Tanzania - Agricultural Export Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 16204-TA IMPLEMENTATION COMPLETION REPORT TANZANIA AGRICULTURAL EXPORT REHABILITATION PROJECT (CREDIT 1891-TA) December 30, 1996 Agriculture and Environment Operations 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 Jun-96 US$1.00 = T Sh623.3 Jun-95 US$1.00 = T Sh604.8 Jun-94 US$1.00 = T Sh516.2 Jun-93 US$1.00 = T Sh395.0 Jun-92 US$1.00 = T Sh 300.0 Jun-91 US$1.00 = T Sh227.0 Jun-90 US$1.00 = T Sh193.2 Jun-89 US$1.00 = T ShI45.0 ABBREVIATIONS AND ACRONYMS AERP Agricultural Export Rehabilitation Project BOT Bank of Tanzania CRDB Cooperative and Rural Development Bank ERP Economic Reform Program GOT Government of Tanzania IPC Investment Promotion Centre MOF Ministry of Finance SPU Special Programs Unit FISCAL YEAR OF BORROWER Government of Tanzania July I - June 30 Cooperative and Rural Development Bank July I - June 30 until June 30, 1994 January 1 - December 31 from January 1, 1995 Vice President C. Madavo Country Director J. Adams Technical Manager S. Ganguly Task Team Leader A. Thompson  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT TANZANIA AGRICULTURAL EXPORT REHABILITATION PROJECT CREDIT 1891-TA TABLE OF CONTENTS Preface ...................................................... i Evaluatio n Su mm ary . . .. ................................................................. PART I: PROJECT IMPLEMENTATION ASSESSMENT I A. Statement/Evaluation of Obbectives B. Achievement of Objectives .2 Summary.2 Agricultural Rehabilitation and Expansio .3 Institutional Development .5 Economic and Financial Objectives .6 C. Major Factors Affecting the Project ..7 D. Project Suusainabbiltty.8 E. Bank Performancee.9 F. Borrower Performance .10 G. Future Operations..1l H. Assessment of Outcome..11 I. Key Lessons Learned.l12 PART II: STATISTICAL ANNEXES Table 1: Summary ofAsssseesmtt.13 Table 2: Related Bank Loans/Credits1 15 Table 3: Project Timetable1 15 Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual.16 Table 5: Key Indicators for Project Implementation.1 6 Table 6: Key Indicators for Project Operation.1 6 Table 7: Studies Included in Project.1 7 Table 8A: Project Costss 18 | This documcnt has a restricted distribution and may be used by recipients only in the performance of theirl offioEcial duties. Its contents may not otheruise be disclosed wiihout World Bank authorization.l F . B r o e e f r a c ..............................................................  Table 8B: Project Financiig .......................................... 19 Table 9: Economic Costs and Benefits .......................................... 20 Table 10: Status of Legal Covenants .......................................... 21 Table 11: Compliance with Operational Manual Statements .......................................... 24 Table 12: Bank Resources: Staff Inputs .......................................... 24 Table 13: Bank Resources: Missions .......................................... 25 APPENDICES A ........................................... Aide Mmoire B ........................................... B orrower's Contribution to the ICR C .......................................... AERP Portfolio Tables -T  Preface This is the Implementation Completion Report (ICR) for the Agricultural Export Rehabilitation Project (AERP) in Tanzania, for which IDA Credit 1891 -TA in the amount of SDR 22.0 million (US$30.0 million equivalent) was approved on 29 March 1988, and became effective on 14 November 1988. The Mid-Term Review took place in May 1995 and the credit was closed on 30 June 1996 at which time a balance of SDR 9.2 million was cancelled. This ICR was prepared by a mission' from the FAO/World Bank Cooperative Programme, on behalf of the Africa Technical Division Agriculture Group I (AFTA I) of the World Bank. Preparation of the ICR commenced with the mission's visit to Tanzania from 25 August to 14 September 1996. Comments were provided by Tony Thompson, James Coates, Fred King and Jeri Larson (AFTA 1) and Jacomina de Regt, AFTSI. This report is based on material gathered from the project files and supervision reports, as well as discussions with Bank and Government of Tanzania (GOT) staff associated with the project. In addition, the mission undertook intensive discussions and field investigations with staff of the implementing agency, the Cooperative and Rural Development Bank (CRDB) now privatised and known as CRDB (1996) Limited, and sub-borrowers of the credit for export crop rehabilitation purposes. CRDB staff contributed to the preparation of the ICR through the arrangements for field visits and meetings, and provided useful comments on the mission's Aide-Memoire. The ICR was reviewed by Sushma Ganguly, Technical Manager, Agricultural Operations 1, Eastern and Southern Africa. 'The mission comprised Tim Lamrock (FAO TCIR - Economist/Mission Leader) and Peter Beeden (Agronomist - Consultant). The ICR for the Cashew and Coconut Tree Crops Project (Cr 2050-TA) was prepared during the same mission.  I  IMPLEMENTATION COMPLETION REPORT TANZANIA AGRICULTURAL EXPORT REHABILITATION PROJECT (AERP) (Cr. 1891-TA) Evaluation Summary Introduction The Agricultural Export Rehabilitation Project was conceived in the mid-1980s at a time when the financial sector in Tanzania was performing poorly due to lack of competition, excessive interest rate controls and minimal bank supervision. Credit was directed on the basis of Government priorities without regard to creditworthiness, and banks were convenient agents of fiscal policy. Non-bankable activities were financed, with a misallocation of resources to the parastatals and agricultural co-operatives. This resulted in a large accumulation of non-performing loans (around two-thirds of the total) and by 1988 the banking system was technically insolvent. At that time, there was a critical shortage of foreign currency to import the machinery and consumables necessary for Tanzania's traditional export agricultural industries. The project formed part of the World Bank's development strategy for Tanzania and was designed to support the Government of Tanzania's (GOT) objectives for the provision of medium- and long- term assistance for the rehabilitation and expansion of agricultural enterprises. The institutional component also addressed a major GOT objective; namely the strengthening of the agricultural lending capability of financial intermediaries, in this case of CRDB. The project was signed in August, 1988 and became effective in November, 1988. Project Objectives The objectives of the project were (i) to promote agricultural exports in the traditional crop sector (coffee, tea, sisal, cashewnuts, cotton and tobacco) through a medium/long-term credit for rehabilitation, thereby increasing foreign exchange earnings; (ii) to provide greater emphasis to market forces to improve the efficiency of resource allocation and, (iii) to advance institution building by strengthening the capability of the Cooperative and Rural Development Bank (CRDB) to engage in medium- and long-term lending in the agricultural sector. Implementation Experience and Results The first subloans under the project were approved by May 1989, and disbursements for the lending program commenced in August, 1990. A total of 23 sub-projects were financed under the project. ii  Approved loans from CRDB to sub-borrowers under AERP totalled US$ 15.4 million or 57% of the projected level of US$ 26.7 million. A recognised deterioration in performance of the loan portfolio, coupled with the real isation that CRDB and not the Government, was liable for any losses incurred, led IDA to advise CRDB in February 1994 to discontinue the processing of new loans, and to review each loan on a case-by-case basis before further disbursements were made. No new loans were approved after December 1993, and disbursements on existing commitments effectively ceased by the end of 1994. The Mid-Term Review took place in May 1995 and the project closed on 30 June 1996. US$ 12.1 million of the original US$ 30 million (the equivalent of SDR 9.2 million of the original SDR 22 million) was cancelled. Achievement of Objectives The project partly achieved its development objective of foreign exchange lending to export- oriented agricultural investment, and the project provided a valuable adjunct to the Investment Promotion Centre (IPC) scheme established by GOT in 1990 which was designed to attract investment to Tanzania by offering tax exemptions on imported capital items. The project was closed at the end of June 1996, in line with the schedule foreseen at appraisal, although the credit disbursements to agricultural producers did not reach the anticipated level. The intention to on-lend to co-operative unions and parastatals did not materialise, largely due to their lack of credit worthiness. Consequently all approved sub-loans were to the private sector. The project was expected to lend to enterprises involved in all the traditional export crops (cotton, tea, sisal, coffee and tobacco). However, due to lack of liberalisation in the marketing of some of these crops, the coverage of the project was widened to include non-traditional export enterprises. The institutional strengthening objectives were not met. The impact on the restructuring of CRDB and on its overall financial and organisational management was negligible and by 1992 institutional change aspects of the project were left to the Financial Sector Adjustment Credit, which was dealing with all financial institutions, including CRDB. Considering the failure of the previous IDA efforts to restructure CRDB (Credit 987-TA), and a relative lack of success in this area of the subsequent IDA-supported Financial Sector Adjustment Credit (FSAC - Credit 2308-TA), it is perhaps unreasonable to have expected significant improvements without the accompaniment of serious political and CRDB management support for such changes. In the event, it required the enactment of the legislation associated with the Banking and Financial Institutions Act (1991), the involvement of DANIDA in CRDB's affairs from late-l 992, and the strengthened supervision of banks by the BOT, for significant achievements to be made in the restructuring of CRDB. Sustainability The project is unlikely to be suustainable2. Under CRDB's present policy of decentralization. it is intended that staff at branch level will monitor and supervise the AERP portfolio, including the restructuring of loans where appropriate. However, the branches have neither the capacity nor capability to undertake these tasks, and unless reinforced with staff experienced in handling major and complex loans, are likely to fail to effectively manage the difficult task ahead of ensuring eventual loan repayments. 2 The Borrower rates the sustainability as likely, a view the Bank does not agree with. iii  The performance of the individual sub-borrowers has been higlhly variable. The floriculture enterprises appear already in profit, while those for sisal, which is of much longer term development, require additional working capital resources to be successful. Tea enterprises are intermediate and could reach profitable levels within the next 2 years. Apart from the 7 sub-borrowers under litigation, there is sufficient equity in the remaining enterprises to provide substantial motivation for their private sector owners. However, CRDB will need to determine rapidly the appropriate approach (loan restructure or the legal option) for the recovery of loans from each sub-borrower. Project Costs Total project costs were estimated during appraisal at US$ 40 million, to be financed as follows: IDA credit US$ 30 million (SDR 22 million equivalent), CRDB US$ 2.0 million and sub-borrowers US$ 8.0 million. Foreign exchange costs were estimated at US$ 28.0 million or 70% of total project costs. Loans were disbursed to 23 sub-borrowers totaling US$ 15.4 million, equivalent to 57% of the projected level of US$ 26.7 million. IDA's disbursements total US$17.9 million. Categorized and itemized project costs are not available from CRDB, despite an external audit of AERP's accounts for the period ending 30 June 1996, and their subsequent certification by the CRDB Board of Directors. There has been no attempt to assess the contribution to total project costs by either the sub-borrowers, CRDB or GOT. Major Factors Affecting the Project Most significantly, the project was negatively affected by the lack of substantial organizational and financial management strengthening in CRDB during the period of loan disbursement, and the lack of progress in building appraisal capacity and capability during the period of activity of sub- loan appraisal and supervision between 1989 and 1994. In addition the project was negatively impacted by both sector and macroeconomic policies during the period of implementation. These tended to make investments less attractive and also affected the viability of the financial system. Bank Performance The project was adequately designed to address the foreign exchange issues that were limiting further investment in the agricultural sector at the time. However, the necessary prerequisites for the reform of CRDB, the major factor for project success, were underestimated. There were several assumptions in the project design which proved to be invalid. A majority of potential borrowers identified in the SAR were co-operative unions or parastatals, recognised as being commercially unviable, yet forming the basis of the estimated credit demand, and with no proposals under the project or other concurrent programmes to make the necessary changes. Over the 92 month project period, only nine supervision missions visited the project. Supervision was satisfactory until November 1991, after which no full supervision was undertaken from headquarters until February 1994. Some oversight of the project was taken on by the resident mission during this period, but in retrospect this informal follow-up, which focused on subloan administration, iv  was insufficient as a majority of sub-loan activities were undertaken during this period, the mid-term review was rescheduled, CRDB restructuring commenced, and the marketing and foreign exchange reforms were put in place. This was a critical time for the project and inadequate levels of monitoring and advice were provided to CRDB. From February 1994 however, substantial efforts were made during and between supervisions to restructure the project, by limiting the exposure of CRDB and assisting in better management of the existing portfolio. Borrower Performance The project had a history of either delayed compliance or noni-complianlce of the assurances and covenanits which were central to the project's design. In addition, there were delays in preparing and submitting reports and accounts that were required by IDA under the project agreements, and these appeared not to form part of the routine management information system required by CRDB. Project Outcome The overall outcome of the project is considered unsatisfactory due to the poor loan repayment record, the uncertain repayment outlook for the tea and sisal sub-loans, the large number of loans now subject to legal proceedings, and the negligible impact on the restructuring of CRDB. Summary of Findings, Future Operations and Key Lessons Learned As part of the continuing Programme of CRDB privatization, the Ministry of Finance (MOF) has agreed to assume responsibility for loan liabilities under AERP, as well as those of the two IFAD- supported lines of credit and the Pilot Credit Scheme under the IDA-supported Cashew and Coconut Treecrops Project that have been implemented by CRDB. The CRDB expects to act as the agent for loan recovery, and details of the Memorandum of Understanding involving the financial arrangements are currently under negotiation. It would be in the Government's interest to ensure that CRDB is adequately rewarded in its capacity as management agent, so that CRDB is provided with sufficient motivation to pursue sub-loan repayments. The key lessons learned from the project were as follows: it was unreasonable for the Bank preparation/appraisal missions to have expected significant restructuring of CRDB without the political will that was not evident until after the enactment of the Banking and Financial Institutions Act (1991); the negotiation of assurances and covenants between IDA and GOT is only meaningful if IDA is prepared to take action in the event of subsequent non-compliance; the disruptive effect of the eventual CRDB restructuring on staff morale, continuity and efficiency was known at appraisal, and this contributed to the poor monitoring and supervision of sub-projects; v  the international technical assistance personnel were directly involved in project implementation and contributed little to the capacity of CRDB, to the extent that CRDB staff derived little benefit from their presence; CRDB was able to make multiple loans to individual sub-borrowers, and to considerably exceed the individual loan limits established at appraisal, with insufficient evidence of management commitment. This contributed substantially to CRDB's exposure to risk of loan default; sub-borrowers with substantial and commercially operating enterprises, require a timely and efficient banking service for effective and profitable business operation. CRDB was not able to provide this service; the mid-tenn review should have been undertaken at its appointed time in PY 3, particularly as the AERP performance had already been rated by IDA as unsatisfactory; supervision missions for agricultural credit programmes should include staff with technical/agronomic skills. so to be better able to assess agricultural businesses; and the absence of headquarters supervision from November 1991 until February 1994, to complement the monitoring being undertaken by the resident mission at a critical period in both sub-borrower development and CRDB restructuring, may have contributed substantially to the poor performance of the project. vi  I i  IMPLEMENTATION COMEPLETION REPORT TANZANIA AGRICULTURAL EXPORT REHABILITATION PROJECT (AERP) (Cr. 1891-TA) PART I: PROJECT IMPLEMENTATION ASSESSMENT A. STATEMENT/EVALUATION OF OBJECTIVES I. The objectives of the project were (i) to promote agricultural exports in the traditional crop sector (coffee, tea, sisal, cashewnuts, cotton and tobacco) through a medium/long-term credit for rehabilitation, thereby increasing foreign exchange earnings; (ii) to provide greater emphasis to market forces to improve the efficiency of resource allocation and, (iii) to advance institution building by strengthening the capability of the Cooperative and Rural Development Bank (CRDB) to engage in medium- and long-term lending in the agricultural sector. 2. The two project components designed at appraisal, comprised: (a) a line of foreign exchange credit (US$ 26.7 million) managed by CRDB and financed under a subsidiary loan agreement between CRDB and the Ministry of Finance, designed to: (i) provide long-term financial support for the rehabilitation and expansion of agricultural enterprises (sub-borrowers) in the traditional export crop sector and to improve the quality of Tanzania's exports; and (ii) provide foreign exchange to finance capital investment and up to three years of the recurrent import requirements (excluding seasonal inputs) of sub-borrowers. (b) institutional building in CRDB (US$ 3.3 million) including technical assistance, staff training, equipment and studies, with a view to: (i) restructure CRDB and transform it into a viable agricultural financial intermediary; and (ii) advance the process of agricultural reform in the sector and enable co-operative unions, parastatal agencies and the private sector to take advantage of the recent liberalisation of agricultural input supply and distribution, export crop marketing and producer pricing. 3. The project formed part of the World Bank's development strategy for Tanzania and was designed to support the Government of Tanzania's (GOT) objectives for, amongst other things, the provision of medium- and long-term assistance for the rehabilitation and expansion of agricultural  enterprises. The institutional component also addressed a major GOT objective; namely the strengthening of the agricultural lending capability of financial intermediaries, in this case of CRDB. 4. The AERP was conceived in the mid- 1 980s at a time when the financial sector in Tanzania was performing poorly due to lack of competition, excessive interest rate controls and minimal bank supervision. Credit was directed on the basis of Government priorities without regard to creditworthiness, and banks were convenient agents of fiscal policy. Non-bankable activities were financed, with a misallocation of resources to the parastatals and agricultural co-operatives. This resulted in a large accumulation of non-performing loans (around two-thirds of the total) and by 1988 the banking system was technically insolvent. 5. The Government introduced an Economic Reform Programme (ERP) in 1986 to improve macroeconomic management, tackle underlying structural weaknesses and encourage more active private sector participation. Economic measures designed to reduce the budget deficit and improve the balance of payments included a substantial depreciation of the overvalued exchange rate, trade liberalisation, the removal of most price controls and the easing of restrictions on the marketing of most food crops. Nevertheless, foreign exchange remained in short supply, severely constraining the essential rehabilitation of export enterprises. 6. Following the Commission of Enquiry of 1988 into the Monetary and Banking System, the Government acknowledged the perverse impact on the sector caused by a lack of competition and Government interference, and through the enactment of the Banking and Financial Institutions Act (1991) permitted the entry of private banks, with an expanded and enhanced supervisory and regulatory control vested in the Bank of Tanzania (BOT). B. ACHIEVEMENT OF OBJECTIVES Summary 7. The project partly achieved its development objective of foreign exchange lending to export- oriented agricultural investment, and a valuable adjunct was provided to the Investment Promotion Centre (IPC) scheme established by GOT in 1990 which provided tax exemptions on imported capital items, and was designed to attract investment to Tanzania. The project was closed at the end of June 1996, in line with the schedule foreseen at appraisal, although the credit disbursements to agricultural producers did not reach the anticipated level. The intention to on-lend to co-operative unions and parastatals did not materialise, largely due to their lack of credit worthiness. Consequently all approved sub-loans were to the private sector. In addition, sub-borrowers were expected from the traditional export crop sector (cotton, tea, sisal, coffee and tobacco) which, in the event, were unable to absorb the available credit, and in 1989 project objectives were widened to include non-traditional export enterprises. 8. The institutional strengthening objectives were not met. The impact on the restructuring of CRDB and on its overall financial and organisational management was negligible and by 1992 institutional change aspects of the project were left to the Financial Sector Adjustment Credit, which was dealing with all financial institutions, including CRDB. Considering the failure of the previous IDA efforts to restructure 2  CRDB (Credit 987-TA), and a relative lack of success in this area of the subsequent IDA-supported Financial Sector Adjustment Credit (FSAC - Credit 2308-TA), it is perhaps unreasonable to have expected significant improvements without the accompaniment of serious political and CRDB management support for such changes. In the event, it required the enactment of the legislation associated with the Banking and Financial Institutions Act (1991), the involvement of DANIDA in CRDB's affairs from late-1992, and the strengthened supervision of banks by the BOT, for significant achievements to be made in the restructuring of CRDB. Agricultural Rehabilitation and Expansion 9. The project sought to provide finance for the capital (machinery, equipment, spare parts and vehicles) and recurrent costs (although not for seasonal costs) required to rehabilitate or expand existing agricultural export enterprises, through lending to co-operative, parastatal and private firms which were creditworthy. In line with the project design of expanding the capability for term lending by existing institutions, 3 financial institutions were identified prior to appraisal, from which CRDB was finally selected as the sole implementing agency for the project line of credit. Under the Subsidiary Loan Agreement between CRDB and GOT in which CRDB assumed responsibility for sub-loan liabilities, CRDB on-lent at a fixed rate of interest of 8% in US dollars for the term of the sub-loan, retaining 6% to cover credit risks, administrative costs, margins and taxes, with the balance of 2% accruing to GOT. Exchange rate risks were bome by the sub-borrower, whose repayments in foreign exchange came from an escrow account established specifically to handle export receipts. 10. An initial pipeline of 38 sub-projects was identified at appraisal of which only five were creditworthy and eventually participated. The co-operative unions and parastatals that were to form a substantial proportion of project beneficiaries, could not be supported due to their ineligibility on the grounds of substandard management and inadequate creditworthiness, largely as a result of their past excessive and non-retumed borrowings. A large number of proposals for rehabilitation in the cotton and coffee industries were also rejected, due to a lack of progress with market liberalisation in the cotton sector. Ginnery operators continued to be prevented from direct export and thus retention of foreign exchange, and the Govemment continued to use coffee revenues as part of a barter exchange for petroleum products. It was therefore necessary to redefine the project's investment focus to include both traditional and non- traditional agricultural exports, as well as start-up ventures, and to revise upwards the loan limits. Following this change in 1989. there was a delay before new, non-traditional sub-projects were designed, appraised and approved, the majority of which were in cut flower and other horticultural production in the Arusha and Kilimanjaro regions. 11. With an initial Project Preparation Facility (PPF) that enabled feasibility studies to be undertaken for those enterprises identified at appraisal as potential borrowers, some 84 sub-projects were put forward for consideration by CRDB3 and either approved (23), rejected on non-viability grounds (27), loans approved but not taken up (7), or were in the pipeline at the time of the closure on new loans in 1994 (27). A considerable proportion of CRDB staff time was devoted to the evaluation and appraisal of new sub-projects that failed to result in further loan disbursements, rather than to the supervision and monitoring 3 Including: tea 12; sisal 10: cotton 18: tloriculture 5; horticulture 9: coffee 9; honey/beeswax 2; tobacco 2; miscellaneous agro-processing 8; aquaculture 3; and miscellaneous 6. 3  of existing loans. This may have enabled approved sub-projects to deteriorate beyond the point at which disbursements should prudently have been curtailed. Although sub-project appraisals were generally of adequate quality, there was a tendency towards over-optimistic yield and production assumptions, with too little attention given to the availability of the working capital contributions of individual sub-borrowers. This was to have important consequences, particularly for the sisal sub-sector. 12. To assist CRDB to fulfil its lending objectives, the project provided resources for internationally recruited technical assistance, which included a project accountant (50 months), financial analyst (33 months) and a project officer (22 months). An industrial engineer was anticipated at appraisal, but was not recruited. Terms of reference of the technical assistance personnel were not available in the SAR, and were not developed by CRDB. The consultant team was involved mainly in sub-project evaluation and appraisal. The project accountant frequently acted for extended periods as the Chief Accountant for CRDB and was unable to develop an effective system for the allocation of project costs and expenditures. There were three incumbents of the Financial Analyst position, including a banking specialist inexperienced in project appraisal and evaluation. 13. Feasibility studies totalling some US$ 539,000, were undertaken through short-term contracts with local and international consultants, probably equating to the 120 months anticipated in the SAR. However, none of the 24 months of short-term consultancies for management assistance to sub-borrowers was undertaken. Adequate vehicles and office equipment were supplied to CRDB as envisaged. 14. Loans were disbursed to only 23 sub-borrowers (Appendix C-I, Table 1), as against some 100 anticipated at appraisal, with total of US$ 17.8 million approved, of which only US$15.4million was disbursed, equivalent to 57% of the projected level of US$ 26.7 million. 15. It became clear as early as 1991 that the sub-loan portfolio was of mixed quality; with an accumulation of interest arrears, apparently without action being taken, as well as principal defaults by many sub-borrowers. In February 1994, IDA advised CRDB to discontinue the processing of new loans, and to review each existing sub-loan on a case-by-case basis before further disbursements were made. Recognising the problems within the portfolio, CRDB complied and no new loans were approved after December 1993. Disbursements on existing commitments effectively ceased by the end of 1994 although some smaller disbursements were made up to the end of the project. 16. It was also at this time that CRDB realised that it, and not the Government, was liable for any losses incurred under the project. The accounts of CRDB never reflected the AERP portfolio, or other special donor-supported credit programs up until this point. 17. The average loan size was some US$ 772,500, considerably in excess of the US$ 500,000 indicated as the maximum size of foreign exchange sub-loan at appraisal. The loan ceiling was subsequently increased as major capital investments were put forward, and there were 4 loans approved of US$ I million or more. In line with the SAR, the purposes of sub-loans advanced included mainly the development and rehabilitation of on-farm infrastructure (agro-processing equipment, growing houses and storage facilities), purchase of field equipment (tractors, implements, graders and bulldozers), and vehicles. 4  18. Sub-loan repayment performance has been highly unsatisfactory (Appendix C-I, Table 2). At project completion, only 8 of the 23 disbursed sub-loans were not in default, and 7 of these were still within the grace period. Seven sub-projects, with loans totalling some US$ 5.2 million, were the subject of legal action by CRDB and of the remaining projects, 15 were at the least in default of interest repayments. The result is a portfolio of mainly non-performing loans that require review and where necessary, restructure or legal action to recover assets. Institutional Development 19. The objective of strengthening CRDB through a restructuring process was not achieved as a result of the project. Institution building was anticipated to derive from the development and implementation of a program to restructure CRDB, and the provision of management assistance and training to strengthen CRDB's project promotion and appraisal capability. As an initial measure, funds from a PPF of US$ 1.5 million were available for studies to develop a Monitorable Action Programme for CRDB's restructure, that included: organisational and financial restructure with a financial plan with target liquidity ratios, rates of return on assets and equity, profit margins, and the development and implementation of strategies for portfolio and credit risk management, including operational policy. These ambitious targets were to be achieved through: a detailed portfolio review by independent auditors to recommend changes in loan loss reserves and an appropriate level of provisions; and a study by banking specialists to develop a phased plan for restructuring CRDB. In addition, a financial information system study was expected to develop an effective system appropriate for CRDB's current and future needs and determine the training requirements of bank staff. There is no evidence that these PPF studies played any useful role in the restructuring of CRDB that has taken place since the early 1990s. 20. The training program for CRDB staff was abbreviated, unstructured, and failed to gain IDA approval. None of the anticipated 60 months of overseas fellowship training in appraisal techniques was provided, and study tours were restricted to some 8 months, rather than the 24 months envisaged at appraisal. On-the-job training was limited to that provided by the technical assistance group, rather than by the University of Dar es Salaam as indicated in the SAR. Computer training to CRDB staff was provided however, and an accounting system was developed by short-term consultants for use throughout CRDB. This system is only now being replaced as CRDB decentralisation is implemented. 21. The project was designed so that the experience gained through the operation of a line of medium- to long-term credit, supported by technical assistance in portfolio management, would filter through to the remainder of CRDB lending operations. This implementation strategy became ineffective with the approval by the IDA supervision of October 1989, of the formation of a special AERP implementation unit to maintain the staff, funds and accounting for AERP activities separate from mainline bank programmes, pending CRDB's financial and organisational restructure. This major shift in strategy may have contributed to two outcomes: first, the failure to transfer project appraisal experience to other banking operations; and second, the misconception that the liabilities incurred under the AERP were not those of CRDB and thus need not be recorded in CRDB's annual accounts. This misconception was not corrected until the Supervision Mission of February 1994, and may have contributed to a lack of attention from senior CRDB management to the discipline required to successfully appraise and monitor the sub- projects. For example, no effective system of sub-loan supervision appears to have been designed or 5  implemented, with the result that the financial condition of some sub-borrowers was allowed to deteriorate alarmingly before being fully appreciated and remedial action taken. 22. The eventual restructuring and privatisation of CRDB led to reduced staff resources being dedicated to AERP matters; with 8 staff in the AERP Unit in 1989 reducing to 4 in 1994. Ten professionals were anticipated at appraisal to be required to manage the AERP issues, in addition to existing staff of CRDB's Planning and Development Department and Development Banking Department. This level of staffing was never achieved. Currently only 2 of the Special Programme Unit (the successor to the AERP Unit in the restructured CRDB), are fully experienced in AERP portfolio management. Economic and Financial Objectives 23. Total project costs were estimated during appraisal at US$ 40 million, to be financed as follows: IDA credit US$ 30 million (SDR 22 million equivalent), CRDB US$ 2.0 million and sub- borrowers US$ 8.0 million. Foreign exchange costs were estimated at US$ 28.0 million or 70% of total project costs. 24. Categorised and itemised project costs are not available from CRDB, despite an external audit of AERP's accounts for the period ending 30 June 1996, and their subsequent certification by the CRDB Board of Directors. Discrepancies exist in the documentation within CRDB, and there has been no attempt to assess the contribution to total project costs by either the sub-borrowers, CRDB or GOT. 25. The overall economic rate of return was not calculated at appraisal. However, the economic justification for the project was that only sub-loans with ERRs in excess of 15% were to be approved. Whilst the sub-loan appraisal process involved an ERR assessment and each satisfied the 15% cut-off point, the sensitivity to major physical and price assumptions was high in many cases. An attempt was made to reassess the ERR of the investments which contributed to on-going businesses but, in these cases, a true assessment of the effect of the project supported investment was not possible for two reasons: (i) it was not possible to determine the extent of the borrower's contribution to the investment and therefore the total sub- project cost; and (ii) further investments had been made since the original project-financed investment, financed from a mixture of external debt, equity, or from internally generated funds, which made it impossible to identify the benefits deriving from the original investment alone. 26. Whilst the impact on1 foreign exchange earnings was not estimated in the SAR, two factors may have depressed the net foreign exchange benefits that were anticipated at appraisal. First, a malority of the individual sub-projects have not been able to generate the foreign earnings as anticipated due to marketing and management problems. This applies particularly to tea and sisal enterprises, for whichi world market prices have decreased. Second, almost one-half of sub-project investments were in floriculture enterprises that have operating costs with a higher contenit of foreign exchange than the traditional export agriculture enterprises. 27. Of the traditional export crops envisaged, only tea and sisal were included in the portfolio with 5 and 8 loans respectively, representing 14% of the total disbursements in the case of tea and 27% for sisal. However, since several loans were given to individual sub-borrowers, these represent only 7 effective borrowers, considerably increasing the exposure of the portfolio to risk of default. Apart from a silgle loan 6 1  that has been repaid in full, all loans are in arrears of interest or capital repayments or both. Loans for sisal rehabilitation are of particular concern, given the labour-intensive nature of the industry, the long period between crop establishment and harvest, and the poor world market outlook. Four sub-loans, totalling US$ 2.5 million were approved for one newly-formed sisal producing company without an agricultural track record, and insignificalit quantities of fibre have been produced from these four estates in recent years. 28. Of the 10 sub-borrowers il the non-traditional export crop category, 4 are producers of cut flowers (mainly roses) in Arusha and Kilimanjaro, for export to Europe, with loans totalliig US$ 5.3 million or 34% of the AERP portfolio. These represent AERP's most successful venture that, with CRDB's aggressive promotion, attracted investors from the much larger flower producing areas of neighbouring Kenya, throughi cheap loans and with IPC tax exemptions as an additional incentive. This potentially profitable sub-sector has a high associated risk however, as sub-borrowers appear to prefer to expand their production rather thaln repay loans. Sub-projects in vegetable horticulture and essential oil production have failed, and those in crocodile farming, seaweed production and fishing, have yet to prove their viability. This demonstrates the inherent risk in bank lending to new enterprises for which venture capital might be a more appropriate source of investnent finance. 29. The project has made a substantial contribution through the increase in local employment opportunities, particularly for women and youth in the sisal, tea and floriculture enterprises (Appendix C-I, Table 3). An estimated 10,000 additional opportunities have been created directly as a result of AERP activities. C. MAJOR FACTORS AFFECTING THE PROJECT Project Management 30. CRDB. Although the restructuring of CRDB was a major objective of AERP and was achieved by means other than through the project, little attention was given at appraisal to the effect on the operation of CRDB in general, and of the AERP Unit in particular, of the restructuring process itself. Over the period 1992 to 1995, the CRDB staff complement was reduced from almost 1,600 to less than 500, and all staff reapplied for positions within the restructured bank, with an inevitable negative impact on staff morale and performance. With management attention directed elsewhere, AERP activities proceeded with little professional guidance or monitoring from senior CRDB personnel. The resulting shortfall in AERP Unit staff numbers and lack of staffing continuity, has contributed to the project's unsatisfactory performance. 31. Sub-project Approval. Considerable delays were experienced in the feasibility, appraisal and approval of loans to sub-borrowers, caused variously by inadequate staff capacity for appraisal and evaluation, and the time consuming process of approval by the Credit Committee, "no objection" from IDA, and final approval by CRDB's Board of Directors. After loan approval, the disbursement procedure also proved cumbersome and time consuming. There were considerable delays in the opening of letters of credit for the provision of capital items, which may have contributed to the ultimate poor performance of some sub-borrowers. Although CRDB's selected external bank recognised the delays and made some efforts to expedite disbursements, this ultimately proved ineffective. CRDB assistance to sub-borrowers in negotiating these procedures appeared minimal, although one workshop for sub-borrowers was held and a 7  simple guideline prepared. However, the performance of CRDB as implementing agency was not aided by the disruption in the early 1990s when CRDB reorganisation occurred and almost two-thirds of staff were retrenched. 32. Sub-Loan Supervision. In many cases there was a lack of clearly defined implementation arrangements in the sub-loan agreements that led to borrowers not being subjected to pressure to follow agreed repayment schedules. A number of borrowers continued to draw down after the expiry of the grace period under their loan agreements, and in some cases the equipment purchased was different from that specified in the proposal. This can be attributed directly to the inadequate supervision of individual sub- projects during the implementationi phase. In many cases, little if any attempt has been made to enforce sub-borrower compliance with loan conditions agreed at negotiation. This may have contributed to a late appreciation by CRDB of poor sub-borrower performance, as regular reporting procedures and CRDB's presence on1 company boards have not been implemented. 33. Delay in Mid-term Review (MTR). At appraisal, the MTR for the project was scheduled for the end of June 1991. This was considered to be essential as particular attention would be paid to the progress of institutional reforms, with a major review setting out an action program for the remainder of the project period. CRDB management requested a postponement of the MTR so that the conclusions of the portfolio audit and assessment of CRDB's financial position to be carried out under the Financial Sector Restructuring Project (FSRP) under FSAC could be incorporated. A second postponement was agreed by IDA to avoid the period of major changes in CRDB which started during 1993. The review was eventually carried out in May 1995, some 4 years after its original schedule and 7 years after project initiation, and far too late to influence the course of the project. 34. Given the risks inherent in the project that were identified at appraisal, the changes to the focus and implementation mechanism of the credit program that had already been necessary during the early stages of the project, and the recognition by World Bank supervision missions that the progress of project implementation was unsatisfactory, the MTR should have taken place as planned. This would have enabled the implementation issues to have been addressed, particularly those relating to the poor performance of the loan portfolio. As it was, the MTR took place after new loan disbursement had ceased, the technical assistance program had been completed and the project was effectively over. Macro and Sectoral Policies 35. The project was being implemented at a time of general deterioration of the Tanzanian economy as reflected in worsening terms of trade, prolonged drought, insufficient producer incentives and resources being allocated to the agricultural sector. There was a deteriorating balance of payments and liberalization of prices and controls were not carried out. Fluctuations in the exchange rate adversely affected CRDB's operations and the financial viability of the subborrowers. Within the agricultural sector, there were continued price controls, marketing restraints and other insufficient producer incentives. Agricultural input supplies were deficient. The cyclical nature of international prices also adversely impacted the producers of primary products. In this environment, shortfalls in project design or implementation were only magnified. 8  D. PROJECT SUSTAINABILITY 36. The project is unlikely to be suustainable4. Under CRDB's present policy of decentralisation, it is intended that staff at branch level will monitor and supervise the AERP portfolio, including the restructuring of loans where appropriate. However, the branches have neither the capacity nor capability to undertake these tasks, and unless reinforced with staff experienced in handling major and complex loans, are likely to fail to effectively manage the difficult task ahead of ensuring eventual loan repayments. 37. Currently only two CRDB staff in the Special Programme Unit (SPU) have been continuously involved in sub-loan appraisal and evaluation. There is little assurance of a continuity of staff involvement in the AERP portfolio for the future. Few other CRDB personnel possess the necessary appraisal skills for large-scale investment programmes, despite the current and on-going staff training programmes associated with decentralisation. In addition, the computer expertise essential for rapid enterprise evaluation is not available at the branch level. 38. The performance of the individual sub-borrowers has been highly variable. The floriculture enterprises appear already in profit, while those for sisal, which is of much longer term development, require additional working capital resources to be successful. Tea enterprises are intermediate and could reach profitable levels within the next 2 years. Apart from the 7 sub-borrowers under litigation, there is sufficient equity in the remaining enterprises to provide substantial motivation for their private sector owners. However, CRDB will need to determine rapidly the appropriate approach (loan restructure or the legal option) for the recovery of liabilities from each sub-borrower, and to pursue the chosen option with vigour. E. BANK PERFORMANCE 39. Project Design. The project was adequately designed to address the foreign exchange issues that were limiting further investment in the agricultural sector at the time. However, the necessary prerequisites for the reform of CRDB, the major factor for project success, were underestimated. The project aimed at the restructuring of CRDB through the provision of technical assistance and studies which, whilst potentially valuable exercises in themselves, needed to be provided in the context of a visible and internally generated motivation for change in organisational behaviour. There was little evidence of this in 1988, although it was subsequently prompted by legislation, Government action and the involvement of DANIDA from 1992. The legal covenants in the SAR that relate to the financial parameters in CRDB's restructure were unrealistic given the prevailing economic and political climate at the time. 40. There were several assumptions in the project design which proved to be invalid. A majority of potential borrowers identified in the SAR were co-operative unions or parastatals, recognised as being commercially unviable, yet forming the basis of the estimated credit demand, and with no proposals under the project or other concurrent programmes to make the necessary changes. Therefore, at project start-up, the identification of other borrowers was immediately necessary, and this contributed to delayed and reduced project impact. The lack of reform in the marketing of coffee and cotton was recognised as a risk, 4 The Borrower rates the sustainability as likely, a view the Bank does not agree with. 9  but free marketing of these commodities could reasonably have been expected by AERP start-up, with their subsequent inclusion as sub-projects. 41. The procedure for disbursements on1 sub-loans was complicated and proved to be a significant drawback to project implementation. Many sub-borrowers apparently did not fully understand the procedures and felt constrained by the conditions. Indeed, several approved sub-borrowers declined to take- up their loans on discovering the complexities of the disbursement procedures. A study on AERP disbursements that was financed in early 1994, recognised the absence of a supervision schedule for disbursement activities for each sub-loan and made several recommendationis including the strengthening of staff at the Foreign Exchange Department in the Azikiwe Branch of CRDB that was responsible for communication with the correspondent bank. Neither this nor other procedural recommendations from the study were implemented. 42. Supervision. Over the 92 month project period, only nine supervision missions visited the project from headquarters. Supervision was satisfactory until November 1991, after which no full supervision from headquarters was undertaken until February 1994. During this period, the internal reorganisation of the World Bank divisions transferred project responsibility from AF6AE to AF2AE, and some oversight of the project was given to the resident mission. However, in retrospect this informal follow-up, which focused on subloan administration, was insufficient as a majority of sub-loan activities were undertaken during this period, the mid-term review was rescheduled, CRDB restructuring commenced, and the marketing and foreign exchange reforms were in place. This was a critical time for the project and inadequate levels of monitoring and advice were provided to CRDB. From February 1994 however, substantial efforts were made during and between supervisions to restructure the project, by limiting the exposure of CRDB and assisting in better management of the existing portfolio. In addition, project performance may have suffered through the absence of a technical perspective in supervision that would have more readily identified problems with sub-loan enterprises. F. BORROWERS PERFORMANCE 43. The project management was entrusted to the CRDB for the implementation, monitoring and supervision of both the institutional building and credit components. There was no steering or co-ordinating committee at senior level, although a Credit Committee (CC) met at irregular intervals until its 26th meeting in March 1994 to consider loan applications and other AERP matters. The CC was chaired by the Managing Director (previously General Manager) of CRDB, with as members: a university economist; an accountant from a commercial company; CRDB staff from Legal Services and Commercial Banking; and AERP Unit staff including the technical assistance group. A single field visit only was made by the CC, in early 1994, so little opportunity was gained of first-hand experience of enterprise management levels and field performance. Effective day-to-day implementation of a major loan portfolio was therefore the responsibility of the Project Co-ordinator, a middle-rank CRDB staff member. 44. At appraisal, an additional 10 professionals were considered necessary to reinforce CRDB's staff for the increased promotion and appraisal activities of AERP. Initially, some 8 professionals formed the AERP Unit; but this figure was often lower and there was little continuity of staff. 10  45. Implementation has been generally unsatisfactory througlhout the life of the project, both for the credit disbursement and institutional building components. Most significantly, the project was negatively affected by the lack of substantial organisational and finanicial management change in CRDB during the period of loan disbursement, and the lack of progress in building appraisal capacity and capability during the main period of sub-loan appraisal and supervision between 1989 and 1994. This was identified as a major risk at appraisal and was to be covered by a Monitorable Action Plan for restructuring, written into the Development Credit Agreement. This was not undertakeni. 46. The project had a history of either delayed compliance or non-compliance of the assurances and covenants whiclh were central to the project's design. However, the covenants relating to the ilstitutional and financial restructuring of CRDB were superseded by the FSAC operating from 1991 to 1993, after which AERP's role in strengthening CRDB was subsumed by the involvement of DANIDA as a major sharelholder of the new CRDB. 47. Many of the legal covenanits relating to CRDB's financial and management performance were not complied with. In addition, there were delays in preparing and submitting reports and accounts that were required by IDA under the project agreements, and these appeared not to form part of the routine management infornation system required by CRDB. Although AERP's most recently audited accounts to 30 June 1996 were certified, the account documentation appears in some disarray. 48. Although the SAR specifically excluded working capital as a credit item for sub-borrowers, the supervision mission of November 1991 agreed that part of the loan proceeds could be utilised for local cost financing, but on a case-by-case basis. In the event, the issue remained unattended to by CRDB, apparently due to foreign exchange complications, and an opportunity was lost to provide an additional service to sub-borrowers and assist those facing local currency constraints. G. FUTURE OPERATIONS 49. As part of the continuing program of CRDB privatisation, the Ministry of Finance (MOF) has agreed to assume responsibility for loan liabilities under AERP, as well as those of the two IFAD-supported lines of credit and the Pilot Credit Scheme under the IDA-supported Cashew and Coconut Treecrops Project that have been implemented by CRDB. CRDB expects to act as the agent for loan recovery, and details of the Memorandum of Understanding involving the financial arrangements are currently under negotiation. It would be in the Government's interest to ensure that CRDB is adequately rewarded in its capacity as management agent, so that CRDB is provided with sufficient motivation to pursue sub-loan repayments. 50. With the transfer of AERP liabilities to MOF, sub-loans are effectively covered by a Government guarantee. Thus the BOT provisions under the Banking and Financial Institutions Act (1991) that relate to those conditions under which sub-loan restructuring can take place, do not apply. Accordingly, CRDB may reschedule individual loans as deemed appropriate. 51. Following requests from several sub-borrowers, CRDB has agreed that sub-loans may be repaid in local currency, assuming the exchange rate at the time of repayment. This may be particularly valuable for producers of sisal, for whom domestic sales of fibre for process and subsequent export, can represent a substantial proportion of their total production. 11  H. ASSESSMENT OF OUTCOME 52. Overall, the project objective of strengthening CRDB's capability to engage in medium- and long-term lending in the agricultural sector has been partially achieved. The objective of promoting the financial and management restructure of CRDB was not successful and was overtaken by subsequent legislation for financial institutional reform, and the involvement of the private sector in CRDB's management. The other objective of export development through credit has been partially achieved to the extent that production and the quality of produce has improved in those enterprises that remain viable. The final outcome may depend ultimately upon CRDB's ability to restructure existing sub-loans to ensure sub- project survival and the subsequent repayment of money borrowed. Support for export rehabilitation and of new ventures has helped in improving rural incomes and employment, and AERP provide a valuable stimulus to the development of the floriculture sub-sector in Tanzania. However, overall outcome of the project is considered unsatisfactory due to the poor loan repayment record, the uncertain repayment outlook for the tea and sisal sub-loans, the large number of loans now subject to legal proceedings, and the negligible impact on CRDB of the project. I. KEY LESSONS LEARNED 53. The key lessons learned from the project were as follows: it was unreasonable for the Bank preparation and appraisal missions to have expected significant restructuring of CRDB without the political will that was not evident until after the enactment of the Banking and Financial Institutions Act (1991); - the negotiation of assurances and covenants between IDA and GOT is only meaningful where IDA is prepared to take action in the event of subsequent non- compliance; - the disruptive effect of the eventual CRDB restructuring on staff morale, continuity and efficiency was known at appraisal, and this contributed to the poor monitoring and supervision of sub-projects; - the international technical assistance personnel were directly involved in project implementation and contributed little to the capacity of CRDB, to the extent that CRDB staff derived little benefit from their presence; - CRDB was able to make multiple loans to individual sub-borrowers, and to considerably exceed the individual loan limits established at appraisal, with insufficient evidence of management commitment. This contributed substantially to CRDB's exposure to risk of loan default; 12  sub-borrowers with substantial and commercially operating enterprises, require a timely and efficient banking service for effective and profitable business operation. CRDB was not able to provide this service; the mid-term review should have been undertaken at its appointed time in PY 3, particularly as the AERP performance had already been rated by IDA as unsatisfactory; supervision missions for agricultural credit programmes should include staff with technical/agronomic skills, so to be better able to assess agricultural businesses; and the absence of headquarters supervision from November 1991 until February 1994, to complement the monitoring being undertaken by the resident mission at a critical period in both sub-borrower development and CRDB restructuring, may have contributed substantially to the poor performance of the project. 13  PART II: STATISTICAL TABLES Table 1: Summary of Assessments A. Achievement of objectives Substantial Partial Negligible Not Applicable (V) (V/) (/) (7) Macro policies L [] E Sector policies I 1LI 0 Financial objectives LI LI -II Institutional development LI LI 11 LI Physical objectives LI LI LI Poverty reduction LI I L I Gender issues LI L I [I Other social objectives -I LI LI [ Environmental objectives LI LI LI [FI Public sector management LI LI FkI L I Private sector development LI LI LI Other (specify) LI LI LI LI B. Project sustainability Likely Unlikely Uncertain (7) (7) (7) 4I ZI L 1 4  Highly C. Bank perfornance satisfactory Satisfactoly Deficient Identification E E E: Preparation assistance 7 z z Appraisal i 7 E Supervision 7 E Highy D. Borrower performance satisfactory Satisfactorv Deficient (/) (/) (/) Preparation F FII Implementation Z] E Covenant compliance LZ E E Operation (if applicable) LliHbly E. Assessment of outcome satisfactory SatisfactQrv Unsatisfactory unsatisfactory 15/) (/) 01) V) 1 5  Table 2: Related Bank Loans/Credits Loan/Credit Title Purpose Year of approval Status Preceding operations 987-TA TRDB (Rural Development) 1980 Completed Following operations 2308-TA Financial Sector Adjustment 1991 Completed Credit Table 3: Project Timetable Steps in Project Cycle Date Planned Date Actual/Latest Estimate Identification Jul-86 Preparation Nov-86 Pre-Appraisal Appraisal Apr-87 Apr-87 Negotiations Jul-87 8-Feb-88 Board presentation Sep-87 29-Mar-88 Signing 16-Aug-88 Effectiveness 14-Nov-88 Midterm review Jun-91 May-95 Project completion 30-June-96 30-June-96 Credit closing 30-June-96 30-June-96 16  Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual (US$ million) FY 88/89 89/90 90/91 91/92 92/93 93/94 94/95 95/96 (US$ million) Appraisal Estimate (cumulative) 1.7 5.7 11.7 17.7 22.5 25.9 28.4 30.0 Actual (cumulative) 0.4 1.5 3.0 5.0 8.1 13.6 17.4 17.9 Actual as % of estimate 25% 26% 26% 28% 36% 52% 61% 60% (SDR Million) Appraisal Estimate (cumulative) 1.2 4.2 8.6 13.0 16.5 19.0 20.8 22.0 Actual (cumulative) 0.3 1.1 2.2 3.7 5.8 10.2 12.5 12.8 Actual as % of estimate 25% 27% 26% 28% 35% 54% 60% 58% Note: At the time of the ICR mission, the reconciliation of the Special Account had not yet been completed Table 5: Key Indicators for Project Implementation Item Unit At Appraisal Adjusted at Actual/ I I MTR Loans Disbursed US$ 24.1 n/a 15.4 Training - Overseas fellowships months 60 n/a 0 - Study tours months 24 n/a 8 Table 6: Key Indicators for Project Operation I. Key operating indicators in SAR/President's Report Estimated Actual n/a n/a 17  Table 7: Studies Included in Project Purpose as defined at Study appraisal/redefined Status Impact of Study Detailed portfolio Recommend changes in Completed Negligible review loss reserves and provisions CRDB Restructuring Phased plan for Completed Negligible study restructuring Financial Information Effective financial systems Completed Negligible Systems study for CRDB's future needs 18  Table 8A: Project Costs Appraisal Estimate Actual /1 (US$ m) (US$ m) Item Foreign Local Total Foreign Local Total Credit Component Costs Investment and Working Capital 24.1 10.0 34.1 15.6 n/a Feasibility Studies for Sub- projects 0.8 0.3 1.1 0.9 0.03 Management Assistance and Training 0.4 0.1 0.5 0.04 n/a Sub-total 25.3 10.4 35.7 16.5 n/a n/a Credit Component Costs Restructuring of CRDB 1.2 0.9 2.1 0.05 n/a Computer Equipment 0.4 0.1 0.5 0.1 n/a Training to CRDB staff 0.3 0.3 0.6 0.1 n/a Technical Assistance 0.7 0.2 0.9 1.1 0.13 Contingencies 0.1 0.1 0.2 Sub-Total 2.7 1.6 4.3 1.4 n/a n/a TOTAL PROJECT COST 28.0 12.0 40.0 17.9 n/a n/a Note: /1 There is no record of local costs incurred either by sub-borrowers, or CRDB except for costs of appraisal and supervision costs by AERP Unit and CRDB staff as set out in the CRDB document "Profitability Analysis for the AERP Programme (1996)". At the time of the ICR mission, the reconciliation of the Special Account had not yet been completed 19  Table 8B: Project Financing Appraisal Estimate (US$M) Actual/Latest Estimate (US$M) Source Foreign Local Total Foreign Local Total IDA 28.0 2.0 30.0 17.9 n/a n/a CRDB - 2.0 2.0 - 0.17 n/a Beneficiaries - 8.0 8.0 - n/a n/a TOTAL 28.0 12.0 40.0 17.9 n/a n/a 20  Table 9: Economic and Financial Costs and Benefits i) Categorised and itemised project costs are not available from CRDB, despite an external audit of AERP's accounts for the period ending 30 June 1996, and their subsequent certification by the CRDB Board of Directors. Discrepancies exist in the documentation within CRDB, and there has been no attempt to assess the contribution to total project costs by either the sub-borrowers, CRDB or GOT. ii) The overall economic rate of return was not calculated at appraisal. However, the economic justification for the project was that only sub-loans with ERRs in excess of 15% were to be approved. Whilst the sub-loan appraisal process involved an ERR assessment and each satisfied the 15% cut-off point, the sensitivity to major physical and price assumptions was high in many cases. An attempt was made to reassess the ERR of the investments which contributed to on-going businesses but, in these cases, a true assessment of the effect of the project supported investment was not possible for two reasons: (i) it was not possible to determine the extent of the borrowers contribution to the investment and therefore the total sub- project cost; and (ii) further investments had been made since the original project-financed investment, financed from a mixture of external debt, equity, or from internally generated funds, which made it impossible to identify the benefits deriving from the original investment alone. iii) Whilst the impact on foreign exchange earnings was not estimated in the SAR, two factors may have depressed the net foreign exchange benefits that were anticipated at appraisal. First, a majority of the individual sub-projects have not been able to generate the foreign earnings as anticipated due to marketing and management problems. This applies particularly to tea and sisal enterprises, for which world market prices have decreased. Second, almost one-half of sub-project investments were in floriculture enterprises that have operating costs with a higher content of foreign exchange than the traditional export agriculture enterprises. iv) The project has made a substantial contribution through the increase in local employment opportunities, particularly for women and youth in the sisal, tea and floriculture enterprises, with an estimated 10,000 additional opportunities created directly as a result of AERP activities. 21  Table 10: Status of Legal Covenants Agreement Section Covenant Status Comments Description Type Credit 2.02(a) Management aspects CP DANIDA Restructuring of CRDB project and strengthening of its overtook agricultural lending IDA's role. capabilities through management assistance and training (schedule 2). Credit 2.01(b) Finance. performance C Complied. Borrower to open and maintain in dollars a special account at BOT in accordance with schedule 3 provisions. Credit 3.01 (b) Flow & util of funds CP Partial Borrower to relend credit compliance. proceeds in accordance with a subsidiary loan agreement entered between borrower and CRDB. Credit 3.05(b) Finance. performance -Counterpart funding CP Complied re.: CRDB to maintain a escrow foreign exchange escrow account but account from its earnings not funded utinder the project in an with the amount equivalent to amount of the provisions created to provision. cover bad doubtful debts. Credit 4.0 1(b) Accounts/audit CD/P Complied Borrower to have project with delay. records and accounts Audits required to be maintained frequently under 4.01 (a) and those overdue for the special account and/or audited and furnished to qualified. IDA not later than six months after the end of the year. Project 2.02 Management aspects C Complied. CRDB to employ consultants withl qualifications, experience and terms satisfactory to IDA. Project 2.03(a) Flow & util of funds PC Partial CRDB to make sub-loans compliance. in accordance with procedures, terms and conditions in schedule I of PA. Project 3,01(b) Flow & util of funds C Complied. CRDB to lhave its records, accounts and financial statements audited and furnished to IDA not later than six moutlhs after each fiscal year. Project 3.02(a) Finance. performance NC Not complied. CRDB to earn for each ot its fiscal vear alter June 30, 1990. an anIual rate of return of not less than 1.2% of thc total average value of its assets in operation. 22  Table 10: (Continued) Agreement Section Covenant Status Comments Description Type Project 3.02(b) Management aspects -Monitoring & report NC Not complied. Belbre March in each of its fiscal years, CRDB to review whether it would meet the requirement under pa3 02(a). based on forecasts prepared by CRDB and acceptable to IDA. Project 3.02(c) Management aspects -Implementation NC Not complied. If any such review shows that CRDB would not meet the requirement in pa 3.02(a) CRDB shall promptly take all necessary actions to meet such requiremenlt. Project 3.03 Finance. performance NC Not complied. After its fiscal year ending June 30. 1990. CRDB to maintain a ratio ot current asscts to currcnt liabilities ot'not less than 1.5: 1. Otherwise it shall take steps similar to those in 3.02(b) and 3 02(c) above. Project 3.04 Finance. performance NC Not complied. For each of its fiscal year after June 30. 1990, CRDB to carn an annual rate of return of not less than 10% on its equity. Otherwise, it shall take steps similar to those indicated in 3.02(b) and 3.02(c) above. Project 3.05 Finance. performance -Implementation NC Not complied. CRDB take steps. as shall be necessary to protect itseilfagainst the risk of loss resulting from changes in rates of exchange betwceen the various currencies used in its operations Project 3.06 Finance. performance C Complied Not later thani September 30, 1988, CRDB to. (a) provide evidence satisfactory to IDA that it has established a sound asset base, and (b) prepare a plan otf actionl i'or reducing the remainiina loan arrears as of June 30, 1987 by 33 1/3% annually. 23  Table 10: (Continued) Agreement Section Covenant Status Comments Description Type Project 2.01(b) Management aspects -Implementation NC Not complied CRDB to carry out implementation program in sche.2 relating to its organizational restructuring; staffing; finanlcial rehabilitation; operating costs and profit margin, establishing a portfolio management and operational policies; information systems; Project Schedule 2, Management aspects CP Partial Not later than 12/31/88, Para 4 compliance CRDB to employ the following professionals: (i) for its Dev. Bank dept., two economists, a statistician/economist. two financial analysts and two agronomists; & (ii)for its Planning. and Dev. Dept., an engineer, a financial analyst. Project Schedule 2. Monitoring & report.-Implementation CP Draft report CRDB to undertake a Para 13 received May major review of overall 1995 project performance and furnish report thereof to IDA by not later than Juiie 30, 1991, and discuss the findings of the review with IDA by not later thain October 1991, and reach agreement on an action program for the remainder ot 'roject Implementationi. Present Status: C - Covenant complied with CD - Complied with after delay CP - Complied with partially NC - Not complied with 24  Table 11: Compliance with Operational Manual Statements Basically, there was compliance with the applicable Bank Operational Manual Statements. Table 12: Bank Resources: Staff Inputs Stage of project cycle Planned Revised Actual l/ Preparation to n/a n/a 113 234.2 appraisal Appraisal n/a n/a 63.1 149.6 Negotiations through 9.7 23.3 Board approval Supervision n/a n/a 129.4 386.0 Completion 2/ n/a n/a 14.0 48.1 1 ~~~~~~~~~~~~~~~~~~~~~~~~~~1 1Including travel costs 2/ Carried out by FAO/CP n/a - Not available - not recorded on Cost Accounting System 25  Table 13: Bank Resources: Missions Performance Rating Number Days in Specialized Staff Implementation Development Stage of Project Month/Year of Field Skills Status Objectives Types of Cycle Persons Represented Problems Pre-appraisal 5/85 5 n/a n/a - - - Appraisal 4/87 4 n/a E, FA, B - - - Supervision 5/89 2 7 FA, FA U U PM, F 11/89 2 25 FA, FA U U PM, F 4/90 2 17 FA, FA S S - 11/90 2 13 FA, FA U U PM 5/91 2 12 FA, FA U U PM, F 11/91 1 14 FA U U PM, F 2/94 4 9 Ops, E, E, FA U U F 5/95 2 5 FA, Ops S S - 3/96 3 28 FA, AE, Ops U U Completion /1 9/96 2 12 AE, Ag Abbreviations: AE = Agricultural Economist; Ag = Agronomist; FA = Financial Analyst; Ops = Operations Officer; B = Banking Specialist; E = Economist. U = Unsatisfactory; S = Satisfactory PM = Project Management; F = Financing 1/ Carried out by FAO/CP 26  APPENDIX A MISSION'S AIDE MEMOIRE  FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS Investment Centre Division TANZANIA: Agricultural Export Rehabilitation Project (AERP) (Cr 1895-TA) Implementation Completion Report Mission Aide-Memoire A. Introduction An FAO/World Bank Cooperative Program (CP) mission1 visited Tanzania from 25 August to 13 September 1996 to prepare the ICR for the above project. Discussions were held with the concerned officials from the Ministry of Finance (MOF), Ministry of Agriculture and Co-operatives (MOAC), Bank of Tanzania (BOT) and staff of the World Bank Resident Mission in Dar es Salaam, including the Task Manager of AERP. The mission worked closely with CRDB (1996) Limited, the restructured Cooperative and Rural Development Bank, which was the implementation agency of the project. Contact was made with the enterprises of 13 of the 23 sub-borrowers of project funds in the Arusha, Kilimanjaro and Tanga Regions. A wrap-up meeting was held on 12 August 1996 in Arusha. The FAOR, Tanzania was kept informed of progress and briefed on the outcome of the mission. The mission would like to express its appreciation for the assistance given by the officials of the Government of Tanzania, and for the cooperation and courtesy of the management and staff of CRDB B. Background The AERP was conceived during the mid-1980s at a time when the financial sector was performing poorly due to lack of competition, excessive interest rate controls and minimal bank supervision. Credit was directed on the basis of Government priorities without regard to creditworthiness, and banks were convenient agents of fiscal policy. Non-bankable activities were financed, with a misallocation of resources to the parastatals and agricultural cooperatives. This resulted in a large accumulation of non-performing loans (around two-thirds of the total) and by 1988 the banking system was technically insolvent An Economic Reform Programme (ERP) was introduced in 1986 to improve macroeconomic management, tackle underlying structural weaknesses and encourage more active private sector  participation. Economic measures designed to reduce the budget deficit and improve the balance of payments included a substantial depreciation of the overvalued exchange rate, trade liberalisation, the removal of most price controls and the easing of restrictions on the marketing of most food crops. Nevertheless, foreign exchange remained in short supply, severely constraining the essential rehabilitation of export enterprises. Following the Commission of Enquiry of 1988 into the Monetary and Banking System, the Government acknowledged the perverse impact on the sector caused by a lack of competition and Government interference, and through the enactment of the Banking and Financial Institutions Act (1991) permitted the entry of private banks, with expanded and enhanced supervisory and regulatory controls vested in the Bank of Tanzania (BOT) C. Project Objectives and Components The objectives of the project were to: (i) promote agricultural exports in the traditional crop sector (coffee, tea, sisal, cashewnuts, cotton and tobacco) and thereby increase foreign exchange earnings; (ii) give greater emphasis to market forces to increase the efficiency of resource allocation; and (iii)advance institution building by strengthening the capability of CRDB to engage in medium and long-term lending in the agricultural sector. The project consisted of two components: (i) a credit component (US$ 26.7 million) consisting of sub-loans to be managed by CRDB, and financed under a subsidiary loan agreement between CRDB and the Ministry of Finance; and (ii) an institutional building component (US$ 3.3 million) for CRDB comprising technical assistance, training, equipment and studies. Although 3 possible implementing agencies had been identified during preparation, the project was implemented solely by CRDB, which had close links with the cooperative unions and parastatals that were expected to be the main beneficiaries of the project To achieve its objectives, the project was to: (i) provide .long-term financial support for investment projects for the rehabilitation and expansion of agricultural export enterprises (sub-borrowers) and for improving the quality of Tanzania's exports: (ii) provide foreign exchange to finance capital investment and up to three years of the recurrent import requirements (excluding seasonal inputs) of sub-borrowers; (iii) restructure CRDB and transform it into a viable agricultural financial intermediary;  (iv) advance the process of agricultural reform in the sector and enable cooperative unions, parastatal agencies and the private agricultural enterprises to take advantage of the recent liberalization of agricultural input supply and distribution, export crop marketing and producer pricing; and (v) provide technical assistance funds for the recruitment of short-term consultants to carry out sub-project feasibility studies and for long-term advisors to support the management capability of sub-borrowers. Total project costs were estimated during appraisal at US$ 40 million to be financed as follows: IDA credit US$ 30 million (SDR 22 million equivalent), CRDB US$ 2.0 million and sub- borrowers US$ 8.0 million. Foreign exchange costs were estimated at US$ 28.0 million or 70% of total project costs. The AERP was appraised in April 1987 and approved by IDA in March 1988. The Credit was signed on 16 August 1988 and became effective on 14 November 1988. The project was intended for a period of up to 8 years. Although appraised and initially supervised by the Southern Africa Department of the World Bank, the Credit was transferred during implementation to the Eastern Africa Department in July 1992. Approved loans from CRDB to sub-borrowers under AERP totalled only US$ 17.8 million, of which only US$ 15.3 million or 57% of the projected level of US$ 26.7 million. A recognised deterioration in performance of the loan portfolio, and the realisation that CRDB and not the Government, was liable for any losses incurred, led IDA to advise CRDB in February 1994 to discontinue the processing of new loans, and to review each loan on a case-by-case basis before further disbursements were made. No new loans were approved after December 1993, and disbursements on existing committments effectively ceased by the end of 1994. The Mid-Term Review took place in May 1995 and the project closed on 30 June 1996. D. Achievement of Project Objectives The project at least partly achieved its development objective of foreign exchange lending (at an interest rate of 8% per annum) to export-oriented agricultural investment. Thus a valuable adjunct was provided to the Investment Promotion Centre (IPC) scheme which provided tax exemptions on imported capital items, established by the Government in 1990 to attract investment to Tanzania Loans were disbursed to only 23 sub-borrowers, due mainly to the limited number of proposals forwarded that were eligible for project support. In addition, commodity prices for traditional export crops had deteriorated, rendering investment in rehabilitation and expansion marginal in some cases. The  cooperative unions and parastatals that had been targeted at appraisal, were unable to be supported due to their ineligibility on grounds of substandard management and inadequate creditworthiness. Thus the sub-borrowers were all companies in the private sector, initially in the tea and sisal industries and latterly in the horticultural sub-sector The project was operational during the period following market liberalisation and reform, and commercial investments, mainly in the tea, sisal and horticulture industries, were possible as a result of freedoms in exports and imports, as well as foreign exchange flows. The very large number of proposals for rehabilitation in the cotton and coffee industries could not be supported however. This was due to a lack of progress with market liberalization in the cotton sector that did not allow export by producers themselves and thus retention of foreign exchange, and the use by the Government of coffee revenues as part of a barter deal for oil purchases. Whilst the impact on foreign exchange earnings was not estimated at appraisal, two factors have depressed the net foreign exchange benefits that were anticipated. Firstly, a majority of the individual sub-projects have not been able to generate foreign earnings as anticipated due to marketing and management problems. Secondly, more than 50% of investments were in floricultural enterprises that have operating costs with a high content of foreign exchange. The project has contributed to a substantial increase in local employment opportunities, particularly in the sisal, tea and floriculture enterprises. The institutional strengthening objectives were not met as a result of the project. The technical assistance concentrated on subloan appraisal and supervision of the AERP portfolio, as opposed to the strengthening of CRDB's overall capacity for medium and long-term lending as the project had intended. The training programme was abbreviated and unstructured, and there were few study tours for CRDB staff and none for sub-borrowers. There was no overseas fellowship training in appraisal techniques. The project had little impact on the overall financial and organisational management of CRDB. Considering the failure of the previous IDA efforts to restructure CRDB (Credit 987-TA), and a relative lack of success in this area of the subsequent IDA-supported Financial Sector Adjustment Credit (FSAC), it is perhaps unreasonable to have expected significant improvements without the accompaniment of serious political and CRDB management support for such changes. In the event, it required the enactment of the legislation associated with the Banking and Financial Institutions Act (1991), the involvement of DANIDA in CRDB's affairs, and the strengthened supervision of BOT, for significant achievements to be made. The mission notes that these events and interventions have resulted in a substantially  restructured CRDB by the time of project completion, more market-oriented and able to survive in a competitive business environment. E. Implementation Experience and Results General Implementation has been generally unsatisfactory throughout the life of the project, both for the credit disbursement and institutional building components. Most significantly, the project was negatively affected by the lack of substantial organizational and financial management change in CRDB during the period of loan disbursement, and the lack of progress in building appraisal capacity and capability during the period of activity of sub-loan appraisal and supervision between 1989 and 1994. This was identified as a major risk at appraisal and was to be covered by a Monitorable Action Plan for restructuring, written into the Development Credit Agreement. The project had a history of either delayed compliance or non-compliance of the assurances and covenants which were central to the project's design. However, the covenants relating to the institutional and financial restructuring of CRDB were superseded by the Financial Sector Restructuring Project (FSRP) under FSAC which operated from 1991 to 1993. Following this, the project's role in strengthening CRDB was subsumed by the involvement of DANIDA as a major shareholder of the new CRDB. The Credit Component The project sought to provide finance for the capital (machinery, equipment, spare parts and vehicles) and recurrent costs (although not for seasonal costs) required to rehabilitate or expand agricultural export enterprises, lending to cooperative, parastatal and private firms which were creditworthy. An initial pipeline of 38 sub-projects were identified at appraisal of which only five were creditworthy and eventually participated. It was therefore necessary to redefine the investment focus to include both traditional and non- traditional agricultural exports, as well as start-up ventures, and to revise upwards the loan limits. Following this change, there was a delay before new, non-traditional sub-projects were designed, appraised and approved, the majority of which were in flower and horticultural production in the Arusha and Kilimanjaro regions. Nevertheless, some 84 sub-projects were put forward for consideration by CRDB1 and either approved (23), rejected on non- viability grounds (27), loans approved but not taken up (7), or were in the pipeline at the time of the closure on new loans in 1994 (27). A considerable proportion of project staff time was  devoted to the appraisal of new sub-projects which failed to result in further loan disbursements, rather than to the supervision and monitoring of existing loans. This may have enabled approved sub-projects to deteriorate beyond the point at which disbursements should prudently have been curtailed. Of the 23 loans disbursed, only 8 are currently not in default of payments, and 7 of these are still in their grace period. Seven sub-projects are currently the subject of legal action, while all but one of the remaining loans require review and where necessary, restructure. Generally, sub-project appraisals are of good quality, although several make over-optimistic yield and production assumptions, and pay too little attention to the availability of the working capital contributions of individual sub-borrowers. Delays were experienced in the feasibility, appraisal and approval of loans to sub-borrowers. After loan approval, the disbursement procedure proved cumbersome and time consuming. There were considerable delays in the opening of letters of credit for the provision of capital items, which may have contributed to the ultimate poor performance of some sub- borrowers. Although CRDB's selected external bank recognised the delays and made some efforts to expedite disbursements, this ultimately proved ineffective. CRDB assistance to sub-borrowers in negotiating these procedures appeared minimal, although one workshop for sub-borrowers was held and a simple guideline prepared. However, the performance of CRDB as implementing agency was not aided by the considerable disruption to staff effectiveness in the period 1991-1995, during which CRDB reorganisation occurred and almost two-thirds of total staff were retrenched. In many cases there was a lack of clearly defined implementation arrangements in the sub-loan agreements that led to borrowers not being subjected to pressure to follow time schedules. A number of borrowers continued to draw down, even after the expiry of the grace period under their loan agreements, and in some cases the equipment purchased was different from that specified in the proposal. This can be attributed directly to inadequate supervision of individual sub- projects during the implementation phase. In many cases, little if any attempt has been made to force sub-borrower compliance with loan conditions agreed at negotiation. This may have led to a late appreciation by CRDB of poor sub-borrower performance, as regular reports and CRDB's presence on company boards has not been enforced. Traditional Export Crops. Only tea and sisal were included in this category with 5 and 8 loans respectively, representing 14% of the total disbursed portfolio in the case of tea and 28% for sisal. However, these represent only 7 effective borrowers, considerably increasing the exposure of the portfolio to risk of default. Apart from the loan to the Mufindi Tea Company Ltd  that was repaid in full, all loans are in arrears of interest or capital repayments or both. Loans for sisal rehabilitation are of particular concern, given the nature of the industry and the poor world market situation. Four sub-loans were approved for a single company without an agricultural track record, and non- significant quantities of fibre have been produced in recent years. Non-Traditional Export Crops. Of the 10 sub-borrowers in this category, 5 are producers of cut flowers (mainly roses) in Arusha and Kilimanjaro, for export to Europe, with loans totalling US$ 7.5 million or 49% of the AERP portfolio. These represent AERP's most successful venture and with CRDB's aggressive promotion, attracted investors from the much larger producing areas of neighbouring Kenya, through cheap loans and the IPC incentive. This potentially profitable sub-sector has a high associated risk, particularly as sub-borrowers appear to prefer to expand their production rather than repay loans. Other sub-borrowers in horticulture, essential oil production, crocodile farming, seaweed production and fishing, appear less satisfactory, and provide a demonstration of the inherent risk in bank lending to new enterprises for which venture capital might be the normal source of investment finance. The Institutional Building Component CRDB institution building was expected in the form of (i) a programme to restructure CRDB, including the development of the restructuring plan and its implementation (ii) management assistance and training for CRDB staff to strengthen CRDB's project promotion and appraisal capability, and (iii) technical assistance to assist CRDB and sub-borrowers to prepare feasibility studies for the sub-projects. As an initial measure, funds from a PPF of US$ 1.5 million were available for studies to develop a Monitorable Action Programme for CRDB restructure including: a detailed portfolio review by independent auditors to recommend changes in loan loss reserves and an appropriate level of provisions; and a study by banking specialists to develop a phased plan for restructuring CRDB. In addition, a financial information system study was expected to develop an effective system appropriate for CRDB's current and future needs and determine the training requirements of bank staff. There is no evidence that the PPF studies played any useful part in the restructure of CRDB that has taken place since the early 1990s. Although some 111 months of technical assistance were provided for both long-term and individual sub-loan feasibility studies, the TA team was involved in sub-project appraisal and supervision and there was little effective staff training and  strengthening of CRDB's overall capacity for medium and long term lending as the project had intended. The project was designed so that the experience gained through the operation of a line of medium to long-term credit, supported by technical assistance in portfolio management, would filter through to the remainder of CRDB lending operations. This implementation strategy was altered almost as soon as the project had started, and in late-1989, a special Project Implementation Unit was created to keep the staff, funds and accounting for AERP activities separate from mainline bank programmes. Whilst the mission recognises the uncertainties concerning CRDB restructuring prevailing at that time, this change may have contributed to two outcomes: firstly the failure to transfer AERP experience to other banking operations; and secondly the misconception that the AERP liabilities were not CRDB's liabilities and thus need not be recorded in the CRDB's annual accounts. This in turn may have contributed to a lack of attention from senior CRDB management to the discipline required to successfully appraise and supervise the sub-projects. For example, no effective system of routine sub-project review and supervision appears to have been designed or implemented, with the result that the financial condition of some sub-borrowers had been allowed to deteriorate alarmingly before the condition was fully appreciated and remedial action taken. The eventual restructuring and privatization of CRDB led to reduced staff resources being dedicated to AERP matters; falling from 8 staff in 1989 to 4 in 1994. Currently only 2 of the Special Programmes Unit (including the Chief Manager) are fully experienced in AERP sub-loan management. It became clear as early as 1991 that the portfolio was of mixed quality. Interest arrears had accumulated, apparently without action being taken, and principal repayments were not being met by sub-borrowers. At project completion, legal action is being taken over debt recovery from 7 of the sub-loans, amounting to some US$ 5.2 million, and 15 of the remaining projects are at least in default of interest repayments. The result is a portfolio of currently non-performing loans. Mid-term Review (MTR) At appraisal, the MTR for the project was scheduled for PY 3, at the end of June 1991. This was considered to be' essential, and particular attention was to be paid to the progress of institutional reforms, with a major review setting out an action program for the remainder of the project period. At the time scheduled for the MTR, CRDB management requested a postponement so that the conclusions of the portfolio audit and  assessment of CRDB's financial position to be carried out under the FSRP could be incorporated. However, as a result of the major changes at CRDB which started during 1993, a further postponement was agreed by IDA, and the review was not carried out until May 1995, some 4 years after its original schedule and 7 years after project initiation. Given the risks inherent in the project that were identified at appraisal, the changes to the focus and implementation mechanism of the credit program that had already been necessary during the early stages of the project, and the recognition by World Bank supervision missions that the implementation progress was unsatisfactory, the MTR should have taken place as planned. This would have enabled the implementation issues to have been addressed, particularly those relating to the performance of the loan portfolio. As it was, the MTR took place after new loan disbursement had ceased, TA had been completed and the project was effectively over. F. Institutional Performance World Bank Performance Project Design. The project was adequately designed to address foreign exchange issues that were limiting further investment in the agricultural sector at the time. However, the necessary prerequisites for the reform of CRDB, the major factor for project success, were underestimated. The project aimed at the restructuring of CRDB through studies and technical assistance which, whilst potentially valuable exercises, needed to be provided in the context of visible, internally generated motivation for change in organizational behaviour. There was little evidence of this in 1988, although it was subsequently prompted by legislation, Government action and the involvement of DANIDA from 1993. The legal covenants relating to the financial and management restructuring are considered unrealistic targets given the prevailing economic and political climate at the time. There were several assumptions in the project design which proved to be invalid. The majority of potential borrowers identified in the SAR were co-operative unions or parastatals. At appraisal these were recognized as being commercially unviable, yet they formed the basis of the estimated credit demand, with no proposals under the project or other concurrent programs to make the necessary changes. Therefore, at project start-up, the identification of other borrowers was immediately necessary, and this contributed to delayed and reduced project impact. The lack of reform in the marketing of coffee and cotton was recognized as a risk, but free marketing of these commodities could reasonably have been expected at project start-up.  The procedure for disbursements on the sub-loans was complicated and proved to be a significant drawback to project implementation. Many of the enterprises apparently did not fully understand the procedures and felt constrained by the conditions. Supervision. The supervision of the project has been satisfactory until November 1991, after which no full supervision was undertaken until February 1994. During this time, the internal reorganization of the World Bank divisions transferred project responsibility from AF6AG to AF2AG, and the task management to the resident mission. However, during this period of project implementation, a majority of sub-loan activities were undertaken, the mid-term review was rescheduled, CRDB was restructured, and the results of the marketing and foreign exchange reforms were in place. This was a critical time for the project and an inadequate level of monitoring and advice were available to CRDB. The mission recognizes the significant efforts made since 1994 to restructure the project, by limiting the exposure of CRDB and assisting in better management of the existing portfolio. Supervision may have suffered through the absence of a technical perspective that would have more readily identified problems with sub-loan enterprises. CRDB Performance The project management was entrusted to the CRDB for the implementation, monitoring and supervision of both the institutional building and credit components. There was no steering or coordinating committee at senior level, although a Credit Committee (CC) met at irregular intervals until March 1994 (26th meeting) to consider loan applications and other AERP matters. The CC was chaired by the Managing Director (previously General Manager), with a university economist, an accountant from a commercial company, CRDB staff from Legal Services and Commercial Banking, as well as AERP Unit staff including the technical assistance group. A single field visit only was made by the CC, in early 1994, so little opportunity was gained of first-hand experience of enterprise management levels and field performance. At appraisal, an additional 10 professionals were considered necessary to reinforce CRDB's staff for the increased promotion and appraisal under AERP. Initially, some 8 professionals formed the AERP Unit; but this figure was often lower and there was little continuity of staff. As mentioned above, many of the legal covenants relating to CRDB's financial and management performance were not complied with. In addition, there were delays in preparing and  submitting reports and accounts that were required by IDA under the project agreements. The most recently audited accounts to 30 June 1996 were certified. G. Future Operations CRDB is in default of the subsidiary loan agreement, commencing September 1995. The Treasury has agreed to assume responsibility for loan liabilities under AERP (as well as those of IFAD, and the Pilot Credit Scheme under the IDA-supported Cashew and Coconut Treecrops Project), and CRDB expects to act as the agent for loan recovery. It would be in the Government's interest to ensure that CRDB is adequately rewarded in its capacity as management agent, so that CRDB is provided with sufficient motivation to pursue sub-loan repayments with vigour. With the transfer of AERP loan liabilities to the Treasury, there is an effective Government guarantee. Under these conditions, the BOT provisions under the Banking and Financial Institutions Act (1991) that relate to conditions under which loan restructuring can take place, do not apply. Accordingly, CRDB may reschedule individual loans as considered appropriate. Several sub-borrowers have requested that loans be repaid in local currency. This may be particularly valuable for producers of sisal, for whom domestic sales of fibre for process and subsequent export, can represent a substantial proportion of their total production. H. Project Sustainability Currently there are only 2 CRDB staff in the Special Programme Unit (SPU) that have been continuously involved in sub-loan appraisal and evaluation. Although other SPU staff are receiving training, there is little assurance of a continuity of staff involvement in the AERP portfolio for the future. Few other CRDB staff possess the necessary appraisal skills for such large investment programmes, despite the current and on-going training programmes at headquarters and branch level. The computer expertise essential for rapid enterprise evaluation is not available at the branch level. Under the CRDB decentralisation policy, it is intended that staff at branch level will monitor and supervise the AERP portfolio, including the restructuring of loans where appropriate. The mission believes that branches have neither the capacity nor capability to undertake these tasks, unless reinforced with staff experienced in handling major and complex loans.  The performance of the individual sub-borrowers is highly variable. The floriculture enterprises are already in profit, while those for sisal, which is of much longer term development, require additional working capital resources to be successful. Tea enterprises are intermediate and could reach profitable levels within the next 2 years. Apart from the 7 sub-borrowers under litigation, there is sufficient equity in the remaining enterprises to provide substantial motivation for their private sector owners. However, CRDB will need to determine rapidly the appropriate approach (loan restructure or the legal option) for the recovery of liabilities from each sub-borrower, and to pursue these with all the vigour at their disposal. I. Main Lessons Learned The key lessons learned from the project were as follows: - it was unreasonable to have expected significant restructuring of CRDB without the political will that was not evident until after the enactment of the Banking and Financial Institutions Act (1991); - CRDB was able to make multiple loans (up to 4) to individual sub-borrowers, and to considerably exceed the individual loan limits established at appraisal without evidence of enterprise profitability or management commitment. This contributed significantly to the bank's exposure to risk; - commercial businesses require a timely and efficient banking service for effective and profitable operation. CRDB was not always able to provide this service; - a mid-term review is an essential tool at which an independent assessment can be made of project progress, and should have been undertaken at its appointed time, particularly as the AERP performance had already been rated as unsatisfactory; - supervision missions for agricultural credit programmes should consider the inclusion of technical; and - the absence of formal supervision from November 1991 until February 1994, at a critical period of both sub-borrower development and CRDB restructure, may have contributed to the poor performance of the project. J. Items Outstanding Closing of the Special Account. CRDB is continuing the process of reconciliation. All accounts should be finalised and closed by 31 October 1996.  Project Costs. CRDB has yet to provide an itemised breakdown of project costs. K. Follow-up On return to Rome by 15 September 1996, the mission will prepare a draft ICR to be forwarded to IDA by mid-October 1996. The Borrower's contribution to the ICR is expected from CRDB and/or the Treasury before the mission leaves Tanzania. 13.9.96  APPENDIX B BORROWER'S CONTRIBUTION TO THE ICR  IMPLEMENTATION COMPLETION REPORT FOR THE AGRICULTURAL EXPORT REHABILITATION PROJECT (CREDIT NO. 1891 TA) CRDB (1996) LIMITED PERSPECTIVE 1.0 PREFACE: This is the Implementation Completion Report (ICR) for the Agricultural Export Rehabilitation Project (AERP) launched in the United Republic of Tanzania; for which a credit in the amount of SDR 22,000,000 (equivalent to US$ 30.0 million) was approved in 1988. Of the total US$ 30.0 million US$ 26.7 million and US$ 3.3 million was for credit and institution building components respectively. The project's credit was closed on 30th June 1996 for all disbursements except for those which IDA had approved prior to that date. As at the close of the project period, US$ 17.8 million of the credit component amount (65%) had been approved by CRDB's Board of Directors. Of the total amount approved, US$ 15.3 million - which is 88% and 57% of the approved and disbursed amounts respectively - had been disbursed. The last disbursement is expected to take place in September 1996. The remaining amount in the credit component is US$ 11.4 million (43%). The Institution Building Component remaining as at the close of the project period amounts to US$ 2.7 million (82%) of the total amount in that component. The remaining amount in both the credit and institutional building components amounting to US$ 12.0 million has been cancelled. 2.0 EVALUATION SUMMARY: 2.1 Projects Objectives: The project objectives, as set out in the Staff Appraisal Report (SAR) are as follows: (a) Providing long-term financial support for investment projects for the rehabilitation and expansion of agricultural export enterprises and for improving the quality of Tanzania's export; (b) Providing foreign exchange to finance up to three years of the annual recurrent import requirement (excluding season inputs) of sub-borrowers; 1  (c) Restructuring CRDB and transforming it into a viable agricultural financial intermediary; (d) Advancing the process of agricultural reform in the sector and enabling the cooperative unions and private agricultural enterprises to take advantage of liberalization of agricultural inputs supply and distribution, export crop marketing and producer pricing, and (e) Providing technical assistance funds for the recruitment of short-term consultants to carry out sub-project feasibility studies and for long-term advisors to support the management capability of sub-borrowers. 2.2 Modification of Objectives: At the inception of the project, it was assumed that funds from the credit components would be disbursed to the traditional crops sub-sector; which included coffee, cotton, sisal and tea. However, that was not the case due to the following reasons: (a) Most of the traditional crops and agro-processing businesses being run by cooperative unions which due to high degree of inefficiency and poor financial position failed to meet the project's conditionalities; and (b) Syndication of coffee by the Government of Tanzania for purposes of oil importation. Slow disbursement resulting from the above reasons prompted CRDB to make marketing to the non-traditional crops sub-sector. The material modification was studied and considered by IDA and incorporated in the project's objectives in early 1990. 2.3 Project's Loan Agreements: To promote achievement of the project's objectives the following loan agreements were signed in the year 1988: (a) Development Credit Agreement between the Government of Tanzania and the IDA; (b) The Project Agreement between the IDA and CRDB; and (c) Subsidiary Loan Agreement between the Government of Tanzania and CRDB; In addition to the legal documents mentioned above, there was a Staff Appraisal Report (SAR) which provided all the project's implementation pre-requisites and criteria. 2  2.4 Evaluation of the Objectives: For continued country's economic growth, the project's objectives were of utmost importance. The same were just complementing the already existing Economic Recovery Program (ERP). The goals were clear, and they were expected to bring about direct response to the country's priority sector, that is agricultural production, especially of export crops. 3.0 IMPLEMENTATION EXPERIENCE AND RESULTS: 3.1 Achievement of project objectives: As a result of implementation of the project the following achievements are registered: (a) According to the 1995 AERP mid-term review findings, the project has had a significant social economic impact on the country by increasing employment to the tune of 2,740 in the agricultural export sub-sector. Majority of the employees in the projects financed under this loan in the non-traditional crops sub-sector are women; (b) New export commodities have been put on the market (Floriculture, Horticulture and Aquaculture) with a great potential to earn vital foreign exchange. Of the project's portfolio, 63% (US$ 8.9 million) is invested in new businesses of export oriented non- traditional crops sub-sector; (c) The 1995 AERP mid-term review revealed a combined annual export business generated by new businesses as a result of the project amounting to US$ 4.0 million. (d) Provision of technical assistance in conducting feasibility studies, appraisal of sub- projects and training of the AERP Implementation Unit (PIU) in administration of sub- loans. By 1993, the PIU had a total of eight (8) full time professionals as required by the SAR. However, most of these professionals have left CRDB; (e) Computer equipments were purchased for computerization of the projects operations and the bank at large. 3.2 Sustainability of the project's achievements: It is likely that the achievements generated by the project in relation to its major objectives will be sustained. This is due to the following reasons: (a) It has been learnt through CRDB's supervision missions and the sub-borrowers' finanicial commitment that the current effective n1on1 traditional sub-projects sector will contilLie providing the current level of employment and otlher economic benefits suchi as 3  generation of foreign exchange through export of the project's products. This is evidenced by the fact that the sponsors are now putting in their own money for expansion, much as the AERP has closed up; and (b) CRDB restructuring through AERP did not achieve much. However CRDB through DANIDA assistance using a different model has carried out its own restructuring in the areas of organizational, operational, financial, human resources and cultural which assures sustainability of the bank and the projects it has financed. There is in place loans and portfolio management tools such as credit policy, operations manual, credit operating manual, authority manual, accounting manuals, and staff service manual. CRDB has also, conducted credit training for all levels of its technical staff; starting from Directors to Credit Officers. CRDB is currently working out guidelines for administration of donor funded projects; for which training will also be conducted. As a result of that, customer services have been brought down to branches where the sub-borrowers are located, there will be uniform way of managing credit operations in all branches, and reporting will be immediate through the decentralized structure. 3.3 Factors affecting achievements of the major objectives: Below are the major factors which affected the project implementation. The same are rated as substantial, partial and negligible: (a) Failure to capture the original customers in coffee and cotton industries as envisaged in the SAR; which slowed down disbursement (Substantial); (b) Failure of the IDA/CRDB to conduct project mid-term review in 1991 as scheduled in the SAR (Substantial); (c) The economic decline of the country reflected by worsening terms of trade, prolonged drought, insufficient producer incentives and resource allocated to agricultural sector (Substantial); (d) Continued inappropriate price policies, reliance on administrative controls for allocating resources and excessively expansionary fiscal and monetary policies (substantial); (e) Deteriorating balance of payment (partial); (f) Deficiencies in the agricultural inputs and implements (Substantial); (g) Fluctuation in the exchange rate, adversely affecting CRDB's operations; and the sub- borrowers by increasing the costs of imported equipments, spares and supplies 4  (Substantial); (h) Chronicle disbursement delays (Substantial); (i) The cyclic nature of commodity prices in international markets for tea and sisal (Substantial), (j) Low knowledge of international market by local producers (partial); (k) Failure to carry out CRDB restructuring earlier as planned (Substantial); (I) Failure by IDA to agree to finance CRDB staff training (Substantial); 3.4 Performance of the IDA and the Treasury: 3.4.1 The IDA Performance: The IDA performance can be said to have been effective in the initial stage of the project implementation; and later on it became passive; only to pick up later when the current Task Manager was appointed. In the initial stage the Bank's supervision team was much concerned about the project performance; especial when they realized diversion from the original plan. Change in the IDA's supervision team in early 1990s, changed even its attention to the project such that supervision visits were substantially reduced. Thus, variances in the project's actual implementation was left to continue contrary to the requirement in the projects documents, especially the SAR. 3.4.2 The Borrower Performance: The borrower (Government of Tanzania) did not implement its commitments on time as agreed during project negotiations. Such commitments included the following: (a) To increase producer prices; (b) To widen quality producer margin; (c) To liberalize the marketing of foodgrain; (d) To liberalize the marketing of the major export crops and dissolve the related marketing parastatals; 5  (e) To permit all private estates and cooperatives to export directly instead of exporting through marketing boards; (f) To restore the role of private sector in commercial agriculture, privatization of several agricultural enterprises (eg coffee, tea, and sisal estates); and (g) Removal of the National Milling Corporation monopoly. 3.5 Assessment of outcome: The AERP project has been operating with a lot of difficulties throughout its existence. The conceived objectives and the implementation whiclh could lead to their achievements was not accorded due attentioni by all parties. Fulfillment of commitments by the parties concerned, was not to the level set out in the project documents. As a result of that the achievement of the intended outcome is rated as unsatisfactory. 4.0 SUMMARY OF FINDINGS, FUTURE OPERATION AND KEY LESSONS LEARNED: 4.1 Summary of Findings: (a) Through implementation of the project, it has been learnt that there were problems in the project original design assumptions. (b) It was originally planned that the project's funds will be disbursed to the cooperative unions which were known to be highly inefficient. This shows that no thorough study was done at the appraisal stage to that effect; (c) Lack of clear action plan and supervision of the CRDB restructuring process made the whole idea abortive; (d) Lack of management and legal input in the project implementation unit (PIU); although the same requirement was planned in the SAR brought about existence of porous sub- loan agreements; (e) The consultants initially attached to the project implementationL unit were ineffective; were not monitored actively so as to measure their performance; (f) Capacity building components failed as IDA constantly turned down CRDB traililig programmes as submitted; and (g) Low knowledge in international banking by CRDB in the initial years and poor cooperation by Citibank NY seriously affected timely disbursements and disturbed 6  borrowers implementation plans. 4.2 Future Operations: Future operations of the AERP project will be characterized by the following factors: (a) Decentralization of the AERP including strengthening of CRDB branch capacities will bring better results in supervision of the projects. (b) Working as agents will put CRDB on alert to follow up and realize the agency fee; (c) Project operations and sustainability will only be achieved if the sub-borrowers will continue supporting the already funded sub-projects; and (d) Export tax recently introduced will act as a disincentive to AERP exporters. To date all sub-borrowers are complaining. 4.3 Key Lessons Learned: In the course of implementing the Agricultural Export Rehabilitation project, the following lessons have been learnt: (a) So much evolutions have taken place during the actual project implementation period such that some basic assumptions have been taken by events, for example while the project targeted traditional export crops, 63% of its portfolio is invested into the non- traditional export sub-sector. This diversion came in the middle of the project implementation due to problems in the original design assumptions; (b) The design of the project was not critically analysed by CRDB management and risk determined;the requirement to keep separate books of accounts as per SAR, this misled CRDB on the risk aspect, and such oversight has costed the bank enormously; (c) Non inclusion of the legal inputs in the project implementation unit brought about existence of poorly done sub-loan agreements. The bank had relied on its legal department which however failed to capture the special characteristics of such a project; as a result most agreements are porous; (d) Diversion into non-traditional crops should have been done along with equipping the 7  project implementation unit with requisite training to be able to monitor projects effectively and be ahead of the sub-borrowers: (e) Giving loans for capital items alone is not helpful if the policy on1 working capital is left vague. (f) Lack of aggressive and effective supervision of sub-projects has resulted in alarmingly high rate of default; the bank has learned this and has already addressed the issue; and (g) IDA and CRDB having failed to appraise regularly the performance of the consultants attached to the project implementation un1it, gave them a room to offer poor services to the bank. From this lesson, IDA and CRDB should in future address this deficiency if the two will undertake any project of same design in future. 8  APPENDIX C AERP PORTFOLIO TABLES  Appendix C-1 Table 1 AERP Credit Component - Disbursements to Sub-borrowers - 31 August 1996 Total Loan Annual Disbursement (US $) Total Loan % of Loan Sub-borrower Approved 1989 1990 1991 1992 1993 1994 1995 1996 Disbursement Disbursed I Ambangulu Estates Ltd 225,231 0 162,437 49,125 11,950 0 0 0 0 223,512 99 2 Arusha Farns Ltd 950,000 0 0 0 609.897 0 13,200 0 0 623,097 66 3 Balangai Estates Ltd 550.000 0 0 0 72,993 115,436 11,620 0 0 200,049 36 4 Centrepoint Investment Ltd 443,416 0 15,460 289,017 18.314 0 0 0 0 322,791 73 5 DamascenaEss. Oils Ltd 550,000 0 0 0 291,553 200,267 18,368 0 0 510,188 93 6 Florissa Ltd 801,001 0 0 0 0 0 796,057 0 0 796,057 99 7 Hortanzia Ltd 1.800,000 0 0 0 0 1,241,983 509,773 24.692 0 1,776,448 99 8 J.Z.Makuri Farms Ltd 70,000 0 0 0 69,427 0 0 0 0 69.427 99 9 Karimjee Agric. Ltd - Sisal 981,000 0 0 15,416 0 816,043 0 143,310 64,325 1,039,094 106 10 Karimjee Agric. Ltd - Tea 757,000 0 0 11,895 0 549.235 775 t18.354 0 680,259 90 11 Kiliflora Ltd 1,793,070 0 0 0 324,494 1,015,713 449,369 0 0 1.789,576 100 12 Kingsway Int.ernational Ltd 215,000 0 0 0 0 22,931 41,650 0 72,795 137,376 64 13 Makinyumbi Estates Ltd 700,000 0 0 0 428.466 0 0 0 0 428,466 61 14 Mamba Ranch Ltd 878,300 0 0 0 0 222,844 555,650 0 99,762 878,256 100 15 Marungu Estates Ltd 463,124 0 0 425,345 17.483 0 0 0 0 442,828 96 16 Mufindi Tea Company Ltd 585,643 0 568.037 0 0 0 0 0 0 568,037 97 17 Prawtan Ltd I/ 0 0 0 14,850 0 0 0 0 14.850 na 18 Ralli Estates Ltd 1,000,000 0 0 277,525 78.586 122,157 80.600 0 0 558,868 56 19 Sumagro Ltd 660,000 0 0 168,237 124.421 92,428 25,109 14.376 0 424,571 64 20 Sunripe Kilimanjaro Ltd 2,600,000 0 0 0 204.003 1,583,373 419,242 0 0 2,206,618 85 21 Tanfarms Ltd 363,549 0 9.240 346,354 0 0 0 0 0 355.594 98 22 Tanzania Flowers Ltd 950,004 0 0 0 0 950,004 0 0 0 950,004 100 23 Tukuyu Estates Ltd 432,720 0 359,572 53,486 23,689 0 0 0 0 436,747 101 17,769,058 0 1,114,746 1,636,400 2,290,126 6,932,414.00 2.921,413.00 300.732.00 236,882.00 15,432,713 87 Note: Derived from CRDB computations. 1/ Prawtan loan was approved but not taken up. Disbursement for feasibility study only.  Appendix C-i Table 1 AERP Credit Component - Disbursements to Sub-borrowers - 31 August 1996 Total Loan Annual Disbursement (US $) Total Loan % of Loan Sub-borrower Approved 1989 1990 1991 1992 1993 1994 1995 1996 Disbursement Disbursed I Ambangulu Estates Ltd 225,231 0 162,437 49,125 11,950 0 0 0 0 223,512 99 2 Arusha Farms Ltd 950,000 0 0 0 609,897 0 13,200 0 0 623,097 66 3 Balangai Estates Ltd 550,000 0 0 0 72,993 115,436 11,620 0 0 200,049 36 4 Centrepoint Investment Ltd 443,416 0 15,460 289,017 18,314 0 0 0 0 322,791 73 5 Damascena Ess. Oils Ltd 550,000 0 0 0 291,553 200,267 18,368 0 0 510,188 93 6 Florissa Ltd 801,001 0 0 0 0 0 796,057 0 0 796,057 99 7 Hortanzia Ltd 1,800,000 0 0 0 0 1,241,983 509,773 24,692 0 1,776,448 99 8 J.Z.Makuri Farms Ltd 70,000 0 0 0 69,427 0 0 0 0 69,427 99 9 KarimjeeAgric. Ltd-Sisal 981,000 0 0 15,416 0 816,043 0 143,310 64,325 1,039,094 106 10 Karimjee Agric. Ltd - Tea 757,000 0 0 11,895 0 549,235 775 118,354 0 680,259 90 11 Kiliflora Ltd 1,793,070 0 0 0 324,494 1,015,713 449,369 0 0 1.789,576 100 12 Kingsway Int.emational Ltd 215,000 0 0 0 0 22,931 41,650 0 72,795 137,376 64 13 Makinyumbi Estates Ltd 700,000 0 0 0 428,466 0 0 0 0 428,466 61 14 Mamba Ranch Ltd 878,300 0 0 0 0 222,844 555,650 0 99,762 878,256 100 15 Marungu Estates Ltd 463,124 0 0 425,345 17,483 0 0 0 0 442,828 96 16 Mufindi Tea Company Ltd 585,643 0 568,037 0 0 0 0 0 0 568,037 97 17 PrawtanLtd I/ 0 0 0 14,850 0 0 0 0 14,850 na 18 Ralli Estates Ltd 1,000,000 0 0 277,525 78,586 122,157 80,600 0 0 558,868 56 19 Sumagro Ltd 660,000 0 0 168,237 124,421 92,428 25,109 14,376 0 424,571 64 20 Sunripe Kilimanjaro Ltd 2,600,000 0 0 0 204,003 1,583,373 419,242 0 0 2,206,618 85 21 Tanfarms Ltd 363,549 0 9,240 346,354 0 0 0 0 0 355,594 98 22 Tanzania Flowers Ltd 950,004 0 0 0 0 950,004 0 0 0 950,004 100 23 Tukuyu Estates Ltd 432,720 0 359,572 53,486 23,689 0 0 0 0 436.747 101 17,769,058 0 1,114,746 1,636,400 2,290,126 6,932,414.00 2,921,413.00 300,732.00 236,882.00 15,432,713 87 Note: Derived from CRDB computations. 1/ Prawtan loan was approved but not taken up. Disbursement for feasibility study only.  Appendix C-1 Table 3 AERP Credit Component - Impact of Sub-projects on Employment Total Loan Sub-project Employment Sub-borrower Disbursed Enterprise Created 2/ 1 Ambangulu Estates Ltd 223,512 tea 80 2 Arusha Farms Ltd 623,097 sisal 0 3 Balangai Estates Ltd 200,049 tea ? 4 Centrepoint Investment Ltd 322,791 sisal 0 5 Damascena Ess. Oils Ltd 510,188 ess.oil 40 6 Florissa Ltd 796,057 flowers 50 7 Hortanzia Ltd 1,776,448 flowers 170 8 J.Z.Makuri Farms Ltd 69,427 prawn fishing ? 9 Karimjee Agric. Ltd - Sisal 1,039,094 sisal 3,000 10 Karimjee Agric. Ltd - Tea 680,259 tea 0 11 Kiliflora Ltd 1,789,576 flowers 750 12 Kingsway Int.ernational Ltd 137,376 seaweed 500 13 Makinyumbi Estates Ltd 428,466 sisal 0 14 Mamba Ranch Ltd 878,256 crocodiles 20 15 Marungu Estates Ltd 442,828 sisal 250 16 Mufindi Tea Company Ltd 568,037 tea 200 17 Prawtan Ltd 1/ 14850 prawn farming 0 18 Ralli Estates Ltd 558,868 sisal 350 19 Sumagro Ltd 424,571 sisal 50 20 Sunripe Kilimanjaro Ltd 2,206,618 vegetables 40 21 Tanfarms Ltd 355,594 sisal 0 22 Tanzania Flowers Ltd 950,004 flowers 265 23 Tukuyu Estates Ltd 436,747 tea ? 15,432,713 5,765 Note: Derived from CRDB, Mid-term Review and mission estimates. 1/ Prawtan loan was approved but not taken up. Disbursement for feasibility study only. 2/ Reflects total enterprise that may include investments other than AERP.    IT C-i l. I jCm Z~~~~~~ ~~4-.J A Ci ) r'1 ~ ~ I H, "..I j_j , t ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I