Ghana - The National Agricultural Research Project, the National Agricultural Extension Project, the National Livestock Services Project, the Agricultural Diversification Project, and the Agricultural Sector Investment Project (sector overview)

Document of The World Bank FOR OFFICIAL USE ONLY Report No.: 22439 PERFORMANCE AUDIT REPORTS AND SECTORAL OVERVIEW GHANA NATIONAL AGRICULTURAL RESEARCH PROJECT (CREDIT 2247 - GH) NATIONAL AGRICULTURAL EXTENSION PROJECT (CREDIT 2346 - GH) NATIONAL LIVESTOCK SERVICES PROJECT (CREDIT 2441 - GH) AGRICULTURAL DIVERSIFICATION PROJECT (CREDIT 2180 - GH) AGRICULTURAL SECTOR INVESTMENT PROJECT (CREDIT 2555 - GH) June 22, 2001 Sector and Thematic Studies Group Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Currency Equivalents Currency Unit = Cedi Annual Average Exchange Rate (Cedi to US$): 1990 326 1994 957 1998 2314 1991 368 1995 1200 1999 2647 1992 437 1996 1637 2000 (mid-October) 5800 1993 649 1997 2050 Abbreviations and Acronyms AAGDS Accelerated Agricultural Growth and Development GNP Gross National Product Strategy GOG Government of Ghana ADB Agricultural Development Bank GOPDC Ghana Oil Palm Development corporation ADP Agricultural Diversification Project GREL Ghana Rubber Estates Limited AfDB African Development Bank GSS Ghana Statistical Service AgSSIP The Bank's Agricultural Services Sub-sector GLSS Ghana Living Standards Measurement Survey Investment Program ICO International Coffee Organization AGSSIP The Government's Agricultural Services Sub-sector ICR Implementation Completion Report Investment Program ICRISAT International Crop Research Institute for Semi-Arid APCC Agricultural Policy Coordinating Committee Tropics APL Adaptable Program Loan IDA International Development Association ARPCC Agricultural Research Policy Coordinating IFAD International Fund for Agricultural Development Committee ITSPM Intermediate Technology Small Scale Oil Palm Mill ASIP Agricultural Sector Investment project M&E Monitoring and Evaluation ASPR Agricultural Sector Performance Review MOFA Ministry of Food and Agriculture BOPP Benso Oil Palm Plantation MTADS Medium-Term Agricultural Development Strategy CDF Comprehensive Development Framework MTR Mid-term Review CG Consultative Group NAEP National Agricultural Extension Project CGIAR Consultative Group for International Agricultural NARP National Research Agricultural Research Project Research NAR National Agricultural Research Station CIDA Canadian International Development Agency NDPC National Development Planning Commission CLW Community Livestock Worker NGO Nongovernmental organization COCOBOD Ghana Cocoa Board NLSP National Livestock Services Project CPM Computerized Project Management OED Operations Evaluation Department CRIG Cocoa Research Institute of Ghana PAD Project Appraisal Document CRI Crops Research Institute PAR Performance Audit Report CSIR Council for Scientific and Industrial Research PCC Program Coordination Committee CWIQ Core Welfare Indicators Questionnaires PIC Project Implementation Committee DANIDA Danish International Development Agency PPMED Policy, Planning, Monitoring and Evaluation DAs District Assemblies Department DFID Department for International Development PU Project Unit DRC Domestic Resource Cost RELC Research Extension Liaison Committee EIU Economist Intelligence Unit RELC Research-Extension Linkage Committee ERP Economic Reform Program SARI Savannah Agricultural Research Station ERR Economic Rate of Return SMS Subject Matter Specialist EU European Union T&V Training & Visit FOB Free on Board TSA Technical Support Agency GAINS Ghana Agricultural Research Information System UNDP United Nations Development Program GDP Gross Domestic Product VIP Village Infrastructure Project GEPC Ghana Export Promotion Council WDR World Development Report Fiscal Year January 1 to December 31 Director-General, Operations Evaluation Mr. Robert Picciotto Director, Operations Evaluation Department Mr. Gregory Ingram Manager, Sector and Thematic Evaluation Mr. Alain Barbu Task Managers Mr. Ridley Nelson and Ms. Nalini Kumar  The World Bank Washington, D.C. 20433 U.S.A. Office of the Director-General Operations Evaluation June 22, 2001 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Reports on Ghanta National Agricultural Research Project (Cr. 2247-GI) Ghana National Agricultural Extension Project (Cr. 2346-GH) Ghana National Livestock Services Project (Cr. 2441-GH) Ghana Agricultural Diversification Project (Cr. 2180-GH) and Ghana Agricultural Sector Investment Project (Cr. 2555-GH) and Sector Overview Attached are Performance Audit Reports for five Ghana rural projects approved between 1990 and 1993 together with a sector overview drawing mainly on the findings of those audits. The five audits are for the following projects: (i) the National Agricultural Research Project (NARP) (Cr. 2247-GH) for which a credit in the amount of SDR 16.3 million (US$22.0 million equivalent) was approved in May, 1991. The project closed in December 1999, more than two years behind schedule. The total disbursement was SDR 16.1 million (US$22.6 million equivalent). (ii) The National Agricultural Extension Project (NAEP) (Cr. 2346-GH) for which a credit in the amount of SDR 21.8 million (US$ 30.4 million equivalent) was approved in March, 1992. The project closed in November 1999, two years behind schedule. The final disbursement was SDR 20.3 million (US$ 28.76 million equivalent). (iii) The National Livestock Services Project (NLSP) (Cr. 2441-GH) for which a credit in the amount of SDR 15.3 million (US$ 22.45 million equivalent) was approved in December, 1992. The project closed in December 1999, one year behind schedule. The final disbursement was SDR 14.88 million (US$ 21.29 million equivalent). (iv) The Agricultural Diversification Project (ADP) (Cr. 2180-GH) for which a credit in the amount of SDR 12.5 million (US$ 16.5 million equivalent) was approved in October, 1990. The project closed in December 1999, one year behind schedule. The final disbursement was SDR 10.2 million (US$ 14.27 million equivalent). (v) The Agricultural Sector Investment Project (ASIP) Cr. 2555-GH) for which a credit in the amount of SDR 15.3 million (US$21.5 million equivalent) was approved in December,1993. The project closed in November 1999. The final disbursement was SDR 15.12 million (US$ 20.93 million equivalent). The Implementation Completion Reports (ICRs) for all five projects are incomplete: they omit some of the basic data. While each individual project had its own specific objectives, the common goal was to accelerate agricultural growth and increase the incomes of smallholders through improvements in productivity and diversification, supported by the improvement of small rural infrastructure. Technology, institutional improvements, increased participation, and training were important means to achieve the objectives. The NARP supported the improvement of the planning, management, budgetary procedures, research- extension linkages and human resources in the research system and invested in selected research programs. The NAEP strengthened the extension services in all 10 regions and in the central ministry, in particular establishing Research Extension Liaison Committees to link with research. The NLSP improved the capacity of the Open Nucleus Breeding Centers, rehabilitated veterinary laboratories, supported the privatization of veterinary services, and supported vaccination and work on tsetse control. The ADP supported diversification into higher value tree and horticulture crops, especially pineapple, This document has a restricted distribution and may be used by recipients only in the performance of their official duties. its contents may not otherwise be disclosed without World Bank authorization.  2 through policy reforms to promote liberalization and privatization, through rehabilitation or planting of coffee, rubber, and oil palm, through feeder roads and through technical support for export crops. The ASIP supported small rural infrastructure including investments in small-scale water schemes, marketing infrastructure, feeder roads, storage and food processing and studies and technical assistance. The CDF approach has initiated a process of donor coordination in the sector, particularly with respect to addressing poverty concerns, and the Bank has played a significant role. However, there is still a long way to go and at this stage it is questionable whether coordination of actual programs has really changed much. So far donors have yet to buy fully into the new Bank project (AgSSIP), approved by the Board in August 2000 but not yet effective, and there remain differences of view and emphasis on the design of support for some sub-sectors. The PAR ratings for the five projects are as follows: National Agricultural Research Project (Cr. 2247). The audit rates outcome as moderately satisfactory versus the ICR rating of satisfactory. The project is not rated fully satisfactory because there were a number of weaknesses including the implementation delays, the lack of transparent lines of management and accountability, and the problems created by delayed and inadequate release of funds which raise questions about the scale of the program. The PAR Institutional Development Impact, Bank and Borrower Performance ratings are the same as the ICR. The PAR however, downgrades sustainability from likely to unlikely. This is common to four of the projects that were audited in this package. The severe shortfall in counterpart funding, followed by the limited availability of funds in the post project period suggests that the borrower was either unwilling to or unable to support the project scale designed in all four cases. National Agricultural Extension Project (Cr. 2346). The audit rates outcome as moderately satisfactory versus the ICR rating of satisfactory. The project is expected to achieve most of its major relevant objectives, but with significant shortcomings. As in the case of the research project there is no change in the Institutional Development Impact, Bank and Borrower Performance Ratings. However, sustainability is again downgraded to unlikely. National Livestock Services Project (Cr. 2441). The audit rated outcome as moderately satisfactory versus the ICR rating of "partial." The project is expected to achieve most of its major relevant objectives, but with significant shortcomings. Institutional Development Impact is rated substantial versus the ICR rating of "partial." There is no change in the Bank and Borrower Performance ratings. As in the case of the other two projects, sustainability is downgraded to unlikely. Agricultural Diversification Project (Cr. 2180). The PAR Outcome, Bank and Borrower Performance ratings are the same as in the ICR. The PAR rates Institutional Development Impact as substantial versus the ICR rating of "partial." Sustainability is rated as non-evaluable versus the ICR rating of likely. This is because some evidence supports a likely rating and other evidence supports an unlikely rating and there is insufficient information to make a conclusive judgment. Unlike the other four projects, a continued flow of resources from the government is not required under ADP to sustain activities begun under the project. Most of the outgrowers schemes in rubber and oil palm and the pineapple growers can be viable, but for the current economic crisis. Agricultural Sector Investment Project (Cr. 2555). The PAR rates the outcome as unsatisfactory versus the ICR rating of satisfactory. By closing, though the project had created a number of physical structures, the quality of construction was variable and their usefulness, limited. Several structures were poorly designed and were constructed in locations where they could not be effectively utilized. Few of them were actually constructed in response to identified community needs. A poor and unrealistic project design that over-estimated the capacity of grass-roots institutions, led to an inefficient use of resources. It resulted in a target driven implementation schedule, and did not allow for development of capacity among the communities at their own pace to manage, implement and maintain their investments. The audit also downgrades both Bank and Borrower Performance to unsatisfactory. Both need to share the blane for the  3 poor and unrealistic design of ASIP. Bank supervision missions, did not stress the need to build capacity in the community, nor did the Borrower bring the issue to the attention of the Bank. The Bank should have put greater emphasis on providing training to beneficiaries. The pressure to lend created an unrealistic disbursement schedule. In common with the other three projects, sustainability is downgraded to unlikely. Institutional Development Impact is rated negligible versus the ICR rating of "unsatisfactory." The audits of the five projects provided the building blocks for the sector overview. The overview finds that growth in the agriculture sector has accelerated from about negative 1 percent in the 1970s, through positive 1 percent in the 1980s, to 2.7 percent in the 1992-1997 period (matching population growth rate), then accelerating to 4 percent from 1996 to 1999 - overall quite an impressive improvement. The government's ambitious goal is now to raise this growth rate to 6 percent and to make more progress on poverty alleviation. But there are a number of economy-wide constraints. A quite unstable macroeconomic situation with a high inflation rate is crowding out the private sector. Decentralization of most services to the District Assemblies, including agricultural extension and small infrastructure, while a sound move for the longer term, is stalled halfway - with roles but not budget transferred - and is presenting challenges including capacity and financial management problems. The main rural sector constraints to achieving the growth and poverty targets include: financial constraints in key areas reflected in the widespread shortage of counterpart funding for projects; insufficient attention to the poorer northern areas where, in particular, the cotton sub-sector needs institutional reform; insufficient acknowledgement in the strategy that expansion of cultivated area will continue for many years and needs to be managed; a land tenure system arguably unsuited to sustainable land management under more intensive conditions; lack of rural finance services; and, lack of a rural sector policy coordination mechanism. The projects themselves offer six main lessons. First, borrower capacity to provide counterpart funding should be projected and realistically assessed, and program design should be adjusted to match the findings of that assessment. Second, there are opportunities for relatively minor adjustments in project strategy and mechanisms which can enhance poverty focus without risk to growth. Third, publicly funded agricultural research needs to focus not simply on yield maximization but on financial profitability, returns to labor, and the low input technology needs of the poorer farmers. Fourth, to improve the efficiency of extension services a phased strategy is needed to increase the role of cost recovery, private sector, community groups and NGOs. Fifth, in supporting community infrastructure investments, the importance of the social change element should not be underestimated, approaches that specifically target the poor are important if community driven programs are to reach the poorest who are generally the least organized. Sixth, the short-term impact of public sector services reform may be negative on the poor. Likely impacts need to be forecast, and, to the extent feasible, addressed through some form of short-term safety net arrangements. The findings of this sector overview point to several recommendations for the Bank's future involvement in Ghana's rural sector: * Once the immediate macroeconomic crisis is resolved, the challenge will be to further increase sectoral growth, to improve public sector efficiency, to enhance poverty focus, and to broaden strategy coordination cross-sectorally toward a rural rather than agricultural focus. * The Agricultural Policy Coordinating Committee should be reactivated, but as a rural coordinating committee rather than an agricultural coordinating committee, to broaden the attention to rural spatial issues, strategy analysis, including cross-sectoral linkages and poverty alleviation. * The Agricultural Services Subsector Investment Project (AgSSIP), awaiting effectiveness, should be re-examined with respect to realism of the implementation schedule, scale of local funding demand, enhancement of the pro poor focus, and better addressing the input supply constraints.  4 * There are opportunities for enhancement of the pro-poor focus of rural interventions in a number of areas including research, the District Assembly's community organization skills, support for Farmer-Based Organizations, livestock breeding, and the institutional reforn of the cotton subsector in the north. * The Bank should revisit, and agree with the leading sectoral donors, on a consultation process mechanism, procedures, and leadership arrangement acceptable to all. Attachment  Contents Principal Ratings ............................... iii Key Staff Responsible .iv Preface.v PART I .1 Introduction .1 Section I Background to the Rural Sector .1 Resources, Comparative Advantage, and Early Government Sector Policy .I Section 11 The New Agricultural Strategy and the Macro and Institutional Environment in the Country .4 Section III Direction of Past Government Expenditure and Audit Findings of Broad Sectoral Relevance .7 Section IV The Future - The Borrower's AGSSIP and Proposed Bank Support 14 The Borrower's AGSSIP Compared With The Bank's AgSSIP .14 The Bank's AgSSIP .15 Status of Donor Coordination .16 Section V Summary Findings and Recommendations .18 Summary Findings .18 Summary Recommendations .1 9 Part II.21 National Agricultural Research Project (NARP) .22 National Agricultural Extension Project (NAEP) .28 National Livestock Services Project (NLSP) .35 Agricultural Diversification Project (ADP) .42 Agricultural Sector Investment Project (ASIP) .48 Basic Data Sheets .55 This report was prepared by Ridley Nelson and Nalini Kumar who audited the projects in October 2000. William B. Hurlbut edited the report. Helen Phillip and Maisha Hyman provided administrative support.  ii  111 Principal Ratings National Agricultural Research Project ICR PAR Outcome Satisfactory Moderately Satisfactory Sustainability Likely Unlikely Institutional Development Substantial Substantial Borrower Performance Satisfactory Satisfactory Bank Performance Satisfactory Satisfactory National Agricultural Extension Project ICR PAR Outcome Satisfactory Moderately Satisfactory Sustainability Likely Unlikely Institutional Development Substantial Substantial Borrower Performance Satisfactory Satisfactory Bank Performance Satisfactory Satisfactory National Livestock Services Project ICR PAR Outcome 'Partial" Moderately Satisfactory Sustainability Likely Unlikely Institutional Development 'Partial" Substantial Borrower Performance Satisfactory Satisfactory Bank Performance Satisfactory Satisfactory Agricultural Diversification Project ICR PAR Outcome Satisfactory Satisfactory Sustainability Likely Non evaluable Institutional Development 'Partial" Substantial Borrower Performance Satisfactory Satisfactory Bank Performance Satisfactory Satisfactory Agricultural Sector Investment Project ICR PAR Outcome Satisfactory Unsatisfactory Sustainability Likely Unlikely Institutional Development Unsatisfactory Negligible Borrower Performance Satisfactory Unsatisfactory Bank Performance Satisfactory Unsatisfactory  iv Key Staff Responsible Department Director Division Chief Task Manager National Agricultural Research Project (Credit 2247) Appraisal E. Lim John Joyce Ian Hill Midterm Olivier Lafourcade Gotz Schreiber Solomon Berkure Completion Peter Harrold Jean-Paul Chausse Solomon Berkure National Agricultural Extension Project (Credit 2346) Appraisal E. Lim John Joyce Solomon Berkure Midterm Olivier Lafourcade James Wright Solomon Berkure Completion Peter Harrold Jean-Paul Chausse Eustacius Betubiza National Livestock Services Project (Credit 2441) Appraisal E. Lim John Joyce S. Berkure Midterm Serge Michailof Jean Paul Chausse S. Berkure Completion Calisto Madavo Jean Paul Chausse Eustacius Betubiza Agricultural Diversification Project (Credit 2180) Appraisal Caio K. Koch-Weser Anand K. Seth Surjit Singh Midterm Olivier Lafourcade Gotz Schrieber Surit Singh Completion Peter C. Harrold Jean Paul Chausse Solomon Berkure Agricultural Sector Investment Project (Credit 2555) Appraisal E. Lim John Joyce P. Werbrouck Midterm Caio K. Koch-Weser Anthony Pellegrini P. Werbrouck Completion Peter Harrold Jean-Paul Chausse Eustacius Betubiza  v Preface This report consists of five project performance audits complemented by a sectoral overview. The audit elements are as follows: * Performance Audit Report (PAR) for the National Agricultural Research Project (NARP) (Cr. 2247-GH), for which a credit in the amount of SDR 16.3 million ($22.0 million equivalent) was approved in May 1991. The project closed in December 1999, more than two years behind schedule. The final total disbursed was SDR 16.1 million ($22.6 million equivalent). An Implementation Completion Report (ICR) was submitted on 06/30/2000. * Performance Audit Report (PAR) for the National Agricultural Extension Project (NAEP) (Cr. 2346-GH), for which a credit in the amount of SDR 21.8 million ($30.4 million equivalent) was approved in March 1992. The project closed in November 1999, two years behind schedule. The final total disbursed was SDR 20.3 million ($28.76 million equivalent). An Implementation Completion Report (ICR) was submitted on 06/30/2000. * Performance Audit Report (PAR) for the National Livestock Services Project (NLSP) (Cr.244 1-H), for which a credit in the amount of SDR 15.3 rnillion ($22.45 million equivalent) was approved in December 1992. The project closed in December 1999, one year behind schedule. The final total disbursed was SDR 14.88 million ($21.29 million equivalent). An Implementation Completion Report (ICR) was submitted on 06/30/2000. - Performance Audit Report (PAR) for the Agricultural Diversilication Project (ADP) (Cr.2180-GH), for which a credit in the amount of SDR 12.5 million ($16.5 million equivalent) was approved in October 1990. The project closed in December 1999, one year behind schedule. The final total disbursed was SDR 10.2 million ($14.27 million equivalent). An Implementation Completion Report (ICR) was submitted on 06/30/2000. * Performance Audit Report (PAR) for the Agricultural Sector Investment Project (ASIP) Cr.2555-GH), for which a credit in the amount of SDR 15.3 million ($21.5 million equivalent) was approved in December 1993. The project closed in November 1999. The final total disbursed was SDR 15.12 million ($20.93 million equivalent). An Implementation Completion Report (ICR) was submitted on 06/30/2000. The report was prepared by the Operations Evaluation Department (OED) based on the Implementation Completion Reports, the Staff Appraisal Reports, the Development Credit Agreements, and review of Bank files. An OED mission traveled to Ghana in September-October 2000 where it discussed the projects with Bank staff, relevant government officials, nongovernmental organizations, beneficiaries, donors, and consultants. The cooperation and assistance of all stakeholders and government officials is gratefully acknowledged, as is the support of the staff of the World Bank Country Office in Accra. This report set out to answer a number of questions of which the following were the most important: * How sustainable were these projects given the apparently persistent counterpart funding problem? * How did these projects interact with the decentralization process in Ghana? * How did they handle the participatory processes?  vi * What has been the poverty impact? How could it be increased? * More broadly for the sector as a whole, what have been the main emerging issues from the project experience, and what are the implications for the future of the rural sector and the appraised, but not yet effective, AgSSIP. The ICRs for these projects were somewhat incomplete, partly as a result of a commendable attempt to do a sectoral overview which, in the event, short-changed the project ICRs themselves and omitted some of the basic data. The borrower's completion reports were more useful and more complete, although not fully covering the required elements of the Bank ICR. Following standard OED procedures, the draft report was sent to the borrower for comments before being finalized. No comments were received.  PART I Sectoral Overview Introduction 1.1 The purpose of the sectoral overview section of the report is to expand on the sectoral background relevant to the five audits, to explore the issue of poverty relevance and impact, and to translate the main individual lessons of experience from the five audits into broader sectoral lessons, focusing particularly on those which have implications for the design of the future program. The sectoral overview does not represent a comprehensive sectoral study-for example, it omits important sub-sectors such as cocoa and forestry, it does not include a public expenditure review, and it has not addressed rural finance or land policy issues. However, with five audits available, it does cover a significant portion of recent sectoral activities and the findings have substantial relevance to the future lending program, particularly since the proposed future lending is largely an extension of these projects. The overview builds on the Region's sectoral overview. In particular, it offers a poverty-oriented view with an element of farm systems perspective, it focuses on decentralization, it addresses donor perspectives, and it offers findings relevant to forthcoming projects. Section I of the Sectoral Overview provides a background to the rural sector in Ghana including a rural poverty profile (expanding on this in Annex A). Section II outlines the government's new agricultural strategy and briefly examines whether the current conditions (macro and institutional environment') are right for the strategy to deliver. Section III examines the direction of government agricultural expenditure and summarizes the main generic findings of the five audits relevant to the sector as a whole -leaving the audits themselves to cover the main project-specific lessons. Section IV discusses the proposed Bank support through the Bank's Agricultural Services Sub-sector Investment Project (AgSSIP) - shortly to be effective - and the status of donor coordination in that context. Section V concludes with summary findings and recommendations. Section I. Background to the Rural Sector Resources, Comparative Advantage, and Early Government Sector Policy 1.2 Agriculture Sector. Agriculture, the dominant sector of the Ghanaian economy, employs 70 percent of the labor force, contributes about 40 percent to GDP, and accounts for more than 50 percent of foreign exchange earnings. The sector is made up of five sub sectors: crops other than cocoa, (61 percent of agricultural GDP), cocoa (14 percent), livestock (7 percent), fisheries (5 percent) and forestry (1 1 percent). Ghanaian agriculture is dominated by family-operated farms that produce about 80 percent of total agricultural production? Ghana has 5.3 million hectares of land under cultivation. This represents 22 percent of the total land area and 39 percent of the total agriculturally useable land area. Less than 1 percent of this area is irrigated. Fertilizer use, at about 4 kilograms per hectare, is one of the lowest in Africa. 1. expanding on this in Annex B. 2. MOFA estimated in 1997 that that the country's agriculture was operating at just about 20 percent of its potential. Draft Accelerated Agricultural Growth and Development Strategy in Support of Ghana Vision 2020.  2 1.3 Agricultural Growth. Despite its importance, agricultural growth has historically lagged behind other sectors. Average annual growth of agricultural GDP rose from about -1 percent in the 1970s, through 1 percent in the 1980s to 2.7 percent in the 1990-97 period (matching population growth), then accelerating to 4 percent from 1996 to 1999. The Government of Ghana's (GOG) objective is to now raise this growth rate to 6 percent - a substantial challenge. Agriculture's contribution to GDP is expected to fall to about 25 percent by the year 2020. The Food Production Index for 1996-98 on a 1989-91 base is impressive at 144. In fact, it is one of the highest in the world over that period. However, this is more a reflection of the poor production levels in the base period than high production levels in the closing period. Ghana has a Gini Coefficient of about 33 percent. About 35 percent of the rural population is below the national poverty line. Over the 1992-97 period, Total Factor Productivity in Ghana's agriculture was estimated at 3 percent, relatively high for Africa. There is therefore still substantial scope to exploit agriculture as a significant factor in broad-based growth. 1.4 Where Does Ghana's Comparative Advantage Lie? I Domestic Resource Cost (DRC) calculations are available only for certain commodities.4 The commodities covered by the analysis exhibit an order of comparative advantage from best to worst (depending on zone, technology, etc.) approximately as follows: yam, plantain, coffee, cotton, maize, cassava, and finally rice. Both cassava and rice have DRCs that are mostly above 1, indicating that the economy is spending more than it gets back with respect to foreign exchange when it produces these commodities. This is especially true for rice. At the other end of the scale, yams, plantain, coffee, and cotton show very competitive DRCs well below 0.5. 1.5 Regional Differences in Production Potential. With a wide range of agro-ecological zones in Ghana, regional differentiation in strategy is important. As might be expected from the land/labor ratios, soils, and climate, the northern areas of Ghana show high levels of cropped area expansion whereas the southern areas show high levels of yield increase. Soil suitability, land availability, and rainfall suggest that the Brong-Ahafo Region and the northern regions offer potential for area expansion in annual crops including cereals, legumes, cotton, and root crops and, indeed, this is where area expansion is taking place. These are also the regions of greatest poverty which suggests that, for poverty alleviation, area expansion will still be an important factor. Main constraints in these northern areas are infrastructure, problems with cotton production and marketing, limited alternative sources of income, and limited technologies for these low rainfall areas. Rural population density, cropping intensity, and rural road density for Brong-Ahafo, Northern, and Upper West regions are well below national averages. Elsewhere, the Westem Region has substantial potential for tree crop area expansion, but the location of infrastructure would need to be carefully planned to ensure protection of the high moist tropical 3. The mission did not address the cocoa subsector since none of the projects audited had a significant direct impact on cocoa, and since it would require a whole study on its own to do ajustice. The earlier Bank funded Cocoa Rehabilitation Project largely achieved its objectives. The policy reform, institutional development, and investments, and the resulting improved farm gate prices, have lifted exports from less than 200,000 tons in 1987 to over 400,000 tons in 1999. However, there remains signifieant unfinished business. The Government is moving cautiously on the privatization front given the initial negative experience of total liberalization in Nigeria, Cameroon and Cote d'Ivoire. 4. The DRC compares the opportunity costs of domestic production to the value added it generates - all calculated at border prices. A DRC of less than I indicates efficiency and international competitiveness. If a DRC is greater than I the economy is incurring costs in excess on what it gains or saves from the production in terms of net foreign exchange. The efficiency of domestically producing a non-tradable good can be estimated by using the price of a tradable substitute as a benchmark. A ranking of DRCs is indicative of relative levels of efficiency of domestic production or of intemational competitiveness.  3 rainforest and steep slopes.5 There is a dilemma with respect to investing in lower poverty, higher potential, higher rainfall areas in the south versus investing in higher poverty, lower potential, lower rainfall areas in the north. While clearly a balance needs to be maintained, given the potential for cotton and the poverty needs of the north, and, the still relatively low per capita investments in that area, the north appears still to be relatively neglected. 1.6 Early Government Sector Policy. In 1983, following a long period of failed interventionist policies, the Economic Reform Program (ERP) was initiated. This aimed at trade liberalization, privatization of public entities, improved collection of revenue, reform of the civil service, tighter budgetary control, and devaluation. Before the ERP, policy constraints in the agricultural sector included grain purchasing and storage restrictions, government marketing control, and low fixed prices in a number of commodities including cocoa, cotton, coffee, and palm oil. The government also controlled input supply. The result of these policy distortions was a staggering decline in the per capita food production index from 100 in 1974-76 down to 62 by 1983. With the ERP, agriculture GNP growth picked up, but significant price and marketing controls remained. To respond to this situation, with support from the World Bank, the 1990 Medium-Term Agricultural Development Strategy (MTADS) sought to put the private sector at the heart of rural development. The strategy prescribed liberalization of fertilizer imports and domestic marketing, phasing out of state-owned enterprises, improved public expenditure coordination, stronger focus on the rural poor and women, promotion of rural finance, promotion of forest protection, enhanced public sector investment in transport infrastructure, market information, and improvements in research and extension. It was within the framework of that strategy that the five audited projects were implemented. 1.7 Bank Support in the 1990s. The Bank program of the 1990s represented a significant increase in agriculture sector lending over the 1980s. Project support since fiscal 1990 has been largely in sector adjustment, technology, infrastructure and market investments, diversification, and environment. As indicated in the Bank's 2000 Agricultural Sector Performance Review (ASPR), issues addressed by these operations included strengthening the Ministry of Food and Agriculture (MOFA); strengthening local institutional capacity through, for example, support for producer associations; policy reform, including liberalization of pricing, marketing, and input supply, resulting in increased private sector involvement; strengthening publicly supported technology development (research and extension); strengthening rural finance; support for rural infrastructure; cocoa sub-sector rehabilitation; diversification into palm oil, coffee, rubber, and pineapple; and improved environmental management. The Bank support has been broadly relevant to the strategies of both the borrower and the Bank although there appear to be some areas that would have warranted more or different attention at the time and new directions that warrant consideration now. We return to this issue in Section II. 1.8 In four particularly critical areas the reforms did not deliver. First, the private response to liberalization was disappointing because of the devaluation of the Cedi and very high real interest rates - unanticipated macroeconomic developments. Second, notwithstanding redirection of expenditures in some areas, the government found it impossible to maintain committed budgeted resources resulting in serious implementation delays with projects. Third, the Agricultural Policy Coordinating Committee (APCC), set up in 1990 to coordinate sectoral policies and to review 5. The Region in its comments re-emphasizes the importance of not advocating expansion of tree crop cultivation in the Western Region at the expense of environmental degradation. We agree with this.  4 budgets of agricultural sector agencies, never functioned and is now moribund (para 1.24).' Fourth, cocoa sector marketing liberalization was not completed. These issues are discussed in more detail later. 1.9 Rural Poverty Profile. Ghana has unusually good poverty data by the standards of most low income countries. Annex A presents an analysis of the poverty profile with particular reference to the rural areas. Briefly, the overall trend in poverty since 1988 has been broadly favorable in Ghana. From 1988 to 1992, poverty incidence and the level of inequality (measured by the Gini coefficient) declined. There was a further decline in poverty levels between 1992 and 1998. GDP is estimated to have grown, on average, by 4.3 percent over the period 1992-98. Based on the upper poverty line of 900,000 cedis, the percentage of the Ghanaian population defined as poor has fallen from 52 percent to about 39 percent, whereas the percentage of the Ghanaian population that is extremely poor (lower poverty line 700,000 cedis) has fallen from about 36 percent to about 27 percent.7 The decline was not evenly distributed geographically, however. The significant reductions in poverty at the national level have been concentrated in four regions: Western, Accra, Volta, and Brong-Ahafo. The Central, Northern, and Upper East regions have actually experienced increases in poverty between these two years. Statistics show that poverty is lowest in Greater Accra and highest in the Upper East Region. The three poorest regions in the country are the Upper West, Upper East, and Northern. Annex A Table A2 shows that these three regions also have the least access to education, health, water supply, transport and marketing facilities and the highest concentration of livestock per household. A modest decline in poverty incidence is reported in the Upper West Region from 88 percent to 84 percent. The Upper East Region, on the other hand, reports the greatest increase in the incidence of poverty. The document Poverty Trends in Ghana in the 1990s also reports that the depths of poverty for those who remain poor has remained stable between the two periods. Section II. The New Agricultural Strategy and the Macro and Institutional Environment in the Country 1.10 The New Agricultural Strategy. The government's national strategy, Ghana Vision 2020, introduced in 1995, is a long-term comprehensive vision for the Ghanaian economy that is directed towards transforming the country from a low-income economy to a medium-income economy by the year 2020. GOG's overall social and economic development strategy includes the following elements: * Sound fiscal, monetary, and other macroeconomic policies. * Strengthening growth in the agricultural sector by opening access to markets and promoting modern farming methods and marketing practices. * Broadening and deepening manufacturing and services to create employment opportunities. 6. The APCC consisted of senior representatives from the Ministries of Agriculture, Roads and Highways, Trade, Industry Science and Technology, Transport and Communication, Local Government and Finance and Economic Planning. Its main objective was to make recommendations to the Secretary of Agriculture on matters affecting the formulation, implementation, monitoring, and evaluation of agricultural policies and programs sector-wide. Budget allocations for all major public sector agencies operating in the agriculture sector were to be reviewed together so as to rationalize sector wide allocations of public financial resources. 7. As is often the case, perceptions of NGOs at the field level differ from the data. The audit mission for the five agricultural sector projects met with NGOs who felt that the conditions of the poor had worsened over the years. 8. The depth of poverty is a measure of its intensity. It is measured by the income gap ratio, which gives the proportion by which the average consumption level of poor households falls below the poverty line.  5 * Human resource development through improved nutrition, education, health, and water and sanitation services. * Improving access to training, financial services, and markets to develop entrepreneurial capacity. * Assisting poor communities by strengthening infrastructure such as roads and communications. 1.11 Within the overall strategy, the MOFA has formulated an Accelerated Agricultural Growth and Development Strategy (AAGDS) which is designed to increase the sector's annual growth from 2-3 percent (1990-99) to an ambitious 5-6 percent. The government's Agriculture Services Subsector Investment Program (AGSSIP) is considered the vehicle for implementing the strategy.9 It is aimed at rationalizing both sectoral public expenditures and the role of MOFA and improving the effectiveness of policy formulation, regulations, and service provisions in the agriculture sector. Within the economy-wide vision, the strategic objectives of MOFA are stated as: * Strengthen institutional capacity for improved agricultural research and development, policy formulation and analysis, technical services delivery, and sound management of programs. * Formulate and pursue policies that are equitabte and gender sensitive in order to promote and increase agricultural productivity. * Promote and provide efficient technical services, and technological and other measures, that would improve and diversify food and agricultural production and processing for domestic and export markets. * Ensure the availability of timely, reliable, and relevant information for strategic planning and efficient management of the sector. 1.12 Current Unstable Macroeconomic Situation. The economic renewal of the 1983-90 period suffered a setback in 1992 when, coinciding with elections, public expenditure financed by borrowing from the banking sector increased substantially following large rises in the salaries of the civil service. This resulted in a large increase in money supply and high inflation - reaching levels of around 30% - 40%. There was an appreciation of the real exchange rate which adversely affected the export sector and the balance of payments. The high interest rates have continued through the 1990s. Given the continuation of the difficult macro-.economic situation, the immediate relevance of the sectoral strategy statements are somewhat questionable for three main reasons. First, the strategy emphasizes the major role of the private sector yet, at present interest rates, the private sector cannot be expected to make significant new investments and indeed private sector operations that borrowed over this period are facing serious repayment difficulties (see audit report for Agricultural Diversification Project). Second, government budgetary resources are being squeezed so that planned public interventions that may have been considered realistic in terms of financing capacity earlier may no longer be so, at least in the shorter term, calling for rethinking of short term public expenditure priorities. Third, the persistent problematic macro-economic situation is impacting on the poor and thus raising the relative importance in the strategy of poverty-focused interventions. Indeed, there may be a case for an interim rural bridging strategy specially designed to get over the immediate hump and emerge prepared for take-off once stabilization is achieved. 9. For a distinction between AGSSIP and the Bank supported AgSSIP project, see Box 1.2.  6 1.13 Institutional Capacity at the Central and District Level. Ghana has only recently (1988) launched an ambitious decentralization program and progress on the decentralization front is likely to have a major impact on the implementation of AAGDS. Annex B describes the decentralization scenario more fully. Briefly, over the period of the five projects, the institutional environment has changed fundamentally although the final stages of this change are still to be accomplished. District Assemblies (DAs) now implement much of the strategy at the district level following the shift to decentralization, but significant capacity problems are taking time to resolve. Several studies'