97403 Strengthening National Comprehensive Agricultural Public Expenditure in Sub-Saharan Africa SIERRA LEONE BASIC AGRICULTURAL PUBLIC EXPENDITURE DIAGNOSTIC REVIEW (2003-2012) FEBRUARY 2015 FINAL REPORT TABLE OF CONTENTS ACKNOWLEDGMENTS.............................................................................................................. 6 ACRONYMS AND ABBREVIATIONS ...................................................................................... 7 EXECUTIVE SUMMARY ............................................................................................................ 9 INTRODUCTION ...................................................................................................................... 15 1. BACKGROUND TO SIERRA LEONE’S AGRICULTURE PUBLIC EXPENDITURE REVIEW...................................................................................................................................... 17 1.1 Methodology and scope of the review............................................................................... 17 a) Definition of the Agricultural Sector.......................................................................................... 17 1.2 Economic, policy and institutional context ..................................................................... 18 a) Rapid overview of the Economic Context ................................................................................ 18 b) Public Financial Management in Sierra Leone ...................................................................... 20 c) Decentralization in Sierra Leone ................................................................................................. 22 1.3 Agriculture Sector : policy framework, institutions involved, past and recent performances .................................................................................................................................... 23 a) Agriculture Sector framework ..................................................................................................... 23 b) Public Institutions in Sierra Leone’s Agriculture Sector ................................................... 25 c) Overview of Projects and Programs Implemented in the Agricultural Sector ......... 28 d) Conclusions of the recent Agriculture Sector Review on the agriculture sector contribution and challenges ................................................................................................................... 29 e) The evolution of the Agriculture GDP in recent years in Sierra Leone: ...................... 29 2 LEVEL OF PUBLIC EXPENDITURE IN AGRICULTURE ........................................... 35 2.1 Projected And Executed Budgets of MAFFS, MLCP, MFMR......................................... 37 a) Budget funded by domestic resources...................................................................................... 37 b) Internally funded decentralized expenditure........................................................................ 42 c) Budget funded by external resources........................................................................................ 44 2.2 TRENDS IN AGRICULTURAL SECTOR EXPENDITURE AND THE MAPUTO TARGET ................................................................................................................................................................ 47 a) A sector predominantly funded by ODA .................................................................................. 47 b) Expenditure levels consistently bellow the Maputo statement of intent................... 48 2.4 Expenditure per capita ........................................................................................................... 52 3. PUBLIC EXPENDITURE IN AGRICULTURE: ECONOMIC AND FUNCTIONAL COMPOSITION, REGIONAL DISTRIBUTION ................................................................... 54 3.1 Objective of the Analysis ........................................................................................................ 54 3.2 Economic Composition of Expenditure ............................................................................ 54 3.3 Functional Composition of Expenditure .......................................................................... 56 a) Functional analysis of Development Partners funded expenditure in the agriculture sector ........................................................................................................................................ 57 3.4 Regional distribution of expenditure................................................................................ 62 a) Centralized versus decentralized spending ................................................................................ 62 b) Regional Allocation of public agricultural Expenditure .................................................... 62 c) Funds devolved to districts............................................................................................................ 63 4. TECHNICAL EFFICIENCY IN THE PREPARATION, IMPLEMENTATION, AND MONITORING OF AGRICULTURE EXPENDITURE ........................................................ 66 2 4.1 Planning and budgeting procedure ................................................................................... 66 a) Overview Of Past And Current Budget Planning Procedures .............................................. 66 b) Quality of annual budget and plans ........................................................................................... 67 4.2 Execution rates .......................................................................................................................... 67 4.3 Management of ODA in the Agriculture Sector .............................................................. 70 a) Fragmentation of Aid, high number if PIUs................................................................................. 70 b) Mis-alignment between national and projects’ budget preparation processes .......... 71 4.4 Resources allocation process, efficiency of spending ................................................. 72 4.5 Revenues and their management ....................................................................................... 74 5. CONCLUSIONS AND RECOMMENDATIONS: How could the quality of spending be strengthened ? .................................................................................................................. 76 5.1 Factors affecting the quality of spending ......................................................................... 76 5.2 Proposed actions for improving the efficiency of public expenditure in agriculture.......................................................................................................................................... 77 5.3 Addressing aid fragmentation ............................................................................................. 79 5.4 Working towards a Sector-Wide Program ...................................................................... 79 ANNEXES Annex 1: Terms Of References Of The Agriculture Sector Public Expenditure Review ..................................................................................................................................................................... 80 Annex 2: Tors And Composition Of The Technical Working Group And Of The Steering Committee............................................................................................................................. 89 Annex 3: MAFFS organogram.......................................................................................................... 90 Annex 4: Ministry of Land and Country Planning organogram ......................................... 91 Annex 5: Data collection strategy .................................................................................................. 92 Annex 6: Level of expenditure in comparison to Maputo Targets (in Billion Leones) ..................................................................................................................................................................... 95 Annex 7: functional classification used by MAFFS under the MAFSAN project .......... 96 LIST OF BOXES Box 1 : Steps taken............................................................................................................................... 16 Box 2: Medium-Term Macroeconomic Framework ............................................................... 19 Box 3: The four strategic aims pursued under the Sierra Leone 2014-2017 PFM reform strategy ..................................................................................................................................... 21 Box 4 : Sierra Leone’s Public Sector Hierarchy ........................................................................ 22 Box 5: Measuring the Local Public Sector Expenditures in Sierra Leone ...................... 23 Box 6: Data collection process. ...................................................................................................... 36 Box 7: limits of functional analysis of agriculture expenditure ......................................... 61 Box 9: MAFFS reporting .................................................................................................................... 73 3 LIST OF FIGURES Figure 1: Historical growth of GDP per capita since independence ................................. 18 Figure 2 : Annual Agriculture GDP growth 1998-2011 (%)................................................ 30 Figure 3 : Evolution of the AgGDP and GDP annual growth over the period 2003- 2012 .......................................................................................................................................................... 31 Figure 4: Evolution of the agricultural Value Added per worker, per ha of arable land and cereal production per 100 inhabitant (1980-2012) ...................................................... 32 Figure 5 : Main organizations conducting agricultural extension in Sierra Leone .... 33 Figure 6. Evolution of the MAFFS provisional and implemented budget from domestic resources in current and constant terms, 2004–2012 ...................................... 38 Figure 7. Evolution of provisional budget from domestic resources in current terms, 2004–2012 ............................................................................................................................................. 40 Figure 8: Evolution of implemented budget from domestic resources in current terms, 2004–2012 ............................................................................................................................... 40 Figure 9: Evolution of the agriculture sector institutions executed expenditure (cumulated, in ‘000 Le) ...................................................................................................................... 41 Figure 10: Compared evolution of implemented budget from domestic resources in current terms, for the agriculture, health and education sector (2004-2012) ............ 42 Figure 11: Local level expenditure (spent budget), by sector (2007-2012)................. 43 Figure 12: Externally funded expenditure (agriculture sector), 2003-2012 ............... 44 Figure 13: Externally funded expenditure (across sectors), 2003-2012 ....................... 45 Figure 14: Comparison of agriculture sector expenditure (various sources of funding and spending agencies, in million Leones) ................................................................................ 47 Figure 15: Evolution of combined agriculture expenditure (% of total) in Sierra Leone and comparison to the Maputo Target. .......................................................................... 49 Figure 16: Trends in Agricultural Sector Expenditure, percent of GDP and AgGDP, 2004-2012 .............................................................................................................................................. 50 Figure 17: Per agricultural capita government expenditure on agriculture by area, 1980 – 2007 ........................................................................................................................................... 53 Figure 18: Evolution of personnel, recurrent non-wage and development expenditure over the period 2004-2012 .................................................................................... 54 Figure 20: Sub-sector contributions to AgGDP (av. 2003-2012) ..................................... 59 Figure 21: Relation between Poverty rates and Agriculture sector district allocations for the 13 rural districts of Sierra Leone ............................................................. 64 Figure 22: Relation between percentage of food insecure households and Agriculture sector district allocations for the 13 rural districts of Sierra Leone ........ 65 4 Figure 23: Actual vs budgeted expenditure 2004-2012 (MAFFS, agriculture sector, Domestic Development Expenditure in the Agriculture Sector). Internal Resources. ..................................................................................................................................................................... 68 Figure 24: Execution rate of budget transferred to the local councils (2007-2011). Agriculture (COFOG) and all sectors. ........................................................................................... 69 Figure 25: Number of projects and average amount disbursed annually by project (in $US) over the period 2003-2012 ............................................................................................ 71 LIST OF TABLES Table 1: List of SLARI supervised agencies involved in agricultural R&D ..................... 28 Table 2: Summary of aggregate expenditure (as a share of total expenditure) for the ministries and entities considered in the AgPE review (years 2011 and 2012) ......... 39 Table 3: Evolution of devolved spending in selected sub-saharian Africa countries 44 Table 4: Share of various sectors in total externally funded expenditure in Sierra Leone (2003-2012) ............................................................................................................................. 46 Table 5: Share of domestic fund in total agriculture expenditure .................................... 48 Table 6: Percentage of total expenditure executed in the agriculture sector (2007- 2012) ........................................................................................................................................................ 49 Table 7: International comparison of budget transfers to agriculture, 2002–2012.. 51 Table 8: Expenditure per capita (USD) in various Sub-Saharan countries ................... 52 Table 9: Share of wages and non wages expenditure in various sub-saharian African countries (2009) .................................................................................................................................. 55 Table 10: Funds budgeted for agricultural R&D in sub-Saharan Africa ......................... 61 Table 11: Centralized vs decentralized spending in various sub-Saharan Africa countries.................................................................................................................................................. 62 Table 12: DP funded project expenditure by region vs population repartition .......... 63 Table 13: District level expenditure in the agriculture, health and education sectors (Le. Millions). ......................................................................................................................................... 63 5 ACKNOWLEDGMENTS This Agriculture Public Expenditure Review analysis was carried out by a team of consultants under the guidance of the Sierra Leonean Ministry of Agriculture, Forestry and Food Security (MAFFS) and the World Bank. The core work was undertaken by Niels Morel (Agriculture Specialist – Consultant), Dr Bob Conteh (Economist – Consultant), and Abdulaye Kamara (Data collection specialist - Consultant). The consultants presented the preliminary findings during a Video Conference organized by the World Bank on October the 9th 2014. The consultants wish to express their gratitude to the Sierra Leonean authorities, the technical and financial partners, and other non-governmental partners for their warm welcome, their availability, and their excellent collaboration. In particular, they wish to Mr. Prince Kamara, National SCP Coordinator and all the members of the Technical Working Group for the support they gave to the work of the Review. The consultants also wish to thank Stephen Mink (Lead Economist, TTL, Word Bank) and Hardwick Tchale (Agricultural Economist, World Bank) for their support, advice, and comments throughout the duration of this work. The cost of this Review was met by the Strengthening National Comprehensive Agricultural Public Expenditure in Sub-Saharan Africa Program, co-funded by the Bill and Melinda Gates Foundation and the Comprehensive Africa Agriculture Development Program (CAADP) Multi-Donor Trust Fund, and executed by the World Bank. 6 ACRONYMS AND ABBREVIATIONS ABC Agriculture Business Centre AFAIP Agriculture, Fishery and Agro-Industry Program AfDB African Development Bank AgGDP Agric. Gross Domestic Product AgPER Agriculture Sector Public Expenditure Review CAADP Comprehensive Africa Agriculture Development Programme CGE Computable General Equilibrium CILSS Comité permanent Inter-Etats de Lutte contre la Sécheresse dans le Sahel COFOG Classification of the Functions of Government CORAF West and Central Africa Council for Agriculture Research and Development ECOWAS Economic Community of West African States ER External Resources EU European Union FAO Food and Agriculture Organization FARA Forum for Agricultural Research in Africa FFLS Farmer Field and Life School FFS Farmer Field School GAFSP Global Agriculture and Food Security Program GDP Gross Domestic Product GIZ German Federal Enterprise for International Cooperation GoSL Government of Sierra Leone IAR Institute of Agricultural Research IDB Islamic Development Bank IFAD Islamic Fund for Agricultural Development IFPRI International Food Policy Research Institute IMF International Monetary Fund IR Internal Resources IVS Inland Valley Swamp JICA Japan International Cooperation Programme MAFFS Ministry of Agriculture, Forestry and Food Security MDA Government Ministries, Departments and Agencies MDG Millennium Development Goal MLCP Ministry of Land and County Planning MFMR Ministry of Fisheries and Marine Resources MoFED Ministry of Finance and Economic Development MTEF Medium Term Expenditure Framework 7 MTI Ministry of Trade and Industry NaCSA National Commission for Social Action NAFSL National Association of Farmers for Sierra Leone NATC National Agriculture Training Center NEPAD New Partnership for Africa’s Development NGO Non Government Organization NRDS National Rice Development Strategy ODA Official Development Assistance PEFA Public Expenditure and Financial Accountability PER Public Expenditure Review PETS Public Expenditure Tracking Survey PRSP Poverty Reduction Strategy Paper ReSAKSS Regional Strategic Analysis Knowledge Support System-West Africa RRS Rice Research Stations SCP Smallholder Commercialization Program SLANGO Sierra Leone Association of NGO SLARI Sierra Leone Agricultural Research Institute TWG Technical Working Group UNDP United Nations Development Programme USAID United States Agency for International Development WB World Bank WDI World Development Indicators WHH Welt Hunger Hilfe 8 EXECUTIVE SUMMARY Despite the civil conflict which affected the country from the early 1990s to the beginning of 2000, Sierra Leone economy has shown significant improvement during the last decade starting in 2003. While agriculture has also shown remarkable improvement, more needs to be done to achieve higher and sustainable level of sector growth. Agricultural value-added has picked up after the end of the civil war and growth has averaged around 6 percent or more per annum over the last few years. Labor and land productivity remain however, leading invariably to depressed rural wages, limiting the ability of the sector to contribute to poverty reduction. Since 2002, with the help of financial and technical assistance from several Development Partners the GoSL has made good progress in strengthening its PFM framework and systems although emerging challenges – such as the rapid devolution process - need to be managed so as to consolidate on these achievements and build on them. As illustrated during the recent Consultative Workshop on Agriculture (Consultative Dialogue on the Path to Prosperity, Nov 2013) chaired by HE. Ernest Bai Koroma, President of the Republic of Sierra Leone, agriculture is increasingly seen in Sierra Leone as a sector involving multiple government institutions and it is in the context of a broader Agriculture Sector Review that the Ministry of Agriculture, Forestry and Food Security commissioned this Agriculture Public Expenditure Review (AgPER). The key objective of the AgPER were to collect and analyze key historic data on public spending on agriculture, examining the efficiency of spending, and identifying areas where additional funds could be applied effectively to achieve national agricultural policy and CAADP objectives. The main findings of this review are the following: Expenditure levels in the sector – while tripling over the period - remain consistently below the Maputo statement of intent Under the Maputo Declaration of 2003, African governments committed to increasing the share of their national budgets devoted to agriculture, with a target of at least 10% so as to achieve agricultural growth of at least 6% per year. While this level of growth was achieved, the analysis of past expenditures shows that combined spending from internal resources and donors on broad agriculture was approximately 4% percent in FY 2004 to 2007, reaching 9% in FY 2008 and 2009 and approximately 6-7% in the recent years. The overall trend in terms of budget allocated to the agriculture sector (which has tripled in nominal terms over the period) however clearly reflects an increased attention paid to the sector during the period of re-construction and concomitant with the Maputo statement of intent. To demonstrate a level of support in line with the Maputo statement of intent, Sierra Leone would have to reach a level of agricultural support in the order of USD 92.7 million per year. The current report shows also that external resources are a very significant portion of the budget allocated to / spent in the agriculture sector (as well as in many other key sectors 9 such as health and education) in Sierra Leone and donor funded expenditure in the agriculture sector have increased very sharply with DP funded annual expenditure fluctuating around 40 million $US since 2010 in the agriculture sector. Local spending have seen a massive increase over the period and this sustained increase while potentially positive, raise important questions in terms of capacity building for the local government officials in charge of administering these resources. The report provides cautionary notes in this regard, recommending that the authorities build more capacity and improve procedure and oversight. Economic and functional composition of expenditure appears broadly adequate given existing constraints but could see improvements An analysis of the economic composition of spending shows (on internal resources) shows that recurrent expenditure (salaries and non-salary recurrent expenditure) represent by far the largest share of total spending (close to 80%) at central level on internal resources. The situation is however not un-common in low income sub-Saharan Africa countries and a positive aspect of how the Sierra Leonean situation has evolved over the last 10 years is the fact that the share of Domestic development expenditure is on the increase: representing less 10% in 2004 they hovered around 20% in recent years. In terms of regional evolution of expenditure, a comparison of the Sierra Leonean situation regarding decentralized spending (on internal resources) puts the country in an intermediate situation between the countries that have already achieved a good degree of decentralization (Ghana for example) and those who did little or no implementation (Togo, Liberia notably). The level of agriculture expenditure in the various regions is consistent with the repartition of the population but do not appear strongly correlated with poverty levels and food insecurity. Technical efficiency in the preparation, implementation, and monitoring of agriculture expenditure could be greatly improved, notably by reducing the fragmentation of ODA. Among the weaknesses of the planning/budgeting process in Sierra Leone is the fact that investment is not sufficiently linked to its recurrent expenditure implications. Also, The budget formulation process is generally weak and would require a substantial overhaul to link policy and planning to expenditures. Changes also need to be made to make it easier to analyze budget data and comparing outturns with budget estimates. Execution rates in the sector have been lower than execution rates of the overall State budget, though they are improving. The MoFED ODA database shows that the number of project in the agriculture sector has constantly increased during the period 2003-2012 while at the same time, the average amount disbursed by project was decreasing. In all logic, the trend should be exactly the reverse, with less and less larger and larger projects. This situation creates a roadblock to the quality of the management of ODA, considerable coordination challenges and reductions in aid effectiveness. 10 In addition, there is clearly a mis-alignment between national and projects’ budget preparation processes (in terms of format and timing) and a large portion of aid is ‘off - budget’. Effectiveness of spending may remain limited Available data do not allow analysis of effectiveness in great depth. Hard evidence on the effects of government spending is not available, as the MAFFS’s monitoring system is not yet operational. However, the recent Agricultural Household Tracking Survey (AHTS, 2011) contains rather detailed information on access to all types of agricultural services at the level of the household. Overall, access to affordable agricultural services remain very limited for the majority of the farmers and efforts done by MAFFS to enhance service delivery needs to be improved. Conclusions and recommendations In addition to increasing the volume of public expenditure to achieve appropriate levels of growth of agricultural production and poverty reduction, the available funds need to be spent in a focused and efficient manner. Level of spending is in effect important and quality of spending is most critical. For better-quality public expenditures in agriculture that generate a greater degree of ownership, incidence, impact, and sustainability, the authors recommend to apply the following policy options and specific recommendations: Actions Responsibility Budget - Establish medium-term expenditure State House, programming and frameworks as a budget planning tool. MoFED, Line level of spending - Increase the resources allocated to the Ministries agriculture sector as the current share of the State budget appears inadequate in terms of the Maputo statement of intent. - Reduce the number of DP funded projects MoFED, and increase size of each projects by more Ministries, DPs systematically using pooled fund (basket (Ag. Sector funds, etc) in order to reduce the Working Group) transactions costs of ODA management; - Include in the budgets of the agricultural sector ministries all agricultural development projects, in particular those of DPs that are currently not monitored by MEF or the Ministry of Planning. - Formulation of project to be made on the Ministries and basis of past experience (evaluations, other institutions capitalization on success and failures). involved in the Ag. Sector, DPs 11 - Establish a mechanism for budgeting for MoFED, operating expenditure on maintenance of Ministries and capital items. other institutions - Divide coherently roles and means for involved in the investment maintenance between central Ag. Sector level and Local Councils - Improve cost accounting at the development Ministries and funded project level to identify the operating other institutions costs incurred by the project management involved in the - Identify project management functions Ag. Sector, DPs which cost could be reduced by pooling resources across projects. Budget - Present commitment and procurement plans MoFED, Implementation and at the same time as the annual budget, and Ministries and Procurement make tender documents available before the other institutions start of the budget year so that calls for involved in the tender can be issued in January. Ag. Sector - Better monitoring of investments made at the level of support services and of beneficiaries, and better monitoring of the regional distribution of expenditures; Decentralization / - Develop and implement a comprehensive Ministry of Local devolution capacity building plan at Local Council Planning, MoFED, level. Ministries and - Transfer staff (and corresponding budget) other institutions from central level payroll to Local Council involved in the payroll Ag. Sector. Monitoring and - Implementation at all levels of an effective Ministries and Evaluation M&E system coupled with an efficient other institutions information archiving and dissemination involved in the mechanism Ag. Sector - Draw up a system for monitoring performance in the agricultural sector with the use of quantifiable and verifiable performance indicators, potentially based on PNDSA indicators - Institutionalize the AgPER process, with light annual AgPE reviews and in-depth reviews every 3-5 years. - Strengthen annual assessments of ministries’ progress, detailing monitoring of actions and activities. - Overhaul MAFFS (and other public institutions reports involved in the agriculture sector) annual reports - Improve information feedback on actual expenditure by all services and make this information available through annual reports. 12 - Ensure the introduction of an effective system for monitoring program budgets in aggregated fashion (not just PIU by PIU). - Produce evaluation reports for all projects. Establish an Evaluation Unit under PEMSD. - Produce impact studies for major projects. Functional - Better functional distribution of Ministries and allocation expenditures is needed. other institutions - The livestock (including veterinary services) involved in the subsector as well as R&D appear to be Ag. Sector largely under-funded and require immediate attention. - Agriculture extension received only 2.11% of the total spent in the sector through the projects, which appears as a low figure in comparison to the central role that extension can play in the dissemination of agricultural technologies. Since private extension services are quasi inexistent, a higher level of spending into public advisory services is necessary, especially in a context where efforts are put into decentralization and devolution Policy Development - Elaborate a strategy for agricultural Ministries and extension so as to include the role of all the other institutions parties involved (MAFFS, private sector, involved in the NGOs, local councils) Ag. Sector - Study how to end the duplications observed Ministries and between existing project and future PIUs. other institutions Give support to pooling resources and involved in the ensuring their ownership by the sectors Ag. Sector institutions. Enact MAFFS ODA management decree / guidelines. Ministries and - Evaluate past expenditure in support to other institutions mechanization and develop a strategic involved in the framework for mechanization development Ag. Sector Working towards a Sector-Wide Program The preparation of the SCP and its subsequent implementation were positive steps towards making donor interventions and GoL spending consistent and coordinated under the umbrella of the NSADP. The challenge now is to maintain the momentum and develop this approach further into a SWAP-like arrangement with ideally pooled funding arrangements for the required financial requirements1. The preparation of this future Sector Wide approach could also be the 1 The present AgPER evaluates future desirable levels of spending at around 100 million USD per year 13 opportunity to translate into concrete decision and actions numerous of the recommendations of this Ag-PER, in particular with regard to Donor Coordination, M&E, strengthening Local Council capacities and alignment of planning procedures. 14 INTRODUCTION 1. As illustrated during the recent Consultative Workshop on Agriculture (Consultative Dialogue on the Path to Prosperity, Nov 2013) chaired by HE. Ernest Bai Koroma, President of the Republic of Sierra Leone, agriculture is increasingly seen in Sierra Leone as a sector involving multiple government institutions. At least five key Ministries were represented at high level at the workshop and the necessity to establish coordination mechanism and effective complementarities at policies and programmatic levels were highlighted repeatedly. The AgPER methodology builds on this approach by reviewing past expenditures at a sectoral level (based on the COFOG definition). 2. It is in the context of a broader Agriculture Sector Review that the Ministry of Agriculture, Forestry and Food Security expressed its desire for an agricultural public expenditure review to be carried out in order to increase the level of performance in the medium term by learning from past experience of the use of public funds. This request was accepted by NEPAD’s Planning and Coordinating Agency. The costs of this Review were met by the Strengthening National Comprehensive Agricultural Public Expenditure in Sub-Saharan Africa Program and co- funded by the Bill and Melinda Gates Foundation and the CAADP Multi-Donor Trust Fund. This program, to be implemented by the World Bank, aims to improve the impact of public resources at the disposal of governments in Sub-Saharan Africa, foster agricultural development, and reduce rural poverty, which in most countries affects the majority of the poor. 3. This Agriculture Public Expenditure Review (AgPER) provides key background information and guidance in this endeavor by presenting and analyzing historic data on public spending on agriculture, examining the efficiency of spending, and identifying areas where additional funds could be applied effectively to achieve national agricultural policy and CAADP objectives. 4. The goals of the Agricultural Public Expenditure Review in Sierra Leone are as follows: o Gain a better understanding of the country's performance in the context of the 2003 Maputo Declaration; o Draw lessons from the past in terms of budget execution in the agricultural sector and identify bottlenecks, inefficiencies, and deviations from goals; o Seek and recommend corrective actions for existing and future programs with a view to improving their impact and making them more efficient and equitable; o Initiate the implementation of the databases and methodology required for conducting similar reviews regularly and thus contribute to the institutionalization of the process; o Help the government establish an environment and capabilities for results-based management, with particular emphasis on improving planning, execution, and monitoring and evaluation; and o Increase visibility for the government and the financial and technical partners over the sector's absorptive capacity so that the decision may be made to allocate more resources to agricultural development. 5. This report consists of five chapters: o The first chapter introduces the strategic and institutional context; o The second chapter studies the level of public agricultural expenditure in Sierra Leone; o The third chapter analyzes the economic and functional composition of public agricultural expenditure (allocative efficiency); 15 o The fourth chapter assesses the technical efficiency of the processes of preparation, execution, and monitoring and evaluation of agricultural budgets; and o The fifth chapter contains our findings and recommendations. 6. In accordance with the terms of reference, the review covers the 10-year period 2003-2012. Box 1 : Steps taken An inception workshop for the AgPE Review was held on Thursday November the 14th 2013 at the Ministry of Agriculture, Forestry and Food Security, chaired by the Honorable Minister. The meeting was attended by 21 participants, representing various administrations (MAFFS, MoFED, SLARI, Statistics Sierra Leone, Ministry of Lands, Country Planning and the Environment, Ministry of Fisheries and Marine Resources) and Development Partners (AfDB, GIZ, + WB consultants). During the workshop, presentations were made by the consultant team, highlighting: (i) the objectives of doing an AgPER; (ii) steps and processes; (iii) data requirement; and (iv) working modalities. The Honorable Minister of Agriculture, Forestry and Food Security nominated during the inception workshop the members of the Technical Working Group to meet regularly with the AgPER consultant team and review / discuss: progress made, difficulties encountered (notably in terms of data collection) and processing / analysis of the data collected. It was agreed that the group could meet on a weekly basis during the time the full team of consultant would be mobilized in Sierra Leone. The first meeting of the technical working group took place on November 19, 2013. Next steps focused primarily on data collection and subsequently data analysis. Progress and difficulties were regularly reported by the World Bank consultant team to the Technical Working Group. Data collection was conducted essentially through individual meetings with information-holders: DPs, GoSL, and NGOs (through the NGO network SLANGO). The AgPER team progressed alongside the team recruited specifically to produce the broader Agriculture Sector Review and information was regularly exchanged between the two teams. A first presentation of the preliminary results of the AgPER was made on October the 9th, 2014 through Video-Conference with the participation of a number of officials from the GoSL including high level officials from the Ministry of Agriculture, Forestry and Food Security and the State House. The AgPER preliminary results were presented alongside a presentation of the draft Agriculture Sector Review Synthesis report. During the Video-Conference, information was also given by Prof. Dr. Monty Jones (State House) on the current Agriculture, Fishery and Agro-Industry Program (AFAIP) being currently launched. 16 1. BACKGROUND TO SIERRA LEONE’S AGRICULTURE PUBLIC EXPENDITURE REVIEW 1.1 METHODOLOGY AND SCOPE OF THE REVIEW a) Definition of the Agricultural Sector 7. In the Maputo declaration in 2003, African governments made the commitment to increase the share of their national budgets going to agriculture. Following this declaration, NEPAD detailed the terms and conditions for calculating the share of agricultural expenditure in national budgets by issuing a methodology document (AU/NEPAD 2005). This document defined effective public expenditures (not allocated budgets) following the United Nation’s Classification of the Functions of Government (COFOG) in the extended agricultural sector, which includes the domains of agriculture (crop and animal productions), forestry (including silviculture other than timber production), and hunting and fishing. Expenditures for applied research in each of these sectors was taken into account. However, expenditures for supplying drinking water and feeder roads as well as purely environmental expenses were not taken into consideration as expenditures, following NEPAD procedures. 8. The AgPER has followed strictly the guidance note issued by the AU-NEPAD in 2005 relatively to the definition of the agriculture sector. Using the COFOG definition of the Agriculture sector and translating it to the Sierraleonese context, the TWG and the consultant team have agreed to consider the following set of data for inclusion in the AgPER work : o Expenditure data of the following Ministries/institutions : MAFFS (401 2 ), MFMR (402), Ministry of Land and County Planning (306), National Commission on Environment and Forestry (410), Sierra Leone Agricultural Research (418), Sierra Leone environmental Protection Agency (420). o Externally funded project as recorded by the MoFED/ Development Assistance Database (categories : agriculture + Natural Resources Management projects if related to agriculture) o District level expenditure (internally funded) o Expenditure by NGOs to promote agricultural development was only considered when they acted as implementing agencies as part of projects included or not in the government budget. 9. In accordance also with NEPAD recommendations (AU/NEPAD 2005), the budgets of public or commercially focused semi-public bodies were not taken into account as only net subsidies transferred to these bodies from the government budget were considered public expenditure. Similarly, private investment, including by producers, was not considered. 10. Expenditure related to rural/feeder roads were not taken into account, again in accordance with the NEPAD guidelines. Taking into account the important role that feeder roads play in agriculture development, it was however agreed initially that the AgPER work could be conducted using a “COFOG+� definition (incorporating rural/feeder roads) in addition to the basic COFOG definition. In practice, however, this was not possible because of lack of access to reliable data on spending on rural roads. It is not impossible however that some of the DP funded projects included in the agriculture sector in this review have executed expenditure relating to rural roads. 2 Codes used in Sierra Leone public accounts to designate expenditure by institution. 17 1.2 ECONOMIC, POLICY AND INSTITUTIONAL CONTEXT a) Rapid overview of the Economic Context 11. Despite the civil conflict which affected the country from the early 1990s to the beginning of 2000, Sierra Leone has shown significant improvement over the years since the end of the conflict in 2002. Economic performance since 2010 has been quite robust as the per capita growth has increased to levels that surpass the levels achieved at any time since independence. Following better spending controls and improved revenue collection, the fiscal deficit is expected to narrow to 3.9% of GDP in 2014. After an estimated leap in real GDP growth, to 17.9% in 2012, due to the onset of iron ore production, growth will remain brisk in 2013-14, at an average of 12.3%, supported by continued expansion in the mining sector. Average inflation is set to moderate, to 12.2% in 2013 and 10.4% in 2014, as domestic agricultural production improves, world food prices fall and the currency appreciates, only partly offset by a reduction in fuel subsidies. The current-account deficit is forecast to shrink from an estimated 19.8% of GDP in 2012 to 7.7% of GDP in 2014 as mining exports increase. Strong export growth means that the trade balance will move into surplus by 2014. Figure 1: Historical growth of GDP per capita since independence Source: Statistics Sierra Leone, IMF and Bank staff estimates. 12. Despite the strong post-conflict growth, significant challenges still remain. First, poverty levels are still high. Poverty is heavily concentrated in the rural and other urban areas outside Freetown. According to the Sierra Leone Integrated Household Survey (SLIHS) conducted in 2011, the poverty rate declined from 66.4 percent in 2003 to 52.9 percent in 2011. The overall reduction was led by strong growth in rural areas, where poverty declined from 78.7 percent in 2003 to 66.1 percent in 2011, yet this figure was overall still higher than urban poverty. Urban poverty declined from 46.9 percent in 2003 to 31.2 percent in 2011, though poverty increased in Freetown from 13.6 percent to 20.7. The causes behind the increase in poverty in Freetown are not completely explained by the available data, and further research is needed to fully understand the dynamics. While the characteristics of poor households varied between urban and rural areas, a common factor was those with less education had higher likelihoods of being poor. This was especially true in urban areas. In addition, those households which were engaged in a non-farm 18 enterprise in urban areas and female headed households were less likely to be poor. In rural areas, households in which the household head’s primary occupation was agriculture, as well as those with smaller landholdings, were more likely to be poor. Box 2: Medium-Term Macroeconomic Framework Economic growth (excluding iron ore) is projected to reach 7 percent in 2017, up from 6 percent in 2013, due to continued public investment scaling up, increased productivity, notably in agriculture, and sustained activity in construction and services. Non-iron ore activity will also benefit from upstream activity in iron ore mining where production (under phase I of the largest mine) is expected to increase through 2015, and level off starting in 2016. Consequently, total real GDP growth is forecast to rise from 13.3 percent in 2013 to 14 percent in 2014 and decelerate to 12.4 percent in 2015 and 5.2 percent by 2017. Inflation is projected to decline from 9 percent in 2013 to 5.4 percent in 2017, on account of continued prudent monetary and exchange rate policies. It is also expected that government supported programs in agriculture would contribute to increased supply of rice and other key food crops, and help dampen food inflation. In addition, monetary policy would be adequately calibrated to contain inflationary pressures, and macro-prudential measures would be geared towards a healthy expansion of private credit. In the external sector, the current account deficit would narrow from about 20 percent of non-iron ore GDP in 2013 to around 7 percent by 2017 as exports are projected to increase, notably in mining and agriculture, while imports would moderate over the medium term, partly reflecting lower FDI flows than in 2011–12. The improvement in the external current account, combined with sustained capital inflows is expected to increase gross international reserves from 3.2 months of imports (excluding iron-ore related imports) in 2013, to 4.1 months by 2017. The government’s medium-term fiscal strategy aims to strengthen revenue collection, improve expenditure management, and reduce domestic debt. The revenue efforts will focus on: (i) broadening the tax base; (ii) reducing customs and GST duty waivers; (iii) combating tax evasion; and (iv) strengthening the National Revenue Administration through administrative reforms and increased use of Information Technology. The projected expansion in economic activity and increased iron ore exports will also support the government’s revenue mobilization efforts. On this basis, revenue is projected to increase from 12.4 percent of non-iron ore GDP in 2013, to 13.5 percent in 2017; with revenue from mining rising from 1.2 percent of non-iron ore GDP to1.8 percent for the period. Total expenditure will decline from 21.9 percent of non-iron ore GDP in 2012 to 19.7 percent in 2013, which reflects the unwinding of one-off expenditure related to elections and emergency programs. Total expenditure is projected to average about 20.5 percent of non-iron ore GDP for 2014–17, with more resources channeled to investment. It is projected that wages and salaries will stabilize at 6 percent of non-iron ore GDP under the ongoing pay reform. Under this strategy, domestic borrowing would gradually decline, allowing the stock of public domestic debt to decline from 12.7 percent of non-iron ore GDP in 2013, to 11.2 by 2018. Extracted from: IMF. Sierra Leone: Letter of Intent, Memorandum of Economic and Financial Policies and Technical Memorandum of Understanding. Attachment 1. 13. With regard to trade and business environment, Sierra Leone conducted and validated an initial DTIS report in 2006, supported by the World Bank. This included an extensive action matrix aimed at prioritizing and sequencing policy reforms and other interventions for 19 mainstreaming trade into national poverty reduction and development strategies. Despite considerable constraints Sierra Leone has achieved significant progress in a range of areas specified in the action matrix, including legislative and regulatory changes to improve the overall business climate and expanding institutional capacity for formulation and implementation of trade policies. Implementation remains however partial in a number of areas, in particular those relating to cross–cutting issues like trade facilitation and logistics, as evidenced by Sierra Leone’s position in the latest Global Competitiveness Rankings which places the country second from last. 14. The United Nations' Human Development Index, which focuses on the challenge of sustainable and equitable progress, ranked Sierra Leone 180 out of 187 countries and territories based on 2011 data. In 2012, Sierra Leone moved up to 177 out of 187 countries. Significant challenges still remain in infrastructure including roads and energy, low public sector implementation capacity, youth unemployment, and high maternal and infant mortality. b) Public Financial Management in Sierra Leone 15. Since 2002, with the help of financial and technical assistance from the European Commission (EC), the UK Department for International Development (DFID), the African Development Bank (AfDB) and the World Bank, the GoSL has made good progress in strengthening its PFM framework and systems. This has resulted in major PFM achievements over the years, including the establishment of the legal and regulatory framework for budgeting, accountability and procurement, implementation of a financial management information system (FMIS) in the Accountant General’s Department in 2005 and its roll-out to several ministries, departments and agencies (MDAs), and the removal of the backlog of annual financial statements (over the last 4 years, the public accounts have been produced within the statutory deadline of 3 months of end of fiscal year). Independent external audit quality and scope has improved, as has timeliness in the completion of audits. Basic budgeting, procurement, accounting, and reporting procedures have been established in local councils. As a result, despite the very low starting point at the end of the civil war, the scores from the Public Expenditure and Financial Accountability (PEFA) assessment for Sierra Leone carried out in 2007 and 2010 show some clear improvements (30 percent of the indicators improved). 16. The country faces significant new PFM opportunities that need to be harnessed and emerging challenges to be managed so as to consolidate on these achievements and build on them. The expected revenue streams from natural resource assets are very material and the A4P highlights the significant funding gap required for the Government to close its infrastructure deficit and cater to other development challenges. This requires the ability to transparently and effectively translate financial resources into human and physical capital, a critical function of the PFM system. 17. Recent observations by internal and external review missions (the World Bank, IMF and EU mid-term review of Integrated Public Financial Management Reform Project) and the Audit Service Sierra Leone reports indicate that areas of major weakness remain. Issues identified include budget credibility and predictability, fiscal management challenges, weaknesses in expenditure control (particularly development expenditure), and low levels of transparency. Added to these are issues related to incomplete collection and reporting of revenues including from the natural resource sectors and, hitherto, the lower than expected tax collections (making the available resource envelope insufficient to fund rising expectations for service delivery); the lack of a robust cash management system as there is no Treasury Single Account; fragmentation between the budget and planning process; recurring off-budget expenditures; and a budget system 20 largely based around inputs rather than achievement of results. These make management of public finances susceptible to allocative inefficiency and distract from the PFM system’s key role of supporting effective service delivery. 18. In order to build on the progress of reforms achieved to date and to respond to the inherent challenges, the GoSL has prepared a comprehensive program of PFM reform strategy. This process is directed by the PFM Reforms (Oversight) Steering Committee and the strategy targets improvements in the quality of PFM. There is a Medium Term Expenditure Framework (MTEF), but in reality, the budget is annual and the medium term forecasts of future expenditure plans are not robust nor are they coherently linked to policy or plans. Box 3: The four strategic aims pursued under the Sierra Leone 2014-2017 PFM reform strategy Budget Planning Comprehensiveness, and Credibility Under this theme, the primary aims are that to establish a credible and stable budget process, particularly to establish a transformational public investment program (PIP) and link investment to recurrent operations and maintenance spending through the MTEF process. Financial Control and Accountability, Service Delivery and Oversight For the next phase of the strategy, the most critical objective is to complete the roll-out of IFMIS to major spending MDAs and bring all CG public accounts--including sub-vented accounts and DP project accounts--on to IFMIS. It is of utmost importance to note at this stage that this will require substantial investment in the ICT infrastructure for the IFMIS. This will be a crucial step toward establishing a TSA, as discussed below. Equally importantly, it will be the basis for a comprehensive budget covering the whole of central government, and eventually general government. Revenue Mobilization Two objectives will be stressed under this theme: (1) Consistently with likely developments in resource-related industries, major efforts are being made to establish more effective tax and control regimes for extractive industries through the Extractive Industries Revenue Act and the Oil Exploration Act. (2) Challenges identified in revenue collection, reporting and reconciliation will be addressed by improved systems and interfaces, which aims at improving the system for recording and reconciling payment and receipts. A new MOU with the banks has already been established. Strengthening Local Governance Financial Management through Local Councils for Effective Decentralization The next phase shall emphasize a transfer of authority to the Councils, combined with effective use of systems to meet both community and national goals of service delivery. These aims will be reflected in the revised legislative framework. A critical objective shall be the consolidation of the implementation of the Petra Accounting Package in all local councils including the real time processing of transactions by selected councils. PETRA-based reports by councils could be consolidated at the national level by the LGFD. Improved sectoral reporting from the Council level should in turn allow redesign of the process of allocating grants and give greater authority to councils over a smaller number of grants. 21 c) Decentralization in Sierra Leone 19. Devolution / decentralization is a strong commitment of the GoSL and has the potential of resulting in more efficient spending in the agriculture sector. Sierra Leone began its transition towards decentralized government in 2002, following the peace process that ended over a decade of civil war and strife. Since 2004, significant progress has been made in devolving public service delivery functions from the central line ministries to the 19 Local Councils. So far however, Local Councils remain seriously constrained by limited devolution of administrative powers, as all ‘local’ public servants -such as teachers and health workers- in practice remain staff of their respective line ministries. In addition, limited local own-source revenues as well as limits on the provision of intergovernmental fiscal transfers constrain the ability of Local Councils to improve the quality of and access to local public services. Box 4 : Sierra Leone’s Public Sector Hierarchy Sierra Leone's public sector hierarchy comprises governmental entities operating at the National, District, and Chiefdom levels. The Local Government Act (2004) established thirteen District Councils, five Town Councils, and the Freetown City Council. The 19 Local Councils are further subdivided into wards, the smallest level of local political organization. Sierra Leone's 149 Chiefdoms overlap with these District Councils, and are governed by Paramount Chiefs and Chiefdom Councils drawn from eligible families. Some governance responsibilities, notably poll tax collection, are shared between councils and chiefdoms, although the precise limits of the authority of Chiefdoms are still being negotiated. Functional Responsibilities of the Local Public Sector 20. Sierra Leone's Local Government Act (2004) provides for the devolution of all aspects of some 80 functions to Local Councils, with funding for these services to be secured through a combination of Local Council own-source revenues and formula-based grant allocations. As of 2011, 47 functions had been partially devolved, including the responsibility for basic education, basic health services, agricultural extension services, water provision, solid waste management, fire protection, and other community-level services. Although the responsibility for these functions has been formally transferred to the Local Councils, the sectoral functionaries at the local level continue to be appointed and managed by the central government. 22 Box 5: Measuring the Local Public Sector Expenditures in Sierra Leone Measuring public sector expenditure is a complex endeavor that must recognize de existence of three types of local expenditures: 1. Devolved expenditures – made by local governments; 2. Deconcentrated expenditures – made by deconcentrated units of the national government; and 3. Other local public sector expenditures – expenditures that are made directly (or through delegation) by the central government for “localized� public services. In theory also, a good measure of Local Public Sector finances should not simply aggregate the different funding streams into a single measure of local public sector expenditures, since different types of political, administrative and fiscal arrangements at the subnational level are likely to result in different levels and quality of public services. In practice, the AgPER work in Sierra Leone has, as a first approach to measuring public expenditure at local level, used the “devolved expenditures� (made by local governments) only. 21. Local Councils remain highly dependent on intergovernmental fiscal transfers. Sectorally earmarked intergovernmental transfers account for 78% of local financial resources; no unconditional grants are provided. Local own-source revenues, including poll taxes, fees, and all other local revenues account for a minute share (about 1 percent) of national revenue collections. An increasing share of national financial resources is dedicated to local, front-line services. Despite these increases, only a small share of these resources is currently managed by the elected Local Councils. Instead, central line ministries continue to retain control over the bulk of localized expenditures, in conflict with the spirit of the Local Government Act (2004). If Sierra Leone wishes to continue its post-conflict progress towards more responsive, more efficient and more equitable public services, policy makers will need to consider moving wage payments and development expenditures associated with devolved functions out of the central ministry budgets, as a first step towards greater responsibility over these expenditures by local councils. 1.3 AGRICULTURE SECTOR : POLICY FRAMEWORK, INSTITUTIONS INVOLVED, PAST AND RECENT PERFORMANCES a) Agriculture Sector framework 22. The policy framework through which Government has been addressing these challenges over the years includes the Agenda for Change (2004-2012) and the Agenda for Prosperity (2013- 2018). In the agricultural sector, the Government has operationalized these development strategies through the National Sustainable Agriculture Development Plan (NSADP) which was developed in line with the requirements of the Comprehensive African Agriculture Development Program (CAADP) compact. The flagship program under NSADP is called the Smallholder Commercialization Program (SCP). The SCP has the overall aim of making agriculture the ‘engine’ for socio-economic growth and development through the development of commercial agriculture. The program has over the last few years been focusing on intensification, diversification and commercialization of smallholder agriculture (through improving value- addition and access to marketing). The priority areas for the Sierra Leone Agriculture Development Strategy, as per the NSADP, include promoting commodity commercialization, 23 rehabilitating and upgrading the agriculture infrastructure, creating incentives for private sector investment, and coordinating the sector’s activities. This is to support the Agenda for Prosperity’s (AfP) sector focus which is to increase agricultural productivity and value� addition. This program was implemented with support through the Global Agriculture and Food Security Program (GAFSP). There are also a number of projects supported by various development partners including the World Bank which are aligned to the objectives of the sector strategy3. 23. Overarching development Framework : o Agenda for Change (2004-2012): Roadmap for the country from 2007 to 2012 to combat poverty as part of its post-conflict peacebuilding process in the aftermath of a decade of civil war; and o Agenda for Prosperity (2013-2018) : Roadmap for the next five years, to guide the country to meet its goal of becoming a middle income country and donor nation within the next 25 to 50 years. Overarching agricultural development Framework : o National Sustainable Agricultural Development Program (NSADP) – 2010- 2030: Broad framework for putting the objectives of the Government’s Agenda for Prosperity into action in the agricultural sector; and o Smallholder Commercialization Program (SCP) – 2010-2014: Flagship program for operationalizing the NSADP – commercializing agriculture through linking large and small farmers to markets. 24. In addition to the above, a number of Strategies (or other document of a policy/strategic nature) have been elaborated in Sierra Leone during the last decade, and notably : o The National Rice Development Strategy (NRDS, 2009) : the goal of the National Rice Development Strategy (NRDS) is to lay out a framework for significant increases in rice production in order to contribute to the improvement of food security and economic development in Sierra Leone. The specific objectives are to: 1. Ensure an increase in the sustainable productivity and production of rice in Sierra Leone 2. Promote appropriate post harvest handling, processing and marketing of rice 3. Develop appropriate infrastructure for rice production and marketing 4. Improve the capacity of stakeholders and institutions involved in rice sector This strategy targets a land area of 830,000 ha and an increase in the average rice yield/ha to 2 mt/ha to realize the government’s goal of rice self-sufficiency. Furthermore, an extension of the area to 1,100,000 ha over the following years, coupled with an increase in the average yield of rice to 4 mt/ha (ranging from 1.5 mt/ha in the uplands to 4.0 mt/ha in the IVS) is expected to result in the production of over 3 million tons of rice in 2018. o The Sierra Leone National Export Strategy 2010-2015 prepared under the leadership of the Ministry of Trade and Industry with the assistance of the Commonwealth Secretariat which presents specific strategies for several sectors/ commodities including : “Agriculture, Cocoa Coffee, Oil Palm, Cashew, Ginger, Chili, Rice, Cassava, Sugar, and Fisheries�. Each commodity export plan is 3 Extracted from the recent ASR 24 associated with an action matrix and Investment Plan. The National Export Strategy promotes a strong private sector participation and addresses issues of competitiveness through its strategic objectives as follows:  improving productivity and enhancing value addition  diversifying the current export base  strengthening the human and financial capacity of all stakeholders along the value chain. b) Public Institutions in Sierra Leone’s Agriculture Sector The Ministry of Agriculture, Forestry and Food Security 25. The Ministry of Agriculture Forestry and Food Security (MAFFS) is the arm of Government, responsible to achieve sustainable food security and reduce poverty through agricultural intensification, diversification and efficient management of the natural resource base. MAFFS has the mandate to support the production of all crops and livestock in an environmentally sustainable manner and to ensure the achievement of Food Security. 26. MAFFS seeks to provide an enabling environment for farmers and by promoting appropriate research, extension, input delivery and market systems. To this end, MAFFS formulates and implements policies, and coordinates, designs and monitors programs for the development of the agricultural sector. It works with the following objectives: o To increase diversified domestic production of food, with a view to achieving food security in the medium to long term; o To increase agricultural productivity, output, rural incomes and employment, while ensuring adequate protection of the environment; o To ensure balanced regional agricultural growth and equitable distribution of income; and o To maximize foreign exchange earnings from the agricultural sector. 27. The Headquarters of MAFFS is in Freetown and the ministry is divided into a professional wing headed by the Director General and an administrative wing headed by the Permanent Secretary. The Honorable minister, assisted by two deputy ministers, is the political head of the ministry4. 28. The office of the Director General is directly responsible for overseeing all activities of Food Security, thereby ensuring that its objectives are achieved. The Director General supervises 4 technical Divisions (Crop, Livestock, Forestry, agriculture engineering, agricultural services) as well as the Planning, M&E and Statistic Division (PEMSD) and the District Director. The later, as a result of the devolution process, liaises with the Local councils and other Non-Governmental Organizations involved in agricultural activities at grass-root level. He is also fully involved in the preparation of activity plans at district level and monitors the use of assets belonging to the ministry. 29. The number of Divisions and their respective scopes of intervention have fluctuated during the period under study (2003-2012). The Ministry has staff and activities at District and local council levels. Despite of the ongoing devolution process, all the technical staff positioned at decentralized level is still on the payroll of the ministry at central level. 4 See organogram in appendix 3 25 30. Based on the COFOG definition adopted throughout this review, all of MAFFS expenditures are to be accounted for in the AgPER. The Ministry of Fisheries and Marine Resources (MFMR) 31. MFMR is in charge of undertaking all fisheries management and development activities. The legal framework for fisheries management in Sierra Leone is the 1994 Fisheries (Management and Development) Act, complemented by the Fisheries Regulations from 1995 and amended in 2007. New Draft Fisheries Regulations were prepared in 2010. The Local Government Act of 2004 regulates the functions of local councils. 32. MFMR is entrusted with an oversight responsibility of making sure that required fishing facilities such as jetties, fish processing facilities, fishing gear are available in the respective fishing communities and monitored to uphold good fishing standards. 33. MFMR’s vision is to ensure the sustainable use and management of the fish resources of Sierra Leone. The policy objectives of MFMR include the following: o improving national nutrition and food security through increased fish production based on responsible fishing and reduction of spoilage o increasing employment opportunities o enhancing livelihoods in fishing communities with emphasis on women and youth o improving skills in the fishing communities and increasing wealth through rational management based on scientific research o promoting marine and inland artisanal fishery for local affordable animal protein production in Sierra Leone o strengthening regional and international collaboration in the sustainable use, management and conservation of fish resources in shared water bodies 34. Based on the COFOG definition adopted throughout this review, all of MFMR expenditure are to be accounted for in the AgPER. The Ministry of Land and Country Planning 35. The mandate of the Ministry as summarized in the National Land Policy (2005) gives it “a central role in the management of state lands, compulsory acquisition of lands for development, surveying and surveyors, mapping, planning, town planning schemes, development standards, building codes, etc., but not registration of deeds’’. 36. The Ministry’s broad policy objectives are to: o Enhance proper land administration and management, land use planning and environmental management; o Lay the foundation for the development of planned settlements all over the country ; o Enhance planning and development of control measures for proper infrastructural development. 37. The respective divisions in the Ministry of Lands are briefly described below: o Administration :This division which performs a significant role in policy formulation, implementation and monitoring has the responsibility of coordinating the daily administrative activities of the three divisions of the Ministry for proper management and equitable distribution of state lands, effective development of control measures and sound environmental 26 management in the country. It also has the responsibility of ensuring that there is compliance with regulations related to the proper management of Lands, Town and Country Planning and Environment in order to achieve the successful implementation of the Ministry’s policy directives. o Surveys and Lands: The division is primarily responsible for the surveying and mapping of all lands and the administration of state lands. The division keeps record of all licensed surveyors and approves all survey plans, both for state and private lands. It is responsible for the effective management of state lands, the setting up of a land management and information system, the establishment of ground control points and the mapping and supervision of all surveying activities throughout the country. o Country Planning: This division is responsible for physical/land use planning, providing guidelines for the growth of settlements within the context of a plan, indicating the nature, form and direction of such settlements. Furthermore, it is also responsible for the preparation, implementation and monitoring of various town planning schemes including urban structured plans, detailed plans, action area plan and subject plans. o Environment: This division is responsible for ensuring that land use occurs in a responsible manner with minimal negative impact on the natural environment. The Environment division is responsible for the coordination of all the environmentally related activities of the Ministry and Local Authorities. It monitors the implementation of national environmental policies, programmes and projects. It is also responsible for international cooperation in the global effort for the protection of the environment. It takes care of all policies, programmes and projects, environmental education and awareness raising, the collection of environmental research and baseline data and information on land management. 38. Based on the COFOG definition of the agriculture sector, the expenditure of the following divisions only are included in the present AgPER: Surveys and Lands division, Environment division as well as a fraction of the administration expenditure. The expenditures on Country Planning are almost all urban related National Committee for Social Action 39. NaCSA was set up by an Act of Parliament in 2001 as successor to the National Commission for Reconstruction, Resettlement and Rehabilitation (NCRR), a ministerial-level government commission that coordinated humanitarian relief and reconstruction assistance and implemented donor-funded projects during and after the war. Community Driven Development approaches supported by NaCSA were introduced in a gradual and politically sensitive way, and therefore as a partnering project which helped the Government to achieve its own goals of community reintegration and decentralization after the war. NACSA has administered programs involving agricultural activities, especially right after the end of the war. NACSA interventions in this sector have however gradually phased out. 40. According to the NEPAD guidance note for agriculture expenditure tracking system in African countries (Sept 2005), if a large project has multi-sectoral objectives including ones related to agriculture, its expenditure should be include in agriculture sector only if 70 percent or more of the cost of the project is justified for or related to agriculture sector. 27 41. According to the information collected by the AgPER team during the data collection phase, for the period considered (2003-2012) while numerous NACSA implemented projects and programs included agriculture sector activities, these activities never represented more than 70 percent of the total cost of the project, and consistent with the COFOG Guidance Note these expenditures are therefore not included in the AgPER. Sierra Leone Agricultural Research Institute 42. SLARI was established in 2007 as the primary national agricultural research institute. It is planned that SLARI will operate eight research centers focusing on various commodities and research themes, but as of 2009 only two of the eight centers were operating; the remaining centers are in the process of being rehabilitated and staffed. Table 1: List of SLARI supervised agencies involved in agricultural R&D 43. Based on the COFOG definition adopted throughout this review, all of SLARI expenditures are to be accounted for in the AgPER. c) Overview of Projects and Programs Implemented in the Agricultural Sector 44. Multiple projects and programs are implemented in the agriculture sectors and more than 15 Development Partners are currently (2012) financing projects / programs. 45. Currently, the 20 years NSADP National Sustainable Agricultural Development Plan (NSADP) is the comprehensive framework to promote agricultural development in Sierra Leone based on the principles of the Comprehensive Africa Agricultural Development Programme (CAADP). The National Sustainable Agricultural Development Plan’s primary goal is to address on a sustainable basis the recurring problem of food insecurity and hunger including disasters and emergencies. In theory at least, all project and programs should align on the structure of the NSADP and contribute to its objectives. In practice, such alignment is often unclear5. 46. The section 4 of this AgPER review provides an analysis of the evolution of the portfolio of projects managed by various government entities in the agriculture sector and draws conclusions and recommendations for a more ambitious rolling out of the Paris Declaration on aid effectiveness in the Sector. 5 For a more detailed description and discussion on Development Partners funded activities, see chapter 4.3. 28 d) Conclusions of the recent Agriculture Sector Review on the agriculture sector contribution and challenges 47. The following is an extract of the Agriculture Sector Review (2014). It summarizes the key recent sector evolution and remaining challenges. 48. “For the last decade since 2003, agriculture in Sierra Leone has shown remarkable improvement, but more needs to be done to achieve higher and sustainable level of sector growth. Agricultural value-added has picked up after the end of the civil war and growth has averaged around 5 percent or more per annum over the last few years. As such, agriculture has maintained its position as the key sector of the economy, contributing nearly half of the Gross Domestic Product (GDP), even after the country experienced a significant surge in mineral output, particularly iron ore, since 2011. 49. As demonstrated by the sector review, agricultural value-added per agricultural worker started improving after the civil war, but is still yet to catch up with the pre-war years. The historical trend shows that since independence, sector value-added had been growing at an average of 0.8% per year until 1980s, and surged to an average of 8.5% between 1981 and 1990 and thereafter declined sharply during the period of the civil war. However, agricultural productivity in terms of yields per hectare has been relatively low (even by sub-Saharan African standards), with cereal yields averaging less than 1.2 metric tones per hectare, and cocoa yields averaging less than 400 kg/ha. 50. The low levels of agricultural labor and land productivity invariably lead to depressed rural wages rendering the sector less competitive and a disincentive especially among the youth who are the most non/under-employed in the country. For the country to achieve the twin goals of reducing extreme poverty and improving shared prosperity, there is need for renewed focus to transform the largely subsistence low-input, low-output agriculture into more productive and commercialized system capable of creating decent gainful employment as a basis for broad-based poverty reduction. 51. In terms of food security, the overall food balance analysis shows that at the national level, kilocalories consumed per capita are on average less than the required amount. Although on average food quantity and kilocalories consumed have been improving, with a cereal-dominated diet, the level still falls short of meeting the recommended intake level of 2100 kilocalories per capita per day. Projecting into the future, based on the 2011 estimated food balance sheet, it means food production has to increase more than the rate of population growth in order to catch up with the shortfall. 52. The main binding constraints in the sector include: low productivity and competitiveness; ineffective delivery of public goods and services leading to low farmer access to technology through public research and extension; low private sector participation in agriculture; low and unreliable sources of agricultural financing; increased land tenure insecurity among investors, low level of farmers’ organization, including the capacity and organization of other stakeholders along the agricultural value-chain; degradation of natural resources and environmental; and inadequate sector coordination, capacity and governance. Many of these constraints were identified a decade ago, and a lot of effort has been undertaken to address them over the years. However, many of them linger on and continue to stifle sector growth.� e) The evolution of the Agriculture GDP in recent years in Sierra Leone: 53. The Maputo declaration was made with a stated objective reaching a 6% annual growth the agriculture sector. In Sierra Leone, the following graph shows the longer term evolution of the Agriculture GDP. 29 Figure 2 : Annual Agriculture GDP growth 1998-2011 (%) AgGDP annual Growth (%) 40 30 20 10 0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 -10 -20 -30 -40 Source: Statistic Sierra Leone in WB-WDI (website accessed) 54. During all this period, agricultural growth contributed to only 45-50% to overall growth in current terms over the period, which is highly insufficient for a sector that supports over 80% of the country’s population. As a result, rural poverty decreased slowly during the period. 55. The next figure focuses on the period considered by this AgPER and shows that annual AgGDP growth evolved during this period (2003-2011) between a minimum of 3.5% and a maximum of 9.5%, averaging at around 7%. One can see also that the AgGDP would have been slightly superior to the GDP growth during almost all the period before that evolution actually reversed in 2011. The latter is essentially attributable to the surge in mining revenues (= non agriculture growth). 30 Figure 3 : Evolution of the AgGDP and GDP annual growth over the period 2003-2012 14 12 10 AgGDP 8 Growth (%) GDP 6 Growth (%) 4 2 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: Statistic Sierra Leon in WB-WDI. 56. Overall, despite a level of spending sensibly below the target set in Maputo, an annual growth of over 6% was achieved for the agriculture sector. While this performance is important and positive, it should be kept in perspective that this has been achieved in a period of post- conflict recovery, when much needed to be re-built. Also, while certainly not negligible, the extent to which agriculture sector public expenditure may have contributed to this growth cannot be estimated. Finally, since 2009, annual AgGDP growth has evolved below the 6% target and a rebound of this growth would be needed in the coming years. a) Production, yields, self-sufficiency, food security 57. The following figure illustrates a long term perspective on the evolution of three important parameters : o The Agricultural Value Added Per Agricultural Worker o The Agricultural Value Added Per Hectare of Arable Land o The Cereal production (in metric tons per 100 inhabitant) 58. Based on existing statistics, one can conclude that the cereal production per inhabitant increased, as well as the value added per ha (with an acceleration since the last years, probably correlated to the global food prices increases), both being positive outcomes. On the other hand, 31 agricultural productivity (expressed in “Agricultural Value Added Per Agricultural Worker) decreased in average over this same period. The combination of these indicators tends also to infer that the production increase (kg of cereal/inhabitant) has essentially been driven by an increase in the area farmed rather than by an increase in productivity (Ag-VA per worker decreasing). This latter fact would confirm the still valid issue of the necessary improvement of access to basic agricultural goods (improved varieties, chemical inputs) and services (finance, extension…) by farmers in Sierra Leone. Figure 4: Evolution of the agricultural Value Added per worker, per ha of arable land and cereal production per 100 inhabitant (1980-2012) 2000 1800 Agricultural Value Added 1600 Per Agricultural Worker 1400 Agricultural Value Added Per Hectare of Arable Land 1200 Cereal production (metric tons per 100 inhabitant) 1000 Linear (Agricultural Value 800 Added Per Agricultural Worker) 600 Linear (Agricultural Value Added Per Hectare of Arable Land) 400 Linear (Cereal production (metric tons per 100 200 inhabitant)) 0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Source: Statistic Sierra Leon in WB-WDI. b) Delivery of agricultural services to farmers 59. As mentioned in chapter 1, sources of information on the delivery of services in the agriculture sector, be it public or private services, are scarce and often unreliable. In particular, MAFFS annual report only offer a partial account of all the activities carried out by MAFFS and likely to generate directly agricultural services or enhance their delivery by third parties. 32 60. The services section of the recent Agricultural Household Tracking Survey (AHTS 6 , 2011) contains however rather detailed information on access to all types of agricultural services at the level of the household, including training and extension services, farmer field schools (FFS), financial services (access to credit), relevant agricultural infrastructure, in particular, drying and storage facilities and finally markets. However, it is practically impossible in general to link the existence / delivery of these services to specific sector expenditure. 61. With regard to agricultural training and extension, extension officers were found to be the most frequently cited source of agricultural information, with 18% of farming households having spoken to an extension officer in the past 12 months. This number varies from 8% in Bonthe to 31% in Kailahun, with 18% being the national average. When asked which organization the extension officer belonged to, the majority of respondents cited MAFFS (39%) while international NGOs (World Vision, CRS) were also frequently cited. The following graph, extracted from the AHTS reports illustrates the situation. Figure 5 : Main organizations conducting agricultural extension in Sierra Leone Source: AHTS, 2011 62. While it is not surprising to see NGOs being active in providing extension services, the fact that MAFFS was cited in “only� 39% of the case, while at the same time MAFFS resources (including DP funded projects) are considerably higher than NGO resources should call to attention (even though the idea of devolving agricultural extension services to third party organization wouldn’t lack sense in theory). 6 The AHTS is based on a sample of 8840 Households 33 63. Disparities across districts were found to be important with regard to farmer field schools (FFS). Kailahun (15%), Koinadugu (14%) and Kono (13%) had the largest proportions of farmers who had been a member of a farmer field school (FFS) in the past month, while this percentage was lowest in Bombali, Kambia, Moyamba and the Western Area Rural (2%), and Port Loko (1%). These important differences should call the attention of the authorities overseeing the sector as it raises questions in terms of equity of the access to extension services by farmers. 64. The AHTS provides also a disaggregated analysis of the access to services across the core crops cultivated in Sierra Leone. Access to extension and information services is relatively homogenous amongst households cultivating food crops (rice, cassava, sweet potato, groundnut and maize). However, there is a greater proportion of households reporting access to extension, training and FFS amongst those cultivating tree/cash crops. Thus, about one quarter of the subset of farmers cultivating cacao, coffee and/or oil palm reported having spoken to an extension officer. 65. The AHTS mentions that the adoption of improved seed varieties remains low: only 2% of rice farming households having ever cultivated one of the NERICA varieties. Here also, there is substantial room to improve the situation with a substantial overhaul needed of the way MAFFS is strategically tackling this issue (see also section 4.4 reporting of the allocation of resources by MAFFS for the procurement and distribution of improved rice seeds pointing to significant inefficiencies). 66. With regard to the use of fertilizer, on average, about 7% of households use fertilizer, a strikingly low figure. These data were collected in 2010, prior to the effective start of the SCP program which has taken the option to set up Agribusiness Services Centers in order to boost access to fertilizers (notably). Elements are however not available yet to evaluate the impacts of this new governmental strategy. 67. The AHTS survey also reports that 51% of households had borrowed money ( access to credit) for any agricultural purpose in the past 12 months. In the majority of cases (56%) the lender was a friend or relative and the analysis of responses confirms the scarcity of lending through the formal credit market. This situation most certainly calls for an increased effort to promote the emergence of reliable and self-sustained sources of credit (such as micro-finance). 68. The AHTS questionnaire also explored the access to drying facilities for cereals. The responses most frequently given were a tarp (used by 31% of the farmers harvesting cereals), a straw mat (22%), the ground (37%), and a cement floor (17%). Since cement drying floors are considered the best drying space for farmers, the figure above would call for an increased use of sector resources to (co-)finance these very simple and small scale (usually) individually managed investments. 69. Finally, the AHTS also reveals that the distance from the community interviewed to the closest market avec in average of about 6.6 miles nationally (8.8 miles for permanent markets) with quite important regional variations. While this distance could seem relatively low, it is nevertheless certainly dis-incentivizing the commercialization of bulky agricultural products. Here again, the situation revealed could call for better strategizing of the investments on rural infrastructures (dimension that is not salient in the current MAFFS plans). 34 2 LEVEL OF PUBLIC EXPENDITURE IN AGRICULTURE 70. Chapter II provides a detailed presentation of the size and structure of public spending on agriculture. The definition of agriculture used follows strictly that of the international COFOG, complemented by the NEPAD’s guidance note7. Agriculture includes crops, livestock, inland and ocean fisheries, forestry, and the management of agricultural land. Excluded are multi-purpose projects unless 70 percent or more of their expenditure benefits agriculture. 71. Downstream functions such as marketing, storage, processing, access roads, and means of transport are on the principle not included in the concept of “agriculture� as defined in technical guidance. However projects in Sierra Leone that include activities to promote agricultural value chains and rural trade were systematically included in the expenditure shown in the following analysis. Throughout this report, spending data are shown in nominal terms only, not deflated. 7 African Union (2005). 35 Box 6: Data collection process. Overall, expenditures (budgeted and actuals) are thought to have been satisfactorily enough captured in this study. Details on the type of analysis to be carried out in the study as well as the source of data for each of the analyses were discussed with the counterpart team at the inception workshop to reach a common understanding of the scope and nature of the study. Required data included: approved budget; actual expenditure; and releases under the main budget line items disaggregated by functional directorates and units. On a case by case basis, templates for the basic data collection were prepared and agreed between the consultants, the counterpart team, and the Technical Working Group. In addition, the consultant team elaborated and discussed with the TWG a “data collection roadmap� which can be found in annex 3 to this report. “On budget� planned and executed expenditure were captured in full detail from printed copies of the “Recurrent and Development Expenditure Actuals� (detailed versions, parliamentary copy) for the years 2004 to 2012 included. Although some inconsistency or errors were spotted in these documents, they have proved to be good sources of information for the AgPER. Development Partner funded expenditures were captured using the ODA database maintained by the Ministry of Finance and Economic Development. While this source of information can ben considered as solid enough (high and consistent number of projects recorded), data only include spent budget (and not planned expenditure), and only aggregate information is available (level of expenditure by project and by year) and it is quite likely that a number of project of regional dimension or implemented directly by multilateral agencies were not recorded. Steps were taken by the consultant to complement this source of information with project specific information extracted from project completion reports or evaluations8. NGO expenditures were captured from the analysis of the annual reports of the MAFFS “NGO Bureau�. However data collection through this office only started effectively in 2011 and data were consistent only for 2012 and only for NGOs in direct contact with the MAFFS. This initiative needs however to be encouraged as a genuine effort is being produced by this office to interact in a constructive and organized manner with NGOs active in the sector and to track expenditure on the basis of the current NSADP (NGOs are asked to report on expenditure by pillar of the NSADP). District level expenditures (using GoSL internal resources) were accessed easily through the MoFED Local Government Finance Department. The level of detail available however did not allow for a functional analysis of expenditure. It was initially planned to devote some time during the consultants’ 3rd mission to Sierra Leone to interact with a sample of the districts. The Mission discussed the matter with the Technical Working Group and agreed that the four districts would be Bo, Kailahun, Moyamba and Port Loko (or alternatively Kambia). Time did not allow however for these on-site interactions to take place. In view of the increasing share of public expenditure transiting through local government office, it can be recommended that future AgPERs in Sierra Leone devote the time needed for these local interactions in order to analyze qualitatively at least the expenditure incurred for the agricultural sector at the district level. 8 The difficulty to access detailed data on DP funded projects calls for attention and agriculture sector GoSL institutions should more systematically keep track of project documents (at least the project document, the project execution report, including financial data, and – if any – the final evaluation of the project) while Development Partners could strengthen their effort to make this information available and archived in a convenient manner at country level. 36 2.1 PROJECTED AND EXECUTED BUDGETS OF MAFFS, MLCP, MFMR a) Budget funded by domestic resources MAFFS Budget: General Overview 72. The budget allocated to MAFFS every year by the Budget Act includes: o The budget of the Office of the Permanent Secretary. Attached to the Office of the Permanent Secretary budget line are two line items: National Farmer Association, Support to Agricultural Financial Management and Support to Agricultural Financial Management (IFMIS). o The budget of the Office of the Director General. A key budget line of MAFFS was until 2005 attached to the office of the Director General: the “food security� budget line (individualized since 2006). This budget line usually represents more than 65% of the recurrent non-wage expenditures of the Ministry and usually largely consists in subsidies over fertilizer and other agricultural equipment and inputs. o The budget of the Technical Divisions. There are five Technical divisions since 2009: Agriculture, Livestock, Land & Water/Agricultural Engineering, Food Security (see above) and Forestry. The Land & Water/Agricultural Engineering division was called simply Land and Water Division up to 2005. The number of divisions has varied over the period considered by the AgPER Work. o The budget of the Planning / M&E / Statistics Division. o Allocations to Extension Services (which started in FY 2010). o Allocations to various “Agricultural Institutions� including: the National Agriculture Coordination Council, the Institute of Agricultural Research, the Rice Research Stations (RRS) and the National Agriculture Training Center (NATC) (which was until 2008 included under the Permanent Secretary Office budget). 73. The Office of the Permanent Secretary, Technical Divisions, PEMSD, the Extension Services and the Agricultural Institutions have separate lines of funding in their provisional budgets for staff expenses (called recurrent personnel expenditures), operating expenditures (called recurrent Other), and their capital expenditures from domestic (called domestic development expenditures) and external resources (called foreign development expenditures). In practice however, the capital expenditures are not attached to any MAFFS division in particular. They consist essentially in a list of agricultural development projects for which the government contribution (counterpart funds) is presented as “domestic development expenditure�. 74. It is likely that a number of MAFFS entities / divisions have also their own resources. Some of the Technical Divisions were reported to have put in place “cost-recovery mechanisms� for services provided (ex.: veterinary services, provision of seeds on cost-recovery basis). It is not clear however if these Divisions have a MoFED-certified advance funds administrator who turns this revenue over to the Treasury. In theory, this revenue should figure in the State budget and represents relatively substantial sums (based on the fact that the food security division budget line has represented 17 billion Leones per year in average for the last two years). It may also well be that a certain percentage of this revenue is placed in a special account that the divisions concerned can use for operating expenses, but these revenues are then not included in the State budget (and probably managed by an accountant assigned to each of these institutions). The recent 37 establishment of the ABCs may have modified this situation but the information available at the time of the completion of the AgPER was too little to ascertain this9. Level and Evolution of the MAFFS Provisional and implemented Budget 75. The MAFFS provisional budget (Figure 1) increased almost 1.5 times in current terms between 2004 and 2012, increasing from 23 to 34 billion Leones, which is an increase of around 50% in real terms. A peak can be observed in 2009 (40 billion Leones), followed by a relative stable trend. The executed budget has followed a similar trend during the same period. This increase in the core of the MAFFS budget over the 10 year period considered obviously illustrates the desire of public authorities to increase support to the agricultural sector. It should be noted however that the MAFFS budget estimates increased however less rapidly than the State’s aggregate expenditure which, over the same period, increased nearly fourfold in current terms. Figure 6. Evolution of the MAFFS provisional and implemented budget from domestic resources in current and constant terms, 2004–2012 40,000 35,000 30,000 Budget (current) Actual 25,000 (current) Actual (constant) 20,000 15,000 10,000 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Expenditure actuals, Parliamentary copy (years 2004 to 2012). Level and Evolution of the Provisional and Implemented Budget of the other ministries and entities having expenditures in the agriculture sector 76. Besides MAFFS, two other ministries have budgeted activities (and expenditures) in the Agriculture Sector, namely the Ministry of Marine Resources and Fisheries and a fraction of the budget/expenditure of the Ministry of Land and County Planning10. In addition, the AgPER has 9 For a more extended discussion on revenue generation at MAFFS, see chapter 4.5. 10 See explanation in chapter 1.3 b) Public Institutions in Sierra Leone’s Agriculture Sector 38 considered the following budget lines / expenditures as pertaining to the Agriculture Sector (based on the COFOG definition) : National Commission on Environment and Forestry (code 410), Sierra Leone Agricultural Research (code 418), Sierra Leone environmental Protection Agency (code 420). Table XX provides a summary table of the aggregate expenditures from these agencies and the MAFFS for the last two years and allows to get a quick sense of their relative importance. Figures 3 and 4 respectively illustrate the evolution of the provisional and implemented budget for these two ministries and the three additional areas entities considered. Table 2: Summary of aggregate expenditure (as a share of total expenditure) for the ministries and entities considered in the AgPE review (years 2011 and 2012) 2011 2012 Ministry of Agriculture, Forestry and Food Security 67% 62% Ministry of Lands and County 6% 5% Planning Ministry of Marine Resources and Fishery 4% 13% National Commission on Environment and Forestry 0% 0% Sierra Leone Agricultural Research 20% 17% Sierra Leone Environmental Protection Agency 3% 3% Source: Expenditure actuals, Parliamentary copy (years 2004 to 2012). Note: 2011-2012 data for “Sierra Leone Agriculture Research� are presented as found in the officially published Government yearly “Actuals� (parliamentary copy) but could not be verified. 39 Figure 7. Evolution of provisional budget from domestic resources in current terms, 2004– 2012 12,000 Ministry of Lands and County Planning 10,000 Ministry of Marine Resources 8,000 National 6,000 Commission on Environment and Forestry 4,000 Sierra Leone Agricultural Research 2,000 Sierra Leone environmental - Protection Agency 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Expenditure actuals, Parliamentary copy (years 2004 to 2012). Note: 2010-2012 data for “Sierra Leone Agriculture Research� are presented as found in the officially published Govt yearly “Actuals� (parliamentary copy) but could not be verified. Figure 8: Evolution of implemented budget from domestic resources in current terms, 2004– 2012 12,000 Ministry of Lands and County Planning 10,000 Ministry of Marine 8,000 Resources National 6,000 Commission on Environment and Forestry Sierra Leone 4,000 Agricultural Research 2,000 Sierra Leone environmental Protection - Agency 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Expenditure actuals, Parliamentary copy (years 2004 to 2012). Note: 2010-2012 data for “Sierra Leone Agriculture Research� are presented as found in the officially published Government yearly “Actuals� (parliamentary copy) but could not be verified. 40 77. The overall trend shows a vigorous increase (+20% annual) for the sector expenditure spent (coming internal resources). One can notice that provisional and implemented budgets have increased in average more rapidly for the MFMR and the MLCP than for MAFFS over the period 2004-2012. Average annual increase in spending on IR (2004-2012) MAFFS 14% MLCP 35% MFMR 68% 78. And as illustrated by the following charts, MAFFS retains the largest share of the sector expenditure (IR). Figure 9: Evolution of the agriculture sector institutions executed expenditure (cumulated, in ‘000 Le) 70,000 Sierra Leone environmental 60,000 Protection Agency Sierra Leone Agricultural 50,000 Research National 40,000 Commission on Environment and Forestry Ministry of Marine 30,000 Resources and Fishery 20,000 Ministry of Lands and County Planning 10,000 Ministry of Agriculture, Forestry and Food Security - 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Expenditure actuals, Parliamentary copy (years 2004 to 2012). 41 Level and Evolution of the Provisional and Implemented Budget of other selected other key ministries (education / Health) 79. The figure 10 illustrates how expenditure (on internal resources) have comparatively evolved for the agriculture sector and for two key ministries (namely Health and Education). Figure 10: Compared evolution of implemented budget from domestic resources in current terms, for the agriculture, health and education sector (2004-2012) 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 - 2004 2005 2006 2007 2008 2009 2010 2011 2012 Agriculture Sector - Cofog (IR) 19,39226,18125,44617,16429,35937,83330,97946,97661,345 Ministry of Health (IR) 29,24437,89744,97815,45632,03854,816104,89134,99140,72 Ministry of Education (IR) 51,41551,41351,12429,94352,28268,11978,20781,361135,82 Source: Expenditure actuals, Parliamentary copy (years 2004 to 2012). 80. The overall trend in terms of budget allocated to the agriculture sector (which has tripled in nominal terms over the period) certainly reflects an increased attention paid to the sector during the period of re-construction and concomitant with the Maputo declaration. This fact must nevertheless be nuanced insofar to the extent that other sectors have also benefited from increased attention. In 2012, MAFFS actual expenditure ranked #10, behind Education, Works, Health, Defense, Energy, Police, Foreign Affairs, Social Security and Office of the President. b) Internally funded decentralized expenditure 81. As Sierra Leone has engaged a vigorous devolution process, expenditure directed at local level have increased significantly over the period considered and agriculture is definitely part of 42 the sectors given priority too with regard to the decentralization process. The figure 7 illustrates the evolution of the decentralized expenditure, sector by sector. While the health sector is evidently the sector that has benefited the most from the devolution process, “agriculture� comes just second, before “education�. In 2012, decentralized expenditure in the agriculture have reached a yearly amount of nearly 14 billion Leones, a relatively significant amount when compared to internally funded expenditure made at central level (around 40 billions) for the same sector. Figure 11: Local level expenditure (spent budget), by sector (2007-2012) 40,000,000,000 35,000,000,000 30,000,000,000 Administra ve 25,000,000,000 Health Educa on 20,000,000,000 Agriculture Rural Water 15,000,000,000 Solid Waste Others 10,000,000,000 5,000,000,000 0 2007 2008 2009 2010 2011 2012 Source: MOFED. Local Government Finance Department. 82. A key feature of local spending is their quite massive increase over the period. Indeed, these latter grew of +237% per years for the agriculture sector (+132 % per year for all sectors). These sustained increases while potentially positive, raise important questions in terms of capacity building for the local government officials in charge of administering these resources. Chapters 3 and 4 provide cautionary notes in this regard, recommending that the authorities build more capacity and improve procedure and oversight. Chapter 5 discusses options for strengthening the technical capacity of the civil service at local level. 83. Similar trends (devolution of funding) have been observed in other sub-saharian Africa countries as illustrated in the following table. In most case , however , the increase in spending by local governments in these countries has not been as strong as the observed increase in Sierra Leone. 43 Table 3: Evolution of devolved spending in selected sub-saharian Africa countries Countries 2005 2009 2011 % Change* Burkina Faso 125 250 274 119% Côte d’Ivoire 72 133 - 85% Ghana* 125 251 333 167% Liberia 1 45 56 1580% Senegal* 190 360 - 89% Togo 15 37 66 352% Source: World Bank, various AgPERs. * over the period 2005-2011 c) Budget funded by external resources 84. External resources are a very significant portion of the budget allocated to / spent in the agriculture sector (as well as in many other key sectors such as health and education) in Sierra Leone. Figure 8 (based on the MoFED ODA monitoring system) shows that donor funded expenditure in the agriculture sector have increased very sharply (more than ten-fold) from 2003 to 2009 from just around 3 million $US annually to more than 50 million $US per year. Since 2010, a more stable situation can be observed with annual expenditure fluctuating around 40 million $US. Figure 12: Externally funded expenditure (agriculture sector), 2003-2012 60,000,000 50,000,000 NRM related 40,000,000 Agriculture expenditure Agriculture 30,000,000 20,000,000 Sum in constant( 2005 &US) 10,000,000 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: MoFED / DAC database11 11 Note: the distinction between Agriculture and “NRM related agriculture expenditure� comes from the MoFED ODA database classification. The sum of the two is assumed to represent the total agriculture (COFOG) expenditure externally funded. 44 85. A similar increasing trend was at play over the same period of time for the overall (all sectors aggregated) externally funded expenditure in Sierra Leone (as illustrated by figure 9). In effect, DP funded agriculture sector expenditure have increased annually by 50% over the period, while this increase was higher for all sectors (+110% annually). Figure 13: Externally funded expenditure (across sectors), 2003-2012 Donor funded expenditure - US$ (2003- 2012). All sectors. 500000000 450000000 400000000 350000000 US$ Current 300000000 Axis Title 250000000 US& 200000000 constant (2005) 150000000 100000000 50000000 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: MoFED / DAC database 86. Regarding the share of agricultural expenditure in total expenditure, the following table shows that they have averaged slightly above of 10% during the period 2004-2012, ranking second behind DP funded health sector expenditure. 45 Table 4: Share of various sectors in total externally funded expenditure in Sierra Leone (2003- 2012) SECTOR Share of total Expenditure (externally funded) Agriculture (Cofog) 10,38% Education 7,18% Health (incl. HIV) 14,95% Road Network 8,36 Source : MoFED / DAC database 46 2.2 TRENDS IN AGRICULTURAL SECTOR EXPENDITURE AND THE MAPUTO TARGET a) A sector predominantly funded by ODA 87. The following figure offers of comparison of agriculture sector expenditure evolution over time, for three types of expenditures : expenditure spent by local and central governments (Internal Resources) and Development Partners funded expenditures. MAFFS expenditure (IR) are individualized in order to better visualize the weight of this key ministry in the total. Figure 14: Comparison of agriculture sector expenditure (various sources of funding and spending agencies, in million Leones) 300,000 250,000 Donor funded expenditure 200,000 Tranfers to District Councils 150,000 Other (COFOG) 100,000 ministries and entities MAFFS 50,000 0 2007 2008 2009 2010 2011 2012 Source: MoFED, expenditure actuals (parliamentary copy), DAC database and author’s calculations. 88. As can be seen, the expenditure financed by donors represents a largely dominant share of total expenditures. These expenses represented 75% of total expenditure over the period and this trend does not appear to vary much during the period. This percentage is generally high, although it is not a characteristic uniquely of the agricultural sector in Sierra Leone (similar percentages are observable in other key sectors). The following table offers a quick comparison with other countries in the sub-region : 47 Table 5: Share of domestic fund in total agriculture expenditure Countries 2005 2009 2011 Sierra Leone 54% 20% 26% Burkina Faso 24 37 48 Côte d’Ivoire 62 Ghana* 73 61 63 Liberia - 21 38 Senegal* 45 41 - Togo 64 Source: Hourticq J., Zorya S. WB. Dar Es Salaam Workshop. 89. Chapters 3 and 4 provide cautionary notes in regard to the weight of development partners funded expenditure in total expenditure, especially in a context / sector where ODA is highly fragmented. Chapter 5 discusses options for improving this situation. b) Expenditure levels consistently bellow the Maputo statement of intent 90. Under the Maputo Declaration of 2003, African governments committed to increasing the share of their national budgets devoted to agriculture, with a target of at least 10% so as to achieve agricultural growth of at least 6% per year. The expenditure target refers to all funding sources combined. While the level of public resource allocated to and spent for the agriculture sector in Sierra Leone has increased considerably over the period considered, it is worth examining briefly here how the level of financing compare with the Maputo target set in 2003. The graph hereafter allows visualizing this evolution over the period 2004-2012. 48 Figure 15: Evolution of combined agriculture expenditure (% of total) in Sierra Leone and comparison to the Maputo Target. 12.00% 10.00% 8.00% Agricultu re Exp. (COFOG) 6.00% Maputo Target 4.00% 2.00% 0.00% 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: MoFED (DAC database and Local Government Finance Department data), Expenditure actuals (Parliamentary copy, years 2004 to 2012). 91. Combined spending from internal resources and donors on broad agriculture was approximately 4% percent in FY 2004 to 2007, reaching 9% in FY 2008 and 2009 and approximately 6-7% in the recent years. 92. A conclusion is that the level of financing for the sector was below the 10% target for the years 2004-2012 included. In average, Agriculture sector expenditures have represented 6.04% of the overall implemented expenditure during this period, all sources of funding taken into account This “below target� level of spending seems to originate essentially from the low level of allocation (and therefore spending) on internal resources to the agriculture sector, in particular at central level. Table 5 below provides a breakdown of the expenditure levels by source of funding and destination over the period 2007-2012. Table 6: Percentage of total expenditure executed in the agriculture sector (2007-2012) YEARS 2007 2008 2009 2010 2011 2012 Transfers to district council (IR) Share of Ag Sector in total 11.10 13.59 21.37 23.47 22.95 22.11 disbursement % % % % % % Dev. Partners funded Expenditure (ER) Share of Ag Sector in total 15.70 14.78 12.15 10.75 disbursement 5.47% % % 7.82% % % Govt Expenditure at central level (IR) Share of Ag Sector in total disbursement 1.99% 2.61% 2.79% 3.32% 3.67% 2.45% Source: MoFED (DAC database and Local Government Finance Department data), Expenditure actuals (Parliamentary copy, years 2007 to 2012). 49 93. One can notice that the agriculture sector has quite systematically benefited from a level of expenditures above the 10% target when it comes to externally funded expenditure. On the other hand, internally funded central level expenditure for the agriculture sector (in other words the budget spent by the agric. sector ministries coming from IR) has systematically evolved well below the 10% target (highest figure being 3.67% in 2011, lowest 1.99% in 2007), while transfer to the district councils earmarked for the agriculture sector always reached more than 10% of the total transfers, reaching 22% in 2012. 94. To demonstrate a level of support in line with the Maputo declaration, Sierra Leone would have to reach a level of agricultural support in the order of USD 92.7 million per year12. It is interesting to compare this figure with the needs as estimated by the Sierraleonese government when the National Sustainable Agriculture Development Plan was elaborated under the CAADP auspices in 2009. The NSADP/CAADP was estimated to require a total outlay of USD 333.5 million for its implementation in the first five years (2010-2015), equivalent to around USD 67 million per year. Similarly, the 2010 elaborated Smallholder Commercialization Plan (SCP) indicates a total provisional implementation cost of USD 402 million over a period of 5 years, equivalent to approximately USD 80 Million per year. 2.3 Share of Agriculture Expenditure Relative to GDP and AgGDP 95. When it comes to agricultural support expressed in shares of GDP and agricultural GDP, Sierra Leone places itself in the mid-range of sub-Saharan African countries and (like many other low income countries) and for obvious reasons far below the level of spending of middle- and high-income countries (Table 6) if we are looking specifically at the support to agriculture in shares of the AgGDP. Figure 16: Trends in Agricultural Sector Expenditure, percent of GDP and AgGDP, 2004-2012 Agric Sector Expenditure as % of GDP Agric Sector Expenditure as % of AgGDP 6.00% Expenditure Share in GDP and 5.00% 4.00% AgGDP (%) 3.00% 2.00% 1.00% 0.00% 2004 2005 2006 2007 2008 2009 2010 2011 2012 Year Source: Source: MoFED (DAC database and Local Government Finance Department data), Expenditure actuals (Parliamentary copy, years 2007 to 2012) and Statistic Sierra Leone (for GDP and AgGDP 12 Based on 2012 total expenditure level. See Table 1. Estimate of implemented agricultural public expenditures using NEPAD’s COFOG methodology, 2007–2012 (in millions $US) for the details. 50 96. Here we can see that to demonstrate a level of support in the lower range of middle- income countries (10% of agricultural GDP), Sierra Leone would have to reach a level of agricultural support in the order of USD 136 million per year. It is interesting to compare this figure with the needs as estimated by the Sierraleonese government when the National Sustainable Agriculture Development Plan was elaborated under the CAADP auspices in 2009. The NSADP/CAADP was estimated to require a total outlay of USD 333.5 million for its implementation in the first five years (2010-2015), equivalent to around USD 67 million per year. Similarly, the 2010 elaborated Smallholder Commercialization Plan (SCP) indicates a total provisional implementation cost of USD 402 million over a period of 5 years, equivalent to approximately 80 million per year. Table 7: International comparison of budget transfers to agriculture, 2002–2012 Share of Share of agricultural Share of agricultural Region/Country agriculture in public expenditures in public expenditures in GDP national GDP agricultural GDP High-income countries Australia 3% 0.30% 10% Canada 2% 0.50% 22% EU 2% 0.70% 28% USA 2% 0.70% 46% Middle- income countries Turkey 13% 2.00% 15% Mexico 4% 0.70% 18% Venezuela 5% 0.50% 12% China 15% 1.20% 8% Brazil 9% 0.70% 8% Russia 6% 1.00% 16% Ghana 34% 1.43% 4.31 Ukraine 12% 1.30% 11% Low-income countries Sierra Leone (avg. 48% 1.72% 3.70% 2007-2012) Uganda 32% 1.50% 5% Tanzania 45% 1.20% 3% Ethiopia 44% 2.70% 6% Kenya 29% 1.30% 4% Cameroon (2012) 21% 0.90% 4.10% Burkina Faso (2004– 33% 2.70% 8.20% 2011) Senegal (2009) 14% 2.80% 20.40% Côte-d'Ivoire 23% 0.60% 2.50% Guinea (2003–12) 22% 1.50% 6.80% Togo 41% 1.90% 4.60% 51 Note: the data shown refers to different years within the 2002 and 2011 period, depending on the country. Source: World Bank 2010a; and various authors of country specific AgPER country reports 2.4 EXPENDITURE PER CAPITA The level of agriculture public expenditure per capita in Sierra Leone can be considered as relatively low in absolute terms and in comparison to other sub- Saharan countries. Table 8: Expenditure per capita (USD) in various Sub-Saharan countries Countries 2005 2009 2011 Sierra Leone 1.75 13.11 8.61 Burkina Faso 11.2 20.8 22.0 Côte d’Ivoire 12.3 19.1 - Ghana* 13.0 25.8 33.9 Liberia 0.7 4.7 10.0 Senegal* 29.8 51.1 - Togo 4.5 11.0 19.3 * Off-budget expenditure not included. Source: Hourticq J., Zorya S. WB. Dar Es Salaam Workshop. 97. The figures above are illustrating clearly how scarce resources are and therefore how important an efficient use of these resources is in this context. One should note here that in overall, low levels of expenditure in the agriculture sector has been a common characteristic of many sub-Saharan countries for the last decades. The following graphs offer an interesting comparison of long terms trends in terms of public agriculture expenditure across continents. Over the period 1980-2007 studied, spending on agriculture per agricultural population has increased in all regions except Latin America and the Caribbean (LAC) and Sub-Saharan Africa (SSA). Even more striking, the level of expenditure per capita in sub-Saharan Africa is significantly lower than that observed in Latin America for example. 52 Figure 17: Per agricultural capita government expenditure on agriculture by area, 1980 – 200713 Source: Sarah K. Lowder and Brian Carisma, FAO. 2011. 13 Source: Financial resource flows to agriculture. A review of data on government spending, official development assistance and foreign direct investment. Sarah K. Lowder and Brian Carisma ESA Working paper No. 11-19. Dec-11. Agricultural Development Economics Division. Food and Agriculture Organization of the United Nations. www.fao.org/economic/esa. 53 3. PUBLIC EXPENDITURE IN AGRICULTURE: ECONOMIC AND FUNCTIONAL COMPOSITION, REGIONAL DISTRIBUTION 3.1 OBJECTIVE OF THE ANALYSIS 98. In this chapter, we analyze the economic expenditure to determine whether available resources are allocated in a way that maximizes desired outcomes, and whether inefficiencies in allocations are identifiable. The aim is to answer the questions: “Are the agricultural expenditures that are being financed best suited to achieving the country’s agricultural policy goals? Is the spending sound?�. Allocative efficiency will primarily be determined by the efforts made in budget preparation to align the budget with sectoral goals. Allocative efficiency will be analyzed as follows: o Economic breakdown of expenditure: wage expenditure, non-wage recurrent and development expenditures; o Functional breakdown of expenditures: distribution among the various subsectors (agriculture, livestock, research, etc.); and o Regional distribution of expenditures: we analyze expenditure allocation by region to determine whether allocations are consistent with indicators such as the size of the rural population, spatial distribution of poverty and food insecurity. 3.2 ECONOMIC COMPOSITION OF EXPENDITURE 99. The analysis starts with the expenditure of the three ministries (MAFFS, MMFMR and MLCP) and the three other institutions identified as being part of the agriculture sector (COFOG). The overall budget as set forth in budget laws and actuals reports provides the following picture of the economic composition of their expenditure over the period covered by this review. Figure 18: Evolution of personnel, recurrent non-wage and development expenditure over the period 2004-2012 100% 90% Domestic 80% development Expenditure 70% 60% Non salary non interest 50% recurrent expenditure 40% 30% Personnel Expenditure 20% 10% 0% 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: expenditure actuals, Parliamentary copy (2004-2012) 54 100. As can be seen, recurrent expenditure (salaries and non-salary recurrent expenditure) represent by far the largest share of total spending (close to 80%) at central level on internal resources. Personnel expenditures are relatively stable, absorbing in average 40% of the annual budget (IR) of these institutions. This situation is not uncommon in low income Sub-Saharan Africa countries. Both wage and non-wage shares of total expenditure in the agriculture sector vary from one country to another as illustrated in the table below : Table 9: Share of wages and non wages expenditure in various sub-saharian African countries (2009) Countries Share of wages in Share of non-wage expenditure (IR) recurrent exp. In total 2009 exp. (IR) 2009 Ghana 22 60 Liberia 48 83 Senegal 6 39 Togo 17 73 Sierra Leone 35 56 Source: Hourticq J., Zorya S. WB. Dar Es Salaam Workshop. 101. A positive aspect of how the Sierra Leonean situation has evolved over the last 10 years is the fact that the share of Domestic development expenditure is on the increase: representing less 10% in 2004 they hovered around 20% in recent years. This leaves however very limited margin of action for these institutions to play their role as the majority of their budget is used to cover recurrent expenditures. 102. Limiting the analysis to expenditure funded on internal resources and spent by central level administrations is however very partial as the majority of expenditure are actually funded by external resources in the sector (see Chapter 1). However, it has not been possible for the consultant team to collect a detailed enough database of project expenditure over the period covered by the review (see 3.3.2. for a limited only analysis). Some of the project expenditure are labeled as “foreign development expenditure� in the yearly “actuals� books, and it would suggest that it is essentially capital expenditure when actually projects are funding significantly for operating expenses and personnel. It is clear that the personnel costs and operating expenditures paid for by project funds therefore supplement ministries’ official resources including by financing salaries (technical assistance) and non-wage recurrent expenditure. 103. Another area that would need further analysis is the economic composition of expenditure devolved to local government. Here again unfortunately, not enough detail was available for the consultant team to conduct such an analysis. 55 Economic Composition of the externally-funded projects’ Budgets 104. Running an analysis of the economic composition of externally funded projects in the agriculture sector in Sierra Leone remains a challenge. Firstly, the data collection phase of the AgPE study did not allow for the collection of a comprehensive set of project expenditure data set. The archival system within the administrations of the sector is very weak and actually not in place. It has been even difficult to access this information from the ongoing projects. This is certainly an area that needs further attention from the GoSL and the DPs. In many cases, external PIUs were formed for the purpose of administering and implementing these projects but once these projects are completed, these PIUs were disbanded and the information is simply lost. 105. The other reason why it is difficult to perform such an analysis comes from the fact that the vast majority of the externally funded project budgets and accounting systems are not actually organized by economic type of expenditures but rather following an activity based system of budgeting with some functional breakdown encompassing: components, sub-components, activities, technical assistance, etc. Even though the economic nature of some budget lines can be relatively easily identified (if one can access detailed accounts of these projects expenditure) such as the cost of TAs, it is nearly impossible to differentiate operating costs, from investments, etc. 106. Ideally, all these projects should adopt an accounting system where activity based planning and budgeting is the norm. And reporting should be executed along a double system allowing to track each expenditure according to the functional and economic category it belongs to. If a transition to program budgets is confirmed in Sierra Leone, it would be desirable to adopt a budgetary nomenclature that allows a more precise analytical accounting of expenditure for externally funded project by economic category, namely personnel costs, operational expenses, and investment. It would also be useful to begin making a distinction between investments made on behalf of government entities (e.g., office building, vehicle purchases, computer equipment, provision of training, etc.), and investments made in the field (e.g., productive and social infrastructure, training and equipment of beneficiaries, etc.) in order to measure more precisely the impact and effectiveness of programs. A more precise accounting of capital formation at the different levels should also highlight the issue of recurring costs of operation and maintenance — which has been mostly neglected until now—for equipment made available to government entities, public infrastructure implemented, and infrastructure transferred to beneficiaries. 107. On average more than two-thirds of allocated costs are borne by projects and do not officially appear in the ministries’ budgets. This poses two fundamental questions: o First, this situation prevents these ministries from effectively managing their recurrent costs since they lack a clear picture of those costs and have only a few levers for controlling them, which may allow them to take advantage of economies of scale. Truthful budgeting needs to be reestablished in order to lay a foundation for an effective and sustainable approach within the sector in collaboration with the DPs; o Second, this situation begs the question of the sustainability of actions: What happens to the provisions needed to supervise and maintain the investments made once the project that supported them comes to an end? At present, this question does not appear to be addressed systematically. 3.3 FUNCTIONAL COMPOSITION OF EXPENDITURE 108. The analysis of the functional composition of expenditure in the agriculture sector looks first at the functional breakdown of expenditures i.e. their distribution among the various 56 subsectors (agriculture, livestock, research, etc.). Ideally, the objective of such an analysis would to measure how far the government has ensured that its policy intents and priorities, which are normally explicitly stated in its policy document, are reflected in annual budgets. However, the budgeting / reporting process is not based on the current policy / programmatic framework14 (the NSDAP essentially) making it actually impossible to track expenditure by Program and Sub- Program of the NSADP. a) Functional analysis of Development Partners funded expenditure in the agriculture sector 109. The functional analysis of expenditure is hereafter conducted by comparing the sub- sectoral composition of the expenditure to the actual value added created by each of these sub- sectors. The analysis is based on data assembled by the MAFFS in 2011 with support from a FAO funded Food Security project (called MAFSAN). The objective of this project was to evaluate and track expenditure at country level on food security. 110. The data were extracted directly from the project website (at www.mafsan.org) and sorted to suit the AgPE needs and methodology (in particular the COFOG perimeter). It assembles expenditure data coming from 65 projects that were active during the period 2008- 2011 (4 years). The portfolio includes projects funded by bilateral and multilateral donors, as well as NGO and projects implemented by UN organizations. 111. In total, the data assembled on the 65 projects represent a volume of expenditure of over 140 million $US, a figure that is quite consistent with the figures available through the MoFED ODA database. The expenditures in agriculture over the 2008–2011 period were classified according to the MAFSAN project classification. The following pie-chart provides a quick picture of the functional composition of the DP funded expenditures by main categories: 14 There is to our knowledge one exception to this situation. The MAFFS NGO bureau actually request NGOs to report on their expenditure using a format that follows the NSDAP structure. This is an initiative to encourage and that should be applied systematically to every DP funded project in the Agriculture Sector in Sierra Leone. 57 Figure 19: functional composition of Agriculture Sector spending (external resources) Fishery Forestry development development (Subsector) (Subsector) 3% 0,07% Food aid/Food security programmes Agricultural water (Subsector) resources 15% Agricultural (Subsector) 5% development (Subsector) 43% Agricultural policy and Agricultural administrative services management (Subsector) (Subsector) 23% 6% Agricultural Agricultural Agricultural Agricultural research land resources inputs education/trainin (Subsector) (Subsector) (Subsector) g (Subsector) 0,64% 0,03% 3% 2% Source: MAFFS/MAFSAN database15 15 The “agriculture development sub-sector� (as defined by the MAFSAN/MAFFS team) is a diverse collection of themes including support to Livestock, Livestock, Integrated agriculture projects, Agriculture infrastructure, etc. and totals 43% of the expenditure. A more fine-tuned analysis would be therefore required to understand their signification. 58 112. The next pie-chart shows the relative weight of various sub-sectors in the overall Agriculture GDP, in average, over the period 2003-2012. Figure 20: Sub-sector contributions to AgGDP (av. 2003-2012) Sub-sector contributions to AgGDP (av. 2003-2012) Fishery 17% Forestry 15% Crops 63% Livestock 5% Source : World Bank, WDI. IMF. Food Security and Agriculture Services 113. The two major areas of expenditure funded by Development Partners are : o “Agricultural services� (which include according to the MAFSAN classification : Agricultural co-operatives, Agricultural financial services, Agricultural services, Plant and post-harvest protection and pest control, Livestock/veterinary services) and; o Food Aid / Food Security (15%). 114. The relatively high level of expenditure on “agricultural services� is probably reflecting the effort put by the MAFFS in developing agriculture service centers through the ABCs. Effort put on food security and notably food aid were also significant over the period and essentially carried by the WFP in the form of food aid. 59 Irrigation / Agriculture Water 115. Agriculture Water Resources sub-sector accounted for around 5% of the donor funded expenditure during the period and this refers essentially to projects targeting the development of rice cultivation in the inland valley swamps. Livestock / Fishery 116. Expenditure related to supporting the livestock sector, estimated at USD 2,449,296 from 2003 to 2011, or 2% of total expenditures may be considered relatively low considering the role that livestock plays in food security, savings, and income generation. More preoccupating is the amount spent on Veterinary Service which by all standards can be considered suboptimal: less than USD 200,000 in total over the 9 years considered (less than USD 20,000 per year). This amount is also much lower than the sector’s weight in the agricultural economy (5% of agriculture GDP in average during the period 2002–2011). 117. Similarly to the livestock sub-sector, the fishery sub-sector has accounted for only 0,2% of the total donor funded expenditure, a figure in sharp contrast with the contribution of the sub- sector to the Ag-GDP (16% in average over the period 2001-2012). Agriculture policy / administrative management 118. Expenditures related to agriculture policy development and administrative management have accounted for 6% of the total agriculture expenditure, reflecting the effort put by donor and the GoSL in shaping a vision for the sector (substantial policy work has been undertaken since 2003, see chapter 1) and re-building the capacities at Central Level notably through trainings. It is uncertain to what extent this figure includes or not the projects’ administration costs. Agriculture extension 119. Agriculture extension (including District extension services, Agricultural extension, Radio, TV, electronic extension messages) received only 2.11% of the total spent in the sector through the projects, which appears as a low figure in comparison to the central role that extension can play in the dissemination of agricultural technologies. Agriculture Research Expenditure 120. The MAFSAN database identifies only very few DP-funded expenditure on agriculture research (USD 900,000approximately, over a period of 4 years) and this does probably not provide with an accurate picture of the recent (though modest) efforts put by development partners in developing agriculture research in Sierra Leone. A recent publication derived from an IFPRI managed program (ASTI 16 ) provides a probably more accurate picture of the R&D agriculture public expenditure in Sierra Leone. According to the authors and the database available online at www.asti.cgiar.org/data, the following trend emerges: agricultural research and development (R&D) expenditures in Sierra Leone would have more than doubled between 2001 and 2009 in response to the end of a decade of civil war and efforts to reconstruct the country’s agricultural R&D system and donor support has been increasing since 2007. 16 ASTI /IFPRI / SLARI. Recent Development in Agricultural Research. Country Note : Sierra Leone. 2010. 60 121. Based on the ASTI/IFPRI/SLARI data, agriculture research expenditure in Sierra Leone would have reached USD 5.1 million in 2009, a significant increase in comparison to 2001 (USD 2.8 million). This amount of USD 5,1 million represented 0,38% of agriculture’s share of GDP and 0.21% of GDP overall for the year 2009, percentages which are still far below the objective of earmarking 1% of (Ag)GDP for applied (agricultural) research set by the Executive Council of the African Union in its Khartoum Decision regarding science and technology in 2006. The following table provides elements to compare the Sierra Leonean situation with other Sub- Saharan African countries. Sierra Leone actually ranks among the Sub-Saharan Africa countries that have allocated the least amount of resources to agricultural R&D. Table 10: Funds budgeted for agricultural R&D in sub-Saharan Africa Total amount Amount per Year studied % of GDP (USD capita (USD) millions) Malawi 2007 180.1 12.9 1.70 Uganda 2007 359.8 11.6 1.10 South Africa 2007 4976.6 102.4 1.05 Kenya 2007 277.8 7.4 0.48 Senegal 2008 99.0 8.0 0.48 Tanzania 2007 234.6 5.8 0.48 Gabon 2008 78.7 58.3 0.47 Ghana 2007 120.1 5.0 0.38 Zambia 2008 55.3 4.6 0.37 Mali 2007 37.4 3.0 0.28 Mozambique 2007 42.9 2.0 0.25 Nigeria 2007 583.2 3.9 0.20 Burkina Faso 2004–2011 8.1 0.5 0.10 Togo 2005–2010 2.0 0.4 0.07 Sierra Leone * 2009 5.1 0,38 0,21 Source: AU/NEPAD 2010; World Bank 2012; Authors’ calculations for Sierra Leone. * data source : ASTI/IFPRI/SLARI database 122. It is anticipated that agricultural R&D capacity will increase in the coming years as more SLARI research centers open. Also, the WAAP, a large scale World Bank funded regional project funding agricultural R/D is currently active in Sierra Leone and one of the projects boosting SLARI’s capacities, notably by synergizing its research programs with other western African countries. Box 7: limits of functional analysis of agriculture expenditure Analyzing the functional composition of expenditure coming from internal resources in the agriculture sector, there again, the task is not easy with the sources of information currently available. The composition of goods and services utilized for agriculture is not transparent. The chart of accounts includes a few object codes that are overly broad. This includes code 2393 for agricultural expenses for example which consumed 62 percent of actual spending on goods and services by MAFFS in 2006, 50 percent in 2007, and 67 percent in 2008. This makes it difficult to know how much was spent on agricultural inputs such as seed rice and the conduct of annual crop surveys to more accurately determine output and yields. In addition, the picture remains incomplete without an accounting for how the local councils have contributed to the sector. At present, the IFMIS does not record Local Council spending by object code. 61 123. Conclusion on the functional allocation of expenditure : the analysis conducted on the composition of spending shows that a significant share of the resource are spent through public expenditure programs on private goods, in particular fertilizers, seeds and agricultural machinery while on the other hand key public goods such as agricultural research appear to be significantly underfunded. While the focus on delivering agricultural inputs in the direct post-war context can be understood as a way to boost production rapidly and assist smallholders impacted by the conflict to restore their productive capacities, the effects of such incentives may be counterproductive on the longer term. More than ten years after the return of peace in the country, a refocus of MAFFS expenditure programs on the delivery of essential public goods appears to be desirable. 3.4 REGIONAL DISTRIBUTION OF EXPENDITURE a) Centralized versus decentralized spending 124. Analysis of the data collected during the AgPE review shows that decentralized expenditures have increased dramatically starting from 2007. The agriculture sector has largely benefited from this important administrative reform, being the sector that has received the highest amount of devolved funding after Health and before Education. The following table provides elements of comparison of the Sierra Leonean situation regarding decentralized spending (on internal resources). It puts the Sierra Leone in an intermediate situation between the countries that have already achieved a good degree of decentralization (Ghana for example) and those who did little or no implementation (Togo, Liberia notably). Table 11: Centralized vs decentralized spending in various sub-Saharan Africa countries. Burkina Ghana Liberia Togo Sierra Faso Leone (2012) HQ 51% 100% 70% Technical Directorates 21% 28% Total Central 92% 72% 100% 98% 81% Regional & Local 8% 27% 0% 2% 19% Offices Source: Hourticq and Zorya, and MoFED for Sierra Leone. b) Regional Allocation of public agricultural Expenditure 125. The regional allocation of all the agriculture public expenditure cannot be ascertained in Sierra Leone. For one, MAFFS (and other Ministries) does not report on the geography of its expenditure. At best, some rather partial information is available in the MAFFS annual reports but this concerns more activities usually without indication of budget spent. For example, simply knowing the regional distribution of subsidized seeds or fertilizers across the country is not possible. 62 126. Donor funded initiatives also rarely report on the geographical localization of expenditure executed. Even though the districts of intervention of each projects can usually relatively easily be known (by reading the project document), annual technical and financial reports rarely provide a clear picture on what was spent where. In our case however, an inventory of project expenditures realized by the MAFFS under the MAFSAN project allows us to provide some analysis of the regional allocation of donor funded agricultural expenditure. This analysis will however only show regional allocations and not expenditure at district level. 127. The following table presents the repartition of project reported expenditure by region and the repartition of the population (2010 estimate). Table 12: DP funded project expenditure by region vs population repartition Southern Western Eastern Northern Share of DP funded expenditure 26% 5% 24% 45% Share of the total population 20% 28% 21% 31% Sources: MAFFS/ MAFSAN data, author’s calculations and Statistic Sierra Leone. 128. One can notice here that the western region has received less funding (5%) comparatively to the size of its population (28%). This situation can however be considered as relatively normal since a very large share of the western area population is urban (Freetown). For the other regions, repartition of project funding seems in relatively good accordance with population size. c) Funds devolved to districts 129. Sierra Leone's Local Government Act (2004) provides for the devolution of all aspects of some 80 functions to Local Councils, with funding for these services to be secured through a combination of Local Council own-source revenues and formula-based grant allocations. The following table, provides figures on cumulated district based agriculture spending (IR) per district (in Million Leones) in Sierra Leone for the years 2007 to 2012 (+ equivalent data on health and education for reference) : Table 13: District level expenditure in the agriculture, health and education sectors (Le. Millions). AGRICULTURE COUNCIL HEALTH EDUCATION (COFOG)* Bo District 3,217 2,665 1,116 Bo City 735 8,581 1,373 Bombali District 4,028 2,713 1,384 Makeni City 796 7,569 912 Bonthe District 1,586 1,393 744 Bonthe Municipal 509 1,585 399 Freetown City 597 32,669 4,439 Kailahun District 3,270 6,193 1,293 Kambia District 3,040 5,080 1,202 Kenema District 4,556 3,136 1,372 Kenema City 765 7,886 1,095 Koinadugu District 5,493 5,392 1,120 63 Kono District 3,788 2,028 824 Koidu/New Sembehun City 705 3,284 860 Moyamba District 3,978 5,012 1,335 Port Loko District 4,404 9,695 1,807 Pujehun District 2,426 4,457 919 Tonkolili District 3,984 6,543 1,684 Western Area Rural District 1,568 5,533 992 Source: MOFED and Author’s calculation. * “Agriculture COFOG� combines Agriculture and Marine Resources and Fisheries except for FY '12 (Marine Resource and fisheries pertaining to ‘others’ and not dissociable in the data received) 130. When calculated per inhabitant and per year, these allocations can be relatively variable per district. In average, allocations where around 1,818 SLL per inhabitant and per year (urban average: SLL 546; rural average: SLL 2,146) for the agriculture sector, with the maximum observed for Koinadugu district (SLL 3,086 per year per inhabitant) and the minimum for the City of Freetown (SLL 109 per year per inhabitant), the lowest allocated to a rural district being SLL 1,346 for Kailahun. 131. In theory, the size of the allocations granted to district councils is formula-based. This formula takes into account the size of the population, the urban/rural ratio, the poverty rate (and other parameters). The following graph brings together data on allocation and data on poverty rates. One can see indeed a moderately strong relationship linking allocations the poverty rates for rural districts (the relative exception being Koinadugu). Figure 21: Relation between Poverty rates and Agriculture sector district allocations for the 13 rural districts of Sierra Leone Relation between Poverty rates and Agriculture sector district allocations for the 13 rural districts of Sierra Leone 5.50 Koinadugu Agriculture local council expenditure 5.00 (cumulated 2007-2012, usd) 4.50 4.00 Kono Moyamba 3.50 Bonthe Bombali 3.00 Kambia Kenema 2.50 Bo Port Loko Tonkolili Pujehun Kailahun 2.00 Western Area 1.50 Rural 1.00 40 45 50 55 60 65 70 75 80 Poverty rate Source: MoFED data (for district council budget) and latest poverty data 64 132. The next graph shows the relationship between food insecurity rate (% of household food insecure) and allocations granted to local councils. Here again, one can see a relationship between the two but agriculture sector allocations are not very significantly higher in districts where food insecurity rates are much higher. Figure 22: Relation between percentage of food insecure households and Agriculture sector district allocations for the 13 rural districts of Sierra Leone Relation between percentage of food insecure households and Agriculture sector district allocations for the 13 rural districts of Sierra Leone 6.00 Agriculture local council expenditure (cumulated 2007-2012, 5.00 Koinadugu 4.00 Kono Moyamba Bombali Kenema 3.00 Bonthe Tonkolili Kambia usd) Bo Pujehun Kailahun 2.00 Western Area Port Loko Rural 1.00 - 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 Percent households food insecure Source: MoFED Data and WFP 2011 for food insecurity data 65 4. TECHNICAL EFFICIENCY IN THE PREPARATION, IMPLEMENTATION, AND MONITORING OF AGRICULTURE EXPENDITURE 133. Technical efficiency is primarily determined by the effectiveness of the process of preparing and implementing the budget. The most common approaches for assessing technical efficiency include an alignment between the funds allocated in the budget and their actual use, the budget implementation rate, and the cost-effectiveness of programs implemented. 4.1 PLANNING AND BUDGETING PROCEDURE a) Overview Of Past And Current Budget Planning Procedures 134. The financial year runs from January the 1st and budget preparation is driven by the Ministry of Finance and Economic Development. Guidance provided is quite detailed as according to the MAFFS Planning division, the budget circular (transmitted usually in June) indicates budget ceilings by Divisions for each Ministry, raising the question of the margin of manoeuvre that line ministries have to make the structure of their budget evolve. Cabinet is not formally involved in the preparation of the Budget Call Circular and in general there are only few opportunities for Cabinet officials and Parliamentarians to review policy options at a strategic level before MDA budget proposals are completed. 135. The Budget Call Circular provides a calendar for the preparation of the MTEF and Budget as a combined process. An MTEF thus exist but in practice the budget is annual and the medium term forecasts are only partially used as a reference. Budget approval usually occurs before the end of the FY. Budget approval has been on time these recent years (2010-2014 at the exception of 2013, as 2012 saw elections), but substantial delays were not an exception during the previous years. 136. Among the weaknesses of the planning/budgeting process in Sierra Leone is the fact that investment is not sufficiently linked to its recurrent expenditure implications. The MoFED has recently created a Public Investment Unit which could potentially improve this situation but no such division/ unit exist within MAFFS and the linkages between investments and recurrent expenditures are indeed weak in the MAFFS budget. Box 8: Extracts from “Sierra Leone. Public Expenditure Review. October 28, 2010� Report No. 52817-SL. PREM 4. Africa Region. World Bank. On public investments : “The laws and regulations governing public investment are reasonable but they focus on details at the expense of strategy, cost-effectiveness, and accountability. At present, there is little required from ministries and other Government institutions, or from donors, with regard to monitorable outcome and progress indicators, or maintenance requirements after a project is completed. Moreover, only a small portion of project proposals are carefully appraised by Government officials. As a consequence, the Government has occasionally been unable to attract funding from interested donors due to a lack of fully worked out project documentation. Since most of the development budget is financed externally, Government officials often rely on the appraisals done by donor agencies without developing Government capacity for project appraisal.� 66 b) Quality of annual budget and plans 137. While all MDAs in the agriculture sector have such strategic information as mission statements, division and program purposes, strategic overview, key policy developments, and key performance indicators (notably through the NSADP), linking policy objectives to activities, funding sources and estimated costs, remain some of the key challenges. Objectives are rarely presented in a specific, measurable, achievable, realistic, and time-bound way. The budget formulation process is generally weak and would require a substantial overhaul to link policy and planning to expenditures. Changes also need to be made to make it easier to analyze budget data and comparing outturns with budget estimates. 138. One could also advocate for a strategy- and results-based approach of budgeting instead of a means-based one in the sector and it should help resolve a good number of the problems described above. Nevertheless, this would entail a substantial capacity building effort as such an operation is of little interest unless it truly allows all levels of MAFFS to use it. In each division at least, a small team responsible for leading the planning operations within their division based on Result Based Management principles (and including M&E indicators) should be intensively trained. This same team would later be responsible for monitoring the programs and drafting performance reports. 139. In the event that a sectoral approach for agriculture were adopted for budgeting purposes, needed would be a clearer definition of the various participants’ responsibilities, improved compliance to the budget schedule, the organization of an Agriculture Sector budget seminar before the distribution of the budget guideline letters, and the creation of specialized inter- ministerial committees serving as a driving force for the entire process. 4.2 EXECUTION RATES Internal resources / Central level 140. As can be seen from the graph bellow, the execution rate for the projected budgets of the group of ministries and institutions in the agricultural sector came to 79% over the period considered in average. This rate is not excellent but the figure shows that the trend is gradually improving. One can see also from the graph bellow that it is for Domestic Development Expenditure that execution rates have been the lowest although here again the situation is improving. 141. For the MAFFS itself, the overall average execution rate for the same period is somewhat better as it comes to 85%. Nevertheless, execution rate for Development Expenditure is low (54% in average over the period), a point that critically deserves attention. It shows that planned expenditures are rarely implemented fully and deviations from budgetary allocations are often significant notably for domestic development expenditure at least. For 2012, the execution rate was very high. 67 Figure 23: Actual vs budgeted expenditure 2004-2012 (MAFFS, agriculture sector, Domestic Development Expenditure in the Agriculture Sector). Internal Resources. 140% 120% Overall expenditure 100% (Agric. COFOG) Domestic 80% Development Exp.(Agric COFOG) 60% MAFFS (overall) 40% Overall State 20% Budget 0% 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Expenditure actuals, parliamentary copy (years 2004 to 2012) 142. Overall, execution rates in the sector have been lower than execution rates of the overall State budget. One must take into account the general context into which these implementation rates are discussed: overall state expenditure exceeded 100% during years 2009 to 2012 included. According to the latest PEFA17 assessment, this deviation between actuals and budgeted reflects a period in Sierra Leone initiated by optimistic growth estimates made in 2010. Much of the over- expenditure was accounted for by domestically financed capital investment programme (in particular road building), which was of multiyear nature. It was not until 2013 that expenditures were reined in. Devolved expenditure (through Local Councils) 143. The figures provided by the MOFED LGFD (Local Government Finance Department) are showing extremely good execution rates for devolved budget. 17 Public Expenditure and Financial Accountability (PEFA) Assessment in Sierra Leone Central Government Systems, 2010-2012. Coffey international. 68 Figure 24: Execution rate of budget transferred to the local councils (2007-2011). Agriculture (COFOG) and all sectors. 105% 100% 95% All sectors Agriculture (COFOG) 90% 85% 80% 2007 2008 2009 2010 2011 Source: MOFED. LGFD and author’s calculation. Note : (1) Agriculture COFOG combines Agriculture and Marine Resources and Fisheries (2) For 2007, only Q1 and Q2; For 2008 to 2010 (Q1 to Q3) ; For 2011 (Q1 to Q4) 144. However, these figures must be taken with caution as they only refer to execution rate over the Grants actually transferred to the local councils and not the budgeted Grants. In effect, actual Grants as Share of the Budgeted Grants were only 40% in 2007 and 81% (27% and 70% for Agriculture). This means that in fact, transfers to local councils have also generally been well below targeted levels, reducing service delivery for all devolved functions. These deviations are largely explained by overall resource availability but not only. The planned expenditure may also not be in line with the capacity. 145. Again, it must be stressed here that the devolution process is a challenging one in every aspect and requires a tremendous capacity building effort. In the agriculture sector, the devolution process is still at an early stage of its effectiveness. In respect of devolved functions, staff devolved form the MAFFS are still paid by the central government even though Councils may pay for additional staff from their own source revenue (but in practice, these revenues are very limited). This situation can potentially create tension18 between local officers of the line ministers and the Local Councils which can undermine planning and budgeting for the devolved functions. 18 MAFFS officials, because they are still paid by the central government perceive the central level as their legitimate authority whereas in all logic they should be under the authority of the Local Council. 69 146. According to interviews done for this Report, very few officials within the sector have a comprehensive and detailed understanding of the list of functions specifically devolved to the Local Councils in Agriculture (17 in total, out of the 80 devolved in total). In practice, planning and budgeting are not prepared with explicit references to this list of devolved function. Targets sets do not appear to be linked either. An Australia-funded, FAO-executed project is currently assisting MAFFS in strengthening Local Councils’ planning capacities. While this initiative is timely and can be commended, the effort deployed seems very limited in comparison to the size and complexity of the challenge. Externally funded expenditure execution rates 147. There is currently insufficient knowledge of the effective disbursement rate of the projects to conduct such an analysis. Each project PIU may have this information available, but collecting and aggregating this information at sector level has proved difficult and in practice impossible (high number of PIUs, non-responsiveness, absence of archival system, etc.). The difficulty was raised by the consultant team during AgPER Technical Working Group meetings but none of the participants could find a practical solution. This proves only that the development partner funded projects “owners� (MAFFS, MFMR, etc.) do not have in effect enough ownership on the management of the various projects. This is a situation that needs to be addressed and remediated. 4.3 MANAGEMENT OF ODA IN THE AGRICULTURE SECTOR a) Fragmentation of Aid, high number if PIUs 148. As highlighted in the second chapter of this Public Expenditure Review, externally funded projects represent the largest share (around three quarters) of the resources available to and spend in the Agriculture Sector. The quality of the processes used to manage ODA in the sector is thus essential to the technical efficiency of the agricultural spending. The situation is unfortunately quite far from being optimal. 149. A first structural obstacle to the quality of the management of ODA (including consolidated M&E, oversight, overall planning with verification of the coherence of externally funded project plans with IR funded plans) in the sector is the number of ODA funded projects. 150. This results in considerable coordination challenges and reduces aid effectiveness by: (i) presenting additional challenges to harmonizing and aligning aid, which results in rising transaction costs; (ii) creating wasteful duplication and overlap in the delivery of aid; (iii) and causing competition for scarce skills at country level. 151. MOFED communicated data (ODA database) shows that the number of project in the agriculture sector has constantly increased during the period 2003-2012 while at the same time, the average amount disbursed by project was decreasing. It should be noted however that the analysis should not be based only on the average disbursement per project but also on the distribution of disbursements by the projects. Such an approach shows that – especially in the recent years - a larger share of the disbursements was made by a few “large� projects (such as the SCP) with a multitude of small disbursing projects. The issue then is whether the small projects need to stand alone. 152. The MAFFS has the willingness to reduce the number of PIUs and has recently taken the initiative to examine the feasibility of having only one single centrally managed PIU for all the 70 DP funded projects (“One PIU initiative�). Such a change would need careful planning and cannot be achieved within a shorter timeframe. However, there are certainly economies of scale achievable by limiting the number of PIUs and centralizing or putting in common some of the functions that (sometimes costly) PIUs specialized staff are performing (such as : M&E, procurement, mobilization of audits, etc..). Also, one can only recommend here that the subject of reducing the number of PIUs and increasing the size of each ODA funded project be taken on board very systematically by the Agriculture Sector donors working group. Figure 25: Number of projects and average amount disbursed annually by project (in $US) over the period 2003-2012 2,500,000 50 Average 45 amount disburse 2,000,000 40 d per project 35 1,500,000 30 Project count 25 1,000,000 20 15 Linear (Average 500,000 10 amount disburse 5 d per project) - 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: MoFED ODA database and author’s calculation b) Mis-alignment between national and projects’ budget preparation processes 153. Another critical dimension and actually weakness in the way ODA is managed in the Sector refers to the way planning and budgeting are done for external resources and internal resources. In effect the two processes are “mis-aligned�. Firstly, their calendars are not the same. While the calendar for the preparation of the Government budget is clear, uniform and well known, donors and projects have their own requirements and deadlines. The first practical consequence is that it is impossible to have a meaningful dialogue on what can IR cover and what should externally funded project cover, and finally to check that the internally and externally funded budget-plans complement each other. The multiplication of “project steering committees� does actually little to ensure this complementarity and coherence. 71 154. In effect also, a large portion of aid is ‘off-budget’. According to the 2008 Survey19, almost half of the total aid coming into the country was ‘off-budget’, i.e. not recorded in the national budget, while roughly 80% of aid does not pass through treasury. This leaves much uncertainty about the extent to which aid is conforming to GoSL priorities and leads to weak national ownership of some programs. It also creates budgetary difficulties, by way of unforeseen recurrent-cost implications. 155. The “misalignment� of donor project dates and processes is however not a reason why donor funding should not be better incorporated in the national budget. Every donor project has a project document signed-off by the Government with (in most cases also) counterpart funding identified, and they are usually multi-year. There is no strong reason why the information should not be captured in the Government budget. With a small ODA management Unit, MAFFS and others could do a better job at this. 156. Another dimension of the current mis-alignment between national and projects’ budget preparation processes concerns the processes and formats used for preparing and presenting plans and budget which also can differ largely from one project/donor to another and in general are simply not coherent with the way the government is planning and budgeting the use of its internal resources (means based budgeting vs result based budgeting). While this is certainly a challenge and an endeavor that transcends the agriculture sector, it would be advisable to progress toward a better coherence both in terms of timing and formats in order for planning and budgeting to be more coherent. Here again, the task could be taken up by the Donor Agriculture Sector working group to explore the feasibility of progressing rapidly in such a direction. 4.4 RESOURCES ALLOCATION PROCESS, EFFICIENCY OF SPENDING 157. For lack of time, the resources allocation processes have not been the subject of much exploration and discussion between the consultant team and the government counterparts during the AgPE review. The subject is nevertheless of importance as the quality of the resource allocation has self-evident implication on the quality of service delivery in the sector. The topic is approached here by giving a synthesis of the findings of a recent review that has looked specifically at this subject in Sierra Leone in the Agriculture Sector. A much more detailed account is accessible in the review itself which has been made available to the public (Audit Service Sierra Leone website). 158. The Government of Sierra Leone has in effect used Public Expenditure Tracking Surveys to ascertain the flow of funds between central government and frontline service units as well as assessing local people’s perceptions of government services. Public Expenditure Tracking Surveys have generally been undertaken by the Economic Policy and Research Unit of the Ministry of Finance in the early 2000 and more recently Performance Audits have been produced by “Audit Service Sierra Leone� a service belonging to the Official Sierraleonan Audit Institution. 159. A recent (201220) Performance Audit conducted in the Agriculture Sector examining the procurement and delivery by the MAFFS of rice seeds to farmers concluded of : o Serious irregularities in the procurement process o The absence of established criteria for allocation to district 19 OECD (2008). 2008 Survey on Monitoring the Paris Declaration. Making Aid More Effective by 2010. Country Chapter – Sierra Leone. 20 Performance Audit Report on the Distribution of Agricultural Inputs. ASSL. January 2012. 72 o The lack of adequate control (quality of the seeds procured, targeting, cost / seed recovery) 160. The same report examined the delivery of mechanization services through MAFFS acquired tractors. The government of Sierra Leone has in effect bought a number of tractors that are made available to farmers. The total number of MAFFS tractors operating in 2009 was 92 (including 10 new “Libyan� tractors). These tractors are retained and maintained by MAFFS district offices/workshops. A further 263 tractors procured by the government were distributed under a hire-purchase scheme in 2010. This last acquisition represents a substantial expenditure (author’s estimate: 3 millions $US, no official figure could be found). 161. The ASSL report concluded that the tractors acquired were deteriorating and that the number of acres ploughed per year was decreasing. Furthermore, criteria for the allocation of the most recent batch of tractors (the 263 units acquired in 2010) to the districts were unclear and tractors were in general underutilized. 162. While the conclusions of these reviews point clearly to the need to improve resources allocation processes within the sector and the quality of oversight / control over spending, it is actually difficult if not impossible to go much further into the analysis of the causes of such a situation. For one, an analysis of the efficiency with which government services in agriculture are provided is prevented by the paucity of data on costs and quality of services. This is perhaps to be expected given low levels of capacity within government to effectively monitor and account for expenditures. However, as regularly pointed by the PETS and Audit reports, efficiency (or value- for-money) is an issue that MAFFS should seek to address if it wishes to strengthen its future claim on public resources. A principal means of achieving this is systematically to record information on the costs of service delivery and outcomes achieved. This information should figure more systematically in MAFFS annual reports. Box 9: MAFFS reporting Monitoring information being scarcely available in the Sector, the consultant team has tried to use the option of assessing the quality and effectiveness of expenditure by scrutinizing the annual MAFFS reports. MAFFS annual reports for the years 2009 to 2012 were collected and read for this purpose 21. The reports refer to a large number of activities, but it is not possible to pinpoint activities financed primarily by allocations from the internal budget from many of those that were supported by donors and realized in cooperation with MAFFS. The coverage and scope are often not indicated, and explanations of the rationale or analyses of effects are not mentioned. Also, it does not reveal how these were financed, nor does it indicate how many farmers had contact with extension or technical services staff. It is in a sense surprising to see that so much skilled and costly resources (notably TAs) are allocated to project management and supervision and possibly reporting (a too high level of resources according to the Ministry itself) while at the same time, a comprehensive quality annual report is not available. Underlying this issue, an activity monitoring system is clearly necessary, as this will also enable production of reports that reflect the efforts undertaken and their outcomes in terms of publicly funded agricultural services made available to farmers. 21 Note that MAFFS annual report for the years prior to 2009 could not be collected for lack of reliable archival system. 73 4.5 REVENUES AND THEIR MANAGEMENT 163. There are also problems with procedures for accounting for revenues. MAFFS receives revenues from various sources (forestry concessions, hire of mechanical equipment, sale of fertilizer, etc.), which are not transferred to the consolidated account. Some funds (e.g. forestry) are kept in separate accounts that are not audited. An external audit has not yet been undertaken. Here again, the main sources of information on the subject are the reviews executed by the “Audit Service Sierra Leone� which have looked specifically at the collection and management of revenues within the MFMR and MLCP. The paragraphs that follow present selected extracts of these reports. For the fishery sub-sector: The following is an extract of “Audit Service Sierra Leone Performance Audit Report On Generation And Collection Of Revenue In The Fisheries Sector September 2012�. “MFMR receives their major funding through budgetary allocations from the Ministry of Finance and Economic Development (MOFED). The revenue it generates through Licences, Royalties, Transhipment fees, Fish Health Certificates, Discharge Fees, Fines, Sales of Confiscated Catch, Clearance Fees, Export Fees and By-Catch charges is paid to NRA that pays it into the government’s consolidated fund. A total of Le 11,332,983,340 was collected in 2010 and Le 12,798,458,432 collected in 2011. During the course of the audit, we arrived at the following findings: o The councils have not established and implemented adequate monitoring and controls over fish revenue collection , o Fisheries observers on board the licensed vessels often find themselves in a weak position in relation to the captain of the ship when trying to enforce the correct fishing practices. o Patrolling of Sierra Leone’s fishery waters has not been effective Monitoring of fishing activities by the Ministry has not ensured good fishing standards. MFMR was unable to provide the auditors with documents showing yearly revenue generated and collected from fishing activities for 2007 and 2008. Such documents were only produced for 2009 and 2010. Requested lists of registered fishing companies and the type of fishing vessels operated by such companies could be provided for 2011, but not for the period 2007-2010. In the district councils visited (Moyamba, Pujehun and Western Area Rural), documents showing the number of boats, licensed in the period 2007-2011, were not maintained even though the strategic plans included a programme for data collection of fishing boats. In order to improve on this situation we recommend that the following issues be addressed: MFMR should approve and issue the new Fisheries Regulations as soon as possible and complete the drafting of a bill for the amendment of the 1994 Fisheries Act. The Local Councils should ensure that licensing fees are collected annually in accordance with pre-defined and sustainable tariffs. To facilitate this, databases with details of canoes, fees, revenue collected, etc. should be developed and monitoring by M&E officers improved. To promote the independence of the fisheries observers, MFMR should ensure that their salaries are paid by them and not by the fishing companies. MFMR should also consider how back up support to the fisheries observers could be improved when captains do not want to obey the observers.� 74 For revenue generated from land allocation: The following is an extract of “Audit Service Sierra Leone. Performance Audit Report On The Allocation Of State Lands By The Ministry Of Lands, Country Planning And The Environment December, 2013�. “The purpose of this Performance Audit was to assess the efficiency and effectiveness of the Allocation of State Lands by the Ministry of Lands, Country Planning and the Environment (MLCP&E). The Ministry is charged with the responsibility of allocating all state lands. This report focuses on the processes and procedures in the allocation of state lands. It was observed that for the construction of a dwelling house, a fixed price of one million Leones (Le1,000,000) per annum had to be paid as lease rental for the first three years of ownership regardless of the size/acreage or location of the land; and some Lessees were offered more than the three town lots - the maximum allocation of land to an applicant. The fixed charge of Le1,000,000 for leases irrespective of the size and location of the land had the potential to affect the amount of revenue generated by the Ministry. o The Ministry could not provide any evidence of an established committee as recommended in the Marcus Jones Commission’s report and accepted in the Government White Paper for the allocation of state lands. In addition, there was no evidence that applicants were interviewed before they were allocated state lands for the period 2010 to 2012. o The Ministry could not produce any evidence of a structural plan on the allocation of state lands to include infrastructural facilities like good road network, markets, water supply, electricity, etc. o The Ministry failed to provide evidence of a database or land register for state lands in the Western Area and the reserved lands in the Regions.� 164. Revenue generation from the lease of larger tracks of land to qualified agribusiness investors is a current practice promoted in Sierra Leone that could potentially yield positive prospects both in terms of agricultural production (notably for export crops) and revenues (from the land leases). However, in order to attract investors, a ‘special set of incentives’ has recently been approved which may limit substantially the capacity of the GoSL to generate revenues from the lease of lands. The major incentives includes exemption on: o corporate income tax ‘up to 2020’ [i.e. for 10 years] for companies in the tree crops sub-sector (such as oil palm and sugarcane). This compares to the standard corporate income tax rate of 30 per cent. o import duty on raw materials and agricultural inputs, compared with the standard 3 per cent. 165. A recent report22 produced by NGOs states that “an estimated USD 188.1 million or USD 18.8 million a year on average over a 10-year period in tax revenue will be foregone by the government as a result of special tax deals with Addax Bioenergy, SAC and Goldtree Ltd�. While the team of consultants has not been able to verify this information, this is certainly an area where policy dialogue is needed in order to maximize the potential gains from this type of policy in a context where the demand for leasable land for cash crops growing is sustained. 22WHO IS BENEFITTING? Social and economic impact of three large-scale land investments in Sierra Leone: a cost-benefit analysis. 2013. Report commissioned by Action for Large-scale Land Acquisition Transparency (ALLAT). 75 5. CONCLUSIONS AND RECOMMENDATIONS: HOW COULD THE QUALITY OF SPENDING BE STRENGTHENED ? 5.1 FACTORS AFFECTING THE QUALITY OF SPENDING 166. First of all, attention should be drawn on the following elements which are currently having a negative influence on the quality of spending in the agriculture sector : o Reporting is weak: due to an incomplete M&E system, making it difficult to track progress, identify failures and re-allocate efforts during planning exercises. It is hardly possible to identify corrective measures when clarity is so difficult to reach on the current level of performance of the public administrations active in the sector. Furthermore, this absence of clarity bears the risk of creating an atmosphere of distrust between (and within) the GoSL and the Development Partners. o High fragmentation of aid resulting in high transaction costs and inefficient use of Government scarce resources. This is one of the key areas on which the GoSL must accelerate the current efforts and where the DPs have a crucial role to play. A complete reversal of the current trend (increasing fragmentation of aid in the sector) is a mandatory condition for improvements to be achieved in the efficiency of the use of the sector resources. The international debate and initiatives on the subject (Paris Declaration, Accra Agenda for Action, etc.) give an adequate framework for all parties to act. o Permanence of a substantial off-budget component: related to the above, the budgeting process needs to be strengthened in order for the share of Off-budget resources to decrease rapidly. Here again, DPs have a role to play, and can notably systematically synchronize the project planning/budgeting with the GoSL system. o Low numbers / qualification / motivation / stability of staff: the deficit in capacities is still a major issue within the administrations active in the agriculture sector. This challenge is being currently exacerbated by the devolution process which places increasing responsibilities in the hand of the local councils. o Maintenance and operation cost and mechanisms generally overlooked for all types of capital expenditure (public service equipment, public infrastructure and investments transferred to beneficiaries): issue of recurrent cost traceability in budgets. o Lack of Knowledge management / evaluation: when projects end, a substantial part of the (mostly individual) experience gained in terms of agricultural achievements (and under-achievements) is lost and not transformed into institutional knowledge. The institutions of the sectors must strengthen their capacity to become “learning organization�. This supposes better reporting (see above) but also more systematic transparent and external evaluations of activities implemented and teams dedicated to capitalizing on lessons learned and transferring / applying knowledge to subsequent project and activities. A number of activities of the MAFFS remain un-evaluated (e.g. mechanization). 76 5.2 PROPOSED ACTIONS FOR IMPROVING THE EFFICIENCY OF PUBLIC EXPENDITURE IN AGRICULTURE Actions Responsibility Budget - Establish medium-term expenditure State House, programming and frameworks as a budget planning tool. MoFED, Line level of spending - Increase the resources allocated to the Ministries agriculture sector as the current share of the State budget appears inadequate in terms of the Maputo Declaration. - Reduce the number of DP funded projects MoFED, and increase size of each projects by more Ministries, DPs systematically using pooled fund (basket (Ag. Sector funds, etc) in order to reduce the Working Group) transactions costs of ODA management; - Include in the budgets of the agricultural sector ministries all agricultural development projects, in particular those of DPs that are currently not monitored by MEF or the Ministry of Planning. - Formulation of project to be made on the Ministries and basis of past experience (evaluations, other institutions capitalization on success and failures). involved in the Ag. Sector, DPs - Establish a mechanism for budgeting for MoFED, operating expenditure on maintenance of Ministries and capital items. other institutions - Divide coherently roles and means for involved in the investment maintenance between central Ag. Sector level and Local Councils - Improve cost accounting at the development Ministries and funded project level to identify the operating other institutions costs incurred by the project management involved in the - Identify project management functions Ag. Sector, DPs which cost could be reduced by pooling resources across projects. Budget - Present commitment and procurement plans MoFED, Implementation and at the same time as the annual budget, and Ministries and Procurement make tender documents available before the other institutions start of the budget year so that calls for involved in the tender can be issued in January. Ag. Sector - Better monitoring of investments made at the level of support services and of beneficiaries, and better monitoring of the regional distribution of expenditures; Decentralization / - Develop and implement a comprehensive Ministry of Local devolution capacity building plan at Local Council Planning, MoFED, level. Ministries and - Transfer staff (and corresponding budget) other institutions 77 from central level payroll to Local Council involved in the payroll Ag. Sector. Monitoring and - Implementation at all levels of an effective Ministries and Evaluation M&E system coupled with an efficient other institutions information archiving and dissemination involved in the mechanism Ag. Sector - Draw up a system for monitoring performance in the agricultural sector with the use of quantifiable and verifiable performance indicators, potentially based on PNDSA indicators - Institutionalize the AgPER process, with light annual AgPE reviews and in-depth reviews every 3-5 years. - Strengthen annual assessments of ministries’ progress, detailing monitoring of actions and activities. - Overhaul MAFFS (and other public institutions reports involved in the agriculture sector) annual reports - Improve information feedback on actual expenditure by all services and make this information available through annual reports. - Ensure the introduction of an effective system for monitoring program budgets in aggregated fashion (not just PIU by PIU). - Produce evaluation reports for all projects. Establish an Evaluation Unit under PEMSD. - Produce impact studies for major projects. Functional - Better functional distribution of Ministries and allocation expenditures is needed. other institutions - The livestock (including veterinary services) involved in the subsector as well as R&D appear to be Ag. Sector largely under-funded and require immediate attention. - Agriculture extension received only 2.11% of the total spent in the sector through the projects, which appears as a low figure in comparison to the central role that extension can play in the dissemination of agricultural technologies. Since private extension services are quasi inexistent, a higher level of spending into public advisory services is necessary, especially in a context where efforts are put into decentralization and devolution Policy Development - Elaborate a strategy for agricultural Ministries and 78 extension so as to include the role of all the other institutions parties involved (MAFFS, private sector, involved in the NGOs, local councils) Ag. Sector - Study how to end the duplications observed Ministries and between existing project and future PIUs. other institutions Give support to pooling resources and involved in the ensuring their ownership by the sectors Ag. Sector institutions. Enact MAFFS ODA management decree / guidelines. Ministries and - Evaluate past expenditure in support to other institutions mechanization and develop a strategic involved in the framework for mechanization development Ag. Sector 5.3 ADDRESSING AID FRAGMENTATION 167. This report points to the high level of fragmentation of aid within the sector as a key finding of the review and highlights the necessity to address this challenge as a priority for MAFFS. Various modalities to reduce aid fragmentation exist, ranging from consolidating only selected functions of the project implementation units (such as M&E, procurement, resource management) to the creation of a single project implementation unit for all donor-funded assistance exist. The current option studied by MAFFS – called “one PIU initiative� - to create one single (large) PIU able to manage all (or at least) most projects owned by the Ministry could serve this purpose and the report recommends to pursue this objective with a step by step approach. 168. Initially, a quick scoping study could look at the current roles of existing PIUS and evaluate for which functions/roles, duplication/fragmentation of efforts appears the most obvious and desirable to eliminate. Aggregating PIU functions at the level of a central Unit should then be undertaken function by function, starting with PIU functions selected for the ease with which centralizing the function is the most obvious (readiness, existing capacities, acceptability to the development partner, etc..). Such phased approach is recommended as the best way not to overburden MAFFS planning staff with this effort consuming process. It is also recommended not to try to merge existing PIUs but to negotiate with DPs on a case by case the centralization of certain PIU functions during the project formulation process of future projects. 5.4 WORKING TOWARDS A SECTOR-WIDE PROGRAM 169. The preparation and of the SCP and its subsequent implementation have positive step towards making donor interventions and GoL spending consistent and coordinated under the umbrella of the NSADP. The challenge now is to maintain the momentum and develop this approach further into a SWAP-like arrangement with ideally pooled funding arrangements. 170. The recent Consultative Workshop on Agriculture (Consultative Dialogue on the Path to Prosperity, Nov 2013, Bintumani Hotel) chaired by HE. Ernest Bai Koroma, President of the Republic of Sierra Leone, has demonstrated that agriculture is increasingly seen in Sierra Leone as a sector involving multiple government institutions and led launching of the preparation of the AFAIP. 79 171. A Sector Wide Approach will require an approved sector strategy (NSADP could be considered as such or at least a good basis), donor commitment to adhere to the agreed priorities documented in the sector strategy, a costed implementation plan, a common system for monitoring results, a forum for regular dialogue and appraisal of results (achievements and failures), and strong national leadership. Several of these elements still need to be practically defined in order to be implemented. 172. In addition, a future SWAP would require intensive consultation and discussion to become a tool to guide resource allocation and activity planning, and to avoid creation of a “wish list� of items. Financial requirements shall not be overambitious and the present AgPER as well as a Sector-MTEF could be used as tools to evaluate and describe future desirable levels of spending (evaluated currently at around 100 million USD per year23). 173. 174. The preparation of this future Sector Wide approach could also be the opportunity to translate into concrete decision and actions numerous of the recommendations of this Ag-PER, in particular with regard to Donor Coordination, M&E, strengthening Local Council capacities and alignment of planning procedures. Annex 1: Terms Of References Of The Agriculture Sector Public Expenditure Review 1.0 Introduction 1.1 The Sierra Leone Agricultural Sector Review will be conducted within the framework of Agricultural Sector Review (ASR). The overall objective of the review is to enhance the efficiency, effectiveness and equity of public expenditure in the sector in line with agricultural programs in order to meet the national development objectives. The review is intended to provide evidence-based recommendations that will address, inter alia, budgetary planning, budget execution, and accountability in the agriculture sector, the creation of a reliable data base, and more effective intra- and inter-sectoral coordination and alignment towards the sector strategy. It is also aimed at stimulating larger donor resource allocations, and enhanced harmonization and alignment of resources behind the Sierra Leone’s agricultural development strategy. 1.2 The main methodology will be the basic agricultural public expenditure diagnostic review (Basic AgPER) which is intended to compile basic expenditure data and to assess the extent to which expenditures reflect and contribute towards the stated national sectoral priorities. It also aims to draw lessons from the analysis that will be used to enhance the efficiency and effectiveness and impact of public expenditure in the agriculture sector. 2.0 Scope of the Review 2.1 As in any public expenditure reviews, the exercise is essentially retrospective, aiming to assemble a report of all public expenditures in the agriculture sector with reasonable accuracy, over the past 5-10 years. It would assess expenditure trends and 23 See Chapter 2 : “To demonstrate a level of support in line with the Maputo declaration, Sierra Leone would have to reach a level of agricultural support in the order of 92.7 million $US per year � 80 composition at central and lower levels of government. The latter would be particularly important for Sierra Leone given the decentralization of the agricultural extension services to the District Assemblies. It is also particularly important to understand the recurrent budget in the sector since the structure, financing and governance of the recurrent budget will determine the sustainability of investments that are made on the capital (development) budget. The analysis would need to capture supplementary budgets and expenditures, as well. Definition of Agriculture Sector 2.2 The sectoral scope of the study will be guided by the classification of the functions of government (COFOG) system, even though the classification may not be actually in place in Sierra Leone’s classification of Government functions. This means that while the study will focus on the Ministry of Agriculture, Forestry and Food Security, it may also involve related ministries, especially Ministries responsible for Natural Resources, Energy and Environment. The analysis should however exclude non- agriculture related expenditures such as Rural Development, trunk roads, health and education. The study will present expenditure data in two ways: for analytical purposes it will use COFOG, but for the purpose of presenting results and engaging in dialogue on recommendations, it will adhere to the local budget nomenclature and structure. 3. Implementation 3.1 The study will examine budget implementation both in terms of the way in which budget expenditures are executed as well as sector outcomes. An understanding of the budget planning and “flow of funds� aspect of budget implementation is important as a source of recommendations to improve budget planning and execution. Budget outcomes in the sector will be reviewed in terms of major sector performance indicators as part of the analysis of consistency between budgets and sector priorities and the identification of major funding gaps. Institutional Context 3.2 The Ministry of Agriculture, Forestry and Food Security (MAFFS) will be the centre of the review exercise, together with the Ministry of Finance and Economic Planning (MFEP). However, as stated above, other entities that perform public sector roles within the COFOG boundaries of “agriculture� will have to be included. An understanding of the links between central government and lower levels of administration will be necessary so as to provide insights into “flow of funds� and “use of funds� issues, especially in cases where fiscal decentralisation has been introduced. Methodology 3.3 The dual starting points for the Basic AgPER will be: • To formulate, in close collaboration with counterparts in the Ministries of Agriculture and of Finance, a sound descriptive overview of the public expenditure budget in 81 the agriculture sector and its links with the macro economy, the fiscal context and Ministry of Finance processes. A key element of this will be to identify and agree upon sources of sector performance and budget outcomes that will serve as the focus for the study; and  To assemble a synthetic set of budgets for analytical purposes (from all the concerned ministries, agencies and other entities) that cover public expenditure in the sector for the previous 10 years. Sectoral Context 3.4 The main elements of the sectoral context and the potential information sources will include, but not be limited to: • Main macro issues affecting the sector [Sierra Leone Agricultural Sector Policy and Programs and other existing sector studies]; • Budget planning instruments that influence resource allocations and management of expenditures in the sector; • Role of agriculture in national development [Agenda for Change, Proposed New Agenda for Prosperity, National Sustainable Agriculture Development Program, Smallholder Commercialization Program); • Sector performance [based on existing studies, for example, CAADP “stocktaking�]; • Institutional mapping in the sector including intra- and inter-sectoral coordination mechanisms [Agricultural Sector Working Groups]; • Political economy including recent or imminent changes in government, the role of parliament in resource allocation and oversight, the roles of CSOs and the private sector, and any inherent policy biases; • Role of public and private sectors in agriculture and their classification in the national budget; • Assessment of major expenditure trade-offs between sectors and between types of expenditure within agriculture (e.g. subsidies). Scale and Characteristics of the Sector Budget 3.5 The principal task of the study will be to put together a comprehensive set of public expenditure budgets for the agriculture sector, preferably covering a period of 10 years. This will be the data set for the analysis of the budget, and identification of gaps and inconsistencies. It will also shape the description of the budget, its particular features and the changes that have occurred over time. When determining sectoral and total government expenditures, a number of conventions should be followed in particular: (a) excluding public enterprises and financial institutions except for any subsidy or capital support payments made by government; (b) include any special funds; (c) only include in the sector expenditure any multi-sector projects in which over 70% of expenditure falls within agriculture; (d) for the sake of simplicity, assess expenditures on a “cash� rather than “accrual� basis. 82 3.6 Expenditure level and trends: The elements of the sector budget will be compiled to include all relevant ministries and other entities and include: • Share and trend (over 10 years) of approved budgets (inclusive of any mid -year revisions) and actual expenditures in agriculture with respect to total government expenditure [budget documents]; • Share and trend (over 10 years) of approved budgets and actual expenditure in agriculture with respect to agriculture GDP [budget documents, macro reports];  Share of expenditure by level of government, and when available, disaggregated budgets at lower administrative levels; impact of changes in fiscal decentralisation on sector budget [budget reports]; • Per capita expenditures and trends; rural-urban comparisons; comparisons with other countries [budget documents, demographic data]; • Alignment of budgets with stated policies and priorities to include underinvestment and mis-investment [broad judgment on underinvestment by comparing the share of agriculture in the budget with its contribution to GDP; mis-investment by comparing the share of agriculture in the budget with the shares of other sectors and against indicative shares shown in the National Development Strategy documents etc]; • A data set of off-budget expenditures by donors and NGOs should be constructed, focusing on public good-type activities and excluding any private sector-type support, identifying main types of expenditure and sub-sectors; alignment with national priorities; steps to bring “on budget� [data from interviews with DPs using a standard questionnaire on commitments and disbursements; broad judgment on alignment]; • Private sector investment (including small farmers, traders, small- medium- and large-scale agribusiness) and foreign direct investment (FDI) in the sector , so as to determine leverage impact of public expenditure; judgment on extent to which public expenditure “crowds out� private investment [existing studies, if they exist; recorded FDI statistics; Ministry of Industry and Trade]. 3.7 Expenditure composition: Broad examination of the structure of expenditures covering: • Capital (development) vs recurrent (revenue) budget balance; recurrent budget items in development budget; trends (over 10 years) [budget documents]; • Composition of recurrent budget in particular wage vs non-wage, allocation to SOEs; international comparisons [budget documents, studies]; • Comparison of expenditures on public as opposed to private goods and services; trends (over 10 years); implicit trade-offs [definition of public and private goods and services]; • Expenditure composition at lower administrative levels [budget reports, studies]; 83 • Alignment of policies and expenditures focusing on major expenditures or key areas; identification of any obvious discrepancies [budget documents, national policy and strategy documents]. 3.8 Financing the budget: Identifying the main sources of financing the budget including: • Government Central and local government financing volume and share of total [budget documents];  Revenue-generation sources within the sector; mechanisms for reporting and utilization of revenue (ear-marking etc); stability of revenue collection;  Public sector borrowing generated by the sector;  Seasonal revenue generation variations and impact on expenditure patterns;  Public financing of private goods and services;  Fiscal decentralisation mechanisms in terms of revenue generation and central government grants; • Donors Loans and grants volume (both commitments and disbursements) and share of total; and  Scale of off-budget financing [budget documents; may need data set generation]. 3.9 Share of total expenditure financed by donors  Extent of harmonization of donors funding [donor data, donor coordination mechanisms and code of conduct];  Trends in financing (over 10 years) [budget documents];  Extent to which expenditures form part of a programme-based approach (PBA) or SWAp with donor funding;  Extent of public-private partnerships (PPP) identifying different types, financing and cost recovery mechanisms, assessing their cost effectiveness and strategic role. 3.10 Subsidies and cost recovery: Identifying the main subsidies and sources of cost recovery, their incidence and impact including:  Levels and trends in major subsidies [budget documents];  Policies and mechanisms to manage and monitor subsidies;  Political and economic justification for subsidies [policy statements, studies];  Main areas where cost recovery is operated; level, adequacy and incidence;  Role of donors in subsidy operations. Budget Performance The synthetic budgets assembled and descriptive framework will be used to assess various indicators of budget performance. 84 3.11 Flow of funds: A detailed assessment of many flow-of-funds issues will normally necessitate undertaking a specialized Public Expenditure Tracking Survey (PETS). However, in the case of the Sierra Leone agriculture public expenditure review, the flow of funds will be assessed qualitatively in terms of the ways in which budget resources are channeled to operating units within the sector and different administrative levels so as to determine whether this constitutes a serious impediment to budget performance. The qualitative assessment of flow of funds will include:  MFEP rules and procedures for the release of funds;  Timing of fund release through the year, related to seasonal patterns of expenditure by different units; impact of delayed release and end-of-year effects;  Procedures for modifying budgets during the year including moving resources from one budget line to another (virement) and its impact on overall expenditure;  Procedures for delegating budget responsibility (AIEs) to different operating units within MAFFS;  Procedures for delegating budget responsibility to lower administrative levels and impact of fiscal decentralisation;  Assessment of the quality of outputs and outcomes at decentralized levels vs central level. 3.12 Public financial management: An assessment will be made of the basic public financial management capacity in agriculture sector and its entities at various levels, within the context of the overall status of PFM in the country. This will specifically include:  Gap between planned budgets, approved budgets and actual budget outturn; changes over time and reasons;  Value for money, indicators of unit costs of delivering specific services and scope for cost-cutting measures;  Accountability of budget holders and sanctions. 3.13 Institutional arrangements: An assessment will be made of the institutional mechanisms in place to ensure consistency between stated policies and strategies and budgets, including:  Extent to which a results-based budget system has been implemented overall and specifically within the agriculture sector;  MoF procedures within the budget cycle from planning, through budget implementation, to budget management; description of the processes and qualitative assessment of extent to which they constitute an impediment to effective budget processes;  MoF budget preparation guidelines, effectiveness of prioritization criteria and process of issuing budget ceilings;  Decentralized budget processes, allocation of conditional and unconditional block grants; 85  Systems for evaluating absorptive capacity and effectiveness of outputs at sector level;  Mechanisms for coordinating inter- and intra-agency programmes and budgets within MAFFS;  Role of the legislature/parliament in approving budgets, reviewing performance; forms of accountability. 3.14 Expenditure composition: The study will draw attention to areas where the sector budget appears to be unbalanced including:  Balance between development and recurrent budgets;  Balance between wage and non-wage expenditures;  Distortions to salary and incentive structures resulting from incentive salary payments to staff on development projects under project implementation units (PIUs) and regular (recurrent budget) civil service pay scales;  Areas where private goods and services are being delivered with public funds, proposing priorities for disengagement from certain services;  Identification of mismatches between national priorities and actual expenditures at decentralized levels; reasons for the mismatches and proposals for rectifying them. 3.15 Financing: Proposals for enhancing the financing of the sector budget will cover:  Ways of enhancing the local level financing of the sector budget;  Areas where revenue generation and cost recovery could be enhanced;  Ways of overcoming seasonal revenue generation constraints on expenditures so as to provide greater stability and predictability in the flow of funds;  Assessment of reliance upon donor funding and specific areas where this is excessive; assess degree of harmonization and alignment by donors and propose ways to enhance volume, coherence and predictability of donor funding. 3.16 Budget performance: The study will identify areas of inefficiency in budget implementation, including:  Ways of improving the budget cycle and flow of funds so as to enhance the effectiveness of annual work plan and budget processes;  Identification of areas where public financial management is weak, such as procurement, audit, monitoring and accountability; however, many of the issues related to PFM are likely to be economy-wide rather than sector specific;  Institutional weaknesses in terms of intra- and inter-sectoral coordination, decentralized budget operations and systems for oversight. 4. Sources of Data and Information In view of time and resource constraints, official published budget-related documents will form the primary source of data and little or no generation of original data from studies will be possible. Wherever possible, existing special studies on the sector as well as 86 unpublished material should be used to add depth and documentation to the formal data collected. Where such sources are used, they should be validated and fully referenced. Focus group discussions in the key ministries (MAFFS, MFEP) will be an invaluable source of information when formulating a sound descriptive overview of the public expenditure budget in the agriculture sector. 4.1 Official sources of data will include:  Published budget estimates  Published reports on budget outturn  Electronic data from MAFFS and MFEP  Published reports and statistics from MAFFS and associated agencies such as the Ministry of Industry and Trade. 4.2 Studies and “Grey� sources:  Donor reports and interviews on “off-budget� expenditures  Sub-sector project appraisal documents for more detailed information on focus areas and expenditures  Special surveys or evaluation reports  CAADP Country Roundtable reviews and reports  Sector project evaluations (ICRs) or mid-term reviews 4.3 Sources for sector outputs and outcomes:  MAFFS studies and reports  Project appraisal documents, mid-term reviews, ICRs  Focus group discussions with project managers and teams  Donor evaluation reports  Beneficiary impact assessments 5. Process The main stages in the study process will include: 5.1 Stakeholder briefing: All key stakeholders must be briefed on the purpose and proposed outputs of the study and agree on the main milestones and the timeline. Agreement should also be reached upon the definition of agriculture to be adopted and the institutional scope of the study. 5.2 Inception workshop: A formal workshop for all key stakeholders should be scheduled within two weeks of the commencement of the study to present and discuss an Inception Report that includes:  the revised terms of reference for the study  queries and outstanding issues with respect to the study  agreed milestones and timeline for implementation  designation of counterpart staff 87  definition of the institutional scope for the study The workshop should solicit the support of government officials in facilitating access to published and electronic expenditure data including items outside the regular MAFFS budget, and from DPs to provide details of their off-budget programmes. 5.3 Technical workshop: Within one month, a technical workshop should be scheduled to:  Present the framework for the formulation of the synthetic set of sector expenditures  Discuss the descriptive overview of public expenditure in the sector, agree priority focus areas and identify key issues, and  Discuss and schedule the technical working papers that should be prepared 5.4 Draft report workshop: A formal workshop should be scheduled after five months to discuss the key conclusions and recommendations of the study presented in a draft report. 6. Timeline 6.1 The proposed timeline for the agricultural public expenditure review is 2-3 months. The study is expected to be completed within the period of the overall ASR, which is 4-5 months. The entire study will be undertaken in three phases, starting with the planning and inception phase, the detailed diagnostic studies phase and the final drafting phase. In all the phases, the international and national consultants will work closely and report their progress to the MAFFS Coordination team, including the DPs. 7. Consultant Qualifications 7.1 The agricultural public expenditure review will be conducted by a consultant team comprising a national and non-national expert with at least a Masters Degree in Economics/Agricultural Economics, with no less than 10 years undertaking consultancy in agricultural sector analysis and expenditure reviews, preferably with demonstrable experience working with Sub-Saharan African countries. Extensive analytical and writing skills are a must. Both experts will work closely with at least one or a team of counterpart from the budget unit of MAFFS/MFEP. 8.0 COORDINATION OF THE STUDY The agricultural public expenditure review as well as the overall ASR will be coordinated by the Ministry of Agriculture, Forestry and Food Security (MAFFS). Correspondence should be addressed to the following: Mr. Prince Kamara SCP National Coordinator Ministry of Agriculture, Forestry and Food Security Youyi Building, Brookfields, Freetown Tel: +232 76 657 660 Email: Prince Kamara kotho2013@yahoo.com 88 Annex 2: Tors And Composition Of The Technical Working Group And Of The Steering Committee 89 Annex 3: MAFFS organogram 90 Annex 4: Ministry of Land and Country Planning organogram 91 Annex 5: Data collection strategy TYPE OF DATA / LIKELY SOURCE OF DATA EXPENDITU INFORMATION INFORMATION COLLECTION RE NEEDED STEPS / STRATEGY ON GoSL planned All budgets as Chief Accountant One to one BUDGET and executed voted, allocated, offices of all meetings expenditures revised and involved Ministries between at Central executed, from (All Ministries consultants and Level 2003 to 2012 /institutions/Agenci concerned (disaggregated, es having activities officers. electronic within the scope of versions) the Agriculture Regular reporting Sector.) + MOFED / to working group BUDGET BUREAU on progresses and difficulties. If all information is not available in electronic format, we may hire “data “ persons. Note 1 : SOE (information needs Ministries One to one to be collected but only the net Accountant Office, meetings with flow with public budgets will be SOEs accountants, SOEs / MDAs accounted for) accountants Note 2 : information on resources MDAs, Accountant One to one mobilized/ generated directly by office meetings with Ministries/ technical agencies SOEs / MDAs need to be clarified (ex : delivery accountants of services on a fee recovery basis, taxes, etc.) GoSL planned All budgets as MOFED. Local Govt One to one and executed voted, allocated, Finance meetings expenditures revised and Department. District between at Local Level executed, from councils Chief consultants and 2003 to 2012 for Administrators. concerned Agriculture officers. services (disaggregated, Regular reporting electronic to working group versions) on progresses and difficulties. + on a sample of 2 to 3 rural districts : 92 qualitative assessment (field based work) Government All budgets as Fund One to one institutions voted, allocated, Administration + meetings extrabudgeta revised and MDAs exercising a between ry funds executed, from tutelary over these consultants and : such as 2003 to 2012 funds (Ex : SLRA, concerned Forestry or Ministry of Works) officers. fisheries Fund, Road Cross-check funds. They information can be through meetings financed with donors in through own case some of self generated these funds revenues by manage donor an act of law funding. or an executive Regular reporting branch to working group decision. on progresses and difficulties. Partially Donor funded All project MOFED / STEP 1 : establish ON projects. documents, Multilateral project a first list of all BUDGET (Projects including division. agriculture expenditures financial MAFFS / MDAs, projects (based are reported agreement. + projects bureau. on MDAs but not annual / final PIUs records) executed completion Donors through the reports including STEP 2 : assess IFMIS expenditures data comprehensivene system) (electronic ss with donors version, For multi- disaggregated) STEP 3 : collect sectoral over the last 10 data information projects , data years. (MDAs, PIUs, collected only + project Donors) if more than evaluation report 70% of the when available budget is/was spent on agriculture. OFF NGO projects Project MAFFS / MDAs NGO Step 1 : BUDGET : all document, Desk consultant + agriculture project (do other ministries technical staff projects completion/interi have NGO desks ?) will be mobilized 93 implemented m report, + NGOs to collect a list of during the including reports projects last 10 years on budget implemented in SL. execution + during past 3 project years and evaluation report consolidate when available. information available at MAFFS NGO desk level Step 2 : quality and comprehensivene ss of the information will be assessed Step 3 : Direct consultations with NGOs will allow collection of missing information Donor funded Project Technical Agencies, - collect project. document, donors, GoSL (for information from Technical / project MOUs) resources UN agencies completion persons / MOUs / implemented report, including internet on list of projects (ex : reports on budget projects and regional execution institutions research + project leading the projects) evaluation report execution of when available these projects - look for interim / completion reports on internet / in Sierra Leone - contact by email / phone to ask for additional information Other Annual reports MDAs, Development types of (2003-2012), partners, Statistics informatio Stats, Sierra Leone, n Agriculture Sector Reviews, PERs, PEFAs, PETS, … 94 Annex 6: Level of expenditure in comparison to Maputo Targets (in Billion Leones) 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Total MAFFS 0 16,899 23,232 22,499 15,959 25,791 24,214 25,607 31,625 38,050 223,875 Other (COFOG) ministries and Agricultural entities 0 1,804 2,096 2,063 707 2,501 12,519 4,951 14,002 21,829 62,472 (COFOG) Tranfers to District expenditur Councils 945 3,346 5,876 12,974 12,364 13,940 49,446 e Donor funded expenditure 6,419 24,071 21,229 39,451 47,852 111,301 174,414 147,077 161,801 201,269 934,883 Total COFOG- agricultural expenditure 6,419 42,774 46,556 64,012 65,464 142,939 217,023 190,610 219,791 275,088 1,270,676 Central Level spending 598,000 668,100 828,300 915,500 568,794 800,168 976,068 1,433,482 1,541,440 2,106,066 10,435,918 Transfers to Total GoSL Districts Councils 8,248 24,143 27,187 54,801 53,532 63,055 230,966 expenditur Donor Funding e expenditure 32,269 352,935 690,834 535,087 874,580 708,885 1,408,390 1,880,950 1,390,659 1,871,453 9,746,042 Total GoSL expenditure 630,269 1,021,035 1,519,134 1,450,587 1,451,622 1,533,196 2,411,645 3,369,233 2,985,631 4,040,574 20,412,926 Share of agricultural in total 1.02% 4.19% 3.06% 4.41% 4.51% 9.32% 9.00% 5.66% 7.36% 6.81% 6.22% Source: Authors’ calculations based on data provided by: MoFED and GoSL official budget publications. 95 Annex 7: functional classification used by MAFFS under the MAFSAN project Agricultural development (Subsector) Livestock Food crop production Industrial crops/export crops Integrated agriculture projects Farm development Small livestock and poultry projects Agricultural development, activity unspecified Agriculture infrastructure Agricultural education/training (Subsector) District extension services Agricultural extension Radio, tv, electronic extension messages Formal education/training, Agriculture Agricultural inputs (Subsector) Supply of fertilizers Agricultural machinery/equipment Supply of seeds Agricultural inputs, activity unspecified Agricultural land resources (Subsector) Agricultural land resources, activity unspecified Drainage, Agriculture Agricultural policy and administrative management (Subsector) Agricultural policy and administrative management, activity unspecified Institution capacity building, Agriculture Agricultural sector policy, planning and programmes Aid to agricultural ministries Agricultural research (Subsector) All plant research activities Agricultural research, activity unspecified Agricultural services, purpose (Subsector) Agricultural co-operatives Agricultural financial services Agricultural services Plant and post-harvest protection and pest control Livestock/veterinary services Agricultural water resources (Subsector) Irrigation Food aid/Food security programmes (Subsector) Food supply Fishery development (Subsector) Fisheries Fishing policy and administrative management (Subsector) Institution capacity building, Fishing Forestry development (Subsector) Afforestation and reforestation Fuelwood/charcoal 96