP. Presently, there is no system to track FISP spending in any latter is delays in launching the program for the 2016/17 of agro-dealers in the country from 1000 (pilot phase) to one agricultural year or to report on other performance indicators needed to make informed season due to late government funding, which had 5000 (2018, the end of the PRMC review period). And some significant consequences, such as the e-cards the number of trained agro-dealers engaged in input budget and policy decisions. A robust monitoring and evaluation system would promote greater being distributed late (towards the end of December distribution was reported by IAPRI to have increased information dissemination and analysis of program implementation, costs and impacts as an 2016). This meant that farmers were unable to redeem from 422 in the 2015/16 farming season, to about aid to policy and decision makers. The Government of Zambia may wish to consider engaging inputs ahead of the growing season, resulting in lower 1,364 in the 2017/18 season. The system also improved farmers themselves in the monitoring. Participatory monitoring has been used in many sectors maize yields (Mason, et. al.) There were also issues beneficiary targeting by establishing a consolidated and with input cost uncertainty: in the 2015/16 pilot, rising harmonized national database of beneficiaries on the and countries as a way to increase social accountability. The Farmer's Report Card developed fertilizer prices due to the depreciation of the Zambian ZIAMIS system, which can help strengthen beneficiary as part of the 2010 World Bank survey offers a practical starting point for Zambia to track the kwacha nearly made the e-voucher less attractive when targeting by the Government for programs besides the FISP's performance. compared to the traditional FISP. The Government had FISP. Through the registration process the system was to top-up the value of the voucher from 1,400 to 2,100 rid of ineligible beneficiaries. In addition, each farmer kwacha, inclusive of farmer contribution of 400 kwacha. was able to redeem inputs on his/her own time as an B. Second, continue to improve selection criteria so that they are meaningful, easy to apply, and All these issues need to be addressed in the future. individual, unlike the ordinary FISP, in which farmers had verifiable. Efforts have been made to improve beneficiary selection through cooperatives, which to collect inputs as a group. And the evidence is that the have addressed the very loose selection criteria in the past which contributed to the FISP being But, notwithstanding these initial problems, there is e-FISP was significantly less costly to administer than the pulled in different directions and diluting its potential strategic impact. New selection criteria and evidence that the program was largely successful in traditional FISP (1,748 ZMW GRZ spending per farmer for many of its objectives. In the 2015/16 pilot, evidence E-FISP compared to 2,057 ZMW per farmer on traditional procedures should be consistent with the objectives of the program and should establish clear, indicates that farmers started to diversify: about 85% FISP). If the Government continued to support 1 million easy to apply, and verifiable beneficiary selection criteria. Problems of late delivery should be of the farming households redeemed their vouchers farmers through e-FISP, it would save over ZMW 309 addressed, as should gaps between farmers' expectations and actual deliveries. for fertilizer and maize seed, while the remaining 15% (US$ 14 million) annually over the traditional FISP. C. Third, put in place rules to ensure transparent and competitive procurement procedures for purchases of inputs to supply the DIS, which would improve efficiency and reduce opportunities 34. During the transition to full reliance on and in particular, to include more appropriate for corruption and diversion. e-vouchers, and in recognition of different levels fertilizers in areas of acidic soils, and to give of supply network development in different farmers more choices of crops. Even in these D. Fourth, review (and revise as necessary) regulations associated with inputs, in order to ensure that areas, it has been necessary to temporarily split areas, pilot voucher schemes could be initiated in farmers are getting the products and quality they are expecting (e.g., through truth in labeling the FISP into two sub-programs with different ways that would attract private input suppliers and requirements), but that their options are not restricted by regulatory requirements for lengthy and implementation arrangements. In areas where would eventually allow a full transition to vouchers expensive testing for new varieties and products. In some countries, such testing requirements private networks are severely constrained or instead of direct supply. impose onerous costs and serve as a strong deterrent to transfer of technology from abroad electronic infrastructure is inadequate, direct and its development domestically. supply through something like the existing DIS 35. In non-remote areas, the e-voucher system may still be required in the short term. approach should immediately and fully replace However, input packages would need to be the conventional DIS. This would presumably adjusted to better reflect the needs of each area, build on the e-voucher pilot, making appropriate 16 17 PART 3 | RECOMMENDATIONS ON COMPLEMENTARY MEASURES TO ENHANCE GRAIN PRODUCTION AND MARKETING 37. The proposed changes to the FISP including 40. Priority should be given to implementation of the use of the e-voucher system, and other the Zambian Grain Information System (ZAGIS), enhancements, are necessary but insufficient and a regulatory framework for warehouse to increase and maintain competitiveness of the receipts. Lack of reliable information on grain prices maize sector. Reforms to the FISP will produce and other market conditions is another problem a supply response that will increase the volume that is suppressing private sector development. This of grains in commercial channels for domestic would include a system to track grain stock levels markets and exports. Additional public goods and and locations in real time, which would help both policy adjustments are required to further reduce the private sector and the government respond production and marketing costs to ensure that more effectively to the needs of consumers the reforms have maximum impact. The budget and producers. The ZAGIS would probably be savings generated from the Food Reserve Agency organized most effectively as a public-private and FISP reforms can be used to finance the partnership. A system of warehouse receipts, necessary agricultural public goods. as done in many countries, supports the fluid operation of grain markets and improves access to 38. The complementary public investments credit. The development of this system requires an such as access to irrigation, and expanding appropriate regulatory framework which, among the use of resilient seeds, include numerous other things, clearly defines the rules governing measures to reduce production costs, and its operation including ownership rights over the improve productivity. While a full listing of such stored commodities. This is currently lacking in productivity-enhancing investments is beyond Zambia but should be high on the policy agenda. the scope of this brief paper, high priority items would include upgrading research and extension 41. Investments in reducing marketing costs aimed at increasing adoption rates and efficiency are also needed, particularly road upgrades. of input use, investments in sustainable land Overall, Zambia does not compare well to other management (to harvest water, to increase soil countries in Africa with respect to road networks productivity and in particular yield response to (table below). For example, only 25 per cent fertilizer, and to adapt to climate change), and of unpaved roads in Zambia are in good or fair land registration (to improve security of tenure condition compared to 58 per cent in the middle- and thereby promote investments). income African countries. Priority should be given to an improved maintenance of existing PART 3 39. Reforms in the Food Reserve Agency are roads, for example through private management likewise necessary to improve the functioning (with output and performance-based road of grain markets. The policies and actions of the contracts) paid by the public sector (public- RECOMMENDATIONS ON FRA clearly have a major impact on many aspects private partnerships) as implemented under the of grain markets. As has been the case with many Agricultural Development Support Project. It will COMPLEMENTARY MEASURES TO ENHANCE agencies with similar structures and mandates not only reduce marketing costs but also promote in other countries, the performance of the FRA private jobs in rural areas. Better connecting rural GRAIN PRODUCTION AND MARKETING has been problematic in a number of ways, and areas with “high agricultural potential” to trunk a multi-faceted reform program is necessary roads and consumption centers is also needed, to make this sector efficient and competitive through investments in new rural roads, after throughout the value chain. The functioning of feasibility work ensures that there would not be the FRA and options for reforming it are outside overinvestment. Investments in rail will also help the purview of the current report but are treated to increase competition with truckers and reduce in a separate parallel piece. transportation prices overall. 18 19 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM ANNEX A1 | TURKEY'S SECTORAL TRANSFORMATION THROUGH ADOPTION OF BETTER INSTRUMENTS OF SUPPORT ANNXES Table 6: Zambia’s road indicators benchmarked against Africa’s low- and middle-income countries Middle- Resource income Unit rich Zambia countries Annex A1: Turkey’s Sectoral Transformation through Paved road density km/1,000 km2 of arable land 97.6 56.3 146.8 Adoption of Better Instruments of Support Unpaved road density km/1,000 km2 of arable land 128.2 95.0 257.8 GIS rural accessibility % of rural population within 2 km from all-season road 19.7 16.8 22.9 Overengineering of network % of main road network paved despite low traffic volumes 15.0 65.0 20.0 As in Zambia currently, Turkey in the late 1990s and • Drastically reduced transfers to SOEs and early 2000s experienced severe macroeconomic the agricultural service cooperative unions Paved road traffic Average annual daily traffic 1,408.2 736.6 2,558.3 imbalances. Also in common with Zambia, (ASCUs). The ASCUs were not genuine Unpaved road traffic Average annual daily traffic 54.2 45.2 14.9 subsidies to the agricultural sector were large farmer cooperatives, but rather agents of the Paved network condition % in good or fair condition 67.9 83.0 82.0 and growing, exacerbating the problem, so government, run in a top-down manner. The Unpaved network condition % in good or fair condition 61.4 25.0 57.6 agricultural policy reforms were necessarily part reforms reduced the inefficient interventions Perceived transport quality % firms identifying as major business constraint 27.4 10.6 4.8 of the solution. While the reform program was not of SOEs and forced restructuring of ASCUs to easy to design and implement – neither politically make them real member-operated coops. The Source: Gwilliam and others 2008, derived from AICD national database downloadable from http://www.infrastructureafrica/aicd/tools/data nor technically – it not only helped prevent a program provided technical assistance and a Note: GIS = geographic information system. meltdown of the economy, but also set the stage revolving fund to help ASCUs restructure and for long-term agricultural growth. This is the story to pay severance for the estimated 12,000 42. The competitiveness of the maize sector (IAPRI, 2019). Rationalization and reform of all these of how that happened. employees that were redundant, out of depends on consistency and liberalization of requirements to make it easier and cheaper to buy, 16,500 total staff. policies affecting agricultural trade, both domestic sell, and transport grains would make for a more Turkey suffered from a series of economic and international. For all types of agriculture efficient market. crises in 1999–2001, with hyperinflation and • Provided grants to support farmers in commodities and farm inputs, Zambia imposes total GDP contraction of 4.7 percent in 1999 transitioning away from crops that had been and encounters a wide variety of requirements that 43. In addition to the burden on trade created by and 7.5 percent in 2001. Agricultural policy was heavily supported (hazelnuts and tobacco) but make international trade difficult and expensive. the regulatory requirements and levies, a second a significant contributor to the economic mess. had support drastically reduced. Phytosanitary and other trade rules do serve set of constraints relates to Zambia’s long history All major products had price supports of some legitimate purposes in terms of protecting human, of explicit agriculture trade restrictions and trade kind, through purchases by state-owned entities • Provided all farmers with an area-based animal, and plant health, yet in practice, there are bans. This is particularly relevant to maize, which (SOEs) agricultural sales cooperative unions (decoupled) support payment (Direct Income frequent institutional overlaps whereby traders are is inherently vulnerable to drought and subject to (heavily political and grossly inefficient), and/or Support, DIS) to cushion the shock of the required to obtain certificates for the same or similar many political pressures as the main smallholder tariffs on imports. Virtually all significant inputs removal of other supports. things from different agencies and to submit their crop and daily staple for almost the entire nation. were subsidized, with the highest subsidies for goods for duplicate inspections. In some cases, the Despite the seeming benefit of trade bans, credit and fertilizer. Fiscal transfers to agriculture As a result of these policy measures, the Producer certificates traders are required to obtain do not even these policies are seldom successful and have amounted to around 3 percent of overall GDP, Support Estimate (a measure of support used by correspond with actual buyer requirements, thereby several long-term costs for poverty reduction, making agriculture a huge fiscal burden, and the OECD to indicate how much support is given leading to complaints from the private sector that job creation, and competitiveness. Export bans, moreover, the high prices of basic foods taxed to agricultural producers) fell from 24 percent the process is more about raising revenue for the especially in the presence of large buffer stocks, poor consumers. Despite this enormous volume of total agricultural production value in 1999 to certifying body than actual trade facilitation or safety. are not justifiable and do enormous damage to of support, agricultural growth was very low 10 percent in 2002. A measure of trade policy Such costs can be a particular disadvantage to small the country’s reputation as a reliable exporter. In and declining in the 1990s because the support support, the Nominal Protection Coefficient, fell traders who often lack the necessary economies addition, the trucking industry will not expand mechanisms were so inefficient. from 1.24 to 1.10. Consumers benefited greatly, as of scale and information needed to navigate the their fleet if the trade policy remains inconsistent. the Consumer Subsidy Estimate (how the OECD procedures and participate in formal sector markets. The Government needs to make a credible Starting in 2000, the government implemented measures the benefit or tax on consumers from The market is further burdened by various levies commitment to not restrict or tax exports, with an economy-wide reform program, including big measures supporting agriculture) fell from –22 and paperwork requirements for moving grain perhaps some clearly defined exceptions in case changes in the agricultural support programs. percent (that is, an implicit tax on consumers of from one jurisdiction to another within the country of a true emergency. The changes: 22 percent) to –8 percent. 20 21 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM ANNEX A1 | TURKEY'S SECTORAL TRANSFORMATION THROUGH ADOPTION OF BETTER INSTRUMENTS OF SUPPORT The fiscal savings over time from reduction of the inefficiencies and leakages in the previous around US$90 per hectare. This had to be adjusted 106,935 dispersed parcels covering 65,700 hectares each of the transfers is displayed in Figure A1.1. system had been so large, the loss in actual periodically to keep them close to this level in real into 50,625 parcels, all with road access. As shown in Table A1.1, the overall savings to transfers to farmers was estimated at only around terms, but the real value did change somewhat the Government budget amounted to about US$2.78 billion. Of this, the new DIS payments over time, as the adjustment was not exact. In the What were the results of the reform program for US$4.28 billion per year. However, because compensated for US$ 1.25 billion. short term, the DIS replaced about 40 percent of agricultural growth? The immediate impact of the the income that farmers lost from other aspects of removal of subsidies and protection was a decline the reform program. Like PROCAMPO – a similar in product prices of 13 percent on average, and Figure A1.1. Fiscal savings over time from a reduction in transfers program in Mexico -- this helped make it politically a similar percentage increase in input prices. Not feasible for the government to implement the surprisingly, crop production and value added reforms. Also similar to PROCAMPO, econometric fell (Figure A1.2), but not by much (only about 2 3,500 estimates showed that each $1 in DIS generated percent). This was followed by a rapid recovery 3,000 around $2.5 in agricultural value added for farmers. and acceleration of productivity and growth. US$ Thousands 2,500 2,000 Value-added per agricultural worker (a good 1,500 As a side benefit, land registration for the DIS measure of incomes of workers) increased from 1,000 system created a National Registry of Farmers and a bit over $4,000 before the program to just under 500 completed the land cadaster, setting the stage for $7,000 about 12 years later (Figure A1.3). Cereal 0 future EU accession (which did not happen for yields per hectare, which had been stagnant all 1999 2000 2001 2002 unrelated reasons), as well as creating prerequisites through the 1990s, increased rapidly—about 50 for modern banking and real estate systems. It also percent in this period (Figure A1.4). The same is Transfers to agricultural SOEs Transfers to ASCUs Direct output subsidies facilitated a program of land consolidation of isolated true of food production, which increased by close Input subsidies Credit subsidies Direct income support parcels of land, which eventually consolidated to 45 percent (Figure A1.5). Table A1.1. Scope and incidence of agricultural subsidy phase out in Turkey, 2002 compared to 1999 (US$ millions) Figure A1.2 Agriculture value added, Turkey, 1990–2012 (constant 2005 US$) Transfer/subsidy Decline in Change in agricultural Efficiency 6E+10 fiscal cost transfers to farmers gain 5E+10 Transfers to SOEs 3,063.0 -1,026.0 2,037.0 4E+10 Transfers to ASCUs 814.2 -325.7 488.5 3E+10 Direct output subsidies -23.6a 23.6 0.0 2E+10 Input subsidies 290.2 -290.2 0.0 1E+10 0 Credit subsidies 1,387.6 -1,110.1 277.5 1990 2 4 6 8 2000 2 4 6 8 2010 12 Subtotal 5,531.5 -2,728.4 2,803.0 Source: World Bank staff estimates from Government statistical data. Direct Income Support -1,246.7 1,246.7 0.0 Total 4,284.8 -1,481.8 2,803.0 Figure A1.3. Agriculture value added per worker, 1990–2012 (constant 2005 US$) Source: World Bank staff calculations. 8,000 7,000 The DIS system was begun in 2001/02, following payments were made to them or to their tenants. 6,000 a pilot land registration program in 2000/01. The That is, if payments were made to tenants, the 5,000 pilot led to the decision to make payments to landowners would simply charge higher rent. All 4,000 3,000 landowners, rather than users, since it revealed agricultural land was eligible for DIS payments, 2,000 that payments to users was simply infeasible. It and the program covered about 2.75 million 1,000 was also clear that the landowners would be the farms. Payments were denominated in Turkish 0 main beneficiaries, regardless of whether the lira and were originally set at the equivalent of Source: World Bank staff estimates from government statistical data. 22 23 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM ANNEX A2 | SMARTER SUPPORT TO AGRICULTURE: DECOUPLED DIRECT PAYMENTS Figure A1.4 Cereal yields (kg/ha), 1990–2012 Annex A2: Smarter support to agriculture: Decoupled direct payments 3,500 3,000 2,500 In the 1980s, when it became widely recognized support payment a farmer receives, income 2,000 that policy instruments such as input subsidies, increases by $1.50 – $2.60. 1,500 output price supports, and trade policy were 1,000 500 relatively inefficient mechanisms for supporting • Assist in budget planning; the budget 0 agriculture, and that they also created large required is known in advance and does not distortions in global trade, some important depend on factors that cannot be forecast countries began to move away from these with certainty, such as production or input Source: World Bank staff estimates from government statistical data. mechanisms. This shift was motivated partially purchases by farmers. by multilateral trade negotiations, which correctly Figure A1.5 Food production index, 1990–2012 (2004–06=100) viewed these mechanisms as antithetical to • Are classified as ‘Green Box’ subsidies under the goal of reducing barriers to international WTO rules because they cause minimal 140 trade, and therefore disciplined their use. It also distortion and are generally acceptable in free 120 stemmed from the desire of national governments trade agreements. 100 to get more for their money. The United States 80 60 and EU reformed their support systems, reduced • Have the co-benefit of presenting a good 40 the use of these mechanisms, and began to use opportunity to improve the land registration 20 ‘decoupled’ payments to farmers (payments that and cadaster systems on which they depend. 0 are not linked to input use and output production). The payments in most cases are determined by the • Can reach small, poor farmers who benefit area farmed in some baseline period. Importantly, little, if at all, from traditional input and Source: World Bank staff estimates from government statistical data. they do not require farmers to continue to grow output-based subsidies, because they use any particular crop to receive the payment. very limited purchased inputs and/ or do not market their output. Decoupled support programs have important advantages over mechanisms that provide support • Can be used to encourage good farming through input subsidies, output price supports, or practices. For example, payments can be tied trade policy. These programs: to actions by the farmer that have positive externalities, such as environmental benefits. • Do not encourage farmers to grow products This has been done in many countries in one that they cannot grow at a competitive way or another. cost, or for which there is no market. They thus promote much more efficient use of This same model has been adopted in one form a country’s limited resources of land, labor, or another in other countries, including Mexico’s capital, and other inputs. PROCAMPO program and Turkey’s Direct Income Support (DIS), which is described in more detail • Provide a greater impact in relation to in Annex 1). This reform greatly assisted Mexico spending ex post analyses found an income in negotiating the North America Free Trade multiplier of 1.5–2.6 for decoupled payments Agreement with the US and Canada, and in Turkey in Mexico’s PROCAMPO program and a was a key element in the country’s overall reform similar impact for Turkey’s DIS program. This that prevented an economic meltdown and set means that for every dollar of government the stage for a period of high growth. One lesson 24 25 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM REFERENCES REFERENCES from Mexico and Turkey is that reforms may be There will always be concerns about the more politically feasible (the ‘spoonful of sugar distributional aspects of these kinds of programs— helps the medicine go down’ principle) with partial for example, the fact that large farmers get higher compensation of the losers. Mexico’s PROCAMPO payments. These concerns can be vitiated to program had this explicit purpose. It replaced very some extent by designing the program so that inefficient subsidies with a less distortionary and overall payments to a farmer are capped, and/ Arndt, C., K. Pauw, and J. Thurlow. 2015. “The Jayne, T.S., D. Mather, N.M. Mason, and J. Ricker- more efficient ‘decoupled’ area-based payment or by graduating the payments, so that smaller Economy-wide Impacts and Risks of Malawi’s Gilbert. 2013. “How Do Fertilizer Subsidy system to compensate farmers partially for the farmers get higher payment per hectare. In any Farm Input Subsidy Program.” American Journal Programs Affect Total Fertilizer Use in Sub- income lost when the CONASUPO price supports case, a well-designed decoupled payment system of Agricultural Economics 98 (3): 962–80. Saharan Africa? Crowding Out, Diversion, were abolished. has less regressive distributive effects than input and Benefit/Cost Assessments.” Agricultural subsidies or output price supports and is much Burke, W.J., T.S. Jayne, and N.J. Sitko. 2012 Economics 44(6):687-703. Another useful lesson is that the legal framework more efficient. (March). “Can the FISP More Effectively should allow payments to be capitalized and Achieve Food Production and Poverty Jayne, T. S., D. Mather, N. M. Mason, J. Ricker- pledged as collateral, and/or allow farmers to Before rolling out this kind of support mechanism Reduction Goals?” Policy Synthesis No. 51, Gilbert, and E. Crawford. 2015. “Rejoinder to the choose to get several years of payments in a nationwide, some countries have piloted it in Food Security Research Project – Zambia, Comment by Andrew Dorward and Ephraim lump sum. PROCAMPO began to offer this option limited areas, though some have not. If the Ministry of Agriculture and Cooperatives, Chirwa on Jayne, T. S., D. Mather, N. Mason, after recognizing that its omission undermined Government of Zambia would consider running Agricultural Consultative Forum, Michigan and J. Ricker-Gilbert. 2013. ‘How Do Fertilizer farmers’ ability to make productive investments. a pilot program, a first step would be to define State University. Lusaka. Subsidy Programs Affect Total Fertilizer Use in Decoupled support programs should also issue the criteria for eligibility. Some potential criteria, Sub-Saharan Africa? Crowding Out, Diversion, Chinsinga, B., and C. Poulton. 2014. “Beyond payments early enough in the crop year to allow for example, could include crop(s) produced by and Benefit/ Cost Assessments.’ Agricultural Technocratic Debates: The Significance farmers to use them to buy inputs. the farm, size of farm, or level of sophistication Economics 44 (6): 687–703.” Agricultural and Transience of Political Incentives in the of the farm. It is important to note, however, that Economics 46 (6): 745–55. Malawi Farm Input Subsidy Programme.” It is important for all potential beneficiaries to connection of the support to particular crops Development Policy Review 32 (S2): 123–50. Jayne, T.S., N.M Mason, W.J. Burke, and J. understand the program, so careful attention eliminates much of the advantage of decoupled Ariga. 2016. “Input Subsidy Programs in Dorward, A., Chirwa, E., (2011). The Malawi needs to be given to information dissemination. payment. If, for example, the payments were Sub-Saharan Africa: A Synthesis of Recent agricultural input subsidy programme: Long after PROCAMPO started, surveys indicated only available to maize farmers, they would be Evidence.” International Development 2005-6 to 2008-9. International Journal of that many producers still incorrectly believed ‘decoupled’ from input use, but would still be Working Paper 145. Michigan State Agricultural Sustainability 9(1): 232-247 that they needed to continue planting staples to coupled to outputs. This would be equivalent to University, East Lansing, Michigan. receive payments. This perception was contrary a substitution of a cash payment for the crop- Fan, S., A. Gulati, and S. Thorat. 2008. Investment, Kilic, T., E. Whitney, and P. Winters. 2015. to one of the program’s main goals, which was specific input subsidy. While this would enhance Subsidies, and Pro-poor Growth in Rural India. “Decentralised Beneficiary Targeting in to encourage diversification. Later still, when the farmer welfare, and could potentially have some Agricultural Economics, 39(2): 163–170. Large-Scale Development Programmes: program’s rules were changed to allow payments of the benefits of decoupled payments – e.g., Fisher, M., and V. Kandiwa. 2014. “Can Agricultural Insights from the Malawi Farm Input Subsidy for meeting environmental goals, another survey it could be used to encourage environmentally Input Subsidies Reduce the Gender Gap in Programme.” Journal of African Economies found that farmers were generally not aware of friendly farming practices – it would not encourage Modern Maize Adoption? Evidence from 24 (1): 26–56. the change, which prevented them from taking farmers to diversify and produce products that Malawi.” Food Policy 45 (May): 101–11. advantage of this opportunity. they can produce most efficiently. Kuteya, A. N., C. Lukama, A. Chapoto, and V. Goyal, A. and J. Nash. 2017. Reaping Richer Malata. 2016 (October). “Lessons Learnt from Returns: Public Spending Priorities for African the Implementation of the E-voucher Pilot”, Agriculture Productivity Growth. Washington, Lusaka: Indaba Agricultural Policy Research DC: World Bank and Agence Francaise de Institute Policy Brief No. 81. Developpement. Levine, N. K. 2015. “Do Input Subsidies Crowd In Indaba Agricultural Policy Institute. 2019 or Crowd Out Other Soil Fertility Management (December). “Zambian Maize Outlook and Practices? Evidence from Zambia.” Michigan Regional Analysis 2019/ 2020”. Issue 5. State University, Lansing, MI. (http://www.iapri.org.zm) 26 27 ZAMBIA'S FARMER INPUT SUPPORT PROGRAM AND RECOMMENDATIONS FOR RE-DESIGNING THE PROGRAM Marenya, P., and C. Barrett. 2009. “State- Fertilizer Affect Smallholder Maize Productivity Conditional Fertilizer Yield Response on and Production?” Selected paper presented Western Kenyan Farms.” Am. J. Agr. Econ. at the Agricultural and Applied Economics 91(4):991-1006. Association and Western Agricultural Economics Association Annual Meeting, Mason, N.M., and T.S. Jayne. 2013. “Fertiliser San Francisco, CA, July 26–28. (Nationally Subsidies and Smallholder Commercial representative cross-section; stochastic Fertiliser Purchases: Crowding Out, Leakage frontier model.) and Policy Implications for Zambia.” Journal of Agricultural Economics 64(3):558-582. Pan, L., and L. Christiaensen. 2012. “Who Is Vouching for the Input Voucher? Decentralized Mason, N.M., T.S. Jayne, and R. Mofya-Mukuka. Targeting and Elite Capture in Tanzania.” World 2013. “Zambia’s Input Subsidy Programs.” Development 40 (8): 1619–33. Agricultural Economics 44(6):613-628 Ricker-Gilbert, J., and T. S. Jayne. 2011. “What Mason, N. M., T. S. Jayne, and N. van de Walle. Are the Enduring Effects of Fertilizer Subsidy 2013. “Fertilizer Subsidies and Voting Patterns: Programs on Recipient Farm Households: Political Economy Dimensions of Input Evidence from Malawi.” Staff Paper 2011– Subsidy Programs.” Selected paper presented 09, Department of Agricultural, Food, at the Agricultural and Applied Economics and Resource Economics, Michigan State Association’s(AAEA) and Canadian Agricultural University, East Lansing, MI. Economics Society Joint Annual Meeting, Washington, DC, August 4–6. Schuman, I. and S. Mink 2021 (forthcoming). Assessment of the Design and Delivery Mason, N. M., and M. Smale. 2013. “Impacts Mechanisms of E-Voucher Schemes in of Subsidized Hybrid Seed on Indicators of Agriculture in Sub-Saharan Africa. World Bank. Economic Well- Being Among Smallholder Maize Growers in Zambia.” Agricultural Westberg, N. B. 2015. “Exchanging Fertilizer for Economics 44 (6): 659–70. Votes?” Working Paper 12/2015, Norwegian University of Life Sciences School of Mason, Nicole M., Auckland Kuteya and Hambulo Economics and Business, Ås, Norway. Ngoma. 2018. Recent Innovations to Zambia’s Farmer Input Support Program & Insights on World Bank. 2008. World Development Report Short-run Effects. International Conference of 2008: Agriculture for Development. Washington, Agricultural Economists, Vancouver, Canada, DC: World Bank. July 30. https://www.canr.msu.edu/fsg/ World Bank, FAO, WFP, and IFAD. 2012. publications/presentations-pdfs/Zambia%20 “Operational Plan for Making Adjustments to FISP%20innovations%20&%20prelim%20 the Food Reserve Agency and the Farmers’ results%20on%20short-run%20effects%20 Input Support Program: Recommendations -%20ICAE%20-%202018_07_30%20-%20 to the Government of Zambia.” final%20-%20web-updated.pdf Zorya, S., I. Ekanayake, and I. Shuker. 2020. “Food Morris, M., V. Kelly, R. Kopicki, and D. Byerlee. Reserves in Zambia: How to Use Them Better 2007. Fertilizer use in African agriculture: for Poverty Reduction and Diversification?” Lessons learned and good practice guidelines. internal Powerpoint of Agriculture and Rural Washington, D.C.: World Bank. Development Department of the Africa Namonje-Kapembwa, T., T. S. Jayne, and R. Region of the World Bank. Black. 2015. “Does Late Delivery of Subsidized 28