WORLD BANK TECHNICAL PAPER NO. 41 8 WTP418 Work in progress for public discussion September 1998 Agriculture Sector Programs Sourcebo ok F-_~~~~~~~~~~~~ ,. 1z kdgeau ~ dC -?~~~~~~~ ;- ~ ~ 0j, itf.-! azze~~~~~~ Okd-(-, andA Lo.,,ai, ,@ RECENT WORLD BANK TECHNICAL PAPERS No. 340 Rutkowski, Changes in the Wage Structure during Economic Transition in Central and Eastern Europe No. 341 Goldstein, Preker, Adeyi, and Chellaraj, Trends in Health Status, Services, and Finance: The Transition in Central and Eastern Europe, Volume I No. 342 Webster and Fidler, editors, Le secteur informel et les institutions de microfinancement en Afrique de l'Ouest No. 343 Kottelat and Whitten, Freshwater Biodiversity in Asia, with Special Reference to Fish No. 344 Klugman and Schieber with Heleniak and Hon, A Survey of Health Reform in Central Asia No. 345 Industry and Mining Division, Industry and Energy Department, A Mining Strategyfor Latin America and the Caribbean No. 346 Psacharopoulos and Nguyen, The Role of Government and the Private Sector in Fighting Poverty No. 347 Stock and de Veen, Expanding Labor-based Methods for Road Works in Africa No. 348 Goldstein, Preker, Adeyi, and Chellaraj, Trends in Health Status, Services, and Finance: The Transition in Central and Eastern Europe, Volume II, Statistical Annex No. 349 Cummings, Dinar, and Olson, New Evaluation Proceduresfor a New Generation of Water-Related Projects No. 350 Buscaglia and Dakolias, Judicial Reform in Latin American Courts: The Experience in Argentina and Ecuador No. 351 Psacharopoulos, Morley, Fiszbein, Lee, and Wood, Poverty and Income Distribution in Latin America: The Story of the 1980s No. 352 Allison and Ringold, Labor Markets in Transition in Central and Eastern Europe, 1989-1995 No. 353 Ingco, Mitchell, and McCalla, Global Food Supply Prospects, A Background Paper Preparedfor the World Food Summit, Rome, November 1996 No. 354 Subramanian, Jagannathan, and Meinzen-Dick, User Organizationsfor Sustainable Water Services No. 355 Lambert, Srivastava, and Vietmeyer, Medicinal Plants: Rescuing a Global Heritage No. 356 Aryeetey, Hettige, Nissanke, and Steel, Financial Market Fragmentation and Reforms in Sub-Saharan Africa No. 357 Adamolekun, de Lusignan, and Atomate, editors, Civil Service Reform in Francophone Africa: Proceedings of a Workshop Abidjan, January 23-26, 1996 No. 358 Ayres, Busia, Dinar, Hirji, Lintner, McCalla, and Robelus, Integrated Lake and Reservoir Management: World Bank Approach and Experience No. 360 Salman, The Legal Frameworkfor Water Users' Associations: A Comparative Study No. 361 Laporte and Ringold, Trends in Education Access and Financing during the Transition in Central and Eastern Europe. No. 362 Foley, Floor, Madon, Lawali, Montagne, and Tounao, The Niger Household Energy Project: Promoting Rural Fuelwood Markets and Village Management of Natural Woodlands No. 364 Josling, Agricultural Trade Policies in the Andean Group: Issues and Options No. 365 Pratt, Le Gall, and de Haan, Investing in Pastoralism: Sustainable Natural Resource Use in Arid Africa and the Middle East No. 366 Carvalho and White, Combining the Quantitative and Qualitative Approaches to Poverty Measurement and Analysis: The Practice and the Potential No. 367 Colletta and Reinhold, Review of Early Childhood Policy and Programs in Sub-Saharan Africa No. 368 Pohl, Anderson, Claessens, and Djankov, Privatization and Restructuring in Central and Eastern Europe: Evi- dence and Policy Options No. 369 Costa-Pierce, From Farmers to Fishers: Developing Reservoir Aquaculturefor People Displaced by Dams No. 370 Dejene, Shishira, Yanda, and Johnsen, Land Degradation in Tanzania: Perception from the Village No. 371 Essama-Nssah, Analyse d'une repartition du niveau de vie No. 372 Cleaver and Schreiber, Inverser la spriale: Les interactions entre la population, l'agriculture et l'environnement en Afrique subsaharienne No. 373 Onursal and Gautam, Vehicular Air Pollution: Experiencesfrom Seven Latin American Urban Centers No. 374 Jones, Sector Investment Programs in Africa: Issues and Experiences (List continues on the inside back cover) WORLD BANK TECHNICAL PAPER NO. 418 Agriculture Sector Programs Sourcebook Nwanze Okidegb)e and Associates The Worlhi Bank Waflhington, D.C. Copyright 1998 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing September 1998 Technical Papers are published to communicate the results of the Bank's work to the development community with the least possible delay. The typescript of this paper therefore has not been prepared in accordance with the proce- dures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data in- cluded in this publication and accepts no responsibility for any consequence of their use. The boundaries, colors, de- nominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissem- ination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Drive, Danvers, Massachusetts 01923, U.S.A. ISSN: 0253-7494 Cover photo: "Senegal. Harvesting rice in the Casamance Region." Ray Witlin, World Bank. Nwanze Okidegbe is principal investment specialist in the Rural Development Department of the World Bank. Library of Congress Cataloging-in-Publication Data Okidegbe, Nwanze, 1947- Agriculture sector programs: sourcebook / Nwanze Okidegbe and associates p. cm. - (World Bank technical paper; no. 418) Includes bibliographical references (p.). ISBN 0-8213-4323-8 1. Agricultural development projects-Planning. 2. Agriculture- Economic aspects. I. Title. II. Series. HD1415.0395 1998 338.1'8-DC21 98-30553 CIP CONTENTS FOREWORD ..............................................................................v ABSTRACT ..................................... vi ACRO NYMS AND ABBREV IATIONS ..................................... vii ACKNOWLEDGMENTS .................................... viii 1 INTRODUCTION .................................... 1 Main features ................................2 Why develop an agriculture sector programr? ..............................2 Organization of the sourcebook ..............................4 2 ORGANIZING FRAMEWORK OF AGRICULTURE SECTOR PROGRAMS ..................................5 Sound macroeconomic and sector policies ..................................5 Coherent sector framework ..................................5 Institutional reforms ..................................7 Public expenditure review ................................. 10 Building consensus within government ..................................11 Fostering donor collaboration ................................. 12 3 PLANNING AND PREPARATION OF AGRICULTURE SECTOR PROGRAMS .16 Agriculture sector program preparation .18 Program features .23 Assembling and disseminating draft agriculture sector program reports . 25 4 IMPLEMENTATION ISSUES .28 Management .28 Monitoring, evaluation and reporting .34 Agriculture sector program progress review .37 5 CONCLUSIONS ................................................... 38 BIBLIOGRAPHY ................................................... 76 ANNEXES 1 Agriculture Sector Program Preparation Steps ............................................ 40 2 Possible Agriculture Sector Program Key Performance Indicators .................... 43 iii 3 Stakeholder Participation ................................................ 50 4 Economic Analysis of Agriculture Sector Programs .......................................... 56 5 Agriculture Sector Programs and Agricultural Technology ............................... 62 6 Conducting a Public Expenditure Review: Tips ................................................ 68 7 Cross-Cutting Issues ............................................... 73 BOXES 1 Agriculture sector programs and the private sector .............................................. 3 2 Consultative group meetings ............................................... 14 3 Task force appointments ................................................ 17 4 Regional Unit for Technical Assistance: Successful multidonor hub ................. 23 5 Adaptable program lending ............................................... 25 6 Disseminating agriculture sector program documents at provincial and district levels ................................... 26 7 Harmonizing implementation arrangements ................................... 29 8 Vicious circle of improving financial management ................................... 30 9 Procurement arrangements agreed under Ghana Health Sector Investment Program .31 10 Flow of funds arrangements agreed under Ghana Health Sector Investment Program .33 ANNEX BOXES A3.1 Agriculture sector program stakeholders in Kenya .51 A4.1 Zambia Agriculture Sector Program: Economic analysis .57 A4.2 Ghana Basic Education Sector Investment Program: Economic analysis . 59 A6.1 Public sector functions .69 ANNEX TABLES A2.1 Sectorwide indicators ..................... 43 A2.2 Extension ..................... 44 A2.3 Research ..................... 44 A2.4 Animal production and health ..................... 45 A2.5 Fisheries ..................... 45 A2.6 Farmland management ..................... 46 A2.7 Irrigation ..................... 46 A2.8 Farm power and mechanization ..................... 47 A2.9 Marketing and trade ..................... 47 A2. 10 Land administration ..................... 48 A2.11 Rural finance ..................... 48 A2.12 Forestry .... 49 A6.1 National expenditures in the agricultural sector: Example ................................. 71 iv FOREWORD Sector or subsector programs offer a way for countries to improve their public sector performance and the effectiveness of their sector development activities. They do this by addressing many of the problems which have impeded good performance in the past: an absence of a coherent sector strategy, inappropriate policy and institutional frameworks, misallocation of public resources and fragmentation and duplication of projects carried out by donors acting with little consideration of what others are doing. Until recently, there was little practical experience with sector and subsector programs. But with increasing evidence that these programs can effectively promote agricultural development, many countries are seeking advice on how to get started. This sourcebook provides practical guidance to policymakers and program planners in countries and to development agencies on how to carry out or support an agricultural sector program - from initial discussions, to developing the program concept, to securing broad stakeholder involvement, to implementation. It explains the reasons why countries may wish to undertake a sector or subsector program and describes the conditions which can help assure the program's success. It provides details on how to formulate a sector strategy, bring about institutional reforms and focus on core public sector activities based on a detailed public expenditure review. And it suggests ways of encouraging donors to work together under the leadership of the partner country government. The document has been produced in partnership with the United Nations Food and Agriculture Organization (FAO). It has arisen from a shared commitment of both the World Bank and the FAO to find ways of making development programs more effective and sustainable. The two agencies have now agreed to work together through the Cooperative Program to provide support to partner countries who are undertaking agriculture sector programs. Alex F. McCalla, Director and Chair, Sector Board Rural Development Environmentally and Socially Sustainable Development Network v ABSTRACT Increasingly, countries are undertaking agriculture development programs that cover the entire sector or a subsector. They are doing so to improve the allocation of public resources and the effectiveness of their sector development activities. Agriculture sector programs are country programs, initiated, planned and implemented by a country. Donors may be involved in supporting the program through a variety of instruments. But they must support government priorities as contained in the program, rather than promoting their own agendas. This eliminates the fragmentation, duplication and lack of country-ownership common under the old paradigm. This sourcebook provides practical guidance to policymakers and program planners in countries and to development agencies on how to carry out or support an agricultural sector program - from initial discussions, to developing the program concept, to securing broad stakeholder involvement, to implementation. It explains the reasons why countries may wish to undertake a sector or subsector program and describes the conditions which can help assure the program's success. It provides details on how to formulate a sector strategy, bring about institutional reforms and focus on core public sector activities based on a detailed public expenditure review. And it suggests ways of encouraging donors to work together under the leadership of the partner country government. vi ACRONYMS AND ABBREVIATIONS ASP agriculture sector program CG consultative group ERR economic rate of return ESP environmental sector program FAO United Nations Food and Agriculture Organization NGO nongovernmental organization NPV net present value SIP sector investment program SPA Special Program of Assistance vii ACKNOWLEDGMENTS The report was a joint effort by the World Bank and the Food and Agriculture Organization of the United Nations. It was prepared by a team led by Nwanze Okidegbe (World Bank, Rural Development Department) and included Christopher Gerrard and Daniel Sellen (World Bank, Economic Development Institute), Michael Wales and Alice Carloni(Food and Agriculture Organization, Investment Center). Each of the team members made contributions to the report, which was compiled by Nwanze Okidegbe. Chaim Helman and Naren Chanmugam (Consultants) carried out most of the research work, and Mr. Helman contributed significantly to the writing of the report. Wendy Ayres edited the report. Peer reviewers were Hans Binswanger, Gershon Feder, Sushma Ganguly, Jacob Kampen, David Nielson, Gerhard Tschannerl, and Stephen Weissman (World Bank); Gordon Sithole (Zimbabwe Ministry of Land and Agriculture); Adalbert Engel and Werner von der Ohe (German Agency for Technical Cooperation); and Stephen Jones (Oxford Policy Management); and Henk Moll (Wageningen Agriculture University, Netherlands). viii 1 INTRODUCTION Agriculture sector or subsector programs (ASPs) offer countries ways to improve their public sector performance and the effectiveness of their sector development activities. An ASP can be defined as the sum total of a country's medium-term (4-5 years) public program for agriculture as a whole, or a particular subsector. An ASP can be thought of as a "time slice" of the government's long-term (15-20 years) development program in agriculture. The program is initiated, designed and executed by government, so the exact activities it covers vary by country. Countries will choose the sequencing of particular activities depending on the depth of their private sector, the status of their policy frameworks and their institutional capacity to plan and implement development programs. A key characteristic of an ASP is that it is a country's program: it is initiated, planned and implemented by a country. It is not a lending instrument or a program of a development agency, and donors do not decide what is included or excluded from the program. Donors may of course be involved in supporting the program by providing technical assistance or financial aid. But countries may choose to implement the program relying solely on their own resources. Donor-support of ASPs differs from conventional donor support in a major way: with ASPs, donors support government priorities as contained in the program, rather than promoting their own agendas. This eliminates the fragmentation, duplication and lack of country-ownership common under the old paradigm. This sourcebook is provides practical guidance on how to plan, prepare and implement agricultural sector programs. It will be of interest to policymakers, program planners, government officials, local government staff and politicians, and staff of the World Bank and other development agencies. It provides a range of options to deal with issues that may arise in planning, preparing or implementing ASPs, from building internal consensus among government officials, to developing the basic program concept, to seeking donor support, and ensuring broad participation of stakeholders.' Whenever possible, a variety of approaches are presented as ways to solve problems. It is intended that this sourcebook will be updated periodically as more experience with ASPs becomes available. A stakeholder is an individual or group who is affected by the outcome of a proposed action, either negatively or positively, or who can affect the outcome of a proposed action. Stakeholders may include beneficiaries, government agencies, donors, nongovernmental organizations and other private sector groups. Main features ASPs have several common elements:2 * They are based on a coherent sector policy framework. The policy framework should include well-articulated sector strategies, sound macroeconomic and sector policies, and a medium-term public expenditure plan based on a comprehensive public expenditure review. These should be formulated in consultation with the stakeholders. * Local stakeholders are fully in charge. Local stakeholders plan, prepare and implement ASPs. Countries determine the type of expertise and technical assistance they need and are responsible for obtaining it. * All main donors supporting the sector are encouraged to collaborate in support of the overall program. This avoids fragmentation and duplication of development assistance to the sector and ensures the soundness and integrity of the comprehensive sector program. * As far as possible, countries should encourage the main collaborating financiers to adopt common implementation arrangements in the programs or subprograms they support. This minimizes conflicting procedures among donors and reduces administrative complexities for the country in implementing its agriculture activities. Why develop an agriculture sector program? Country drives the agenda, not donors. It is important for all countries -industrialized and developing - to channel scarce resources to priority activities that generate positive development impacts. Unfortunately, countries that rely heavily on donor assistance often have difficulty doing so, as donors take the lead in identifying, designing and implementing projects, pursuing their individual objectives with little collaboration with government or other donors. Furthermore, even if individual donor projects are well- designed, the multiplicity of donors, objectives, management systems and procedures makes it difficult for sector managers to absorb and manage aid efficiently (in 1992 according to the minister of agriculture, Zambia had some 200 agricultural projects underway).3 Although donors acted with good intentions, the approach did little to help most countries develop capacity or a sense of ownership of development programs. Many donors now recognize that the old approach resulted in fragmentation and often a waste of resources. 2 See Peter Harrold and Associates (1995), and Nwanze Okidegbe (1997). 3 The large number of visits by donors also strains a country's resources. Meetings with donors reduce the time top officials have to manage the sector. 2 Improve the allocation of resources. In many countries, the agriculture sector has suffered from decades of excessive and misguided government intervention. With the latest wave of economic reforms there is a general awareness that governments should concentrate on core activities in areas where the private sector cannot perform effectively. Often this does not include the traditional type of development activities and projects. Indeed it is likely that in the future, government will be involved mainly in creating enabling environments within which the private sector can thrive. During the transition period, however, many governments will need to continue performing functions that in the future will be better handled by the private sector. Because of their emphasis on policy reforms, ASPs provide the opportunity for systematic and vigorous examination of the public expenditure program as a whole, and public sector involvement in the agriculture sector. However, although the underlying principles of ASPs are straightforward, it is not always obvious what the role of government should be. For example, while it is widely accepted that government should not be directly involved in providing credit and instead have a regulatory and enabling role, there is no similar consensus regarding its role in providing extension and research services, especially for small farmers. It is up to countries to decide the appropriate roles for government on the basis of their own analyses and particular set of circumstances: decisions cannot be imposed from the outside. In many cases, countries will not be able to make all the changes they deem necessary during a short span of time and will need to take a longer- term perspective. Because ASPs deal comprehensively with the policies, institutions and public investments in the sector, they have considerable promise for addressing the constraints which have hobbled agriculture development. Important challenges for policymakers and practitioners include how to develop ASPs that provide effective and nondistortionary support to agricultural entrepreneurs, especially small producers, with minimum disruptions in the delivery of public services (Box 1). Box 1 Agriculture sector programs and the private sector Although the public sector manages most ASP expenditures, private sector actors are the main beneficiaries of the program: small farmers, agriculture input suppliers, traders and processors who have been operating in a distorted policy environment of price controls, inefficient government provision of financial, inputs and marketing services, and corruption. Experience has shown that this group of entrepreneurs is likely to function more effectively in a liberalized economy if a supportive policy environment is in place. Such an environment may include apart from favorable policies, infrastructure (roads and communication), support services (extension and research), and private financial, input and marketing services. ASPs generally support the creation or strengthening of such environments to the benefit of small agricultural producers. 3 Comprehensive ASPs require capacity is not available in all countries. Some countries may lack capacity to plan and execute the program: experience to date shows that this is the major obstacle to ASP success. Countries lacking capacity should devote the first the initial phase of the program to reforming institutions and building capacity. Capacity for financial management and monitoring and evaluation is especially important. Countries should also keep their early programs relatively simple, focusing on a few core activities, and undertaking further tasks as their capacity expands. Organization of the sourcebook The sourcebook consists of five chapters and seven annexes. Chapter 1 provides a general background on ASPs, and outlines the main principles. Chapter 2 describes the organizing framework for ASPs. Chapter 3 describes the ASP planning and preparation processes, and provides practical advice for the practitioner. Chapter 4 discusses implementation issues. Chapter 5 provides some conclusions. 4 2 ORGANIZING FRAMEWORK OF AGRICULTURE SECTOR PROGRAMS While the scope and nature of the programs differ by country, all ASPs involve common themes. Sound ASPs are based on (a) a coherent policy framework that identifies and addresses key macroeconomic and sectoral policy issues (there are no major policy distortions or at least measures are in place to redress existing distortions); (b) an institutional framework that is developed on the basis of a detailed institutional assessment of key institutions, including capacity building, and institutional reforms and rationalization; (c) an affordable and sustainable sector or subsector public expenditure program that improves the efficiency and impact of the public investments in the sector; and (d) government commitment to the success of the ASP. Sound macroeconomic and sector policies ASPs are not likely to achieve their development impact without sound macroeconomic and sector policies. Therefore ASPs are appropriate only in countries that are committed to broad macroeconomic reforms.4 Macroeconomic policies should ensure limited budgetary and balance of payments deficits, low inflation (less than 20 percent per year), positive real interest rates on deposits and loans and minimal distortions in the foreign exchange markets. Sector policies should minimize distortions in input and output markets. ASPs also do better if complemented by a vibrant and strong private sector that operates within a supportive policy environment. The following are examples of guiding principles that are conducive for agricultural development: * There is competition in output and input markets. * Activities are environmentally sustainable. * Stakeholders participate in all major policy decisions. * There is equitable access to resources. * Foreign exchange, trade, and tax regimes are not biased against the rural sector. * Government will not crowd out activities which might be undertaken by the private sector. Coherent sector framework The ASP must be anchored on and be developed within a coherent sector framework. A coherent sector framework involves a broad and clear long-term (15-20 years) vision of the future of the sector, includes guiding principles which shape all subsequent decisionmaking, and provides a strategic direction that brings together the actors in the sector. An ASP which covers only a part of the sector must also be grounded on the 4 Of course such policy reforms may be undertaken independently of an ASP. But an ASP can serve as a catalyst to bring about policy reforms as part of a comprehensive reform package. 5 sector strategy, and may be considered a step in a sequence of actions to be pursued over time. Sector strategy. A sector strategy is a plan for getting from where the sector is to where the sector needs to go - it is the means for achieving sectoral objectives. A sector strategy must describe both the processes and outcomes required, and must be in accord with the guiding principles. For example, activities might include processes that would lead to fiscal decentralization, liberalization of agricultural marketing, unification of the agricultural extension service, or promotion of community-based natural resource management. Sector policy. The specific items contained in the sector policy framework will vary between countries but may include some or all of the following: * Liberalization of agricultural markets: establishing market-based prices of inputs and outputs, reducing (or eliminating) export taxes, lifting trade restrictions and privatizing input supply and marketing parastatals. * Privatization of rural finance activities and support for the creation or strengthening of sustainable credit instruments to improve the access to credit by small farmers and the poor (for example, joint liability saving and credit groups). * Thorough scrutiny of all subsidies (for example, on credit, fertilizer, marketing and input supplies), and gradual abolition of most of them. * Support of activities that increase employment, reduce poverty and minimize gender biases. * Regulations and other actions (for example, research and extension activities) aimed at ensuring an adequate level of environmental management and natural resources conservation. The policy framework in countries that suffer from frequent droughts may emphasize the development of production technologies that would minimize susceptibility to drought. * Channeling a portion of agricultural research funding through competitive grants to improve cost-effectiveness of research activities. * Regulations and other activities related to land reform or land registration, or both. * Gradual but decisive steps towards decentralization of sectoral institutions. Pace ofpolicy reforms. Implementation of policy changes is likely to require substantial reforms. While some of these may be instituted rapidly, others are best introduced gradually to avoid creating service gaps that may have a negative impact on production, 6 employment and, especially, on the poor. For example, abrupt withdrawal of government from input supply and marketing functions before the private sector has had time to enter, could be disastrous and result in reduction in the cultivated area or leave farmers without essential inputs or a marketing outlet for their products. The importance of the correct policy environment should not be underestimated, especially where bad policies inhibit the uptake of good technology. Interventionist policies and regulations that protect inefficient public sector producers and organizations can be very expensive in terms of lost potential. It has been estimated that the delayed release of a rice variety known as Kalinga III in India, due to the failure of the public sector varietal release mechanisms, has resulted in over US$300 million of annual benefits forgone. Institutional reforms Institutions should be thought of in the general sense as rules, as opposed to the narrow sense of specific organizations or functions. Institutional reforms (changing the rules of the game) are necessary because prevailing rules may be inappropriate, ineffective or absent. Rules may be inappropriate in that they provide incentives for suboptimal coordination or supply. For example, a legally-enforced state hierarchy for agricultural marketing may crowd out private traders. Rules may be ineffective. They may provide incentives for the ideal kind of coordination but they may be impartially applied or ineffectively enforced. Rules may be absent. For example, markets rely on a set of rules to function properly (well-defined property rights, contract enforcement, competition laws, and the like). Institutional constraints. The past patronizing and interventionist environment, which was characterized by the donor-driven project approach, has created administrations that often suffer from: * Obsolete and unclear definition of sectoral objectives * Inadequate consultations with the clients (farmers) * Deficient or misguided incentives * Poorly motivated staff * Unclear job descriptions * Poor management practices * Over staffing, low productivity, and low pay * Persistence with functions that could be handled more efficiently by the private sector or civil society * Strong vested interests in maintaining the status quo. Substantial reforms will be required to overcome these institutional constraints and transform agriculture into a dynamic sector driven by market forces. This process takes 7 time. Indeed governments may wish to focus the initial phase of an ASP on accelerating institutional reforms and building capacity. Reforming institutions is not a trivial problem. Substantial institutional reforms, required to implement ASPs, take time. Reformers should keep the following in mind. * There are no blueprints. Blueprint thinking occurs whenever policymakers or experts propose uniform solutions to a variety of problems, usually based on one or more successful examples. Therefore, planners of ASPs must be wary of copying programs developed in other countries. * Institutional change is path dependent. The effectiveness of an institution depends on the overall institutional environment. Path dependency means that one has to start where one presently is, and this set of coordinates will differ from country to country. The development path of new institutions is dependent on the skills of the people who have and will work within them. Therefore, institutions created in the past impose constraints on institutional change in the present. Similarly, institutions created today will impose constraints on institutional change in the future. * Reformers must be cognizant ofparadigm paralysis. Paradigms are ways of seeing the world, which provide useful benchmarks for analyzing inputs, but also blind us from seeing and accepting new ideas. Often, those who have invested in the old paradigm have difficulty in seeing and recognizing the virtue of new paradigms to solve old problems, and are likely to resist reforms. * Changing rules is difficult because there are layers of interdependent rules in government. Operational rules influence (and are influenced by) governance rules, which are in turn conditioned by constitutional rules. * There are both collective action and rent-seeking problems involved in changing institutions. Groups which stand to benefit from the reforms - smallholders - are numerous and unorganized. Because it is difficult for them to act collectively, their influence over the reform process may be small. In contrast, powerful vested interests - typically, the potential losers - are often successful rent-seekers and can manipulate the reforms in their favor. Governments are vulnerable to rent- seeking (lobbying by powerful groups for special privileges) and may not be able to provide efficient and equitable rules. Process principles. Instead of blueprints, the ASP should emphasize principles of process, which: Identify how institutions affect people's behavior 8 * Assess the costs and benefits of reform and status quo (identify winners and losers) through stakeholder analysis * Mobilize the groups that would benefit from reform * Involve different stakeholder groups in the process (participatory rule-making processes may mitigate the problems of collective action and rent-seeking) * Keep the reform process transparent. While reforming institutions, the government must establish its credibility with respect to proposed institutional reform. It must also win support of the economic elites whose support is necessary for the reforms to succeed. Credibility can be achieved by coherence (ensured by consistent application of principles), transparency (ensured by open books and other institutional arrangements), and accountability (ensured by high quality reporting, as in the public expenditure review). Institutional assessments. Recognizing the notion of path dependency, reforms would be facilitated by a comprehensive assessment of the present situation. Such an exercise may be undertaken as one of the generic studies that start at the beginning of ASP preparation. This would entail a detailed assessment of the existing situation, including structure, personnel management, physical and financial resources and legal aspects. Approaches to institutional assessments. A variety of options are possible. Following Pinto and Mrope (1995), there are four broad categories of institutional assessment. The functional approach examines the institution's capacity to perform core organizational functions such as policymaking, policy implementation, human resource and operations management, coordination, monitoring and evaluation. The strategic approach focuses on guidance systems, such as mission, legal framework, structures, roles, objectives, plans, resource allocation, and distribution of power and authority. The environmental approach focuses first on the needs of the public that justify the institution's existence, whose demands can be used to assess the outputs of the institutions, which may then be linked to the inputs studied in the two previous approaches. The institutional economics approach also focuses on the outputs, and works backwards through patterns of interaction, rules in use, and nature of the goods and services provided. Capacity building and institutional reforms. Ideally, the institutional assessments would identify key reforms needed to improve institutional performance. They would also identify what capacity is required to plan and implement an ASP, and how to strengthen capacity if existing capacity is inadequate. Building capacity of local institutions to plan and implement ASPs is critical to their success. Where ASPs face problems, it is generally because too little attention was paid to building implementation capacity, especially for financial management and monitoring and evaluation. In some countries, the initial phase of an ASP may have to be focused on implementing institutional reforms and building essential capacity - necessary first steps. 9 Public expenditure review ASPs cover all the activities in the sector or subsector. To the extent that the ASP is subsectoral, it is essential that it is compatible with the overall medium-term expenditure plan for the agriculture sector. The public expenditure review should be used to critically assess the rationale for public sector intervention in agriculture and its consistency with the medium-term expenditure plan. This is important for many reasons. First, it could help to improve the allocation of resources: often the overall level of spending would be sufficient to stimulate growth but the latter may not materialize because of deficiencies in the allocation and utilization efficiency of both capital and recurrent expenditures. Second, it could rationalize spending on operations and maintenance: the effectiveness and sustainability of public expenditures on operations and maintenance of investments are likely to improve if they are met from the government's own budgetary resources. Third, it could facilitate transfer of resources between and within sectors, given the fungibility of resources between sectors. It is therefore desirable that a public expenditure review for agriculture be a part of a general public expenditure review covering the entire public expenditures in the country. However, carrying out a full review of a country's total expenditures may not always be feasible. Fourth, it could help bring about an efficient long-term sector budget allocation system that is sustainable and reduces the risk that the government is unable to meet its future budget commitments. Elements ofpublic expenditure review. The analysis should cover past spending patterns and suggest ways of reallocating expenditures in the future to achieve the strategic sectoral objectives. Overall, the following elements could be considered when analyzing the level and composition of public spending in the sector': 3 The role of the government versus the private sector. This is the principal criterion governing the choice of programs for public sector involvement in the agricultural sector. This requires assessing the determinants of demand and supply to identify the nature and magnitude of specific market failures, and hence the appropriate form of government intervention or regulation, financing or outright provision. It is equally important to determine the core functions of the public sector in agriculture. * An analysis of the impact of key programs on the poor. This is important to identify programs which help reduce poverty in a cost-effective manner. The relative allocations within and across programs should achieve the mix of corresponding sectoral outcomes that maximize social welfare. Consequently, the relationship between program expenditures (for example, rural investment funds) and their outcomes (for example, increased investment by the rural poor) should be the basis for allocating resources to achieve socially desirable outcomes. * The allocations for capital and recurrent expenditures should be analyzed in an integrated manner. This addresses the shortcomings of traditional capital-led 5 Based on Pradhan, 1996. 10 budgeting, which often leaves insufficient funds for recurrent costs, and which results in the crowding out of nonwage operations and maintenance by wage expenditures. This would require eliminating the distinction between capital and recurrent expenditures that donors often make against financing recurrent costs of country programs. It would also require an assessment of civil service overstaffing and the adequacy of civil service pay. Carrying out the public expenditure review. In many countries, the effort to carry out public expenditure reviews has been led by donors, especially the World Bank. As a result, countries neither see the full value of public expenditure reviews nor have the capacity to conduct one themselves. While assistance from donors will likely be required for some time until capacity is fully developed, countries should take the lead. Annex 6 provides tips on conducting public expenditure reviews. Building consensus within government The primary supporters of ASPs are generally senior officials of the ministry of agriculture. They need the support of other key ministries (ministry of planning and ministry of finance) to plan and execute an ASP. However, sharpening priorities and the requirement for coordinated donor support (not through a multiplicity of projects) is bound to generate opposition from those who are not likely to be in priority areas or those who benefit from project proliferation. Senior officials of ministries in countries that depend heavily on donor assistance may have some or all of the following concerns: * The ASP may reduce total donors' support to the sector. * Donors acting as a collaborative group are likely to have greater power to impose their will on government, and may be tempted to use it. * The ASP may reduce competition among donors and the ministry's capability to take advantage of such competition. * Government officials gain some benefits from the multiplicity of donors' projects (for example study tours, vehicles) that provide them compensation and make it more acceptable to work for low wages. The ASP would eliminate the fragmentation and thereby significantly reduce such benefits. In response, the advocates of a coordinated program approach may counter with some or all of the following (depending on the situation in the specific country): A sound ASP is likely to increase the likelihood of success and generate support of local stakeholders and donors. 11 * The fragmented, donor-driven project approach has contributed to duplication of efforts, lack of ownership and poor performance, and the displacement of qualified local experts with long-term foreign consultants. This has distorted remuneration and incentives systems. An ASP which focuses on the country's priorities are more likely to rectify these problems, and at the same time, improve aid efficiency. * The ASP approach means that the country, not donors, will prepare and manage the development program. The duplication and distortions introduced by donor-driven projects will be gradually eliminated (this has a considerable appeal in most countries). Fostering donor collaboration Fostering donor collaboration is important for countries that rely heavily on donor assistance, and need donor support to fund their ASPs.6 Experiences from many countries show that one of the ways to foster donor collaboration is to address their concerns, starting at the planning stage. Donors may have all or some of the following concerns about ASPs: * An ASP may reduce individual donors' influence and especially their control over the sector investments they finance. This is an outcome some donors may be reluctant to accept, especially those donor officials that are responsible for the sector at the field level. * Many donors are used to channeling their support to a specific geographical area (for example a part or whole of a region) or to specific activities (for example extension, research) and may be concerned about the feasibility of continuing to do so within the context of an ASP. * When one large development institution (such as the World Bank) emerges as the main adviser to the government preparing an ASP, some of the other donors or development institutions may be concerned that they will be marginalized. * Some donors may be concerned that the ASP will not include their projects as national priorities and may call for early closure and the dismissal of their nationals who are acting as long-term consultants before the end of their assignments. * Donors may have doubts about the capacity of government institutions to effectively plan and implement an ambitious sector program. Weak institutional capacity is in fact a major obstacle, threatening the effective preparation and implementation of ASPs in some countries. 6 This discussion is not meant to suggest that ASPs are donor-driven. ASPs are countries' medium- term public program for agriculture (covering the sector or a subsector), which donors are invited to support. 12 All the above are relevant issues that must be dealt with in a transparent fashion. This is particularly important for countries that rely heavily on donor support, and that would benefit greatly from reducing or eliminating the fragmentation and multiplicity of methods and strategies of donors. Obtaining genuine donor collaboration is likely to be a high pay-off process that would justify the considerable time and efforts involved. Donors appear to respond positively when they are kept fully informed and their views are considered. Experience so far has shown that, while some donors may immediately embrace the ASP concept, others may initially adopt the "devil's advocate" or "agnostic" attitude. However, donors' commitment to an ASP may increase as they see progress being made. The ASP preparation team should be sure to treat all donors equally and ensure that no one donor dominates the preparation process. The team should also identify and respond to donors' concerns early in the preparation process. Donors' concerns may be addressed as follows: * Donor's influence and control. Country ownership of programs contributes to the long-term impact and sustainability of investments. Many donors are pragmatic and are willing to relinquish direct controls, so long as the impact of their support is evident and they retain some influence as development partners. What is important, therefore, is to maintain effective dialogue with donors on a consistent basis. * Scope for supporting priority activities. Although donors will no longer undertake individual projects, they should still be able to provide support for their priorities within the program framework. Moreover, the monitoring and evaluation system can provide separate reports for each donor's areas of interest.7 * Closures of ongoing projects. Given the importance of initiating an ASP in a collaborative way, the ministry of agriculture may assure donors at the beginning of ASP preparation that, while an effort will be made to phase most projects into the ASP, ongoing projects are unlikely to be terminated before the planned closing date without the agreement of the concerned donor.8 * Capacity issues. While it is likely that several units key to ASP success may not function satisfactorily (notably financial management), strengthening their capacity - rather than bypassing them - will produce a sustainable solution. A pragmatic approach could be commitment from the government to first deal with capacity concerns and reach agreement with donors to gradually transfer responsibility to local institutions as capacity grows. 7 This may seem somewhat complicated but, in the computer era, can be done easily. s Such assurances should be given by the ASP task force and may be mentioned in the concept document. 13 * Doubts about the feasibility of ASPs. Persuading donors to support the government's agriculture sector program may take time. For a start, it should be possible to explain that past efforts which avoided addressing the constraints of inefficient local institutions by setting up separate project management units have failed. A good ASP can improve the effectiveness of development assistance, since (a) it minimizes government involvement by systematically phasing it out of activities that can be better done by the private sector; (b) it tailors the scope of activities to local capacity (including the use of locally managed consultants and technical assistance); and (c) it fosters the country's responsibility, commitment and ownership of its development agenda, which are necessary to improve the impact of donors' support. Communicating with donors. The country should be proactive in communicating all aspects of the ASP to the donors. There may be several ways to manage the process, for example through the ASP task force or a unit of the ministry of agriculture (the planning division or equivalent). The experience to date is not yet sufficient to conclude which way is better. The important consideration should be to appoint a body that can communicate effectively. However, the use of the task force appears especially promising since it enhances broader participation of key stakeholders. It is important that the team managing the preparation of ASP (the task force) maintain active communication with donors at both headquarters and in the country (Box 2). In addition, ASP issues may be raised at the consultative group (CG) meetings and similar forums. Most donors have desk officers for the countries they assist, and gaining their support would be most important. It would be advisable to maintain regular consultations with all key stakeholders to address issues of ASP design and implementation and donors' collaboration. Box 2 Consultative group meetings Given the high rank of both government and donor participants (the government is usually represented by senior ministers and other senior officials), the consultative group meetings could provide an effective forum to discuss ASP concepts and issues with important stakeholders. This can be done in several forms: . Distributing ASP documents and progress reports to CG participants. . Making presentations on the ASP at the plenary sessions of the CG meetings. * Making presentations during special (and less formal) side meetings that may be especially arranged to discuss the ASP. Donor coordination. Coordination of donor activities is clearly a government role, and arrangements that take the leadership away from the government are inconsistent with 14 ASP principles. However, experience reveals that in many countries donor coordination is quite difficult to achieve. Governments may lack the capacity or experience to handle this task. Or donors may simply be unwilling or unable to harmonize their support due to their own internal procedures. To facilitate effective donor coordination, government and donors should reach agreement at an early stage on how coordination will work. Two options include: a) Government, in agreement with all participating donors, designates one of the major donors to work with government on coordination issues. b) Donors nominate a multilateral agency (for example, the United Nations Development Programme, World Bank, regional development banks or the Food and Agriculture Organization) to work with government on coordination matters. There are no firm rules here and the experience gained so far is insufficient to allow firm recommendations either way. However, the observations suggest that it may be useful for government to select a donor representative to assist with coordination on the basis of personal suitability, articulation, vision and belief in the benefits of the ASP, rather than relying on a particular donor to perform this task. 15 3 PLANNING AND PREPARATION OF AGRICULTURE SECTOR PROGRAMS The initial drive to plan for the ASP may come from any of the stakeholders, especially government, such as officials of the ministry of agriculture. Early discussions are likely to center around the types of issues discussed in Chapter 2, and will form the basis for deciding on the focus and content of the ASP. Gaining the support of the key government ministries and politicians would be essential before full preparations start: ways to achieve that will vary among countries and sectors. It is likely to be easier to gain such support where there has been a positive experience with a sector program. In the Africa region, the sector program concept has been extensively discussed in various fora of the Special Program of Assistance and now has broad support, at least among policymakers and high level managers of development assistance. It is desirable to communicate ASP concepts to other stakeholders at an early stage. Because ASPs introduce significant changes that may affect vested interests at various levels, gaining the support of the various stakeholders will likely involve substantial efforts. The process is likely to be gradual, requiring careful planning. Initiating ASP preparation. A brief ASP concept summary document (2-3 pages), prepared by the ASP advocates within the government apparatus may be desirable to clarify the activities the ASP will include. The concept summary would then be used to build consensus within government and among key stakeholders outside government on the need for an ASP. Once a consensus is reached to move forward, an ASP team prepares a longer concept document to guide the preparation efforts. In countries where ASP-type programs are being implemented in other sectors and the stakeholders appear to be reasonably familiar with the concept, it may be appropriate to skip the summary and focus on the preparation of a concept document. Stakeholder identification and analysis. To ensure ownership, key stakeholders should be involved in the planning of ASPs from the beginning, including stakeholders at the provincial and district levels. A stakeholder analysis can be carried out as the starting point for identifying who has a stake in the development of the agricultural sector. Stakeholders in the agricultural sector include: * Various units and subentities within govermnent (agriculture parastatals, ministry of finance and planning and other ministries that overlap with the agricultural sector) * Primary stakeholders (small farmers, commercial farmers, and disadvantaged groups such as rural landless, women household heads and pastoralists) * Private sector actors (input suppliers, agro-processors, produce traders, farmer associations, cooperatives, farm lobbies, private contractors) * Civil society institutions (traditional chiefs, politicians) * Policy groups (universities, think tanks, NGOs) * Donors. 16 Consultation with stakeholders. Few countries have attempted any systematic consultation with key stakeholders in the agriculture sector because of the perception that such consultation on a national scale would not be cost-effective. There have been three approaches to beneficiary consultation that could potentially be cost-effective: (a) ad hoc consultation by working groups, (b) stakeholder workshops, and (c) village-based diagnostics using participatory rural appraisal techniques. Efforts should be made to involve the private sector, since their participation greatly improves the quality of ASP preparations. However, experience so far suggests that private sector interest often wanes during the often lengthy preparation of ASPs. The challenge is how to keep the private sector involved during the course of preparations. Task force. Given its sectorwide coverage, it is desirable to reach a broad national consensus on the scope and objectives of the ASP. It is equally important that it is prepared with broad participation of key stakeholders. One way to achieve this is for a task force (about 15-20 members) - with representatives from the public and private sector, user groups and civil society - to take responsibility for ASP preparation (Box 3). The task force chairperson should be a capable and well respected person in the country. The principal secretary responsible for the sector generally appoints the chairperson, although the minister may also take this responsibility. The principal secretary generally appoints the task force members. The task force and its chairperson report to the minister.9 Box 3 Task force appointments The importance of appointing suitable persons to the task force cannot be overemphasized. In several cases the main causes for a prolonged (and expensive) preparation process can be traced to poor leadership and indecisiveness of the task force and its chairperson. Task forces with poor leadership sometimes conduct an excessive number of seminars and workshops that appear to be building up national consensus, but are instead leaving participants confused and preparation stalled. Adequate training (for example, of the type provided by the Economic Development Institute of the World Bank) and facilitation of important meetings and workshops could help, but cannot substitute for good local leadership. Countries may request short-term technical assistance to help with this task. Here again the issue of leadership would be important. Strong leaders are likely to guide the consultants while weak leaders may hand responsibility to consultants (or the agency providing them), sacrificing national ownership. The head of the task force, should be a stakeholder with proven leadership qualities and experience in the sector. In order to carry out its functions effectively, the task force may require the support of a small, full-time secretariat. The secretariat would prepare material for the task force meetings, enabling the participants to cover a substantial volume of work in a relatively short period of time and perform other functions related to ASP preparation. This is 9 After the completion of ASP preparation and evaluation, the task force may be assigned the responsibilities of an agriculture sector advisory board that would oversee the implementation of the ASP. 17 likely to facilitate the participation of task force senior officials who would otherwise be too busy to participate effectively. There are at least two main options for the appointment of the secretariat: * Use of consultants (nationals, including seconded staff from the public and private sectors) * Use of a ministry of agriculture planning division (or its equivalent). The use of consultants or ministry staff would depend on local circumstances. Countries should decide on the basis of their individual circumstances. Agriculture sector program preparation Concept document. The first important function of the task force is to prepare a fairly detailed (8-15 pages) draft concept document. The document can serve as the main means of communicating to the principal stakeholders the ASP scope (core functions), evaluation criteria, design features and preparation processes. A special effort should be undertaken not to restrict such communications to the capital city but rather seek the views of stakeholders throughout the country, especially in rural areas. The feedback received should then be used to finalize the concept document - now based on a considerable consensus among stakeholders. Core functions. ASP is a country's medium-term development program in the agricultural sector, covering the entire sector or key subsectors. Its core functions will vary among countries depending on factors such as the role of the private sector in the economy, the extent of market failure, the stage of economic reform and the like. ASP core functions could include but are not limited to: * Policy and legislation * Regulation and quality control * Information services * Provision of extension services to smallholder farmers * Research priority setting * Disease and pest control * Development and protection of the natural resource base * Coordination of sectoral development * Private sector support * rural infrastructure. A note of caution could be helpful to the task force. If the core functions or ASP activities are all encompassing, the preparation teams may become overwhelmed and unable to complete preparations in a timely way, causing stakeholder fatigue and loss of 18 faith, straining the capacity of the country to implement such an ambitious program. The task force could minimize this potential problem by: * Limiting the preparation time to a fixed period (preferably less than twelve months). * Focusing the initial ASP on activities that do not require long gestation periods to prepare. Sometimes, it may be advisable to start with a few core activities and expand the scope as capacity is built up. * Initiate key generic studies as early as possible in the ASP planning cycle and convey the findings and recommendations to the teams preparing the ASP. Evaluation criteria for ASP components. Past experience indicates that preparation teams often submit proposals that require substantial revision. This increases preparation time and cost, and causes substantial frustration. To avoid this, the concept document should include evaluation criteria which can be used to assess the suitability and viability of the proposed ASP activities. These criteria will be discussed and agreed with the government and be used for both task force assessments. It should be made clear from the outset that components that do not satisfy the criteria will be changed to fit the criteria or excluded from the ASP. The criteria should be explained to the preparation teams, and may include: * Scope of the ASP, and appropriate roles of public and private sectors * Assessment of consistency of ASP activities * Assessment of affordability and sustainability * Economic analysis and program viability * Cross-cutting issues * Technical feasibility. Each of these is covered in detail below. SCOPE OF THE ASP, AND APPROPRIATE ROLES OF PUBLIC AND PRIVATE SECTORS. What is the role of agriculture in the economy? What will the ASP cover? What are the respective roles of the public and private sectors? The statement should make clear that activities best handled by the private sector should not be carried out by the public sector. Every activity proposed for inclusion in the ASP should be subjected to the private sector counterfactual test. The key studies that would be essential in justifying the focus and content for an ASP will depend on a number of factors, including the quality of information available on the sector. This will certainly vary between countries. However, key studies may include: * Agriculture sector review. Such a review can help define the sector strategy and its policy framework. It can assess, under various scenarios, the growth and 19 employment potential of the sector. It can help determine the current status of policy adjustments, legislation, and investments in the sector, including any unfinished reforms necessary to realize growth, especially among smallholders. The results of the studies are likely to have an important impact on the coverage of the ASP. * Budget management in agriculture. The budgeting system will have to be reformed in most cases in order to decentralize the flow of funds, minimize the potential for misuse and make it possible for the majority of donors to channel their funds through government. The study should analyze the existing situation and propose alternative ways to organize the financial system in a decentralized fashion. * Organization, manpower, and institutional structure. The study, which may include an inventory of manpower and an analysis of the incentives system, would evaluate the existing situation and propose reforms. The findings would be used to design the necessary institutional reforms, including capacity building initiatives. Other studies required, which the task force could commission, may include: * Management information system study, to identify the most suitable ways to organize the flow of information from the districts to headquarters. e Legal framework study, to assess existing laws, rules and regulations, and their suitability for providing the necessary support to a liberalized production system. The study is expected to highlight missing pieces of legislation and obsolete laws and regulations that may constrain private sector activities. e Privatization study, to assess the potential for privatization of government services, and performance studies for key agricultural services (for example extension and research, monitoring and evaluation system, livestock), if this aspect is not included under the public expenditure review. ASSESSMENT OF CONSISTENCY OF ASP ACTIVITIES. The use of the logical framework may be an important tool for deciding which activities are most likely to contribute towards achievement of ASP objectives. The logframe facilitates result-oriented management, with operations beginning with a set of objectives rather than components. The logframe could be used to present, in a matrix format, the objectives of ASP activities, the outputs required to achieve these objectives and the inputs that are needed in order to get the outputs. The matrix would provide a narrative summary of the quantities expected (of inputs and outputs), verifiable performance indicators and important assumptions (mainly linkage between inputs, outputs and objectives, and risk). It could also help to define activities at various levels (including subsector and sector) in an easy to understand and systematic way, emphasizing the cause and effect linkages. 20 ASSESSMENT OF AFFORDABILITY AND SUSTAINABILITY. It should be demonstrated that the ASP is affordable within the government's budget and can be sustained over time. This is where the public expenditure review becomes essential. The preparation teams should generate data to show that the proposed ASP expenditures are within the historical levels of support for agriculture. EcONOMIC ANALYSIS AND PROGRAM VIABILITY. ASPs require the use of pragmatic methods of evaluation that can contribute to improving both their preparation and implementation. There is not yet sufficient experience to say with certainty which methods of economic analysis are most suitable for a program such as an ASP. It would be appropriate to use standard techniques of economic analysis - economic rates of return, net present values, and least-cost assessments - for reviewing specific investments or activities of the program, such as extension and research services or large irrigation schemes. Other types of evaluation techniques which may also be useful include analysis of fiscal impact, and affordability. The standard sensitivity analyses would be required to test the sensitivity of the results to the assumptions made. Such methods are well documented and their description is not repeated here.' The economic evaluation will complement other component screening tests that will assess sustainability and justification for public sector involvement (see Annex 4 for more on the economic analysis of sector programs). Since an ASP is a sector or subsector development program and the bulk of its benefits are likely to accrue from increased public sector efficiency and increased private sector investments, institutional reforms and investments supported within the context of an ASP should also be evaluated. Sector level analyses should be carried out to estimate the changes in the sector that are expected to take place over time." The analysis is likely to include an assessment of the expected increases in production, employment and growth in general, as well as improvement in social welfare in the sector (extent of poverty reduction, improved gender equity), in a detailed and verifiable way."2 It may also include an evaluation of fiscal impacts, private sector counterfactuals and an analysis of indirect impacts that may result from the fungibility of funds. The assessment would also involve the identification of verifiable key performance indicators that can be continuously monitored during implementation and can therefore be used for early detection of serious deviations. The information generated through the sector level analysis may serve as a baseline for assessing future changes. While the sector level analysis undertaken at preparation would be an ex ante exercise, the sector performance 10 See for example, Squire and Van der Tak, 1975. " It would be important to compare the results obtained in categories of producers, classified by income, farm size and gender in order to determine the distribution of benefits, and assess if target groups are positively affected. 12 For example, assessment of estimated increased production of export and subsistence crops, increased private investment, adoption of improved technologies, and others. Much of the baseline data should be generated under the proposed generic study of agricultural strategy and growth prospects. 21 analysis that would be conducted annually during implementation would provide an ex post evaluation of sector performance and facilitate timely changes where necessary. CROSS-CUTTING ISSUES. These include issues such as environment, poverty and gender. The ASP preparation teams should evaluate the extent to which the program contributes to minimizing environmental deterioration, reducing poverty and improving gender equity. Annex 7 contains suggestions on issues to consider when addressing these cross- cutting issues. TECHNICAL FEASIBILITY. The development objectives of any ASP would most likely be achieved through broad-based productivity gains in agriculture, especially among smallholders. At an early stage of the ASP preparation process it is essential to identify and discuss the technical means by which these gains would be achieved, including changes in incentives to use such technologies. The technical changes to be encouraged may range from wider use of a single production input (fertilizer or a new crop variety) to multiple adjustments in the way natural resources are managed in whole production systems. All such changes are labeled as "technology" and should be addressed in the technical strategy for sectoral development. The ASP document should also include a technical assessment of all activities included in the program, taking into account the agronomic issues such as agroecological patterns, the state of research in generating appropriate technology, issues of dissemination and adoption, acceptable agronomic practices and the like. Annex 5 contains a detailed discussion of technology strategy issues. Task force guidance of preparation. During the preparation period, the task force will monitor the progress made by the preparation teams through regular (weekly or biweekly) meetings, during which each team will report about the progress made and problems encountered. Such meetings will help guide the teams in handling complex issues and adhering to the evaluation criteria defined at the start of preparations. They will also help reveal weak teams, and provide them with assistance or, in extreme cases, guidance in changing their membership. The task force should hold regular meetings ( at least once a month) and the chairperson of the task force should keep the minister and senior staff of the ministry of agriculture informed on the progress of preparation. Preparation support to the taskforce. In some countries the task force may ask donors to assist the preparation teams in the planning, preparation or implementation of their programs. This assistance can be delivered in different ways: * Donor staff or consultants. Donors could jointly carry out studies with governments, (for example, public expenditure reviews or institutional assessments). In such cases, their staff or consultants would work under the direction of government. 22 * Multidonor regional hubs. Multidonor hubs, staffed by regional and international experts, can serve as regional centers of assistance for countries preparing and implementing programs. Their functions may include studies and analytical work, consensus building on policy and institutional reforms, and direct support for program preparation and implementation. Box 4 describes the activities of the Regional Unit for Technical Assistance (RUTA), a multidonor hub for Central America located in Costa Rica. Box 4 Regional Unit for Technical Assistance: Successful multidonor hub RUTA is a cost-effective and client-oriented service unit based in Latin America which assists countries with project preparation and implementation, institutional strengthening, the development of trade policy, and training. The establishment of RUTA, with funding from seven different multilateral and bilateral sources, has helped greatly in addressing the needs of rural people in Central America, and is a major success. RUTA's most important function is to serve as a focal point for partnership for several multilateral and bilateral institutions, and borrowing member governments. Reliance on RUTA has enabled the Bank to meet rapidly growing client demands in Central America at reduced cost. RUTA's non-Bank sponsors - FAO, IFAD, UNDP, Inter-American Development Bank, and bilateral donors (including Norway, which has contributed expertise on gender issues) - stress its value in strengthening their partnership with the World Bank. The Central American governments, in turn, are appreciative of the client- orientation and direct attention received from RUTA. Several governments are undergoing economic and social reforms and are seeking guidance or technical assistance from RUTA. Program features An important characteristic of ASPs is their flexibility. ASPs have no standard design that is appropriate for all countries and sectors. Policymakers and planners develop their ASPs according to the particular circumstances prevailing in their country. Some countries have found it helpful to harmonize implementation arrangements and to build into the program monitorable indicators of program impact. Common implementation arrangements. Harmonization of implementation arrangements among donors supporting an ASP facilitates implementation by eliminating costs, delays and sometimes confusion that governments experience in trying to meet the demands of different donors. Donor coordination, while helpful, is not enough. Ideally, donors would all adopt common procedures, including pooling funding with common planning, budgeting, accounting, and audit control procedures. In response to governments' concerns, some donors are now discussing ways to apply common procedures, but the universal use of these measures are likely to take time. 23 There is anecdotal evidence that people preparing an ASP do not deal with the issue of common implementation arrangements until actual implementation issues arise. Often the government does not consult with the collaborating donors until the program is about to go into implementation. This is too important an issue to ignore until the end of the preparation process. Rather, harmonization of implementation arrangements should be made an essential part of donor support. The first step is to consult with collaborating donors to identify the implementation procedures that could be readily harmonized and those that require more time. General rules on how to proceed should be studied, including making individual donor requirements less binding. These rules should then be factored into the ASP and specified in the program documentation. The detailed steps involved in implementation are generally laid out in the government's implementation manual. Impact evaluation. Key performance indicators are useful for evaluating the impact of components of an ASP as well as sectorwide performance. These indicators are expected to measure the impact of the program on the whole sector. They should track major changes in the aggregate economic and institutional changes associated with the ASP, including the extent to which growth in output can be attributed to better policies and management, efficiency gains in the administration of the public expenditure program and improved delivery of public goods and services. In addition, key indicators are also essential in measuring changes in farm household incomes and welfare and tracking the effectiveness of public expenditures in reducing poverty. Program planners should develop a set of performance indicators for the program and the associated quantitative and qualitative measurement techniques. These should be discussed and agreed with the key stakeholders. To make them manageable and effective, it may be best to use only few indicators for the program. The specific indicators to be chosen will depend on the types of activities covered by the particular ASP. Annex 2 contains a list of key performance indicators that some countries have found useful. Even well-chosen performance indicators are of little use if countries lack the capacity to monitor them, evaluate program effectiveness, and make necessary changes. Fostering capacity to carry out these functions must therefore be a priority. Matching the program to available resources. It is important to program ASP activities at a fiscally sustainable level from the outset and avoid the submission of inflated proposals that may have to be substantially reduced at a later stage of preparation or during the evaluation process. Accordingly it would be desirable to obtain guidelines from the ministry of finance or planning on the budget framework for the ASP. If firm guidelines cannot be obtained, a best estimate can be made in consultation with the ministry of finance. A notional allocation which is consistent with the medium-term expenditure program should then be provided to the preparation teams. To minimize the risk that the program must be scaled back, the preparation teams should rank the core functions in order of priority and make adjustments to the program to fit available resources. 24 Countries that depend on donors will have to consult with and convince collaborating donors to support the program. Discussions with donors should begin at the early stages of the ASP preparation process so that government can realistically plan the scope of the program. However, while some donors may be able to provide a rough estimate at the beginning of the preparation process of the level of their contributions, many donors will not make firm commitments before assessing the detailed planned program. Moreover, as the budget program cycle of many bilateral donors may not coincide with the ASP's planning horizon, the commitments provided can be regarded as representing an order of magnitude rather than precise figures. Despite their efforts, some governments may be unable to mobilize sufficient resources to fully fund the ASP, jeopardizing its effectiveness. If the funding gap is revealed prior to implementation, governments can adjust the program to match available resources (from domestic or external resources). However, sometimes unanticipated funding gaps arise during implementation, if collaborating donors drop out or experience cuts in their development budgets. To avoid problems during implementation, governments should work with donors to find flexible funding arrangements that would allow donors to scale their contributions up or down during the course of implementation. Some donors already use flexible lending instruments. For example, the adaptable program loan of the World Bank is an instrument that finances long-term programs whose specific elements may change over time (Box 5). Box 5 Adaptable program lending Adaptable program loans (APL) have been developed in recognition that programs may take years to implement, and that the specific activities pursued may change over time. Sustained effort, flexibility, strong partnerships and continuity are required to implement long-term programs successfully. APLs provide funding for long-term development programs, based on criteria for the program that include agreement on long-term development objectives, sound economic and technical justification, viable financing plans for well-defined priority activities, and clearly defined and measurable milestones and performance indicators with explicit linkages to development objectives. APLs allow the Bank to scale back or increase its support to the program in accordance with prevailing conditions and experiences or to allow other partners to participate. Lending instruments, such as APLs, help countries to avoid the problems that can arise when there are lags between the end of one project and the beginning of another. They can also enable countries plan their long-term development agenda with the knowledge that funding will be available without unnecessary interruptions. Assembling and disseminating draft agriculture sector program reports Typically, the draft ASP reports are submitted to the task force through the secretariat on completion of the preparation team's report.'3 The secretariat will review the report for 13 The procedures outlined in this section provide general guidance only, and should be adapted to the special situation prevailing in each country. 25 consistency with the ASP concept document and the evaluation criteria, and submit it to the task force with comments and recommendations. A joint task force and preparation team meeting may then be convened to discuss the team report, to be finalized with task force guidance on changes that need to be introduced (if any). When this process has been completed for all the teams, and their revised reports submitted, the secretariat will prepare a comprehensive report that will cover all ASP activities, and prepare a summary of the report, translating it, if necessary, to local languages. Local stakeholders from both the public and private sectors (including NGOs) and from national to district levels, will be asked to comment on the draft report and its key elements (Box 6). The comments will be incorporated by the secretariat in the preparation report before it is finalized. The final report, endorsed by the task force, is then submitted to the minister of agriculture for government endorsement, and thereafter sent officially to the donors with an invitation to jointly support the program. Box 6 Disseminating agriculture sector program documents at provincial and district levels So far there has not been a satisfactory experience with communicating ASP documents beyond the ministry of agriculture and donor circles. There is a need for an effective communication strategy. Four main ASP documents that should be disseminated to stakeholders in the rural areas are: the concept document, the proposed ASP program, the approved agriculture sector program, and the implementation manual. Experience has shown that if the dissemination exercise is not programmed well in advance, it may not take place at all. The ASP may then be reduced to an exercise that has been conducted largely at the capital with little concern for field realities, and is likely to be criticized by local stakeholders as well as donors. Consequently, ASP success may be put at great risk before implementation has begun. To prevent this, the task force should plan the dissemination process early in the preparation process. There are several ways to organize the dissemination: a) The task force may send some of its members to discuss the ASP, preferably at the concept document stage. b) Local NGOs or consulting firms may be hired to undertake the dissemination. c) A combination of the above. Preevaluation and evaluation. Donors providing significant funds in support of ASPs will generally wish to evaluate the program before making firm commitments to it. Donors may evaluate the program individually or with other collaborating donors, which may lower costs and promote cooperation. The evaluation is usually conducted in two stages: preevaluation, and evaluation. During the preevaluation, the donors systematically assess the proposals submitted by government using the agreed evaluation criteria, and recommend changes where necessary. Government and donors then agree on the time required to implement the recommendations. The evaluation follows after the agreed changes are made, and is used by donors to identify the program activities they would like to support. Following the evaluation, each donor prepares the documentation 26 it needs to satisfy its own authorities, describing in detail the activities it wishes to finance. The preevaluation and evaluation processes validate to stakeholders and donors that the program is economically and technically sound. The information generated can help preparation teams refine their proposals for ASP activities, eliminating those that do not meet the tests of economic and technical soundness. Stakeholders are also likely to use the information to develop a strong monitoring and evaluation system that will facilitate a reliable ex post assessment of ASP results and continuous improvement of its activities. As ASP operations expand, alternative methods are likely to be developed and be available to stakeholders. 27 4 IMPLEMENTATION ISSUES Management ASP implementation is likely to be affected by the quality and capacity of management at all levels, from district to national. It is important to build capacity to plan and implement the program from the very beginning of program preparations. Governments need to institutionalize mechanisms for generating feedback from different client groups as opportunities in the sector change, and from farmers on the quality, affordability and relevance of services provided. There are a number of options for generating such feedback, including beneficiary assessment, participatory rural appraisal, systematic client consultation, and participatory monitoring and evaluation. Both sound management and stakeholder participation in ASP implementation can be achieved by: * Involving the stakeholders in ASP management in an advisory role. e Appointing competent managers with demonstrated abilities to manage personnel and budgets, who easily consult with clients and subordinates, who can delegate responsibilities, and who set a personal example of efficiency and integrity. - Holding training sessions for staff. It is important that staff at all levels receive training to familiarize them with ASP principles and implementation issues (the need for client consultation, and special attention to gender issues, management, personnel and incentives). The content of the course may be adjusted according to the concerns of the staff taking the course (for example managers, research and extension personnel, livestock specialists, and others). The training courses would typically last 2-3 days. * Initiating the process of integrating ongoing projects into ASPs. ASP management will be easier if donors adopt common implementation arrangements (Box 7). Often, several donors will support an ASP, each with specific requirements for channeling funds, procurement and reporting. Most donors recognize that harmonizing implementation arrangements will help facilitate the execution of the ASP. An agreement in principle on common implementation arrangements may be reached during the preparation or the evaluation stages. However, actually implementing the new arrangements may not be easy and will require persistent efforts. Common implementation arrangements are especially important for financial management (budgeting, accounting, auditing), procurement, and for monitoring, evaluation and progress reporting. ASP advisory board. Experience so far has demonstrated the importance of involving the stakeholders in the implementation of ASPs. One way of doing this is to establish a sector advisory board or similar body to overview implementation. The sector advisory 28 board may be the same size as the task force, have a similar distribution of members (by stakeholder affiliation) and a similar appointment pattern.14 It may or may not include donors at the prerogative of the country. However, if donors are not included, arrangements must be made to brief them regularly on ASP progress and constraints and provide them with a regular opportunity to express their views. Box 7 Harmonizing implementation arrangements The adoption of common implementation arrangements is likely to be a slow process. The implementation manual should highlight the proposed implementation arrangements. Donors are likely to agree to harmonize some activities but resist harmonizing others, until it is clear that the ASP is successful. Donors are most likely to agree to common reporting and auditing functions, and least likely to agree to harmonize procurement and pooling of funds. With the Mozambique health sector investment program donors first harmonized consultant terms of reference, and are now considering adopting common accounting, procurement and management procedures. It is anticipated that, as their confidence and experience with sector programs increases, donors will accelerate their harmonization of implementation arrangements. Financial management. Effective, transparent and reliable management of funds is crucial to the success of ASPs. Stakeholders have been increasingly sensitive to issues of misuse of funds, their delivery to target destinations, and the capability of government officials to demonstrate full transparency and accountability. It is recommended that ways are found to establish a strong financial management system (Box 8). The financial management system should be staffed with skilled and competent financial experts who can inspire confidence among stakeholders. Countries sometimes invite foreign experts to assist them until they build up capacity to manage the system themselves. It is easier to convince donors to channel funds through the country's financial management system, once the government can demonstrate that the country's financial management system performs well. The smooth implementation of an ASP will require the timely flow of funds to the district level. This will most likely occur if government delegates responsibility accountability for financial management to the local level and local staff have the capacity to effectively manage their resources. Thus, it would not be sufficient to have a strong central unit at the national level: there must also be suitable units in each district to account for and control funds for district activities. Detailed proposals that stipulate implementation steps should be specified during ASP preparation. A major training effort of staff would be required before and during implementation. A gradual computerization of the financial 14 Members of the task force may become the members of the sector advisory board. Alternatively, some or all the members may be newly appointed. 29 system would also be required. Consequently, the transition from an existing to a new system may take some time. Box 8 Vicious circle of improving financial management Donors who believe that capacity is weak sometimes refuse to channel funds through the government. However, they usually recognize that the ad hoc arrangements they adopt are unsustainable in the medium and long-run and are willing to channel funds through a government unit provided it works well. Yet, unless donors are willing to use a unit established by government, it cannot gain experience and demonstrate its ability to work effectively. The way out of this vicious circle may be to agree that, during the period in which a strong financial management unit or system is being organized, donors continue to use their established channels. Donors would transfer responsibilities to the financial management system only once they are convinced that the system is functioning satisfactorily. The financial management system would gain experience with the management of donors' funds by initially managing the funds of donors that are obliged (by internal regulations) to channel funds through the government (for example, World Bank, International Fund for Agricultural Development, and African Development Bank). It would therefore be advisable in certain cases to contract experienced consultants to design and initially implement a decentralized financial system, and provide training to staff who will eventually take over the operation of the system. Such a decentralized system would require effective checks and balances in order to minimize opportunities for misuse of funds and corruption. Where the ASP is not the first sector or subsector program in the country, the design should take into account the financial arrangements being implemented in other programs and, if suitable, adopt a similar structure. It may also be desirable to standardize to the extent possible the structure of the financial system of the various ministries that are active in the rural areas of the country. The financial management system would also require adequate audit arrangements. Government and donors should be able to reach agreement on standard audit requirements acceptable to both, probably involving the use of independent auditors. National procurement systems. Appropriate procurement procedures for goods, works and technical assistance are essential for effective implementation of an ASP (Box 9), as for any development activity. Traditionally, each donor has had specific requirements and procedures for procurement. However, when many donors are supporting a program, having individual procedures makes procurement cumbersome and can delay implementation. It is therefore important that government work with the donors to develop a national procurement system that is acceptable to all collaborating donors. As most donors have general (not country-specific) requirements, any agreements reached under one program can be used as the basis for follow-up programs. 30 Box 9 Procurement arrangements agreed under Ghana Health Sector Investment Program The Ghana health sector investment program is financed by the Government of Ghana and its cooperating partners (Danish International Development Agency (DANIDA), Department for International Development (DFID), International Development Association, United Nations Children's Fund, World Health Organization). The following are the agreements reached on procurement: * A procurement unit is to be established within the ministry of health. * Until such time that government's procurement procedures satisfy the requirements of the individual cooperating partners, all donors will follow the guidelines set by the IDA. * The government is to appoint an independent procurement review agent agreeable to the cooperating partners. All procurement decisions not subject to prior review may be subject to review by the independent agent, who will report findings to the government and the cooperating partners. * Government may procure goods directly from United Nations agencies in accordance with United Nations guidelines. * IDA will evaluate all contracts awarded through international competitive bidding and the health account. In turn IDA will consult with the rest of the cooperative partners. Thus, IDA is the primary contact between the government and donors. Experience so far with sector or subsector programs shows that common procedures are feasible, although full harmonization will take some time to achieve. What is important is that all participating donors are mindful of the problem and start working together to find a solution. The following should be considered in addressing the issue of common implementation arrangements for procurement: * Legal framework. An official procurement code must be in existence for the implementation of this initiative. Such a code would establish institutional procedures, define thresholds of authority, and contribute to overall transparency of procurement practices. * Contracting to the private sector. Where local capacity for managing a national procurement system does not yet exist, a substantial share or all of the procurement burden could be contracted out to the private sector until the necessary capacity is built. However, there is not yet much experience with such arrangements. * Standardized documentation. Early agreement on standardization arrangements would likely improve transparency and reduce delays. * Procurement review and accountability. Clear directives must be given on financial ceilings for procurement of goods. Additionally, allocation of responsibilities to the district, regional and central tender boards will have to be agreed, including post- review of procurement actions by a third party. 31 The following are some tips for handling procurement matters: * The progress with common implementation arrangements for procurement made under previous (especially the latest) sector or subsector programs should be assessed. These can serve as the minimum common implementation arrangements for procurement under the ASP being considered. * It is important to ensure that the list of items proposed for procurement annually consists of items required to achieve ASP development objectives - it must not be just a shopping list. Therefore, the sector advisory board in collaboration with the donors should maintain strict criteria for screening the procurement proposals. * To maintain the flexibility required for a comprehensive program like an ASP, it is important to agree that the investments and procurement requirements decided during preparation and endorsed at ASP evaluation, are indicative only - made largely in order to estimate the program's cost. A reassessment of the specific requirements must take place each year at the time budget proposals are assessed. - Although introduction of common implementation arrangements may facilitate procedures, this may not be sufficient to streamline procurement. Often the responsibility for procurement delays lies with subsector managers, who are not presenting their requirements with adequate justification or do not submit the specifications required to complete the tender documents. Consequently the training requirements of managers concerned with procurement must be carefully considered. Flow of donor funds. Timely flow of funds from collaborating donors are essential for smooth implementation of the ASP. Flow of funds are linked to procurement as donor funds are usually disbursed against expenditures as they are incurred. However, once qualified expenditures have been made, an appropriate system for reimbursement must be designed as a part of the financial management system. Decentralization may delay flow of funds, since much of the documentation required for submission of requests for fund release will come from lower level governments. Therefore designers of the financial management system should develop user-friendly procedures for release of funds, and persuade donors to agree to them. Commnon implementation arrangements for donors' funds would facilitate the process: these have already been agreed upon in some sector programs (Box 10). Annual work plans and budget. The preparation process described in Chapter 3 would produce an indicative plan of ASP activities for a period covering 4-5 years. That plan needs to be reviewed and revised in the light of experience in the course of annual budget preparation exercises. In accordance with government budget procedures, annual work plans and budget estimates have to be approved by the government and incorporated in the national budget. Assuming decentralization, preparation of work plans and budget 32 estimates for the ASP would usually start at the district or provincial level and be aggregated at the central level. This is an important exercise, as the work plans and budget estimates approved by government (usually ministry of finance) will determine the resources available for the following year's activities. Box 10 Flow of funds arrangements agreed under Ghana Health Sector Investment Program * Using the standardized financial statements and supporting documentation, requests for withdrawals will be submitted in accordance with each of the individual cooperating partners' requirements. * For annual disbursements, the government and cooperating partners will agree on a procurement schedule by September 30, based on a consolidated sector budget plan and the previous year's performance. Quarterly progress reports and financial statements will be provided by government to the cooperating partners. * The health account will be maintained in a commercial bank and operated by the financial controller of the ministry of health. As a prerequisite to disbursement, budget and management centers throughout the various levels will have to show that they have a strengthened financial management capacity. * Government, with the approval of the cooperating partners, will appoint an independent auditor to oversee the program funds. The following actions could improve the effectiveness of the preparation of the workplans and budget estimates: * Review the lessons from past years to assess the need for changes in activities and methodologies, using data and analyses generated by the monitoring and evaluation system. 3 Issue clear preparation guidelines including budget envelopes and eligibility criteria. * Train the people who are responsible for preparing the workplans and budgets in activity-screening techniques, such as the logical framework, to minimize inclusion of activities that are not essential for the achievement of ASP development objectives. The draft workplans and budget estimates are reviewed by the sector advisory board and finalized. Countries seeking donor support then send them to collaborating financiers for funding. Donors are more likely to commit to funding the annual program if they are consulted while the workplans and budget estimates are being developed. 33 Monitoring, evaluation and reporting In general, good monitoring and evaluation systems can help management with early identification of inadequate performance - such as deviations from plans, client dissatisfaction and weaker than estimated linkages between inputs, outputs and objectives. Monitoring and evaluation systems may also highlight the actions that need to be taken in order to improve performance. However, monitoring and evaluation in many developing countries remain inadequate. This is because two key prerequisites for successful monitoring and evaluation systems are seldom satisfied: * A capable monitoring and evaluation team that can produce timely and reliable results and guide managers on corrective actions (the team must have the managers' confidence); and * Managers who are both capable and willing to use the result of monitoring analyses to improve decisionmaking and have the courage to face disappointing results and take action to correct them. Reasons for poor performance of monitoring and evaluation systems also include the proliferation of different monitoring and evaluation approaches and methodologies in the same country and sector; the unsustainable use of long-term foreign experts (when they leave, the system declines); and a chronic gap between the collection and processing of data due to unclear objectives and inadequate processing facilities."5 A monitoring and evaluation system for an ASP will likely have good prospects for success if: There is an implicit recognition by the government that due to the ASP's scope and complexity, the most capable managers are required. * The government and stakeholders, including donors, require that a strong monitoring and evaluation system be installed to provide them with continuous information on ASP progress and constraints, which will also allow program managers to respond quickly to problems. is Field surveys have suffered from the inclusion of too many questions that generate data having little (if any) operational use on one hand, and the absence of key questions on the other hand. The first deficiency can be corrected by screening every question proposed by asking: "What operational use will be made of the data collected?" and "If the question is omitted, what would be the operational consequences?" If the answers indicate minimal operation use and minimal harm as a result of exclusion, the question should be excluded. To ensure that important questions have not been omitted, it is important to assess (before survey questionnaires are finalized) the source of data required to estimate the key performance indicators and the logframe's verifiable indicators, the input-output linkages, and the statistical analysis methods that would be employed. 34 * The sector advisory board is interested in and demands timely and high quality monitoring and evaluation results. Composition of a monitoring and evaluation system. The monitoring and evaluation system of a typical sector or subsector program is likely to consist of the following main components: * Management information systems as the main source of data; * Sector performance analysis as the means for evaluating the quantitative and qualitative performance of the sector; and * Quarterly and annual progress reports. Comparisons of plans with actual performance, evaluation of key performance indicators and trend analyses would be used to assess deviations from plans and validity of important assumptions. Management information systems. These should be designed in collaboration with managers to provide them with the means to monitor the progress of activities by comparing planned (as recorded in the work plans and budgets) and actual physical and financial inputs and outputs. This can expedite the undertaking of timely corrective actions where warranted. The management information system may be computerized to various degrees, consistent with local capabilities (a gradual approach would be appropriate). Sector performance analysis. As it covers the entire sector or subsector, assessment of ASP impact would require the development of a suitable framework for evaluating the sector's overall performance, termed here "sector performance analysis." The purpose of the sector performance analysis is to review and analyze, both in absolute terms and in comparison with previous years, what happened in the sector during the year being reviewed, and why. The analyses include changes in the cultivated area, production of key crops as well as total production, employment, incidence of poverty, and the like. Furthermore, using a combination of quantitative and qualitative methods (for example multiple regression to estimate production functions, as well as systematic client consultation methods) the analysis is expected to capture the main sources of variation and explain the main reasons for the changes identified. For example, sector performance analyses may try to quantify (using statistical methods) the impact of variable weather conditions, increased adoption by producers of new technologies, investments, availability of credit, policy changes, or external shocks (for example, decline or rise in the price of an important export crop). The findings are summarized in interim and final reports and, following discussions at the sector advisory board and ministry of agriculture, distributed to all stakeholders and acted upon. 35 Measuring the impact on poverty of program activities is an essential element of the monitoring and evaluation system. The methodology used would be country specific, and rely on ongoing surveys to the extent possible. Such surveys may include household budget surveys, labor force surveys, agricultural surveys, demographic surveys and other social surveys. A number of international institutions have implemented national surveys. They include the Demographic and Health Survey of the United States Agency for International Development, the Inter-Agency Food and Nutrition Surveillance Program of the Food and Agriculture Organization, surveys by the United Nations Children's Fund and the World Health Organization, and poverty and vulnerability mapping by the World Food Programme, together with a number of country-specific household surveys supported by bilateral donors. The World Bank's initiatives include the living standard measurement surveys and the social dimensions of adjustment surveys. Sector performance analysis management. To ensure that the investigators are able to conduct studies free of political or administrative pressure, it would be desirable to contract the sector performance analysis to an institution that is independent of government, such as a university or research institute."6 However, the operational interpretation of the results and corrective actions must be discussed by the sector advisory board and undertaken by the government (planning division or equivalent). The timing of sector performance analysis results is crucial. It is important that they are available when the annual work plans and budgets are being assessed and changed. In addition, it is important to convey the results and recommendations to stakeholders before the commencement of the annual review, held to discuss and endorse the annual work plan and budget proposals. Progress reports. The monitoring and evaluation system, designed primarily to serve sector managers and stakeholders (including donors), is expected to produce a large volume of data. In order not to overwhelm managers with excessive details, the data need to be organized and summarized in a user-friendly way. The reports should be demand-oriented, and include information which various levels of management require to improve decisionmaking. Accordingly, reports designed for higher-level managers (minister, permanent secretary, directors) may be appropriately summarized (to ensure that busy officials read them). Donors should be able to specify a uniform report format that will satisfy their requirements. However, it is anticipated that the data collected for ASP monitoring and evaluation will be sufficient for this purpose. 16 This will also be consistent with the private sector orientation of ASP: government should be engaged only in activities that cannot be undertaken by the private sector. In the case of monitoring and evaluation and the sector performance analyses, this suggests that all studies and surveys be contracted to the private sector. However, the supervision, operational interpretation of the findings, and recommendations should be the responsibility of the govermment (for example a planning division in the ministry of agriculture). 36 Agriculture sector program progress review An ASP's progress review requires special arrangements and broad collaboration among stakeholders. It is different from the progress review of projects, carried out by donors through short visits by headquarters staff. Since the ministry of agriculture delegates implementation responsibilities to its district staff, it needs information, collected on a continuous basis, about the program's impact in the field. The local stakeholders and the donors require similar information to assess the impact of the ASP. It may therefore be possible to extend the concept of common implementation arrangements to the supervision process, which could comprise: * Joint progress review (country and donors), held once or twice a year, involving all the key stakeholders; and * Local technical team that could be engaged full-time to provide support to the field staff according to pre-arranged schedules. The team would prepare periodic reports on its findings and share them with the stakeholders. Such reports could also serve as inputs to the joint progress reviews. Joint progress review. One or two joint government and donors progress reviews will usually be held every year. The timing of one of the main joint progress reviews should be synchronized with the schedule of the budget preparation process. Its main purpose will be to review the ASP activities for the following fiscal year as presented in a draft work plan and budget. Such a review will make extensive use of the monitoring and evaluation findings, especially the findings of the sector performance analysis and the recommendations of the sector advisory board regarding corrective actions. This joint progress review would provide an opportunity to critically assess ASP progress - its impact on agricultural gross national product, employment, poverty reduction, gender issues, efficiency of public sector services and the progress with the privatization. It would also provide an opportunity to reassess the basic assumptions used during preparation and evaluation and introduce necessary changes. In the course of the joint progress review, progress with procurement and flow of funds as well as the performance of the financial system are likely to be subjected to special scrutiny. In addition, the review team will also analyze the data on the progress of field activities. Based on the conclusion of the joint progress review, the ministry of agriculture will prepare a revised work plan and budget. 37 5 CONCLUSIONS Many governments, who are attempting to make their public sector activities more efficient and effective, are considering, or actually undertaking, agriculture sector or subsector programs. ASPs differ from traditional donor-driven projects in a major way: they are countries' programs - countries initiate, plan and implement them. ASPs are not lending instruments or programs of a development agency. Often donors choose to support ASPs because they recognize that country ownership and capacity are key to long-term development. But countries may choose to undertake ASPs that rely wholly on their own resources. This sourcebook is intended to provide guidance to countries dealing with a variety of issues encountered while planning and implementing such programs. ASPs are all different, reflecting varying conditions in countries. Still, they all have some cornmon elements: * They are based on a coherent sector policy framework. * They cover core government activities, but are flexible enough to be scaled up or down, depending on the level of resources available at any particular time and the rnanagerial capacity to implement the program. * Local stakeholders are fully in charge. * All main donors collaborate in support of the overall program. * As far as possible, main financiers adopt common implementation arrangements. Experience to date shows that a major obstacle to ASP success is a lack of country capacity to plan and execute the program. Capacity for financial management and monitoring and evaluation is especially important. Countries lacking capacity should focus initially on reforming institutions and building capacity, as part of their overall agricultural expenditure program. They should also keep their early programs relatively simple, focusing on a few core activities, and undertaking further tasks as their capacity expands. Experience also reveals that in many countries donor coordination is quite difficult to achieve. Governments may lack the capacity or experience to handle this task. Or donors may simply be unwilling or unable to harmonize their support due to their own internal procedures. However, there are examples of mechanisms that government could use for donor coordination. The Special Program of Assistance for the Africa region, a forum for donors to discuss ways to coordinate their assistance, provides a model for governments. Aspects of ASPs requiring further observation and experimentation are: 38 * Communicating the ASP effectively to stakeholders and beneficiaries in rural areas without making the preparation period unacceptably long. * Developing effective common implementation arrangements. * Building a sound monitoring and evaluation system, especially the sector performance analysis and associated reports, so that it can replace the conventional and economic sector work undertaken by the World Bank and other donors. * Implementing effective decentralization. * Maintaining effective task forces and sector advisory boards, without raising ministry of agriculture concerns. * Fostering donor collaboration for ASP support under the leadership of government. 39 ANNEX 1 AGRICULTURE SECTOR PROGRAM PREPARATION STEPS Initiating the agriculture sector program Initiating discussions. The initial idea of undertaking an ASP may come from government officials, civil society representatives, or private sector actors (including members of NGOs). If the people who initiate discussions are from outside the ministry of agriculture, they will have to convince the senior officials of the ministry officials (minister, permanent secretary, or other key officials) of the advantages of the program. They will also need to respond to questions and concerns, so it is important they are knowledgeable of the advantages and disadvantages of ASPs and of experiences inside and outside the country. If the response is favorable, ASP proponents may propose appointing of one or two people to prepare a brief draft concept summary paper. Gaining donor support. In countries which depend heavily on donors, ministry of agriculture officials talk individually with the donors they expect would support the program. Appointing the task force. The minister of agriculture appoints a task force of 15-20 members representing key local stakeholders. The task force is assisted by a secretariat and reports to the ministry of agriculture. Early activities Preparing the concept document. The task force prepares a draft concept document for the ASP, including objectives, scope and guiding principles (including evaluation criteria for ASP components). Local stakeholders discuss the concept document and provide comments. The task force finalizes the document. Launching the generic studies. Government and stakeholders agree on which studies to undertake during discussions of the concept document. Generally, consultants or universities carry out the studies. Appointing preparation teams. After determining how many preparation teams are needed (4-6 teams, each with 5-7 members), the task force selects team members and the chairpersons (generally members of the task force). The principal secretary or minister of agriculture endorses and announces the appointments. To the extent possible, team members should come from the various local stakeholder groups. Training. The task force and members of the preparation team participate in training exercises. Training often includes study tours of countries implementing ASPs (or sector 40 or subsector programs); a workshop to initiate ASP preparation"7; and brief training courses for interested task force and preparation team members'8 on preparation techniques (economic analysis, use of the logical framework and key performance indicators, procurement, and the like). Gaining consensus Startup workshop. The task force informs the preparation team of the criteria that will be used (by both task force and the donors) to evaluate proposed ASP activities and components, and explains the evaluation criteria for procurement. Submitting proposals. Preparation teams submit to the task force secretariat outlines of the methodology they propose to follow during preparation, timetable for preparation to completion, and cost estimates. After reviewing and commenting upon the ASP contents, the secretariat passes them to task force members. The preparation teams should be sure that both the time frame and the cost are consistent with the guidelines. Disseminating ASP information to stakeholders. The ASP process and guiding principles are disseminated to local stakeholders throughout the country. This can be accomplished through a series of one-day workshops with beneficiary representatives. The task force and preparation teams are briefed on the feedback. Informing donors. A small team (which may include the minister or permanent secretary of the ministry of agriculture) presents the proposed ASP to donors. The task force chairperson and the minister or other senior official also presents the proposed program to donors at the (usually annual) consultative group meetings. These documents may include the concept document, recent progress reports and reports on the findings of the generic studies. Briefing local stakeholders. During ASP preparation, the task force holds monthly or biweekly briefing sessions with the minister of agriculture and senior ministry staff, and monthly (or more frequently if requested by donors) meetings with stakeholders and donors. In addition, it may be desirable for the task force secretariat to circulate a progress report (1-2 pages) to stakeholders. This will help generate early buy-in by local stakeholders and donors. Submitting draft reports of generic studies. The people charged with undertaking the studies submit their draft reports to the task force secretariat. The findings are communicated to task force and preparation team members in special seminars or in seminars arranged for other purposes, depending on the circumstances and the importance of the study results. 17 This typically lasts about one week. It should be professionally facilitated. Is The courses would last 3-4 days, and cover topics on request. 41 Submitting draft preparation proposals. Each preparation team submits draft proposals to the task force secretariat. The secretariat reviews the proposals for consistency with the specified criteria and passes them on to the task force with its comments. The task force meets with the preparation teams to suggest changes, which the preparation teams incorporate in their final reports. The task force secretariat compiles the draft preparation report. Disseminating ASP preparation reports. A series of workshops are held where task force, preparation team members and other stakeholders (including beneficiaries, government and donors) throughout the country discuss the ASP contents. The scope and content of the ASP are amended if necessary in accordance with the feedback. Issuing the final preparation report. The chairperson of the task force submits the final preparation report to the minister of agriculture, who then seeks official government approval. Donors receive copies and are invited to attend a joint donor meeting to evaluate the proposals. Preevaluation and evaluation Organization of the evaluation process. The donors, the ministry of agriculture and the task force meet to discuss the timing and organization of the preevaluation and evaluation process. Preevaluating the ASP. Each preparation team presents its section of the ASP (including the policy framework) at a seminar and responds to donor's questions. At the conclusion of the preevaluation stage, stakeholders and donors provide feedback and make recommendations for improving the program. The preevaluation process varies in length, depending on the complexity of the program, but typically lasts 2-3 weeks. The task force and preparation teams make the desired changes. Based on experience to date, this may take 2-3 months, or even longer if the changes are substantial, and require frequent consultations. The donors receive an invitation to attend the final evaluation of the ASP. Evaluating the ASP. This process is usually shorter than the preevaluation process because most of the outstanding issues have been resolved. The discussions focus on the scope and technical feasibility of the program, including ways to finance them. 42 ANNEX 2 POSSIBLE AGRICULTURE SECTOR PROGRAM KEY PERFORMANCE INDICATORS Below is a menu of potential key performance indicators that could be used for evaluating the overall impact of an ASP or its sub-program. These indicators are merely suggestions. The final choice of indicators depends on the development objectives and components of an ASP, and should be decided in consultation with the stakeholders. Table A2.1 Sector and subsectorwide indicators Unit Status Current Previous Year Year 1. Agricultural GNP/total GNP Percent 2. Rural poor/all poor Percent 3. Agricultural export/total export Percent 4. Value of agricultural export US$ 5. Agricultural production index US$ 6. Food production index Number 7. Non food production index Number 8. Food production per capita index Number 9. Area under major crops Hectare 10. Incidence of drought Amount of rainfall 11. Yield of major crops index Percent of normal unit 12. Income per farm house hold US$ 13. Incidence of malnutrition Number 14. Adoption of improved technology Percent 15. Beneficiary satisfaction with government Percent agricultural services 16. Value of output per farm household US$ (by farm size) 17. Value of output per hectare of cultivated land US$ 18. Fertilizer consumption Tons 19. Food aid (cereals) Tons 20. Environmental degradation (incidence of Percent soil erosion) 43 Table A2.2 Extension Unit Status Current Previous Year Year 1. Adoption rates of key technologies Percent 2. Percentage of farmers effectively reached by Percent extension 3. Index of beneficiary satisfaction Number 4. Number and quality of requests from farmers Number 5. Change in access, affordability, relevance and Percent use of information 6. Profitability of selected farming activities and US$ different farm types 7. Farmers using conservation farming methods Percent (soil degradation measures,extension coverage) 8. Farmers per extension officer Number Table A2.3 Research Unit Status Current Previous Year Year 1. Applied experiments that have resulted in Number extension messages 2. Percentage of resources devoted to different Percent crops 3. Failed research experiments Number 4. New varieties released Number 5. Number and quality of scientific laboratories Number supporting research 6. Ratio public/private expenditure in research Ratio 7. Share of public research contracted out Percent 8. Input of stakeholders to priority-setting Number and types of inputs 9. Researcher contact with farmers and Number extension agent 10. Farmers adopting new varieties Number 11. Ratio of female to male researchers Percent 44 Table A2.4 Animal production and health Unit Status Current Previous Year Year 1. Livestock mortality and fertility (calving) rates Percent 2. Availability and use of preventive services Percent 3 Need for, availability and use of curative Percent services 4. Number of genetically improved animals Number 5. Livestock covered with compulsory Percent vaccinations 6. Private and public clinics Number 7. Milk yield per location Tons 8. Availability of feed resources Number 9. Profitability of selected livestock activities US$ 10. Slaughter facilities Number Table A2.5 Fisheries Unit Status Current Previous Year Year 1. Acquaculture farmers Number 2 Total income from fish marketing US$ 3. Fish protein as percent of daily protein intake Percent 4. Value of fish exports US$ 5. Hatcheries Number 6. Fishing boats Number 45 Table A2.6 Farmland management Unit Status Current Previous Year Year 1. Farmers' associations for soil conservation Number, type 2. Government soil conservation programs Number, type 3. Farmers using conventional tillage methods Number 4. Farmers using animal draft power/tractor Number 5. Land under conventional tillage Hectares 6. Land under animal draft power/tractor tillage Hectares 7. Land under cover crops Hectares 8. Land under agroforestry Hectares 9. Land degradation Percent Table A2.7 Irrigation Unit Status Current Previous Year Year 1. Average irrigation water delivery per hectare Cubic meters of irrigated land per hectare 2. Area irrigated index Percent 3. Percent of farmers irrigating more than 50 Percent percent of their land 4. Land and labor productivity on irrigated areas US$ per day _ 5. Performance of collection of water charges: Percent ratio of water charges collected 6. Marginal cost of water US$ per cubic meter 7. Marginal cost of new irrigated land US$ per hectare 46 Table A2.8 Farm power and mechanization Unit Status Current Previous Year Year 1. Farmers (and area cultivated) using Number draft power for soil preparation (hectares) 2. Farmers using mechanical Number cultivation/instruments Table A2.9 Marketing and trade Unit Status Current Previous Year Year 1. Intertemporal price variation Percent 2. Marketing margins Percent 3. Farmers' access and use of market information Number 4. Value of traditional and nontraditional exports US$ 5. Percentage of households participating in Percent output and input markets 6. Share of government owned storage capacity Percent 7. Length of rural roads Kilometers 47 Table A2.10 Land administration Unit Status Current Previous Year Year 1. Farmers' perceived security of tenure Percent 2. Number of farmers who want land title to gain Number access to credit 3. Incidence of soil degradation Percent 4. Percentage of female farmers who have title to Percent their land 5. Is there market for land? Number of transactions 6. What is the legal constraint to transferring land? Existing law Table A2.11 Rural finance Unit Status Current Previous Year Year 1. Number of farmers participating in savings Number schemes or having a formal savings account 2. Ratio of savings to total loan portfolio Ratio 3. Percentage of rural population with access to Percent credit 4. Average real on-lending rate on the agricultural Percent credit portfolio 5. Number of new loans to intended farm Number households 48 Table A2.12 Forestry Unit Status Current Previous Year Year 1. Change in the rate of deforestation Percentage change 2. Supply of categories of forestry product Tons 3. Change in forest area by type Percentage change 4. Change in export earnings from forest Percentage products change 5. Change in forest areas under protected status Percentage change 6. Increase of area under watershed management Percentage change 7. Increase in demarcation of lands and resources Percentage for indigenous groups change 49 ANNEX 3 STAKEHOLDER PARTICIPATION Stakeholder analysis A stakeholder is an individual or group who is affected by the outcome of a proposed action - negatively or positively - or who can affect the outcome of a proposed action. Ideally, a stakeholder analysis is done very early in the ASP planning process, to identify who has a stake in the development of the agricultural sector. Stakeholders in the agricultural sector may be classified as follows: Primary stakeholders * Small farmers * Commercial farmers * Disadvantaged groups (landless, women household heads, pastoralists). Public sector * Ministry of agriculture central departments (working group members) * Ministry of agriculture decentralized staff (regions, provinces, districts) * Agricultural parastatals (commodity boards, and others) * Ministry of finance or planning, or both (in the gatekeeper role) * Ministries whose mandate partially overlaps with that of ministry of agriculture (environmental agency, and ministries of natural resources, local government or rural development). Private Sector * Seed companies, input suppliers * Agro-processors (industrial, artisanal) * Produce traders (exporters, importers, wholesalers, petty traders) * Producers' groups (farmer associations, cooperatives, farm lobbies) * Consulting firms (engineering, management, studies, surveys) * Private contractors (equipment, earth moving, and the like) Civil society institutions * Chieftainship * Elected representatives (district and village councils) * Traditional village social institutions (elders, savings groups, exchange labor) * Community-based organizations (farmer groups, women's groups, youth groups, common interest groups) 50 * Universities, agricultural colleges, policy institutes and think tanks. Others * Donors (multilateral, bilateral) * Nongovernmental organizations To identify and analyze which categories of stakeholders are important, the ASP preparation team should ask these questions: What issues are at stake for this category of stakeholder? How important are agricultural policies and services for the livelihoods of the stakeholders? To what extent does the success of the program depend on these stakeholders? How much influence does this stakeholder have over the program? Stakeholder analysis starts with listing all stakeholders in the sector. Primary stakeholders should be identified by geographic zone, farming system, gender and wealth. Analysts then identify the interests at stake, determine what impact the ASP will have on the interests of various stakeholders, assess the roles of various stakeholders in furthering sector objectives, and judge the extent to which various stakeholders influence sector programs. The data are used to identify which stakeholders are most important for the success of the program, which are most influential, and which are most likely to be bypassed unless special efforts are made to reach them. This exercise also helps to ensure that a wide range of stakeholders are involved in the ASP preparation process (see Box A3.1). Box A3.1 Agriculture sector program stakeholders in Kenya In Kenya a workshop to present the agricultural sector review, agriculture strategy and ASP concept documents to a group of private sector, farmer organizations, NGOs and govermment representatives identified key stakeholders as: * Farmers and their organizations: Smallholders and commercial farmers, farmers' groups (male and female), cooperative societies, apex and national farmer organizations * Public sector: Ministry of agriculture, other line ministries, treasury, ministry of planning and national development, department of personnel management, parastatals, area development authorities and local government including provincial and district administrations * Private sector: Traders, exporters, agro-industries (including small-scale traders and agro- processors); * Policy group: Academic institutions, other institutes and think tanks (such as KENGO), elected representatives at the national and local levels, financial institutions and NGOs. Stakeholder participation 51 Participation is a process through which stakeholders influence and share control over development initiatives, decisions and resources. Stakeholder participation can take different forms, ranging from information-sharing, to consultation, collaboration and empowerment. Stakeholder participation can significantly improve program design and sustainability. In Kenya and Lesotho, the involvement of stakeholders at the regional level highlighted the need for differentiated strategies for high and medium potential zones and for mountain areas versus lowlands. In Zambia, involving district staff in program implementation has led to improved performance at the field level. In Senegal, consultation with regional political groups is helping government understand and respond to the concerns of potential opponents to the new policies. In Lesotho, consultation with small farmers has produced useful feedback on price liberalization and grazing fees. Stakeholder involvement in program preparation Representation in the ASP task force and working groups. The principal justification for stakeholder involvement in the planning and preparation of ASP is to ensure that the program is locally owned. Stakeholders participate in the preparation of the ASP primarily through a task force, consisting of a steering committee and thematic working groups. The task force is supported by a secretariat which is usually comprised of national consultants or senior staff of the ministry of agriculture or agricultural policy and planning unit. Experience suggests that it is critical for the implementing institutions to be involved in the preparation of the activities that they will be called upon to implement in order to avoid implementation problems later. While appointing an autol amous task force operating outside government has the advantage of being less subject to political pressure, it also has the disadvantage of potentially weakening the sense of ownership by the implementing agency. Selection and composition of ASP working groups. In Zambia, the agricultural sector task force included members of farmer organizations, individual farmers, representatives from the agricultural industry, financing institutions, and representatives from the University of Zambia and other academic institutions. Donor agencies, NGOs and representatives of private agro-chemical and seed companies attended periodic reviews. It is desirable to have significant private sector representation on the task force. However it has sometimes been difficult to keep private sector representatives actively involved over an extended period of time due to other commitments. Experience in Mozambique suggests that the involvement of the private sector in ASP preparation through an appropriate forum may be more effective than including them on task forces and working groups. 52 Civil society representation is also very useful. Experience in Senegal suggests that task forces comprised exclusively of government staff tend to be excessively dominated by the vested interests of the ministry of agriculture. Groups with members from outside the government were much more likely to look carefully at government involvement in particular areas and embrace a new approach to stimulating the agricultural sector. For example, the initial Senegal agricultural sector investment program which the agricultural planning unit proposed was unacceptable because it called for a significant increase in spending in contradiction with the government's stated policies to reduce government spending. Therefore, a strategy group comprised of independent representatives was established and given the mandate to develop a new agricultural sector strategy. It is chaired by a retired minister of planning and its members are drawn from research institutes, universities, NGOs, the National Farmers' Union and the National Council of Rural Consultation, an apex organization of regional farmer associations. It is suggested that project designers consider using professional facilitators to conduct the major workshops held during the strategy formulation process. Without professional facilitation, workshop plenary sessions tend to be one-way information-sharing sessions dominated by senior ministry staff. When working groups are organized along departmental lines, the existing hierarchical relations sometimes makes comnmunication difficult, because junior staff are often reluctant to put forward ideas that might differ from those of the department head. One way of addressing this is to ensure that all working groups have some outside members. ASP consultation process. It is critical that program designers work closely with beneficiaries in formulating the program to ensure that it reflects their interests and concerns and is sustainable. The process may take any of several forms: a) Thematic working groups may consult with stakeholders on an ad hoc basis. b) Formal workshops with farmers and other stakeholders can be held. c) Analysts can visit villages and quickly assess demand and concerns using participatory rural appraisal techniques. People preparing ASPs have relied mainly on the first approaches. Donors preparing specific projects sometimes used participatory rural appraisal techniques. Each approach has advantages and disadvantages. Ad hoc consultations are generally issue-oriented and well-linked to the formulation process, but the process may be unsystematic and the people consulted unrepresentative of farmers as a whole. Workshop consultations may be dominated by farmer-elites, leaving poor farmers with little voice. In addition, beneficiaries may see little connection between the workshop consultations and the ultimate decisions made. Participatory rural appraisal techniques ensure that disadvantaged groups such as women and landless express their priorities, and empower communities to analyze their own problems and select their own solutions. But they have the disadvantage of taking more time and money than the other approaches. Implementing agency staff should participate in carrying out participatory rural appraisals 53 because direct contact enhances their accountability to clients. ASPs should use any information developed through this technique wherever possible. Better to start at the grassroots or national level? There have been two approaches to the timing of beneficiary consultation. In the first, consultations begin at the grassroots level, and resulting programs and priorities are presented to governments at the regional and then national levels. In the second, a program is devised at the national level, and later presented to lower level decisionmaking bodies for comment and endorsement. In francophone countries the agricultural sector policy and strategy documents have generally been formulated at the central level and later discussed with local government representatives, district councils and farmers through a series of workshops at different administrative levels. Role of donors. Ideally, ASP preparation is country driven, with donors playing advisory roles on demand. Donors then do not participate in the planning process, but meet at least once a year to exchange views during joint government/donor review meetings. However, donors do sometimes involve themselves in the planning process for various reasons. Small donors may be concerned that a large multilateral development agency, such as the World Bank, involved in the planning process, will influence the program to their disadvantage. Small donors may then be reluctant to collaborate with others in the program, and integrate their projects into the ASP. Stakeholder involvement in program implementation Governments need regular information on implementation progress and program impact. They also need to know which farmers are being reached by government services, and farmers' opinions of the quality, affordability and relevance of services provided. There are several options for generating necessary information: * Beneficiary assessment. Governments systematically survey a large number of clients (at least 1,000) to learn their views about the services provided. The approach was developed by the World Bank. * Participatory rural appraisal. Analysts visit representative villages to talk to farmers about their concerns, proposed solutions to their problems, and development priorities. The studies are carried each year as the basis for preparing a demand-driven annual work plan and budget. * Systematic client consultation. Government holds a series of workshops with different types of stakeholders (private sector, farmers, policy group, and others) on an annual basis. 54 Participatory monitoring and evaluation. Clients (beneficiaries, other stakeholders and negatively affected people) are actively involved in assessing benefits (qualitative and quantitative). In Kenya, a proposal has been prepared to institutionalize beneficiary assessment and systematic client consultation during implementation of the ASP. In Mozambique, program designers are planning to establish a forum in which government can talk with clients. Government is encouraging farmers to organize representative associations. In Lesotho, most districts are using participatory rural appraisal techniques through headman village workshops. In Zambia, rather than working through the existing district development committees on agriculture and rural development, ASP designers established separate district agricultural committees comprised of farmers, representatives of relevant government ministries, NGOs and financial institutions. This has raised questions about possible duplication of efforts. Costs of stakeholder participation Stakeholder participation requires both money and time. While the costs of consultation vary depending on the methods used, typically this would account for no more than 10- 20 percent of total cost of local preparation. Governments are only marginally concerned with the costs of consultation, since they are generally funded by donors. However, stakeholder participation is worth extra time and money because it results in more relevant programs, greater ownership and better implementation. 55 ANNEX 4 ECONOMIC ANALYSIS OF AGRICULTURE SECTOR PROGRAMS Various studies have shown that the quality of economic analysis performed before undertaking a program is positively correlated with subsequent performance during implementation (Belli and Pritchette, 1995). Economic analysis of a sector program provides both government and donors with a coherent framework for evaluating the validity of the program and determines the extent to which it contributes to a country's economic well-being and social welfare. Economic analysis of a sector program differs from that of projects, by looking beyond individual rates of return of stand-alone projects. The entire sector's performance must be assessed, and in the context of the country's overall macroeconomic scheme. This may require the use of a combination of quantitative and qualitative methodologies that draw from both the standard rate-of-return calculations and from other tools of analysis. Thus the analyses may include evaluation of fiscal impact, fungibility and sustainability, private sector counterfactuals, and monitoring of key performance indicators. The analyses may also be complemented with qualitative techniques such as rapid participatory appraisals of beneficiary groups. Information for the analysis will come from a number of sources including public expenditure reviews, economic and sectoral analyses, and other studies or surveys. Public expenditure reviews are an integral part of the preparation process and should be undertaken before the appraisal and financing of the ASP. Public expenditure reviews offer insights into past expenditure patterns and comparisons of budgeted versus actual expenditures. They also investigate the extent to which the previous year's expenditure was directed towards the regions, and whether spending patterns are consistent with long- tern goals. Ideally the public expenditure review would cover expenditures of the whole country. However, macro public expenditure reviews are outside the mandate of sector specialists, so generally analysts will review expenditures of the agriculture sector only. Attention should be paid to ensure that the public expenditure review is comprehensive, capturing all activities within the sector. Thus, private sector involvement and other investments outside of the government have to be included. Economic and sector work, is another essential component of ASP preparation process. ASP fundamental principles require that this work be driven by the country and carried out by nationals with assistance (where necessary) with short-term technical assistance. Recent studies show that the number of economic and sector work studies on a country or sector significantly affect the success rate of conventional projects. Economic and sector work conducted in relation to ASPs must investigate the background of the sector or subsector, identify constraints to growth and provide insights into the direction of future growth. Lessons learned from previous experiences in the sector (country or region) will 56 have to be integrated. As an extension of economic and sector work, a study that looks at the growth potential for the sector (ideally such a study would cover all the sectors and show the linkages between sectors) would be of great value. The assessment of the agricultural growth potential under various scenarios will be helpful in estimating realistic growth rates, and assist in setting priorities for the investment program. Box A4.1 and Box A4.2 provides details on how economic analyses were performed for two sector programs. Box A4.1 Zambia Agriculture Sector Program: Economic analysis The ASP preparation team used an ex ante break-even analysis to determine the minimum growth in net farm income needed to justify the proposed investments under the ASP. Factors that are difficult to measure or quantify factors - rural poverty reduction, better environmental management, food price reduction in the urban areas for nontradables or semitradables - were treated as positive externalities. They also conducted an ex post analysis of the actual rates of return of program subcomponents using monitoring and evaluation data generated during the program's first few years and data from econometric projections for the remaining years of the ASP - an approach that allows evaluation of the impact of the ASP after a lag of only two years. The economic rate of return of the ASP was measured as the change in national aggregate net farm income, after controlling for other factors that may affect net farm income- private investment in agriculture, prices of inputs and outputs, weather, disease outbreaks, policy shifts and other exogenous factors, as well as gender and system of cultivation. Data were obtained from the annual survey of 5,000 households. Because the data collecting agency (the Central Statistical Office) used a different sample every year, it was impossible to do a time-series analysis. A new, appropriately-designed survey, has been adopted. A menu of tools for economic analysis of ASPs Linkage with the macro framework. When evaluating the rationale, feasibility and cost effectiveness of an ASP, it is important to ensure that the program is consistent with and clearly links up with the overall policy and macro framework of the country. Studies show that projects have performed better in countries with low policy distortions (Kauflnann and Wang, 1992 and 1995). In addition, the research shows that as counties moved away from high to low distortion policy frameworks, returns improved significantly. Questions to be addressed include: * How does the ASP fit in with the macro framework? * Do the necessary macro conditions exist? * Why should the public sector be involved in each of the subcomponents? * What are the priority activities? * What are the distortions and impediments, and what policy changes are needed for successful implementation? * Can or should the ASP be used to induce appropriate policy changes? 57 Fiscal impact and cost recovery In many instances, the benefits of government's expenditures are captured by the private sector. Treating public costs and private benefits equally in analyses results in systematic overestimation of traditional economic indicators (economic rate of return, net present value). The dilemma of public costs and private benefits needs to be addressed by placing a premium on public income: the marginal costs of funds should be used to analyze the program's impact on the government's budget. While it is not essential to fully recover the costs of many components in an ASP (for example, research, extension), it is desirable to recover the costs of others (for example, irrigation) from the private beneficiaries. Questions to be asked include: * What components of the ASP have cost recovery mechanisms included? What is the share of cost recovery on investment and on operation and maintenance? * How, and to what extent, are costs recovered from beneficiaries? * How will cost recovery arrangements affect demand for services? * What effect will these measures have on marginalized or poor people? What are the exemptions and safety nets? * What is the fiscal impact of nonrecovered (intentional or otherwise) investment and associated recurrent costs? What measures have been taken to compensate for the shortfalls? Sustainability An ASP includes a number of activities, and it is important that each of them be implemented satisfactorily, from beginning to end. The economic and sector work and public expenditure review may be used to evaluate the institutional and budgetary constraints that may hamper sustainabilty. Issues to address include: * Are the proposed ASP activities affordable? Are they linked to the country's medium-term expenditure plan? * Will they be fully funded throughout the cycle? * How much funding is available from the various sources (government, donors, the private sector)? Will it be enough? Are the proposed components a "wish list?" If so, what are the core activities that should receive priority funding? * Does the necessary institutional capacity exist? If not, what steps need to be taken to rectify the situation? 58 Program impact: Key performance indicators Key performance indicators can provide essential information regarding program implementation and impact. Both planners and stakeholders should be involved in their selection and monitoring. It should be noted that for various reasons - including lags between action and response, misspecified plan of action or external shocks - the key performance indicators may not always elicit the desired information, and may have to be modified accordingly. Monitoring and evaluation of key performance indicators is an ongoing process, and should continuously assess the validity of assumption made in putting together the ASP. Changes in the program can then be made to address deficiencies. Fungibility It is important to assess if components that show high rates of returns may have been funded independently by the private sector or government regardless of donor funding. Thus, funds earmarked for this specific component under a particular program may actually be used to finance some other venture that's a white elephant since financial resources are for the most part fungible. Fungibility can occur at various levels - between sectors and within sectors. Analysis of the public expenditure review, economic and sector work and internal audits will shed light into proposed and actual expenditures. Box A4.2 Ghana Basic Education Sector Investment Program: Economic analysis The objectives of Ghana's basic education sector investment program were to strengthen the quality of teaching and learning, enhance management efficiency, and improve access and participation. The program supported the development of an improved sector and macroeconomic policy framework; capacity building at both central and decentralized levels, through management training and effective use of an improved education management information system, compensatory education initiatives targeted to the poor and to girls; and private sector development by encouraging private sector involvement in financing and running schools, and wider community participation in school activities. The sector program team calculated both private and social rates of return by measuring the mean earnings of subgroups - sorted by gender, poverty status, and locality - by level of education: none, primary and secondary. The analysts found a strong relationship between income and years of schooling. The team also wanted to be sure that the program was sustainable. They made sure that the projected minimum level of government funding was linked to the medium-term expenditure plan, and that the program contains built-in mechanisms for cost-recovery, ensuring that beneficiaries bear some of the costs of education. 59 Counterfactual private sector supply response What would have happened if government had not invested in various components of the ASP? Looking at the individual components, would the private sector have provided all or part of the services? In reviewing the government's sector expenditure program, an overall insight can be gained into where the private sector currently is, and what goods and services it could potentially provide. Sectoral expenditure reviews assist in identifying the areas where the private sector may potentially play a role. Evaluating the nature of the services provided - public versus private - and the willingness of beneficiaries to pay for services, will help in answering the question whether a particular component of ASP should be in the public sector or not. In many countries worldwide, services traditionally provided by government are being moved to the private sector. Traditional economic analysis Economic analyses of conventional projects has involved the calculation of discounted expected present value of the benefits, net of costs. The question is: What is the economic rate of return above the opportunity cost of capital? In evaluating ASPs, this type of analysis should be carried out on individual components that lend themselves to such analysis. Rural roads and irrigation are examples of ASP subcomponents that have clearly identifiable costs and benefits. Questions to be asked include: Are the benefits measurable? Can they be isolated from other influencing factors? Have alternative means of attaining the objectives been identified and evaluated? Are the proper yardsticks being used? Is the opportunity cost of the various factors being considered? Methodologies and issues to consider when conducting economic or cost-benefit analysis include: * Calculation of the net present value (NP V). This is the discounted value of the costs and benefits of an investment over time using an agreed discount rate. * Calculation of the internal economic rate of return (ERR). This is the discount rate at which the present value of costs and benefits are equal, which is then compared to a predetermined cutoff rate of return. * Appropriate use of prices and other economic indicators. Are nominal or real prices being used? Is the opportunity cost of the principal being taken into consideration? Have shadow prices being factored into the calculations? * Sensitivity and risk analysis. This identifies and evaluates the impact of particular factors on a project's or subsector's rate of return (note that all the relevant factors may not be under the control of the implementers). Once the factors that 60 substantially affect the outcome of an initiative have been identified, alternatives have to be identified and contingency or mitigation plans, or both, formulated to offset the risks. It is important to pay attention to realistic estimations of factors that affect costs and benefits of a particular investment (the investment should not be mechanically allocated in a generic manner). Sensitivity analysis identifies factors and events that may potentially eliminate the net benefits of the investment, while risk analysis looks at the likelihood of such events occurring. Beneficiary assessment and impact Participatory rural appraisals provide another tool for analyzing the impact of the program. Participatory rural appraisals are conducted by administering informal questionnaires or asking questions of the intended beneficiaries which can provide information suitable for quick evaluation. Understanding the intended beneficiaries' perceptions and feelings towards products and service delivery is the main objective of this exercise. The information allows ASP designers to modify the program where necessary. 61 ANNEX 5 AGRICULTURE SECTOR PROGRAMS AND AGRICULTURAL TECHNOLOGY Basic principles The development objectives of any ASP will ultimately be achieved through raising productivity and increasing output of the nation's agricultural producers and processors. The technical means by which these gains are to be achieved should be identified and discussed at an early stage of ASP preparation. The technical changes to be encouraged may range from wider use of a single production input (fertilizer or a new crop variety), to multiple adjustments in the way natural resources are managed in whole production systems. All such changes are labeled "technology" in the present context. The sum of the changes which are aimed for is the "technical strategy" for sectoral development. Discussion of technical strategy is likely to be wide-ranging. In almost all countries there are numerous production systems, ranging from those operated by small subsistence farmers, up to large commercial farms or estates. The ASP preparation team will have to weigh the potential contribution of each to diverse sectoral objectives, such as food security, job creation, poverty alleviation or export growth. Throughout this discussion it must be acknowledged that decisions to change production or processing technology will be voluntary, and will be taken individually by large numbers of private decisionmakers. The ASP team's decisions on sectoral policies and investments will be conditioned by the technical strategy the team chooses. These decisions will reflect not only the technology itself, but also the characteristics and circumstances of those who are supposed to adopt it. Key questions Initial discussions will generate only a first approximation of a technical strategy. Some questions that the preparation team may ask include: * Does the assumed technology yet exist? If not, what comparative advantage have national institutions for developing it? * Has the technology been tested and proved appropriate to the agro-ecological zones and categories of users for which it is envisaged? * Is it yet known to the intended adopters? How adequate is the national capacity for transmitting new knowledge? * Is the technology sufficiently attractive, so that the target categories of producers or processors will voluntarily choose to adopt it? 62 Do potential adopters have sufficient access to any additional inputs and any extra cash or other resources which may be needed for adoption? Do adopters have access to markets for additional output? Alternatively (for subsistence producers) does additional output match family needs? The team's initial responses to these questions should focus on whether the envisaged technology falls into the category of private or public goods. Technologies in the private category yield benefits to the provider alone. Examples include a machine or a bag of an agrochemical: the farmer wishing to use them must pay the manufacturer for them, and captures the benefits arising from their use. In general, governments should seek - usually through policy measures - to encourage the private sector to generate, advertise and distribute such technologies, and the beneficiaries to pay for them. By contrast, public goods are by definition nonexcludable: once some farmers learn a new technique, others can copy them without paying the originator of the technology. An example would be improved practices for management of soil moisture or fertility or of a livestock herd. If the government views the adoption of such technology to be in the national interest, it must pay its research and development costs, since the private sector will not do so. Some technologies may have the characteristics of both public and private goods - for example new tillage machinery which simultaneously lowers on-farm crop production costs and brings benefits to resource users outside the farm and to future generations because runoff of rainfall, and consequently soil erosion, is also checked. In these mixed cases a degree of cost-sharing between the public and private sectors may be appropriate. In reviewing strategic options for technological support to the various types of producers, the ASP preparation team should recognize that commercial producers should be less dependent on the public sector for assistance. Commercial producers can finance many of their extension and research needs; extension for estate commodities such as oil palm and sugar can be considered as largely private goods. Government cost sharing with the private sector may be more appropriate for crops such as rubber, coffee, tea or tobacco if grown both by commercial farmers and smallholders. All categories of producers will however require policies which encourage the efficient use of inputs and marketing of outputs. In this context, subsidies will require particularly close attention. No country has yet developed airtight means of channeling subsidies only to small producers, except by direct payments to individuals. An early task for the team will be to review likely technical, social, financial and economic impacts of the technological changes which farmers may be induced to make in response to present or future subsidies or policy reforms. 63 Having decided where, and where not, to commit public resources to the generation, testing and transfer of technology, the government should avoid "crowding out" private operators by simultaneously entering the same technology markets, often with subsidized prices. Government resources should instead focus on providing public goods technologies, for which costs will not generally be recoverable from users. These technologies will typically contribute to overarching sectoral aims, such as long-term sustainability, food security, mitigation of negative environmental impacts, or improvements in the income or living conditions of the rural poor. Typically, also, these technologies involve manipulation of whole production systems and are quite location- specific in their adaptability. They will usually require further testing and adaptation before they can be spread from one location to new agro-ecological zones or categories of users. Only at this stage should the ASP team begin to look at institutional arrangements for technology generation, transfer and uptake. Even if government funding is justified, this does not mean government need implement the program. There is growing evidence that contracting out publicly-funded research and extension, or offering grants for which government organizations and private service-providers can compete on equal terms, can raise the efficiency with which new technology in the public goods category is generated, tested and transferred to users. Agenda for decisionmaking If the assumed technology does not yet exist, the team will need to consider an optimum mix between the following: * Seeking to free-ride on research by neighboring countries or international institutes, through investment in networking, partnerships and staff exchanges; * Establishing a grant scheme for research on the target technologies, for which private and public organizations can compete; * Providing stronger incentives for the private sector to commit its own resources to priority topics; * Strengthening existing public sector research capability. In any case, attention will usually need to be given to improving the way research priorities are set. Priorities should be based on field diagnostic work, and be determined through consultations with a broad group of stakeholders including representatives of the intended users of technologies. The aim of priority setting will be to ensure that only public goods technologies receive public support, and that research targets are fully relevant to the demands or anticipated needs of potential technology users. A judgment will also need to be made on the extent to which basic research should receive public 64 funding, bearing in mind that over the long term new fundamental knowledge is needed to feed a continuing flow of new technology. If new funding mechanisms are to be introduced, eligibility criteria and operative procedures will also have to be worked out, and capacity to administer them put into place. If local adaptation of technology is necessary, the ASP team will need to decide whether the present approach to, and capacity for, technology testing and adaptation is adequate to respond to local variations in the resource base, different circumstances, and constraints and demands of technology users. Resources will often be needed for the following: Training or re-education of researchers and extension agents to help them understand the point of view of potential adopters of technology, and so adjust their local-level priorities, targets, and research and development programs. Creation of capacity for participatory rural diagnostic work. This will help outsiders understand individual production systems and the opportunities and constraints facing potential adopters of new technology. It will also help farmers articulate their demands for new technology. * Financial provision for broad stakeholder - including farmer - participation in diagnostic work, and in planning and implementation of adaptive research programs which respond to its findings. Again, decisions may be necessary on the extent to which the government should implement, as opposed to only fund, location-specific adaptive research. It may be more cost-effective to contract the work to NGOs or producers' associations, even when training or other backstopping measures are needed to increase their capacity. If technology exists but is not yet adequately known to the intended adopters, the ASP preparation team must make decisions on how best, and through whom, to extend existing knowledge. The following will be key questions for discussion. The respective weights to be given to investment in different methods of communication. How much emphasis should be given to written materials versus radio or television broadcasts? Should contacts be through farmers' groups or face-to-face?. The scope for transferring control of extension costs to the intended technology users, or for recovering costs from them. Choices include contracting out extension on behalf of users rather than relying on existing government services, providing farmers with funds to make their own extension contracts with outsiders - whether government or not - or funding the creation or operation of a service of their own. Depending on the beneficiaries' wealth and the proportion of 65 benefits which are private versus public goods, different levels of payment or cost sharing by the technology users will be appropriate. * The scope for experiential learning and enhanced farmer-to-farmer transfer of knowledge. The natural tendency of farmers to experiment and learn from peers may be strengthened, for instance by promoting producer or processor associations or farmer field schools. These approaches provide an excellent means for farmers to understand the basic principles or processes which technological changes seek to exploit. They are particularly useful in adapting technologies to local circumstances and helping whole systems become more productive and sustainable. To assess whether the technology being promoted is sufficiently attractive to adopters is a process which requires interaction between technicians (researchers and extension agents), economists and social scientists on the ASP team. Their contributions need to be founded on detailed knowledge of the agro-ecological zones and producers expected to adopt the technology, usually gained through participatory diagnostic work. * Technicians need to give their best estimates of physical inputs (including labor) and outputs under current practices and under the alternative practices being promoted under the technical strategy. Economists need to convert these figures into monetary values then examine changes in returns to the key factors of production (family labor, land, cash expenditures, water, and others as appropriate). Marginal analyses - incremental return per incremental unit of labor or cash committed - are essential in assessing the potential for adoption of new technologies. * Social scientists need to compare the promoted technologies with others available to the individuals and families concerned. Only when it is shown that returns and risks of technological change compare favorably with alternatives, and that demands on family resources will not jeopardize other priorities, can it be concluded that a technology is likely to be adopted by its intended user. To build capacity to carry out such assessments routinely over the longer term will usually require investments in training of researchers, extension agents and socioeconomic staff. It is the role of the ASP team at the planning stage to identify any policy contradictions or failures in the pricing of inputs or outputs. These should be corrected to make the proposed technical strategy implementable. Answering the two final questions - on access to additional inputs and cash needed for technology adoption, and the adequacy of markets for outputs - should be an integral part of the work of the ASP technology team. Producers themselves, their associations, local administrations, NGOs and the private sector are all likely to have views. The task 66 of the ASP team is to listen to and evaluate this information, deriving from it a list of further prerequisite actions of a nontechnical nature if the technical strategy is to be successfully implemented. These findings will generally then be referred to people handling other aspects of the ASP, such as privatization, credit, rural infrastructure and transport. A multidisciplinary, iterative process It is clear that technology in an ASP cannot be handled by technology specialists alone. To arrive at a realistic and implementable vision of technological change in the agricultural sector, the process must also include inputs from rural sociologists, economists and technology users themselves. The ASP technology team must also have regular contacts with those concerned with overall ASP sectoral policy and government expenditures. With the former it must communicate on the policy issues which it considers actually impede, or could in future accelerate, technological change. With the latter it must regularly check that its proposals for public expenditure in support of technology generation, testing and transfer are consistent with the resources likely to be available. When - as will often happen in practice - the technology team finds its ideas in conflict with policy, likely resource availability or limitations of institutional capacities, it will need to reshape its proposals. The final technical strategy adopted for the ASP will be the result of these successive iterations. 67 ANNEX 6 CONDUCTING A PUBLIC EXPENDITURE REVIEW: TIPS Public expenditure reviews provide data vital to the planning and preparation of an ASP. Often governments have only a vague sense of how much and where funds are being spent in a sector, especially when donors provide a significant proportion of total funding. A public expenditure review allows policymakers, donors and others to understand where money is actually being spent and thereby judge whether spending is in line with sector priorities, or needs to be reallocated to achieve greater efficiencies and equity. The public expenditure review sets limits for total expenditures within a sector by establishing a budget envelope, and gives sector leaders justification to argue for resources from the center. Finally, it provides the opportunity to assess the process by which spending decisions are made. In the context of the ASP, the goal is to be sure actual public expenditures in the agricultural sector match the priorities identified in the sector strategy and the institutional analysis. Role of government A main purpose of public expenditure reviews is to identify activities which the government should be undertaking, those which it should carry out jointly with the private sector or civil society, and those which it should transfer completely to the private sector or civil society. The following provides some examples of different types of activities. Public sector Formulating strategy and policy Translating of strategy and policies into a public expenditure program Issuing regulations to address externalities, common property management, and safety standards Creating and enforcing a legal framework, encompassing property rights, contract agreements and others Preparing for disasters Representing national interests in international trade dealings (not trade promotion) Building capacity so that public service agencies can perform their functions efficiently. Joint Planning and construction of irrigation schemes Extension services for food crops 68 * Research and training * Collecting and disseminating agricultural market data. Private * Operating irrigation schemes * Providing extension services for cash crops * Providing veterinary clinical services * Marketing crops * Providing seeds and fertilizers. * The key questions are: What should the govemrnent be doing in the sector five years from now? How does the country get from where it is now to where it wants to be? Box A6.1 lists appropriate roles for the public sector, which vary according to a country's stage of development and the depth of its private sector. Box A6.1 Public sector functions Suitablefor countries with little Appropriate for countries that Suitablefor middle-income private sector development have undertaken signifi cant countries that have undertaken reforms and have emerging significant reforms and have a private sectors vibrant private sector Legal framnework Legal framework Legal framework Regulation Regulation Regulation Quality control and safety Quality control and safety Quality control and safety Information Information Extension and training Extension and training Research Research Livestock production Livestock health and disease Livestock health and Livestock health and disease control disease control control Irrigation Credit Forestry (afforestation) Forestry (afforestation) Input Supply Marketing Commodity boards Subsidies for various purposes Targeted subsidies for social welfare Rural roads and other Rural roads and other Rural roads and other infrastructure infrastructure infrastructure 69 Carrying out a public expenditure review Carrying out a public expenditure review is often a complicated task. Broadly, the process involves identifying what data are needed and collecting them, classifying expenditures by function, tracking budgeted and actual expenditures over time and observing their relationship, establishing boundaries, and making decisions with regard to budgetary allocations. Collect data The first step is to collect several years' data on budgeted and actual expenditures. The scope of the investigation will vary, depending on the country. It will include at a minimum the activities of the ministry of agriculture, and if possible the activities of the entire agricultural sector, or even the entire rural sector. It will also include all expenditures in the sector regardless of source, including expenditures by the ministry of agriculture, regional governments, local residents (which are often in-kind contributions), and donors. Analysts should be careful to include data on capital versus recurrent expenditures, and subsidies for parastatals. Collecting the data may not be straightforward. Both planning budgets and financial records may be incomplete or scattered across offices. Data that do exist may be unreliable, inconsistent or unusable. Government may not wish to share information with donors or with other stakeholders. Sectoral ministries may not wish to participate in the exercise, or may find it difficult to coordinate their efforts with each other or with the ministry of finance. Classi,fy expenditures The second step is to classify expenditures according to function. Two major categories are: * Capital expenditures * Recurrent expenditures . The third step is to track the actual and budgeted expenditures by function. This shows broadly how resources are being allocated in a sector and the extent to which actual expenditures match budgeted expenditures. Table A6.1 provides an example of what public expenditure in the agriculture sector might look like over time. 70 Table A6.1 National expenditures in the agricultural sector: Example (annual averages, constant 1991 US$) 1985-88 1989-91 Billion US$ Percent Billion US$ Percent Public goods 2.0 13 1.9 21 Crop production .055 0.4 .042 0.4 Animal products .025 0.2 .015 0.2 Extension .090 0.9 .081 0.9 Natural resources .260 2.7 .257 2.7 management Research .570 5.8 .546 5.8 Education .180 1.8 .169 1.8 Infrastructure .820 8.5 .794 8.5 Market interventions 12.0 80 5.4 59 Pricing and 2.27 19 1.02 8.9 marketing Marketing boards 1.31 11 .59 4.2 Credit 8.42 70 3.79 36.2 Targeted programs 0.6 4.0 0.9 10.0 Administration 0.4 3.0 0.9 10.0 TOTAL 15.0 100 9.1 100 Source: Authors' calculations Establish a budget envelope The fourth step is to ascertain the overall budget envelope likely to be available to the sector. The ministry of finance should specify how much funding is likely to be available from domestic resources, and donors should indicate how much fhey are intending to contribute and for what activities. Of course, much of the funding is earmarked for specific budget items and subsector programs. Analysts need to ascertain to what degree the available funding is fungible. Finally the ministry or finance and donors must clarify cost-sharing arrangements, and make clear the minimum that communities and local government must contribute to be eligible to participate. 71 Match the expenditure program with the sector strategy The fifth and final step is to bring expenditures in line with the sector strategy and institutional analysis. In general government should not be funding goods and services that the private sector is, or could, be providing. This usually includes parastatal marketing agencies, agricultural banks and public input delivery systems. The team should look carefully at all subsidies, including those for food consumption, and eliminate or reduce those that are poorly targeted or do not serve an important public purpose. Finally, the team should identify the goods and services - funded by the government - whose delivery can be contracted out to the private sector. The first public expenditure review is the most difficult. In the years following, annual programming reviews can generate the type of information that public expenditure reviews provide, but at a much lower cost. Redflags The team should be alert to information suggesting current or future problems with managing the budget process. These "red flags" include: - Central bank overdrafts. This is a way of disguising parastatal losses and delaying budget provisions to cover them. Inadequate debt provisioning both between parastatals and in credit operations. This can be a sign of lack of accountability and transparency, including the potential for insolvency. * )Off-budget items. Multisector expenditures may be recorded in a sector other than the one under examination. For example, expenditures for water management may be recorded in either the irrigation subsector budget of the ministry of agriculture or the budget of the ministry of natural resources. High suppliers' credit. This may mean deferment of losses by parastatals or cooperatives in trouble. * Poor accounting and auditing in any agency that might generate a claim on public resources. 72 ANNEX 7 CROSS-CUTTING ISSUES There are a number of cross-cutting issues that must be included in the planning of agriculture sector programs. The following is a brief discussion of what to consider to ensure that the program addresses issues of environmental sustainability, poverty reduction, and gender. Environment Agriculture and the environment are firmly intertwined. Most agricultural activities have environmental impacts. In turn environmental degradation may threaten current and future levels of production and constrain sustainable agricultural development, thus defeating efforts to reduce poverty. For example, land preparation methods in most developing countries result in severe soil erosion as well as siltation of rivers and dams, and drying of wells. Such deterioration can be expected to worsen in the future, as population pressure on the land increases. Thus, environmental considerations are an integral part in the design of ASPs. Environment is a cross-sectoral issue, since activities in every sector have environmental impacts. One way to deal with environmental issues, especially in countries that have a ministry with broad responsibilities for environment, or that have adopted the program approach in several sectors, is to have an environment sector program (ESP). ESPs have been initiated in Zambia and Malawi. They are a natural outcome of the national environmental action plans prepared in many countries, being implemented with the support of multiple donors. While a parallel ESP can relieve ASP planners of dealing with many cross-sectoral environmental issues (as these will be handled under the ESP), ASP designers must pay full attention to environmental issues arising from agriculture. These include: * Management of irrigation and water resources. How shall runoff, erosion and lowering of the water table be mitigated? What steps shall be taken to ensure equitable distribution of water resources? How are communities being displaced by the construction of dams and other related irrigation structures being compensated? * Fertilizer and chemical use. What is the potential for groundwater contamination? Is the mutation of resistant pests likely? Are alternative approaches, such as agro- forestry and integrated pest management, likely to result in more sustainable outcomes? What are the best ways of storing, handling, applying and disposing of fertilizers and chemicals? 73 * Impact of agricultural practices. What is the impact of cultivation methods on soil erosion, nutrients and structure? To what extent can rotating crops, leaving fields fallow, applying mulch, and planting trees prevent water and wind erosion. * Land tenure and redistribution. What are the systems of property rights (including, distribution, legal issues, accessibility, encroachment, tenure security)? * Rural roads rehabilitation. What steps are planned to deal with drainage and water run-off and externalities associated with construction? * Competing or conflicting interests. What agriculture and other sectors do may have ramifications for the environment. What are the conflicts among these various players within and between the sectors? * Fisheries. What are plans to prevent depletion of stock due to over-fishing? What measures are being taken to prevent water and coastal area contamination? Poverty reduction One of the major challenges facing many developing countries is how to reduce poverty and hunger, especially in rural areas. Although poverty is a cross-cutting theme, empirical evidence shows that nearly three-quarters of poor people in developing countries live in rural areas and depend on agriculture for their livelihoods. It is imperative that the evaluation of any sector development program in agriculture includes an analysis of its impact on poverty reduction. In carrying out such an evaluation, it is essential to understand the composition and dynamics of the poor. This includes an understanding of the constraints and incentives faced by the poor, the relationship between the poor and better-off members of a community, and the ways in which various income groups are likely to respond to government interventions. During ASP preparation, it is important to review laws and regulations that may adversely affect the poor and take steps to amend them. Areas of investigation may focus on constraints to upward mobility; accessibility to credit for those with no collateral; interest rates and repayment schedules and capability to repay by the poor; and land reform, tenure and ownership rights and their impact on the poor. With respect to safety nets, it is important to guide and support farmers caught in the transitional period of policy reforms until they are able to adjust to the prevailing market conditions. Similarly, the welfare of the chronically poor (due to landlessness, handicaps, illness, age, lack of know-how etc.) has to be safeguarded. When considering transfers and safety nets, the target groups must be clearly identified, and cost effective means to providing the assistance have to be devised. In assessing safety nets it is important to identify where public assistance is essential - where traditional family and social 74 networks will not otherwise pick up the slack. If public transfers are to be used, analysts should discuss the steps to be taken to prevent a dependency syndrome from setting in. Poverty profiles can be used to measure and identify the severity and distribution of poverty. These should include analysis at the regional, class (landless or landed), farm size, political, cultural, sociological, and gender levels. In addition, intra- and inter- household analysis may need to be conducted. Resources allocated to help the poor should target the intended beneficiaries and be equitably distributed. For example, across-the-board subsidies for inputs or services (fertilizer, electricity, water) benefit all recipients, including those who are capable of paying for them at market prices. Thus, targeted subsidies or investments on infrastructure, research or training focused on smallholder farmers may be more prudent; Given the vulnerability of the poor to downside shocks, risk analysis is especially important to predict poor people's behavior, and the likely consequences for the poor to policy changes. Gender issues The importance of paying adequate attention to gender issues is widely recognized. Empirical evidence indicates that farm households headed by women are generally poorer, less educated and carry a disproportionate share of the household work. However, women heads of household are also considered to be reliable borrowers and appear to be more responsive when contacted by female representatives in areas like extension, credit and health. Gender-specific issues to be considered during ASP preparation include the following: * Do women have access to land, financial services, and information? * What are farming patterns, and conflicts with domestic functions? * Are extension messages targeted to women? * Is there a gender differential between cash and subsistence cropping? What are its ramifications? * What are marketing opportunities for men and women? Is there labor mobility? * Have economic reforms widened gender differentials? if so, what compensatory steps are being taken? * How do public sector reforms address women's' issues? * How do user fees affect women? * What are the opportunities for schooling for females? What are the literacy rates for women compared with men? * Do women have equal access to health care as men? What are the health care priorities? * How do land and property rights, employment opportunities, family law, financial regulations, and lobbying capabilities affect women compared with men? 75 BIBLIOGRAPHY Belli, Pedro, and Lant Pritchett. 1995. "Does Good Economic Analysis Lead to Better Projects?" HCO Dissemination Note 53, World Bank, Washington, D.C. Bhatia, Anita, and Nwanze Okidegbe. 1997. "Sector Investment Programs: An Update." Paper prepared for the Sector Performance Analysis Donors Meeting, World Bank, Washington, D.C., May 19. Deininger, Klaus, and Nwanze Okidegbe. 1997. "From Adjustment to Sector Investment Program-Based Approaches to Development Lending." Entwicklung and Laendlicher Raum January: 17-20. Devarajan, Shantayanan, Lyn Squire, and Sethaput Suthiwart-Narueput. 1995. "Reviving Project Appraisal at the World Bank." Policy Research Working Paper 1496. World Bank, Washington, D.C. Engle, Adalbert, and Nwanze Okidegbe. 1997. "Summary Reports on the Sector Investment Program in Agriculture." Paper presented at the International Agriculture Sector Investment Program Workshop, February 18, Lusaka, Zambia. Harrold, Peter, and Associates. 1995. "The Broad Sector Approach to Investment Lending: Sector Investment Programs." Policy Research Working Paper 939. World Bank, Washington, D.C. Kaufmann, Daniel, and Yan Wang. 1992. How Macroeconomics Policies Affect Project Performance in the Social Sectors. World Bank Discussion Paper 302. Washington, D.C. 1995. "Macroeconomics Policies and Project Performance in the Social Sectors: A Model of Human Capital Production and Evidence from LDCs." World Development 23: 751-65. Norman, Akbar. 1996. "Sector Investment Programs in Africa: A Review of Implementation Experience." World Bank, Africa Region, Technical Family, Macroeconomics 2, Washington, D.C. Okidegbe, Nwanze. 1997. Fostering Sustainable Development: The Sector Investment Program. World Bank Discussion Paper 363. Washington, D.C. Oxford Policy Management. 1996. "Sector Investment Programs in Africa: Issues and Experience." Paper presented at the Sector Performance Analysis Donors Meeting, World Bank, December 16, Washington, D.C. 76 Pinto, Rogerio, and Angelous Mrope. 1995. Assessing Sector Institutions. World Bank Discussion Paper 297. Washington, D.C. Pradhan, Sanjay. 1996. Evaluating Public Spending. A Framework for Public Expenditure Reviews. World Bank Discussion Paper 323. Washington, D.C. Squire, Lyn, and Herman G. Van der Tak. 1975. Economic Analysis of Projects. Baltimore: Johns Hopkins University Press. 77 Distributors of COLOMBIA GERMANY Tel: (52 5) 624-2800 POLAND Fax: (941) 432104 Infoenlace Ltda. UNO-Vedag ISRAEL Fax: (525)624-2822 Intemational Publishing Service E-niail: LHL@snilanka.net W orld Bank Carrera6 No. 51-21 PoppelsdoderrAlleess YozmotLderature Ltd. E-mail: infotec@dn.net.mx Ul. Piekna31/37 Apadado Aereo 34270 53115 Bonn PO. Box 56055 00677 Warzawa SWEDEN Publications Santale de Bogota, D.C. Tel: (49 228) 949020 3 Yohanan Hasandlar Street Mundi-Prensa Mexico S.A. de C.V Tel: (482) 628-6089 Wennergren-Williams AB Prices and credit ternms vardrfron1 Tel: (571) 285-2798 Fax: (49 228) 217492 Tel Aviv 61560 cuRio Panuco, 141 -Colonia Cuauhtemoc Fax: (48 2) 621-7255 P 0. Box 1305 cosrtry to curtrsj. Consult yrur Fax: (571) 285-2798 E-mail: unovedag@aol.com Tel: (972 3) 5285-397 06500 Mexico, DGF E-mail: books%ips@ikp.atm.com.pi S-171 25 Solna yoa dtt(rtor Cl)e,rrpacrgatCTDORGHN Foe: (97223(5285-397 Tel (52 5) 533-5658 Tel: (46 8( 705-97-50 local distribut(9rbefore placit"gl COTE DIVOIRE GHANA Fax:(9723)Fan: (525) 514-6799 PORTUGAL Fax:we (468)27-00-71 order. Center dEdition et de Difusion Atricaines Epp Books Services R.O.Y lntemational Livrana Podugal E-mail: mail@wwi.se (CEDA) PO. Box 44 PO Box 13056 NEPAL Apadado 2601, Rua Do Canmo 70-74 ARGENTINA 04 BP541 TUC Tel Aviv 61130 Everest Media lntemationalSnrvices(P.)Ld. 1200 Lisbon SWITZERLAND Olcirna del Libro Intemacional Abidjan 04 Accra Tel: (972 3) 5461423 GPO Box 5443 Tel: (1) 347-4982 Librairie Payot Serice Irrsteutionnel Av Cordoba 1877 Tel: (225)24 6510;24 6511 Fax: (9723)5461442 Kathmandu Fax:(1)3470264 (2Les-deMonibenon30 1120 Buenos Aires Fax: (225) 25 0567 GREECE E-mail: royil@netvision.net.il Tel: (9771) 472 152 1002 Lausanne Tel: (54 1) 815-8354 Papasotiriou S.A. Fax: (9771) 224 431 ROMANIA Tel: (4121) 341-3229 Fax: (54 1)815-8156 CYPRUS 35, Sloumara SIr. Palestinin AuthorxtyfMiddle East Compani De Librarii Bucuresti S.A. Faxo (41 21)341-3235 E-mail: olilibro@sadlink.com CenlertorAppliedResearch 106 82 Athens Index Inlomration Services NETHERLANDS Str Lipscani no. 26, sector 3 - Cenrfors Allee Tel: (30 1)364-1826 POB.19502 Jerusalem DeLindeboonvinOr-Publikaties Bucharest ADECOVanDiermenEd4ionsTechniques AUSTRALIA, FIJI, PAPUANEWGUINEA 6CDiogenesStreet Engomi Fax: (30 1)364-8254 Tel: (9722)6271219 PO. Box 202, 7480 AE Haaksbergen Tel: (40 1)613 9645 Ch. de Lacuez 41 SOLOMON ISLANDS, VANUATU, AND PO. Box 2006 Fax: (972 2)6271634 Tel: (31 53) 574-0004 Faa:(40t13124000 CH1807 Blonay SAMOA~ ~ ~~~~~~~~Nioi HAITI Fax: (31 53) 572-9296 TelANFDEAIO a: (41 21) 943 2673 D.A. Inlormalion Senvices Ns ( Culture Diffusion ITALY E-mail: lindeboo@woddonline.nl RUSSIANFEDERATION Fax: (412j9433605 c48 Whneho Road FaxTel: (357 2) 44-1730 5 Rue Capois Licosa Commissionaria Sansoni SPA 9sdaleKlsto