Malawi - Agricultural Marketing and Estate Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 16461 IMPLEMENTATION COMPLETION REPORT MALA WI A GRICUL TURAL MARKETING AND ESTA TE DEVELOPMENT PROJECT (CREDIT 1966-MAI) April 7, 1997 Agriculture Operations Eastern and Southern Africa Africa Region 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 Currencv ULnitt= - Maiawi Kxxacha US$1.00 = MK15.3 (March 1996) WEIGHTS AND MEASURES Metric System FISCAL YEAR OF BORROWER April I to March 31 Vice President: Callisto E. Madavo Country Director: Barbara Kafka Technical Manager: Sushma GanguLly Staff Member: Stanley Hiwa  FOR OFFICIAL USE ONLY ABBREVIATIONS AND ACRONYMS ADD Agricultural Development Division ADMARC Agricultural Development and Marketing Corporation AMED Agricultural Marketing and Estate Development Project CB Commercial Bank CBM Commercial Bank of Malawi DEMATT Development of Malawian Traders Trust EAC Estates and Agro-industries Committee GOM Government of Malawi IACP Industrial and Agricultural Credit Project IDA International Development Association INDE Bank Investment and Development Bank Indefund Small loan subsidiary of 1NDE Bank LGDP Local Government Development Project MBS Malawi Bureau of Standards MC Marketing Committee M&E Monitoring and Evaluation MOALD Ministry of Agriculture and Livestock Development MOF Ministry of Finance MOLG Ministry of Local Government MPS Marketing and Pricing Section MRFC Malawi Rural Finance Company NBM National Bank of Malawi PFI Participating Financing Institutions RBM Reserve Bank of Malawi SACA Smallholder Agricultural Credit Administration SAR Staff Appraisal Report SEDOM Small Enterprise Development Organization SLA Subsidiary Loan Agreement T'his 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.  I I I I i  IMPLEMENTATION COMPLETION REPORT AIMLA WI A GRICUL TURAL MARKETING AND ESTA TE DEVELOPMENT PROJECT (Cr.1966-MAI) TABLE OF CONTENTS PREFACE ................................................i EVALUATION SUMMARY ............................................... ii PART I. PROJECT IMPLEMENTATION ASSESSMENT ................................................1 A. Project Description . B. Achievement of Objectives .2 C. Major Factors Affecting The Project .9 D. Project Sustainability .11 E. Bank Performance .13 F. Borrowers Performance .13 G. Assessment Of Outcome .13 H Future Operations .14 . Key Lessons Learned .14 Part II: STATISTICAL TABLES Table 1.: Summary of Assessments Table 2.: Related Bank Credit Table 3.: Project Timetable Table 4.: Credit Disbursements: Cumulative Estimated and Actual Table 5.: Key Indicators for Project Implementation Table 6.: Key Indicators for Project Operation Table 7.: Studies Included in Project Table 8A.: Project Costs Table 8B.: Project Financing Table 9.: Economic Costs and Benefits Table 10.: Status of Legal Covenants Table 11.: Compliance with Operational Manual Statements Table 12.: Bank Resources: Staff Inputs Table 13.: Bank Resources: Missions APPENDICES Appendix A: Aide-Memoire Appendix B: Borrower Contribution to the ICR Appendix C: Miscellaneous Information Appendix D: Map  I I  Implementation Completion Report Malawi Agricultural Marketing and Estate Development Project (Cr. 1966-AL4I) PREFACE 1. This is the Implementation Completion Report (ICR) for the Agricultural Marketing and Estate Development Project in Malawi, for which IDA Credit 1966-MAI amounting to SDR14.2 million (US$19.8 million equivalent) was approved on December 1, 1988, and made effective on June 9, 1989. 2. The credit was closed on June 30, 1996, six months after the original closing date. It was fully disbursed. The last disbursement took place On August 27, 1996. 3. This ICR was prepared by a mission from the FAO/World Bank Cooperative Program1 on behalf of the Agriculture Operations Division of the Eastern and Southern Africa Department (AFTAI). The draft ICR was reviewed by Ms. Sushma Ganguly, (Technical Manager, AFTAI), Mr. B. Van De Poll (Senior Agriculturist, AFTAl), and Mr. Steven Jaffee (Agribusiness Specialist, AFTAI). The report is based on material gathered during a mission that visited Malawi in April/May 1996, and on information from the project files, supervision reports, the report and findings of the mid-term review, as well as on the findings from field investigations and discussions with Bank and Government of Malawi (GOM) staff associated with the project. GOM's contribution to the ICR is attached as Appendix B of this report. The Borrower also facilitated the preparation of the ICR by arranging field visits and meetings, and by providing substantive comments on the Mission's Aide-Memoire. Messrs. R. Suppa (Economist/Mission Leader, FAO/CP), P.V.A. Rama Rao (Credit Specialist, Consultant).  I I  ii IMPLEMENTATION COMPLETION REPORT MALA WI A GRICUL TURAL MARKETING AND ESTATE DEVELOPMENT PROJECT (Cr.1966-MAI) EVALUATION SUMMARY Introduction 1. Since independence in 1964, the Government of Malawi has regarded the development of the agricultural sector as a core element of its broader development strategy. The World Bank group has been actively assisting this sector and, prior to the Agricultural Marketing and Estate Development Project (AMED), financed 12 operations totaling US$173.3 million for agricultural development in Malawi. The main components of the AMED project were to be the development of a multi-channel system for marketing agricultural produce and inputs, an improvement in the access of small estates to medium-term credit through existing credit channels, and the strengthening of capacity within Govermment in project management and in the management of marketing and pricing policy reform. This project reinforced previous IDA assistance in the sector by introducing improvements in marketing and promoting crop diversification. The project complemented other IDA supported initiatives in agricultural development that focused on research, extension, and rural infrastructure. Objectives 2. The project's objectives were to: (i) improve agricultural marketing and increase rural incomes; and (ii) improve resource utilization and productivity on estates, by providing medium- term credit to intensify and diversify production. The project, designed to be implemented over a seven-year period, included the following components: (i) credit for estate development, agro-industrial sub-projects and a pilot credit scheme for rural merchants; (ii) institutional development; and (iii) market infrastructure development. 3. The actual project cost is estimated at US$25.4 million. The IDA credit of SDR 14.2 million, equivalent to US$19.8 million was fully disbursed. Other disbursements included GOM US$0.4 million, Participating Financial Institutions (PFIs) US$1.8 million, and Borrowers' contribution US$3.4 million. All major legal covenants were fulfilled during project implementation.  Implementation Experience 4. Implementation record. The participating financial institutions (PFIs) lent a total of MK1 6.1 million to eighty-six Borrowers for estate development. The bulk of this lending went to large-scale estates who used the funds for land improvement, livestock, and the development or rehabilitation of on-farm infrastructure. Only seven loans, amounting to MK1.2 million, and constituting only 7.5% of the total loans disbursed under the project, were approved for small estates. The appraisal had envisaged MK24 million, or 33% of total targeted lending, to be channeled to the small estates. The low disbursement of loans to small estates stemmed from a combination of factors including previous high rates of loan defaults from this subsector, the relatively high transaction costs incurred by the PFIs in servicing small estates, the absence of acceptable collateral, and the narrow interest rate spreads stipulated under the project for such loans. The only lending made to small estates using project funds was through Indefund, a subsidiary of the Investment and Development Bank (INDE Bank), which was brought in as a PFI about two years before project closure. 5. With the slow disbursement of the credit to small estates there was concern about the project not achieving its objectives in relation to employment generation and income distribution. Therefore, in 1992 the GOM requested IDA to reallocate SDR 4.2 million to the Smallholder Agricultural Credit Administration (SACA) to finance seasonal credit for smallholders as part of the Government's drought recovery efforts. The DCA amendment which made this re-allocation did not specify any repayment obligations by SACA to the Reserve Bank of Malawi. Very low rates of repayment were achieved on the seasonal loans provided during the 1992/93 season as that year's election campaign heavily politicized rural credit. 6. Project funds were used to finance nine agro-industrial or agribusiness operations, with loans totaling MK7.7 million against an appraisal target of MK12 million. These supported investments in tree crop, oilseed, and dairy processing, and in floriculture. High financial rates of retums were recorded for most of these ventures. These investments contributed to export diversification and to some import substitution. The Small Enterprise Development Organization (SEDOM) piloted a successful lending scheme for private traders. Starting with only six loans worth MK 0.7 million in 1992/93, the scheme expanded to 342 loans, totaling MK 30 million by 1995/96. The loan recovery on this private trader loan scheme exceeded 90% throughout the project period. A planned warehouse certification scheme was not implemented. 7. The objectives that were set for institutional and market development components were substantially achieved. The Pricing and Marketing Section (PMS) played an active role in analyzing and reviewing policy options and advising the Government. The project also provided a vehicle for linking up with other Bank operations to coordinate and continue policy dialogue on market reform and price liberalization. Trading in all crops is now fully liberalized. The sole exception is maize trading where state controls still remain on exports as part of the GOM's policy to safeguard national maize availability. A Price Information System has been installed in the Ministry of Agriculture and Livestock Development (MOALD) which regularly collects and disseminates price information on fiv& food crops and three livestock products. The price data have been used by private traders to guide their decisions in locating their commodity buying or  iv selling operations. The data were also used by MOALD to estimate gross margins as part of the annual price review and price setting exercise. Extensive use of the price data in policy analysis work is limited by sample weaknesses and by the continued understaffing of the section. 8. The project, through MOALD, has trained over one thousand private traders on the fundamentals of commodity trading and business management. Export quality contfol was implemented through the Malawi Bureau of Standards (MBS) with the project funding equipment, TA, training, consumables, etc. The use of the Export Certification Scheme is still limited by inadequate publicity of the facility within the country, and an inability to institute cost recovery pricing due to low volumes of business. In addition, the Export Certification Scheme was conceived and developed as a parastatal-run monopoly service that has failed to adjust in the face of emerging private sector competition. The project constructed twelve out of the planned sixteen markets for local councils. While these council markets have facilitated rural private trade, their long-term sustainability is limited by: the inability of local councils to levy market fees that would at least cover operating costs; the failure and unwillingness of local councils to allow the private sector to provide some of the commercial facilities and installations required in the markets; and, the failure to institute a revolving fund to finance future market construction as designed by the project. 9. Key factors affecting project achievement. The main factor that explains the initial project implementation delays was the lack of coordinated project management. There were two parallel monitoring committees, one each for marketing and credit, with no unity of command, and without senior representation on either committee. This resulted in a lack of guidance for most of the implementation and monitoring activities and explains the project's failure to lend to small estates as planned in the SAR. 10. The AMED project was conceived at a time when the policy and legislative framework was characterized by extensive state controls of all agricultural pricing and marketing activities, sector credit rationing and interest rate caps, general distrust of the private sector, and dominance of input and commodity markets by parastatal monopolies. In the early years of the project several policy reforms were introduced, yet a substantial movement away from state-directed markets did not occur until the last three years of the project. This, itself, generated a period of transition in which private agents, parastatal enterprises and financial institutions began to explore new opportunities and adjust to lost privileges. Many of the reforms in policy and legislation have not been supported by timely publicity and training, resulting in continued misperceptions and mutual private-public sector mistrust. 11. Three developments, outside the control of the project, had a great influence on the project's performance and outcome. First, a more pronounced supply response from policy reforms and the resources provided under the project was constrained by two major droughts, in 1991/92 and 1994/95. These droughts reduced the trading base for private crop traders and reduced farm incomes. Second, a radical change in the country's political system accelerated the implementation of policy reforms, yet at the same time increased the demands on already constrained financial and advisory services. Third, a breakdown in credit discipline occured  v during the political transformation which, in 1992/93 resulted in widespread default on SACA seasonal credit. 12. Sustainability. The policy reforms implemented under the project have a high staying power. The reforms have been reinforced by the effect of complementary SALs/SACs and the change in the political system in the country to one that is more supportive of the private sector. The PFIs will be able to maintain the flow of credit to the large estate sector with their own resources. The experience that the PFIs gained under project will assist in future lending activities, especially in emphasizing the importance of competent loan appraisal and effective loan supervision. The sustainability of providing loans to small estates and private traders is mainly dependent on the implementation of additional measures by GOM and the Reserve Bank and on the financial services provided by the Malawi Rural Finance Company under the Rural Financial Services Project. Indefund and SEDOM now have the capacity and expertise to lend to small estates and private traders. However, the line of credit extended to these two PFIs under the project did not create a revolving fund. This has resulted in a reduction of the amount of resources available to be onlent to small estates after the project closed. 13. The market infrastructure component could be sustainable with a sound policy and operational back-up by GOM. There is still an outstanding need to:- (i) establish a revolving fund from the loan repayment proceeds, perhaps through the Development Fund for Local Authorities, that would finance future market construction and rehabilitation; (ii) devolving greater authority for setting market fees to local authorities; and, (iii) allowing greater private sector participation in the provision of marketing facilities in rural/council markets (e.g. weighing scales, trading stalls, shop space). The operation of MBS and the export certification scheme will be sustainable with improved commercialization and marketing of the services. It is highly likely that the service would be viable if privatized as this would avert the higher overheads associated with a parastatal operation. 14. Bank and Borrower performance. Bank performance was generally satisfactory, throughout the project cycle. The Bank mounted a total of nine supervision missions during the project. These missions provided timely and effective guidance to the GOM and other project implementing agencies. There were three changes in project task management, however. In the course of project supervision efforts were made to strengthen the policy reform and trade liberalization process and to develop and refine key project monitoring indicators. Indefund was also brought in as a PFI to enable lending to small estates. The reallocations of funds to SEDOM and SACA, at various points, were also timely interventions designed to meet effective credit demand and to support the broader objective of raising rural incomes. This project would have benefited from more frequent Bank supervision and from a mix of skills in the supervision missions that also included credit/microfinance and agricultural marketing. The Borrower's performance was generally satisfactory and responsive except in the areas of staffing of MPS and in finding solutions for facilitating the flow of credit to small estates. The GOM's commitment and performance in the area of market reforms was satisfactory. 15. Overall project outcome. Overall, the project outcome can be considered moderately satisfactoiy. The project achieved its objectives in improving agricultural marketing and  vi developing a multi-channel marketing system as part of market reform policy . The other objective of estate development through credit has been partially achieved to the extent that some large estates have intensified and diversified their production. The project had limited success in promoting exports and diversification though it helped in creating and sustaining rural employment and incomes. The PFIs financed a series of investments geared toward tree crop or food processing. The project contributed substantially to price and marketing policy reforms and the establishment of private trader micro-lending as a viable commercial undertaking. It is noted, however, that the performance of components for market infrastructure, institution building, and credit to small estates was not satisfactory and will require further efforts to be considered reasonably successful and sustainable. Summary of Findings, Future Operations and Key Lessons Learned 16. Findings. Private traders emerged and increased their activity in rural markets as a result of the reform policy facilitated by the project. The existing financial institutions have established on-going commercial relations with a segment of the estate sub-sector and with selected agro- industrial firms. The financial institutions experienced problems lending to the small estate sub- sector, and cut back on such lending. 17. Future operations. GOM is continuing with its liberalization of policy reforms through the ASP (Cr. 2514-MAI) and other operations. The issue of operational viability of rural council markets will be pursued by GOM through the ongoing Local Government Development Project (Cr. 2646-MAI). An action plan for the recovery of SACA loans has been undertaken by the Malawi Rural Finance Company (MRFC) under the IDA-assisted Rural Financial Services Project yet only a small proportion of this outstanding SACA debt has been recovered. The MRFC has improved the access of small estates to production finance and is developing its capacity and procedures to provide business loans to private traders. Key lessons learned. 18. The project demonstrated the need for effective arrangements for project management and monitoring. It is important that projects be grafted into the existing institutional arrangements for portfolio management within a sector without resorting to the creation of 'independent'units. The implementation of the AMED project was undermined by the involvement of such an implementation unit which lacked the necessary seniority, capacity for decision-making and linkages with the relevant Ministries' overall planning and monitoring activities. 19. While demonstrating the viability of microloans and lending to private traders, the project's experience suggests that commercial banks may not be the most suitable institutions for extending small loans to SMEs and small estates. The experience under AMED suggests that for reaching such clients it is appropriate to work through existing SME lending institutions, as they already have a lending philosophy, staffing, and procedures aimed at extending and managing  vii small loans. The project's experience points to the needed design of revolving fund facilities, with suitable apex institutions, that can continue to finance micro loans after project closure. 20. The project demonstrated the value of client training which helped to impart skills in the preparation and submission of loan applications and in loan management, plus to further familiarize Borrowers with the key (financial) aspects of the businesses foriwhich they are seeking financial support. This is especially important for clients who are borrowing for the first time as routine yet poorly understood procedures may serve as a barrier to accessing loans. At the same time PFIs need to be sensitized to the special features of lending groups, especially those requiring microloans, which require significant departures from conventional lending approaches to include group loans, closer loan supervision, and simplified application procedures. 21. Institutions and policies outside the direct control of the project should be coordinated so that their weaknesses and pace do not undermine the implementation and success of the project. This was brought to bear for the AMED project in the case of the Ministry of Local Government which moved slowly in amending its set of regulations stipulating areas allowed for trading at official minimum prices. The Ministry was also unable to institute a revolving fund that would sustain the financing of local markets after project closure as the AMED was seen as a source of grant funding for the local authorities. Coordination arrangements for multisectoral activities should consider moving towards the creation of "councils" with permanent membership, defined terms of reference, and incentives. 22. In terms of the project design, the barriers to entry into agricultural trade faced by small entrepreneurs should have been examined and reviewed up front. The project played a reactive role in assessing and consulting with Government on policy reformns required to facilitate and support the realization of project objectives. This should, instead, have been a major input at preparation and design, to study and review the extent to which prevailing policies provided an enabling environment for the project, and to recommend the necessary institutional, price, and market reforms.  Project Implementation Completion Report Malawi Agricultural Marketing and Estate Development Project (Cr.1966-AMI) PART I. PROJECT IMPLEMENTATION ASSESSMENT A. Project Description 1. The project had two objectives: (i) To improve agricultural marketing and thereby increase rural incomes. This would be achieved by supporting Government's efforts to develop a multi-channel marketing system; and (ii) To improve resource utilization and productivity on estates. To achieve this objective, the project would provide medium-term credit to finance intensification and diversification of farm production. This activity was mainly targeted to the small estates which had limited access to commercial credit. 2. The project had three components: (a) estate development and agricultural marketing and processing involving; (i) sub-projects to intensify and diversify estate production; (ii) investment in support of agricultural processing and marketing; and (iii) a pilot scheme to finance the short and medium-term credit requirements of rural traders specializing in the purchase, storage and transportation of smallholder crops. (b) private trader training and institutional strengthening in the MOALD and the Malawi Bureau of Standards to improve market policy analysis and develop market services, private including market intelligence, regulatory and quality control systems; and (c) construction and improvement of selected rural markets that were identified as potential growth centers for private trade in agricultural produce.  2 3. The project was an important component of the World Bank's Country Assistance Strategy to support the Government's market reform efforts and to develop a viable rural credit system through existing financial institutions. The project was also aimed at promoting improvements in agricultural sector production and productivity. The institutional component supported improvements in Government's capacity to monitor market developments and to implement appropriate policy reforms. The loan fund for estate development was expected to contribute to a more efficient and diversified production pattern. B. Achievement of Objectives General 4. The objectives of developing a multi-channel marketing system with expanded private trader activity in rural markets, and improving resource utilization and productivity on estates were achieved satisfactorily. Delays were initially experienced in implementing some components, including credit for estate development, credit for rural merchants under a pilot scheme, introduction of export certification scheme, and construction of rural markets. The restructuring of the project in 1992, by expanding project objectives to cover drought recovery efforts, reallocated project funds to SACA to provide seasonal credit to smallholders. The SACA loan scheme experienced massive default in 1992/93. The warehouse certification scheme, designed to provide working capital loans to private trading enterprises against the security of ADMARC warehouse receipts, was not implemented. Credit for small estate development did not grow to the levels that were planned at appraisal. The project was completed at the end of December 1995 in line with the schedule foreseen at appraisal. The extent and quality of implementation of the different project components is described below. Estate Development and Agricultural Marketing and Processing 5. Estate development: The project sought to improve resource utilization and productivity on estate lands by providing medium term credit to intensify and diversify crop production. Three financial institutions, consisting of the two commercial banks and INDE Bank, were chosen as PFIs for the project. The selection of these three banks as PFIs was in line with the project's objective to promote access to commercial credit by small estates. Lending to estates did not take off during the first two years. This delay was due to the problems that the PFIs experienced in developing loan appraisal procedures that complied with IDA requirements. The PFIs also lacked adequately trained staff that would interpret and operationalize the guidelines provided by IDA for this purpose. The project lent mostly to large estates, contrary to the project design which envisioned that lending to small estates would constitute 50% of all loans to estates, or 33% of the total portfolio. The low level of lending to small estates was due to: (i) the high risk associated with lending to small estates as default rates were high in the early years of the project and as these estates lacked acceptable collateral; (ii) the high operating costs associated with loans to small estates; (iii) the unacceptable modalities for lending on the line of credit which stipulated lower rates of interest than the prevailing commercial rates of interest at which the CBs were lending their own resources. With access to adequate liquidity, there was no  3 incentive for the banks to lend borrowed resources; and (iv) the non-availability of suitable staff to supervise small loans in the PFIs. It is also important that the requirement to concentrate lending on small estates was not reflected in the Subsidiary Loan Agreements (SLA) between the RBM and the individual PFIs. Under these circumstances, IDA and GOM decided to stop further lending to larger estates in 1994 and, in the same year, to accept Indefund as an additional PFI that would lend specifically to small estates. Indefund, which offers small investmenft loans, ranging from MK30,000 to MK1 50,000, was able to make only a few loans, totaling MK1.2 million, to small estates. The Fund started making loans to small estates only after the Reserve Bank agreed to a request for a wide interest spread to meet the heavy cost of servicing relatively small loans. 6. The PFIs disbursed a total of 86 loans, amounting to MK1 6.1 million, for estate development. The number of disbursed loans achieved 28% of the target envisioned at appraisal. Most of the loans, 92.5% of the loan amount, were extended by the two commercial banks and INDEBank for larger estates. Indefund accounted for only 7.5% of the total credit that was disbursed to estates. 7. The deregulation of bank lending rates in Malawi gave the PFIs the discretion and flexibility to fix their lending rates according to prevailing market conditions. The project had stipulated an interest rate spread of 5% for loans to large estates and agro-industries and 7% for small estate loans. The PFIs were required to annually review and revise their on-lending rates to be consistent with the project guideline. The RBM was in turn expected to review and revise its own lending rates on a point-to-point basis to reflect the changes in the agricultural lending rates and the maintenance of the stipulated spreads. The interest rates charged by the PFIs increased during project implementation, in line with the prevailing market conditions. 8. The average loan sizes for Indefund and CBs was MKO. 17 million and MKO. 13 million, respectively. The loans, even for larger estates, were relatively small perhaps to avoid the mandated obligation of working out Finance Rate of Returns for loans above MKO.25 million. In line with the SAR, the loans extended to large estates were for land improvement (irrigation), development and rehabilitation of on-farn infrastructure (tobacco curing barns, storage sheds), and purchase of tractors and implements, vehicles, and livestock. All the loans issued by CBM were for the purchase of tractors. National Bank of Malawi, with a total of 29 loans, lent for a wider range of purposes. The loans issued by Indefund covered mainly diversification of tobacco estates to sunflower, soybeans, commercial maize, and livestock. The project did not monitor the extent to which tobacco farms that borrowed for crops other than tobacco actually applied the loan proceeds for diversification. It is likely that these loans also went to finance tobacco intensification given the constraints small estates faced in accessing credit. 9. The recovery rate on the loans extended by INDE Bank and Indefund was satisfactory at about 90%. The NBM had a loan recovery rate of 80% while CBM, despite lending only for tractor purchase, reported a loan recovery rate of only 50%. Despite the mixed achievements in loan recovery, all the four PFIs punctually met their full loan repayment obligations to the RBM.  4 10. Agro-Industrial Sub-Projects. The project envisaged providing medium term credit for the development or rehabilitation of agro-industrial facilities, construction or rehabilitation of storage/handling/grading facilities, and the purchase of transport equipment by the private traders. Nine units, out 17 envisioned at appraisal, were financed under this program; All the loans were for agro-industrial facilities, covering tea, coffee, cotton, vegetable oil, seed, dairy and floriculture. The project extended no loans for milling investments as envisaged at appraisal. The SAR lending targets were based on the assumption of providing the anticipated loan financing that ADMARC milling operations would require for rehabilitation after being privatized. There was no attempt to undertake a wider review of the prevailing investment climate and needs. Lending for milling investments with project resources did not take off as the banks preferred to lend their own resources and ADMARC'c continuing interest in the privatized milling companies gave them access to significant equity funds. INDE Bank, due to its expertise and experience in this sub-sector, extended seven agro-industrial loans while NBM and INDEFUND each financed one project. CBM did not finance any project in this area. 11. The SAR planned to extend short-term working capital loans to private traders against the security of warehouse receipts by ADMARC. This scheme was not implemented despite the wide prevalence of ADMARC storage facilities in rural areas. Its failure was largely due to lack of knowledge and expertise required to design, install, and operate such a scheme. ADMARC also moved slowly on this sub-component as it viewed this as granting an unfair advantage to its competitors. 12. Pilot Credit Scheme for Rural Merchants. The scheme, implemented by SEDOM, was formulated to help small rural merchants to increase their participation in the marketing of farm produce. The project was planned to provide short and medium-term loans to finance most aspects of crop trading, including crop purchase, and investments in storage, processing, and transport. No medium-term loans were extended due to inadequate credit allocation and the perceived lack of capacity within SEDOM to manage such loans and the SMEs lack of acceptable collateral. The financing of working capital loans commenced after a delay of more than two years. This delay was mostly due to a lengthy process of consultation and approvals required of SEDOM's loan procedures and lending methodology. In order to disburse loans under the pilot, SEDOM was required to demonstrate its loan processing procedures and to clear this, along with the first recommended loan recipients, with RBM and IDA. The whole process of review, comment, and approval took two years partly due to a perception by the supervising institutions that SEDOM was not capable of managing this sub-component. The impasse was only resolved in the field when a Bank supervision mission undertook joint review of SEDOM procedures with RBM and recommended that the pilot could proceed. 13. SEDOM expanded its working capital loans significantly from about MKO.7 million to six traders in 1992/93 to a level of over MK30 million, covering 340 traders by 1995/96. 20% of its loans were to female Borrowers. In 1995/96 SEDOM also initiated lending to retail input dealers. Due to its effective lending and supervision procedures, besides provision of essential business advisory services, SEDOM achieved nearly a 100% loan recovery rate.  5 14. SEDOM's successful lending experience led to the reallocation of additional funds for this category. The final allocation of SDR 2.6 million was fully utilized. SEDOM has been repaying its loans to the RBM annually as stipulated in the SLA. Its request to recycle the recoveries for a period of three years was approved by the Reserve Bank to be effective from FY 95/96. SEDOM further requested RBM to establish a loan revolving fund that would finance a private trader lending program in the long term. This proposal, though approved by GOM, has not been implemented due to lack of resources. Previous PFI loan repayments to RBM are not available to finance the revolving fund since they were routinely credited to Government as counterpart funds and spent on public debt relief. 15. Implementation progress of the credit component for estates was slow throughout the project's life. This was primarily due to the reluctance of PFIs to extend credit to small estates as they were perceived to have high risks and transaction costs. Following a supervision mission in November 1991, the GOM and IDA prepared an action plan to improve the project's performance on this component. The main action was to restructure the credit component to allow project funds to be used for incremental seasonal credit for smallholder farmers. Accordingly, SDR4.2 million was reallocated to SACA in 1992. The DCA was amended for this purpose by broadening the project scope and objectives to cover drought recovery efforts. SACA utilized SDR3.9 million out of the total allocation. 16. As indicated in the ICR for the Smallholder Agricultural Credit Project (Cr. 1851 -MAI), indiscriminate lending by SACA following the drought in 1991/92, coupled with political intervention in the recovery process, resulted in a very poor loan recovery performance of 15% in 1992/93. The project also made no attempt to integrate smallholder credit as one of its components, other than acting as an ad hoc and convenient source of funds to finance the shortfall in seasonal credit. No attempts were made to monitor and supervise SACA once the funds were disbursed nor to institute measures in SACA loan approval and management procedures, as indicated in the ICR for SACA, in order to establish and enhance its viability as a lending institution. Institutional Development 17. The Marketing and Pricing Section (MPS). To assist the MPS in fulfilling its policy formulation and market support functions, the project provided resources for technical assistance, which included three person-years of an internationally recruited marketing adviser. The adviser was able to fulfil his TORs except with respect to documenting and operationalizing the warehouse certification scheme. Vehicles and office equipment were provided to MPS as envisaged. The training program covered two 2-year Master's degree studies in economics in the UK and a short course in marketing in Australia. The rural financial specialist, supposed to implement and monitor the pilot credit scheme of SEDOM, was not recruited as adequate in- house expertise was available within SEDOM and MOALD. The proposed organizational set-up of MPS, as foreseen at appraisal, included a total of seven economists but only three were ever actually in position throughout the project. As a result, the MPS was restricted in the services and functions it was able to perform and some activities, e.g. the price and market information system, were only supported marginally. Overall, considerable achievement has been made in the  6 implementation of the mandate for market reforms by MOALD through amending or repealing legislation and policies restricting the participation of private traders in agricultural marketing. The marketing of all agricultural inputs and crops, except maize, has been liberalized, thus effectively ending the ADMARC's monopoly in crop and input trading. Government control in maize trading, mainly exports, is continuing due to the perceived strategic importance of this crop as the principle staple food in Malawi. 18. MOALD regularly collects farm commodity price information for sixteen crops and three livestock products from 45 rural markets. However, analysis and dissemination of price information through mass media is done only for five foodcrops (maize, cassava, rice, groundnuts and beans) and the three livestock products (beef, goat meat and pork) on the basis of price information collected from 18 markets. 19. Marketing advisory did not receive the required attention during the project. At project design, no funds were allocated for market extension on the assumption that this service would be performed by existing ADD credit and marketing officers. These officers were, however, concentrated on the credit activities under SACA. Scope still exists, however, for market extension to be integrated into the overall extension service of the MOALD through existing Subject Matter Specialists and Field Assistants. 20. The factors that constrain the availability of credit to estates, particularly to the small estates sector through the existing financial institutions, should also have been addressed during implementation. The training of small estates on aspects of farm credit management, including record keeping and the preparation of business plans for loan application, would have improved access to credit. In addition, the PFIs required to be sensitized and trained on the unique features of extending small loans which require different procedures and systems from normnal commercial loans. 21. In terms of the project design, the barriers to entry into agricultural trade faced by small entrepreneurs should have been examined and reviewed up front. The project played a reactive role in assessing and consulting with Government on policy reforms required to facilitate and support the realization of project objectives. This should, instead, have been a major input at preparation and design, to study and review the extent to which prevailing policies provided an enabling environment for the project, and to recommend the necessary institutional, price, and market reforms. One issue that remains unresolved is the conflicting interests of private traders, and private sector development in general, with the continued role of ADMARC and the role of Government in an economy that is attempting to liberalize and promote the private sector. 22. Training. The Planning Division of MOALD, in conjunction with SEDOM, INDEFUND, ADMARC, and Development of Malawian Traders Trust (DEMATT), conducted marketing courses for private traders as well as for the staff of MPS and ADDs covering subjects such as general produce marketing, book and record-keeping, accounts and storage. As a result, some 1,130 private traders received training during the period 1989/90 to 1992/93. The project ensured that all the prospective SEDOM beneficiaries, particularly women, were trained in the businesses they wanted to borrow for. All the entrepreneur training courses were conducted by  7 resource persons from DEMAT, SEDOM, and MOALD. This arrangement for recruiting resource persons did not always guarantee the professionalism and competence required for effective business skills training that may have been possible if the training was provided through specialized institutions like the Malawi Entrepreneur Development Institute (MEDI) and the Polytechnic Business School. 23. Export Quality Control. The project provided funds to MBS to procure office and laboratory equipment and vehicles. The installation of laboratory equipment was completed in 1993. Additional staff, including a scientific officer, two technical assistants and two technical officers were recruited. Six months of TA for export quality control was also provided. Under the training component, six MBS officials were sent abroad on study tours to familiarize themselves with export inspection and certification of agricultural products and the management of export quality control systems. The implementation of this component was delayed for about two years as MBS did not fully understand Government's budget procedures and IDA's funding arrangements. IDA disbursement, which was on a reimbursement basis, required that MBS be allocated resources in the Development Budget which would first be spent before claiming for reimbursement. The MBS was not familiar with this requirement and never submitted its financing requirements to be included in the Government Budget. As a result the Government Development Budget made no financial allocation to them during the first two years that would have enabled them to utilize the IDA credit. An additional development that influenced the implementation of this component was the market liberalization program which led to quality control and certification of agricultural exports becoming optional. In these circumstances, the tests conducted by MBS declined from about 4,600 in 1992 to 2,550 in 1994. During the same period export certificates issued declined from 51 to 27. The opening up of in-house laboratory facilities by the tobacco export companies, and the opening of SGS offices in Malawi, further contributed to the declining business of MBS. This has resulted in gross under-utilization of the facilities provided under the project. While MBS had undertaken many steps to publicize its services, a more aggressive campaign for publicity in potential markets, both at home and abroad, is justified. It is also likely that this service would be viable if privatized as this would avert the higher overheads associated with a parastatal operation. 24. Market Infrastructure. Out of the sixteen district markets to be constructed and/or rehabilitated by the project, only twelve were completed, three through LCB and the remaining on force account. All the constructed markets are collecting revenues by charging fees to the vendors although the project did not develop guidelines for fee collection on cost recovery basis or for the management of the markets. A study of six markets conducted by the Ministry of Local Government (MOLG) has revealed lack of financial viability in the operation of half of the markets even under optimal conditions. The poor design and the structural deficiencies in the Lunzu, Chilumba and Lizulu markets contributed to the under-utilization of the facilities. Due to the poor revenue collections MOLG proposed that 50% of the loan to the local authorities be treated as a grant. However, no formal agreement has yet been entered into between the MOLG and the local authorities in respect of these loans. 25. Reaching an agreement on loan repayment by the local authorities is an urgent necessity. The agreement would enable the setting up of a revolving loan fund needed to finance future  8 market development as envisioned by the project. It would also encourage local authorities to take their loan obligations and financial management seriously and encourage them to start managing their revenues. Economic and Financial Objectives 26. As against the estimated project cost at US$28.3 million, the actual project cost was US$25.4 million. The IDA credit of SDR14.2 million, equivalent to US$19.8 million, was fully disbursed. The reduction in the project's cost mostly affected the market construction component where 4 markets were not constructed by project completion due to lack of funds. All other physical targets in terms of procurement, training, and infrastructure were met. The reduction in the project cost resulted from the reallocation of funds to SACA which did not require counterpart contributions from the Government or final Borrower. The disbursements of counterpart funds included GOM US$0.4 million, PFIs US$1.8 million and Borrowers contribution of US$3.4 million. The final disbursement under the IDA credit was made on August 27, 1996, and the credit was closed on June 30, 1996. 27. At appraisal, financial rates of return (FRR) for small estates were estimated to range from 18% to over 50%. The FRRs for agro-industries were expected to be in the range of 20% to 30%. It was assumed that the associated economic rates of return would be at the same level or higher due to the limited number of subsidies and price distortions in the estate and agro- processing sectors. The SLA stipulated FRRs to be calculated for investment proposals requiring loan amounts above US$ 100,000. Furthermore, in order to qualify for bank loans, the sub- projects were required to yield a FRR of not less than 12%. For INDE Bank, whose loans averaged US$180,000, the relative FRRs were reported to be above 50%. The other PFIs did not undertake detailed financial analyses as their loans were less than the SAR stipulated threshold for estimating the FRR. However, on the basis of data provided by INDEFUND, the ICR mission was able to estimate the FRR for selected projects which were also above 50%. 28. The project significantly assisted individual subprojects that contributed to the intensification of tobacco, diversification of agricultural production, and to increased rural incomes and employment. Support to tobacco and diversification also increased exports earnings and direct rural employment. Investments in oil processing and refining resulted in import substitution and forex savings on imports of edible oils. Along the same lines, loan assistance provided by SEDOM to rural enterprises for trading in agricultural produce and inputs is expected to have assisted in generating direct rural employment. 29. As indicated in a study of six markets carried out by MOLG, the rural markets lack operational viability. However, these markets have provided acceptable physical facilities for rural trade in farm produce, inputs, and other food and non-food products. 30. The institutional strengthening program has helped GOM in establishing improved data collection and analysis, and in facilitating the initiation and implementation of appropriate market reforms. The liberalization of farm produce and input trading is the main impact of the reform process which has also generated rural employment and incomes.  9 C. Major Factors Affecting The Project Project Management 31. The project was jointly managed by MOALD and RBM. MOALD coordinated the institutional development component and the pilot credit scheme while RBM was responsible for the implementation of the credit component. Two coordinating committees were also set up to oversee the implementation of the project. The Marketing Committee (MC) was headed by the Principal Secretary for MOALD and the Estates and Agro-Industries Committee (EAC) was headed by the Secretary to the Treasury. The MC and EAC met regularly but their effectiveness was undermined by the following factors: (i) dichotomy in the responsibility for credit component management between RBM and MOF; (ii) lack of continuity in the membership of the committee; (iii) absence of senior members, including the chairmen, in the committee meetings; and (iv) non-involvement of the Monitoring and Evaluation Unit (M&E) of MOALD in the project monitoring activities. 32. As a result, most of the monitoring activities proceeded with little professional guidance and there was no system for giving feedback to the implementing agencies on the progress reports submitted by them. Further, the capability of the M&E Unit, responsible for the mid-term evaluation and completion report, was impaired by the insufficient flow of project data and information from the implementing agencies and the MCs. Indicators for monitoring the project, outlined in the SAR and refined during supervision missions and during the MTR with Government's agreement, were never systematically applied to the project as the M&E unit was fully occupied with monitoring ADDs and sectoral performance. 33. The implementation and management of the project suffered from the lack of unity in command. In a project with multiple objectives, cutting across several departments of the government as well as different parastatal institutions, the absence of a single controlling and coordinating point contributed to start-up delays experienced by the project in the sub- components for credit to estates, pilot loan schemes for traders, export certification, and in market infrastructure. Other components such as the warehouse certification scheme failed to be implemented. In addition, issues that emerged during implementation e.g. inability to provide credit to small estates or the loan requirements of SEDOM clients, were not addressed effectively. The World Bank supervision mission in April 1991 rated the performance of the project as unsatisfactory because of the slow progress in lending by CBs, lack of progress in lending to traders and in market development. All these were caused by problems in project management.  10 External Factors 34. Reforms in policy and legislation, as well as publicity and training in support of liberalization reform, were not properly phased and linked with the provision of financial services and infrastructure required in a liberalized environment to facilitate entry into production an enable the planned intensification and diversification of farm produfction and trade. Three other developments, outside the control of the project, have greatly influenced the project's performance and outcome: (i) A more pronounced supply response, as a result of the project, was constrained by two droughts, in 1991/92 and 1994/95, that reduced crop output and also reduced the trading base for private crop traders and farmers" incomes. One of the major implications of the drought was to reduce farmers' ability to repay seasonal loans which undermined repayments to loans extended under this project and by SACA; (ii) A radical change in the country's political system also ushered in a period of accelerated implementation of policy reforms that led to increased demands on already constrained financial and advisory services. (iii) The breakdown in credit discipline leading to the massive and widespread default on SACA credit. MOALD 35. Small estates development: The country had about 3,500 small estates in 1982 and the SAR envisioned that 285 would access loans under the project. However, only seven small estates were able to get loans. The project failed to address the factors that constrain the availability of credit to small estates. In the absence of any such exercise, the implementation of an important component, crucial to the development of the agricultural sector and for rural income distribution, did not proceed as planned or required. There is a possibility that training of small estates on aspects of farm credit management, including record keeping and the preparation of business plans for loan application, would have improved access to credit. In addition, the PFIs required to be sensitized and trained on the unique features of extending small loans which require different procedures and systems from normal commercial loans. In this context, access to credit by small estates should have improved if the PFIs recognized their training needs, the need for closer supervision, the option of group loans to improve viability by reducing supervision expenses and as a substitute for collateral. Ironically, many small estates received credit under the leakage of subsidized SACA credit. 36. Drought recovery mechanism. Under-utilization of funds by estates was the main reason for reallocating project funds in favor of SACA in 1992. There was no exercise to inquire into the causes for the slow disbursement of loans to estates and to suggest possible solutions. It was important to provide drought relief to smallholders. However, similar support was also necessary for the small estate sector that was not accessing any formal credit. In addition, there was the need for investments in infrastructure and irrigation to improve the production and productivity of estate lands following the drought. The reallocation of funds was implemented at a time when the PFIs had geared up to expand their loan financing to estates. The decision to reallocate the project funds to SACA is reported to have been taken without consulting the PFIs. In fact, a GOM inter-ministerial meeting, held in August 1992, advised against the proposed reallocations. However, it is undeniable that Government made a request to effect the  11 reallocation. Support to smallholder farmers had a stronger social and economic justification given the numbers of people who participated in SACA credit and the impact that credit would have on use of fertilizer and improved seeds, and on household food security and income. 37. Warehouse Scheme. The Warehouse Scheme, to extend working capital loan finance to agro-industrial enterprises, was not implemented. This is because the institutional, legal, and operational procedures for a warehouse certification scheme, to be implemented through ADMARC, were not developed as required during the project. The TA to advise on this component was never hired and IDA did not provide close guidance during supervision on options for revitalizing this component. A proposal by GOM, supported by IDA, for ADMARC to lease or rent out storage space to private traders in rural areas failed to take off due to the high rental charges that ADMARC demanded. Given ADMARC's opposition to all marketing reforms during this period, and the high rentals that it demanded for storage space, it is doubtful that this activity would have had any chances of success even if it were implemented. RBM 38. PFIs lacked experience in lending to small estates. In addition, the small estates were unable to develop and submit bankable and viable sub-projects. These problems were not adequately reviewed by RBM which also lacked capacity and experience in managing such a line of credit. The Apex Unit was entrusted with this responsibility without any capacity building to enable them to manage this component effectively. Financial Management 39. The various departments and institutions participating in the project have not been able to maintain, reconcile, and prepare project accounts for audit purposes. This was due to the lack of adequately qualified accounting staff to perform these tasks. Further, the project audits themselves were generally delayed, in part due to inadequacies and delays in the receipt of basic accounts and reports and also due to the lack of adequate staff within the audit department. The final audit, for April-June 1996, still remains to be completed. IDA had originally agreed with the Borrower that this audit would be combined with the April 95 - March 96 Audit. However, Government accounting could not accommodate an audit report that overflowed into a second financial year. Government will present a separate final audit, covering the three months from April-June, 1996, by April, 1997. D. Project Sustainability 40. Sustainability. The policy reforms implemented under the project have a high staying power. The reforms have been reinforced by the effect of complementary SALs/SACs and the change in the political system in the country to one that is more supportive of the private sector. The PFIs will be able to maintain the flow of credit to the large estate sector with their own resources and the experience gained under project would assist especially in emphasizing the importance of competent reviews of loan applications and effective supervision of loans. Given  12 the high levels of default on loans extended to the estate sector, it is unlikely that the net worth of the participating commercial banks improved as a result of this project. Providing small estates and private traders access to credit is only partly sustainable and dependent on the implementation of additional measures by GOM and RBM. Both Indefund and SEDOM now have the capacity and expertise to lend viably to small estates and private traders. However, the line of credit extended to these two PFIs under the project did not create a revolving fund. SEDOM and Indefund got annual loans through RBM that were repayable at the end of one lending cycle. Because the funds were not revolved, the amount of resources to be lent to private traders and small estates reduced drastically after the project closed. The market infrastructure component would be sustainable with a sound policy and operational back-up by GOM. There is still an outstanding need to: (i) establish a revolving fund from the repayment proceeds of the local authorities that have already borrowed, perhaps through the Development Fund for Local Authorities, that would finance future market construction and rehabilitation; (ii) devolving greater authority for setting market fees to local authorities; and, (iii) allowing greater private sector participation in the provision of marketing facilities in rural/council markets e.g. weighing scales, trading stalls, shop space, etc. The operation of MBS and the export certification scheme would be improved with increased commercialization of MB S's services, a clear trade policy guiding the certification of agricultural exports that is backed by sound training and publicity of such an exercise, and continued Government financial support, in the short-term, that would be required to re-equip and supply the MBS. However, the sustainability of this service, is at best uncertain given the emergence of private sector and international competition in export certification and product testing. The tobacco industry, which was the main client of MBS testing and certification services, installed in-house facilities to do their own testing to avoid the monopolistic rates that were charged by MBS. 41. Consequent to the restructuring of the project in 1992, a major share of the project funds was reallocated to SACA for urgent drought recovery assistance. The sustainability related to SACA lending operations is jeopardized by poor recovery. The MRFC initiated an action plan for recovery of the overdue SACA loans. This has had only a marginal impact as only MK 15 million, out of an outstanding balance of MK254 million, has been recovered. These recovered funds have become part of MRFC's equity. 42. With regard to the institutional strengthening component, the activities of MPS are expected to continue in view of the GOM commitment to the program of trade liberalization, which received a great impetus under the project. The training program for the rural traders also successfully carried out during the project is expected to be built up under the department's activities.  13 E. Bank Performance 43. The AMED project was conceived under the GOM's Structural Adjustment Program. The marketing development part of the project was identified during the negotiations for the IDA supplement to SAL III while the estate development component was based on the identification work undertaken by FAO/CP under the pilot component of the Industrial and Agricultural Credit Project (Cr. 2646-MAI). 44. Nine supervision missions visited the project, nearly at half-year intervals. Of these, five were joint missions covering two to four related projects in the agricultural sector. At each stage, task plans were drawn up by the missions to facilitate timely action on key items. The more important interventions of these missions related to acceptance of INDEFUND as a PFI to extend loan finance to small estates, the reallocation of funds from the estate (small) sector to SACA as part of the drought recovery mechanism, and the reallocation of funds to SEDOM from residual components. Their interventions also facilitated the introduction of price and market reforms by complementing and deepening measures that were initiated under previous SALs/SACs. These reforms had the tangible impact of widely opening up farm produce trade to private traders. Locating the Task Manager for the project in the local Resident Mission facilitated timely support to various project implementation agencies. Overall, the Bank performance is considered satisfactory. Implementation experience has indicated the need and scope for more effective project monitoring which should have been foreseen at the design stage. F. Borrowers Performance 45. The Borrower's performance in the project preparation was constructive and satisfactory. During implementation it fulfilled its organizational commitments, except in filling up the MPS vacancies for economists. Its commitment and performance in the area of market reforms was highly satisfactory. However, the Borrower did not effectively address the constraints in the flow of credit to small estates. The loan agreements in regard to rural markets between the MOLG and the local bodies remained to be executed. G. Assessment Of Outcome 46. Overall, the project outcome can be considered moderately satisfactory. The project objectives of improving agricultural marketing and developing a multi-channel marketing system as part of market reform policy were substantially achieved. The other objective of estate development through credit has been partially achieved to the extent that several large estates were able to intensify and diversity their production. However, large estates, which were the main loan recipients under the project, would have accessed loan finance for diversification/intensification without the AMED project. The project had limited success in promoting exports and diversification though it did contribute to increased rural employment and incomes. The project contributed significantly to price and marketing policy reform and in the establishment of private trader micro-lending as a viable commercial undertaking. It is noted,  14 however, that the performance of components for market infrastructure, institution building, and credit to estates was not satisfactory and will require further efforts to be considered reasonably successful and sustainable. H. Future Operations 47. It is the GOM's intention to continue the implementation of the market reforms relating to further liberalization of commodity markets and pricing reform. MOALD intends to propagate these policy reforms through a national campaign to sensitize the public. The GOM agreed to fill the MPS vacancies with a view to establishing the capacity required to undertake competent analyses of price and marketing policy and to provide timely market and price information to private traders. GOM will pursue the issue of operational viability of rural council markets through the on-going Local Government Development Project (Cr. 2646-MAI). I. Key Lessons Learned 48. The project demonstrated the need for effective arrangements for project management and monitoring. It is important that projects be grafted into the existing institutional arrangements for portfolio management within a sector without resorting to the creation of 'independent'units. The implementation of the AMED project was undermined by the involvement of such an implementation unit which lacked the necessary seniority, capacity for decision-making and linkages with the relevant Ministries' overall planning and monitoring activities. 49. While demonstrating the viability-of microloans and lending to private traders, the project's experience suggests that commercial banks may not be the most suitable institutions for extending small loans to SMEs and small estates. The experience under AMED suggests that for reaching such clients it is appropriate to work through existing SME lending institutions, as they already have a lending philosophy, staffing, and procedures aimed at extending and managing small loans. The project's experience points to the needed design of revolving fund facilities, with suitable apex institutions, that can continue to finance micro loans after project closure. 50. The project demonstrated the value of client training which helped to impart skills in the preparation and submission of loan applications and in loan management, plus to further familiarize Borrowers with the key (financial) aspects of the businesses for which they are seeking financial support. This is especially important for clients who are borrowing for the first time as routine yet poorly understood procedures may serve as a barrier to accessing loans. At the same time PFIs need to be sensitized to the special features of lending groups, especially those requiring microloans, which require significant departures from conventional lending approaches to include group loans, closer loan supervision, and simplified application procedures. 51. Institutions and policies outside the direct control of the project should be coordinated so that their weaknesses and pace do not undermine the implementation and success of the project.  15 This was brought to bear for the AMED project in the case of the Ministry of Local Government which moved slowly in amending its set of regulations stipulating areas allowed for trading at official minimum prices. The Ministry was also unable to institute a revolving fund that would sustain the financing of local markets after project closure as the AMED was seen as a source of grant funding for the local authorities. Coordination arrangements for multisectoral activities should consider moving towards the creation of "councils" with permanent membership, defined terns of reference, and incentives. 52. In terms of the project design, the barriers to entry into agricultural trade faced by small entrepreneurs should have been examined and reviewed up front. The project played a reactive role in assessing and consulting with Government on policy reforms required to facilitate and support the realization of project objectives. This should, instead, have been a major input at preparation and design, to study and review the extent to which prevailing policies provided an enabling environment for the project, and to recommend the necessary institutional, price, and market reforms.  I  Part II: STATISTICAL TABLES Table 1.: Summary of Assessments Table 2.: Related Bank Credit Table 3.: Project Timetable Table 4.: Credit Disbursements: Cumulative Estimated and Actual Table 5.: Key Indicators for Project Implementation Table 6.: Key Indicators for Project Operation Table 7.: Studies Included in Project Table 8A.: Project Costs Table 8B.: Project Financing Table 9.: Economic Costs and Benefits Table 10.: Status of Legal Covenants Table 11.: Compliance with Operational Manual Statements Table 12.: Bank Resources: Staff Inputs Table 13.: Bank Resources: Missions   19 Table 1: Summary of Assessments A. Achievement of objectives SubstaiaIl Partial Negligible NotApplicable Macro policies I I F I Sector policies II DC Financial objectives RIl D Institutional development D1 E E : Physical objectives II EF D Poverty reduction D 1mF [ Gender issues L D [2 a Other social objectives n Environmental objectives L L LE Public sector management K] EI ol D Private sector development Z LI l D Other (specify) K] K] OI D  20 B. Project sustainability LikeY Unlikely Uncertain (1) (1) (1) Highly C. Bank 2erfonnance satisfactora Satisfactory Deficient (ii (ti) (fi) Identification II L7 LI] Preparation assistance [j] E] [I Appraisal FII Fi [E Supervision LI] Li FI] HIijhlv D. Borrower performance satisfactory Satisfactor, Deficient (V) (/) (1) Preparation [ [I Implementation El [E FI Covenant compliance Fl Li EI Operation (if applicable) El EI Hiehly HighJy E. Assessment of outcome satisfacto Satisfactoyr Unsatisfactorv unsatisfactory (LI Li) El) [  Table 2: Related Bank Credits Loan/credit title Purpose Year of Status approval . Industria and Agricultural CrediTo help meet the sub-sector's MT credit needs and to test mechanism for 1986 Closed (Lnd.u2i64 6- MA I) channeling such credit through existing banking institutions 2. SAL [If Supplement Expand role of private sector in marketing of smallholder crops; improve 1987 Closed financial performance and operational efficiency of ADMARC; strengthen food security planning capability 3. Smallholder Agricultural Credit Project To transform Malawi's smallholder credit system into a better managed, more 1988 Closed (Cr.1851-MAI) efficient, and financially viable rural credit delivery system December 1994 4. Rural Financial Services Project To improve the access to financial services for the rural sector June 1993 Under (Cr.25 1 3-MAI) implementation  22 Table 3: Project Timetable Steps in project cycle Date planned Date actual/ latest estimate Identification' April/May 1987 April/May 1987 Preparation June 1987 Nov. 1987 Final Preparation/Pre-Appraisal Nov. 1987 Jan-Feb. 1988 Appraisal March 1988 May 1988 Credit Negotiations 1988 4-7 October 1988 Board Presentation Nov. 1988 1 Dec. 1988 Credit Agreement Signing 31 March 1989 31 March 1989 Credit Effectiveness 9 June 1989 9 June 1989 Mid-Term Review June 1992 Dec. 1993 Project Completion 31 Dec. 1995 31 Dec. 1995 Loan Closing 30 June 1996 30 June 1996 " The marketing element of the project was identified during negotiations of the IDA-financed supplement to SAL III. The support for estate development was included during preparation based on identification work undertaken by FAO/CP.  23 Table 4: Credit Disbursements: Cumulative Estimated and Actual FY89/90 FY90/91 FY91/92 FY92/93 FY93/94 FY04/95 FY95/96 ............................................................... (USS million) .......................................................................l Appraisal estimate (cumulative) 3.3 6.0 8.8 11.3 13.7 16.0 18.3 Actual (cumulative) 1.9 3.8 6.8 14.7 16.8 18.6 19.8 Actualas