Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15854 PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL CREDIT PROJECT (CREDIT 1804-MAG) June 28, 1996 Agriculture and Environment Division Central Africa and Indian Ocean Departuent Africa Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS Currency Unit: Malagasy Franc Year US$1.00 Equivalent 1987 1,069 1988 1,407 1989 1,603 1990 1,494 1991 1,835 1992 1,864 1993 1,913 1994 ( 3,402 ( As at June 30, 1994 WEIGHTS AND MEASURES Metric System ABBREVIATIONS APEX Financial Sector and Private Sector Development Project (Cr. 2104-MAG) BTM National Rural Development Bank (Bankin's Ny Tantsaha Mpamokatra) BTM I First Agricultural Credit Project (Cr. 1064-MAG) BTM II Second Agricultural Credit Project (Cr. 1804-MAG) DCA Development Credit Agreement ERR Economic Rate of Return EMSAP Economic Management and Social Action Project (Cr. 1967-MAG) FMG Malagasy Franc FRR Financial Rate of Return PCR Project Completion Report GOVERNMENT FISCAL YEAR January 1- December 31 FOR OFFICIAL USE ONLY The Worid Bank Washington, D.C. 20433 U.S.A. Ofie of the Diror-General Opwaons Evauaton June 28, 1996 OED EVALUATIVE MEMORANDUM ON PROJECT COMLETION REPORT Madagascar: Second Agricultural Credit Project (Credit 1804-MAG) The Madagascar Second Agricultural Credit project, supported by Credit 1 804-MAG for SDR 8.0 million (US$10.0 million equivalent), was approved in FY87. The credit closed on schedule in FY94, and an undisbursed balance of SDR 938,000 (US$1.2 million equivalent) was canceled. The Project Completion Report (PCR) was prepared by the Africa Regional Office. The borrower made minor comments on the draft, which have been integrated in Part I. There is no Part II. The project objectives were to contribute to the institutional development of the state-owned National Rural Development Bank (BTM), and to meet foreign exchange costs of new investments and improvements of existing enterprises. The main project components were: (i) a line of medium-term credit for medium-size enterprises involved in agricultural production and processing; (ii) institutional strengthening, including equity infusion, training, computer equipment and management consultancy services, to assist BTM to recover from a weak financial position; (iii) incremental costs of a pilot credit- management program for smallholders, emphasizing repayments; and (iv) a consultant study to formulate a long-term agricultural credit policy for Madagascar, which among other concerns would identify and propose means to remove the basic constraints to smallholder lending. The project started well, but performance subsequently declined. The medium-term lending operation was implemented on schedule, and most of the investment operations are considered by their owners to have been successful. However, the number of sub-loans was smaller than expected (21 operations), most of the funds went to a few large dairy and aquaculture (prawns) firms, and the PCR asserts that some of these could have secured commercial credit outside the project. Moreover, implementation problems emerged during the eighteen-month political crisis that gripped the country in mid-project, starting in 1991. Repayment rates declined after 1992, and the PCR warns about problems some of these companies continue to face in a deteriorating macro-economic environment. The institutional reforms were largely unsuccessful: plans for organizational changes at BTM and reductions in its work force were opposed by the unions and suspended; customized software for the computerization program was inappropriate and introduced prematurely, disrupting BTM accounts. However, there were improvements in the training program and audit procedures. The provision of additional equity did not reverse the decline in BTM's financial condition. The pilot credit-management program and the rural finance study were carried out satisfactorily, although they underscored the problems in smallholder credit that had disrupted implementation of the preceding (first) IDA agricultural credit project in the early 1980s. | This document has a metritbd distribution and may be used by recipients only In the performanco of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. 2 The PCR considers project outcome unsatisfactory, institutional development modest with respect to specific project activities but negligible in terms of BTM's recovery from technical insolvency, and sustainability of the medium term lending program unlikely. The PCR finds IDA performance deficient during supervision, and concludes that the project should have been aborted at least as early as 1992. OED accepts these ratings, including an overall unsatisfactory rating of Bank performance that is implicit in the PCR text. The major lesson cited in the PCR is that it is difficult to bring about dramatic improvement in the performance of a government-owned bank with poor oversight, weak management capabilities, and political interference in the decision making process. Equity infusions in a poorly managed institution will not work without appropriate institutional and administrative safeguards to protect the banks financial integrity. The PCR is satisfactory. The absence of information on the impacts of individual investments, and loss of contact between IDA and BTM following the closing date, prevent an accurate assessment of the progress to date of the participating enterprises. No audit is planned. Attachment FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL CREDIT PROJECT (CREDIT 1804-MAG) CONTENTS PREFACE ........................... I EVALUATION SUMMARY ........................................................ III PART 1: PROJECT REVIEW FROM BANK'S PERSPECTIVE ....................................................1I 1. PRO1ECT IDENTITY ...I........... . BACKROUND .............................1 3. PROJECT OBJECTIVES AND DESCRIPTION ........................2 4. PROJECT DESIGN AND ORGANIZATION ................................3 5. IMPLEMENTATION ...........................4 . PROJECT RESULTS .............................9 7. SUSTAINABILITY ...............9 S. IDA PERPORMANCE ...1........... 0 9. BORROWER PERFORMANCE ...............11 10. CONSULTING SERVICES ...............11 S1. PROJECT DOCUMENTATION AND DATA . ........................... 11 PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ...................... . .... 13 PART III: STATISTICAL INFORMATION ............................................................... 14 TABLE 1: RELATED BANK CREDITS .......................... 14 TABLE 2: PROJECT nMETABLE .......................... 1 5 TABLE 3: PROJECT COSTS & FINANCINo .......................... 16 TALE 4: DISBURSEMENTS .......................... 17 TABLE 5: SIZE OF SUB-LOANS .........................S. 1 TABLE 6: USE OF SUB-LOANs ........................... 19 TABLE 7: SUB-LOAN REPAYMENTS .......................... 20 TABLE 8: COMPLIA NCE WITH COVENANTS .......................... 2 1 TABLE 9: USE OF BANK RESOURCES .......................... 23 MAP: IBRD No. 21799 ........................... 24 Th docunent has a restricted distribution and may be used by reciplents only in the peztorfonce of their offlchl duties. Its contents may not otherwise be dlscloeed wlihout World Bank authorization. l PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL CREDIT PROJECT (CREDIT 1804-MAG) PREFACE This is the Project Completion Report (PCR) for the Second Agricultural Credit Project in Madagascar, for which Credit 1804-MAG in the amount of SDR 8.0 million was approved on May 21, 1987. The Credit was closed on June 30, 1994, as originally scheduled, at which time the undisbursed balance of SDR 938,000 was canceled. This PCR was prepared by the Agriculture and Environment Division of the Central Africa and Indian Ocean Department, Africa Region. Preparation of this report is based on the final project supervision mission carried out in June 19941, the Staff Appraisal Report, the Credit and Project Agreements, supervision reports, correspondence between the Bank and the Borrower, internal Bank memoranda and information provided for the PCR by the implementing agency (National Rural Development Bank; BTM). BTM's minor comments have been fully incorporated into Part I of the PCR; no formal Government comments were received. I Processing of the report encountered serious delays because of difficulties in collecting the data. PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL CREDIT PROJECT (CREDIT 1804-MAG) EVALUATION SUMMARY Background 1. Agriculture is an important economic activity in Madagascar, contributing about 30 percent to the gross national product and 60 to 65 percent to export earnings. Over 80 percent of the economically active population is dependent on the sector for their livelihood. Agricultural performance had suffered in the 1970s and the early 1980s from ill-conceived Government policies that were aimed at protecting the interests of the politically vocal urban consumers. Farm prices were set at too low a level to make investments in the sector attractive and as a result, sector performance was well below the population growth rate during this period. In the early 1980s, Government faced a fiscal crisis following its heavy reliance on borrowing to finance its ambitious public investment program. In response to this crisis, Government embarked on a structural reform process that included the liberalization of prices. Prices for most agricultural products were liberalized by 1986, with a strong production response. 2. The First Agricultural Credit Project (Cr. 1064-MAG) was implemented during the early eighties. The objectives of this project were to support Government's policy to extend credit to smallholders, with the National Rural Development Bank (BTM) as the principal intermediary, and to strengthen BTM's managerial and financial capability as a rural development bank. The project was clearly grounded in the context of the Government's credit and rural development policies of the late 1970s; their failure thus greatly affected the project's implementation. BTM lacked experience in implementing a rural credit scheme that was primarily aimed at smallholders, and as a result repayments were very low. The rural credit activities were essentially stopped in the mid 1980s and most of the Credit was reallocated to finance the importation of inputs and equipment for resale, with the local currency thus generated used as equity for BTM. During this same period, BTM engaged in large scale lending to parastatal enterprises, under pressure from Government. Many of these borrowers were unable to meet their repayment obligations, forcing BTM into insolvency. The Credit closed on December 31, 1986. iv Evaluation Summary Project Description and Objectives 3. The Second Agricultural Credit Project (BTM II) was designed to move away from smallholder credit, in light of the poor experience under the predecessor project, and focus on lending to medium-sized private enterprises. In 1985, at the time of appraisal BTM II was expected to cost US$19.33 million, to be financed by IDA (US$10.0 million or 52 percent), by BTM (US$6.15 million or 32 percent), and the remainder by private investors (US$3.18 million or 16 percent). There were two main components: (i) medium term credit for enterprises involved in agricultural production and processing (US$15.9 million or 82 percent of project costs); and (ii) institutional development, including computerization, training and technical assistance (US$3.4 million or 18 percent of project costs). The proceeds of the Credit were to be provided to BTM by Government as a grant, in order to strengthen BTM's equity position. Moreover, as a condition of effectiveness, Government was to pay FMG 2.5 billion (about US$3.3 million) to honor its past guarantees on irrecoverable loans to parastatals. Project Implementation 4. Project implementation initially proceeded smoothly, primarily because BTM staff had identified most of the potential loans before the credit became effective. Twenty-one development loans were made, with IDA disbursements for this category amounting to SDR 5.2 million (about US$7.0 million), equivalent to 82 percent of the funds provided under this category (SDR6.5 million or about US$8.7 million). Interest rates were to be positive in real terms and borrowers were given the choice of borrowing in foreign currency, or borrowing in local currency at a higher interest rate, with BTM assuming the foreign exchange risk. In order to prevent a concentration of loans to only a small number of individual borrowers, the ceiling for loans was set at US$500,000, with prior review by IDA necessary for each loan over the equivalent of US$150,000. During project implementation, however, this ceiling was raised to US$1,000,000 in order to allow for the participation in several large investments with a high likelihood of success (dairy production and prawn fishing). Despite the project's implied goal to lend to a large number of medium-sized enterprises, a few large loans accounted for a high percentage of the lending program. The three largest accounted for 51 percent of total lending, while the eight largest accounted for 82 percent. Moreover, the two largest borrowers subsequently merged their operations. 5. Project implementation started during a period when Madagascar was introducing its structural adjustment program (1987), and this helped to create a more favorable economic environment for the project. However, in 1991, when the country was starting to experiment with a more open political system, massive strikes seriously disrupted the economy and paralyzed the political decision making process. Following the political transition, the macro-economic situation had worsened, with an overvalued exchange rate and serious shortages of foreign exchange. This caused a financial crisis that affected the state owned banking system in general, and BTM in particular. For several enterprises that had received project funded loans, these events came at a very bad time, for they had just completed their capital development programs and could not achieve projected sales and cashflows. It Evaluation Summary v therefore became necessary for BTM to delay or reschedule repayments for these loans. Loan repayment data show that repayments have been unsatisfactory: 35 percent in 1991, 116 percent in 1992, due to loan prepayments by borrowers that feared a devaluation of the local currency; 83 percent in 1993 and 54 percent in 1994. A floating exchange rate was introduced in April 1994, that provoked a substantial fall in the value of the FMG (about 50 percent). By this time, however, most foreign exchange denominated loans had either been repaid or converted into local currency loans. The repayment issue was not appropriately followed up by IDA supervision missions. 6. In 1991, Government established a steering committee to oversee BTM's operations and changes were made at the senior management level. Further management changes have been made recently, but BTM remains a weak institution, which is technically insolvent and faces an uncertain future. In 1992, BTM was suspended from participating in the IDA supported Financial Sector and Private Sector Development Project (APEX; Credit 2104- MAG), because of problems related to overall portfolio management highlighted in the 1991 audit results; this suspension remains in effect. Project performance and staff morale were adversely effected by the uncertainties about BTM's future. Preliminary studies designed to prepare for BTM's privatization have been funded by IDA under the Economic Management and Social Action Project (EMSAP; Credit 1967-MAG) and APEX. However, this option should have been more resolutely pursued before the end of the project. A provisional administrator was appointed to BTM in November 1995. 7. The institutional development component of the project was only partially implemented as planned. Major difficulties were encountered in setting up the new computerized accounting system, primarily due to the premature introduction of new systems which had not been adequately tested. Effort has subsequently been expended in rectifying these problems and the computerized accounting system is now operating satisfactorily. Also, the organization and human resources development activities planned under the project had to be suspended as a result of strong staff opposition to change. 8. Actual project costs, excluding the contributions made by borrowers, expressed in US$ were 38 percent below the appraisal estimate (US$12.0 million actual versus US$19.3 million estimated), with IDA funding 80 percent (versus 52 percent planned). Even though no activities were funded under the Credit after 1992, the Credit was kept open until its scheduled closing date, because it was believed at the time that a decision on the privatization of BTM was forthcoming and that the undisbursed balance would be needed to prepare BTM for privatization. The credit closed on June 30, 1994, at which time SDR 938,000 was canceled. Project Results 9. Initial results indicate that this project has been unsatisfactory overall. The medium- term credit component has benefited only a small number of large enterprises, some of which could have obtained the credit from other sources. An in-depth analysis of the impact of the investments financed from the proceeds of the credit cannot be carried out for lack of vi Evaluation Summary pertinent information, as borrowers were not required to maintain separate accounts on the investments made with the loans. BTM did continue to track the overall financial performance of these borrowers. Based on 1993 data, six of the eight largest borrowers reviewed were financially healthy, and BTM predicted continued growth. The institutional development component was only partially implemented; major problems with the computerization program were encountered; and the human resources and organizational activities were suspended. The rural credit study, part of this project, but funded from other sources, was instrumental in helping prepare the Rural Finance Technical Assistance Project (Credit 2459-MAG). This pilot project, which became effective in November 1993, is intended to develop savings and loan associations for small farmers and the rural poor. Sustainability 10. Although a successful outcome to the majority of the physical investments is expected, sustainability of the lending program is unlikely. The level of loan repayments for the medium term credit component have not been satisfactory, and this will prejudice future operations unless repayment performance can be improved. In fact, BTM would have been technically insolvent, at the start of the project, were it not for the massive equity infusion under the Credit. However, this assistance did not correct the many problems that plagued BTM, and it is once again insolvent: the bank's 1994 accounts show it has a negative net worth and it does not meet its reserve requirements Lessons Learned 11. The main lessons learned are as follows: (a) It is difficult to bring about dramatic improvement in the performance of a Government-owned bank with weak management capabilities, political interference in the decision making process and poor oversight. Equity infusions in a poorly managed institution will not work unless they are linked to ownership, and appropriate institutional and administrative safeguards. The privatization option should have been far more vigorously pursued before the end of the project. This lesson has been assimilated in the Financial Institutions Development Technical Assistance Project (Cr. 2497-MAG) that supports improved banking supervision and the privatization of state-owned banks as a priority of future adjustment lending. (b) It is inadvisable to undertake such a credit operation without an overall clear assessment and understanding regarding the evolution of the financial and banking sector. (c) Timely remedial actions need to be taken to address weaknesses in project design and operations. For example, the Credit was not closed in 1991 at the start of the macro-economic crisis, nor in 1993 when it was apparent that all Evaluation Summary vii planned activities had been implemented and that no scope existed under the prevailing circumstances to continue existing activities or execute new ones. PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL CREDIT PROJECT (CREDIT 1804-MAG) PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE 1. Project Identity Project Name: Second Agricultural Credit Project Credit Number: 1 804-MAG RVP Unit: Africa Regional Office Country: Madagascar Sector: Agriculture Subsector: Rural Credit 2. Background 2.1 Agriculture is an important economic activity in Madagascar, contributing about 30 percent to its gross national product and 60 to 65 percent to export earnings. Over 80 percent of the economically active population is dependent on the sector for their livelihood. Agricultural performance had suffered in the seventies and the early eighties from ill- conceived Govern;nent policies that were aimed at protecting the interests of the politically vocal urban consumers; farm prices were set at too low a level to make investments in the sector attractive, and as a result, sector performance was well below the population growth rate during this period. In the early 1980s, Government faced a fiscal crisis following its heavy reliance on borrowing to finance its ambitious public investment program. In response to this crisis, Government embarked on a structural reform process which included the liberalization of prices, the beginning of a privatization program and increased support for the private sector. This program was encouraged by IDA Credits for sector adjustment. Initial results from this adjustment program were promising, with a strong agricultural production response. IDA provided support to the First Agricultural Credit Project (Cr. 1064-MAG) in the early 1 980s to address the lack of access to credit in rural areas. The First Agricultural Credit Project (BTM I, Credit 1064-MAG) 2.2 The Credit supporting this project, in the amount of SDR 8.7 million, was approved by IDA in September 1980. The project was implemented by the National Rural Development Bank (Bankin' Ny Tantsaha Mpamokatra, BTM). The objectives of this 2 Part I: Project Reviewfrom Bank's Perspective project were to support Government's policy to extend credit to smallholders and strengthen BTM's managerial and financial capabilities as a rural development bank. The two main components of the project were to: (i) provide loans to smallholders with under five hectares of land, using local government institutions as intermediaries; (ii) grant credit for experimental projects, including some based on collective farming; (iii) train BTM staff in financial, managerial and computer disciplines; (iv) furnish logistical support to BTM, including mobile banks; and (v) develop a computerized information system. Implementation of the first project coincided with a catastrophic period that was primarily the result of poor macro-economic policies. Sharply reduced demand for agricultural credit caused a collapse in agricultural input supply services. There was also a major crisis in the banking sector that can be attributed for the most part to unwise lending, largely to the public sector. Most of these loans had received Government guarantees, but these guarantees were not honored. Moreover, BTM was inexperienced in lending to smallholders and as a result repayment was low and the project did not achieve its objective of increasing lending to small farmers. During the project period, the number of loans provided to small farmers by BTM did not increase and only two percent of the IDA funds allocated for credit were disbursed. The Development Credit Agreement (DCA) was amended three times to allow the use of IDA funds for financing imports of farm inputs (fertilizers, pesticides and veterinary drugs) and for tractors and farm equipment. Sixty-eight percent of the Credit was used to finance these imports. Actual expenditure on vehicles, computer equipment, consultants and training amounted to SDR 2.06 million compared with SDR 450,000 planned. The closing date was extended twice; the Credit was finally closed in December 1986, at which time the undisbursed balance in the amount of SDR 426,000 was canceled.2 3. Project Objectives and Description 3.1 The second project was designed to be part of an IDA supported broader agricultural reform program, which aimed to reduce state involvement in the economy, to give greater emphasis to market forces to increase efficiency of resource utilization, and to strengthen key institutions, in this case BTM. The project complemented efforts being made under other IDA Credits to encourage the private agricultural sector, notably under the Agricultural Sector Adjustment Credit (Credit 1691-MAG), and the Second Agricultural Institutions Development Project (Credit 1709-MAG). Within the context of this program, the project had the following objectives: (i) to contribute to further institutional development of BTM; and (ii) to meet the foreign exchange costs of new investments, expansion and improvement of existing ventures in agriculture (production and primary processing/marketing). A major emphasis was to be placed on pulling BTM out of its difficult financial situation and putting it on a sounder track. In this, it aimed to further build on the institutional strengthening activities financed under the first project (primarily portfolio and financial management) and bolster BTM's sagging equity by passing on the proceeds of the Credit as a grant to BTM. The project was conceived to assist BTM to achieve a full financial recovery and contribute to the Malagasy agricultural sector by extending credit to productive undertakings. 2 Project Completion Report: "Madagascar, First Agricultural Credit Project (Cr. 1064-MAG)," Report No. 7768 Part I: Project Reviewfrom Bank's Perspective 3 3.2 The project was to be implemented over a seven year period from 1987 to 1994. The two principal components were as follows: (a) Medium term credit for agricultural production and processing enterprises (including livestock, fisheries and forestry): Loans were to be provided by BTM for a maximum of 15 years to finance the foreign exchange costs of investments, mainly for machinery, equipment and spare parts. Loans were to have a minimum size of US$20,000 and a maximum of US$500,000. Interest rates were to be positive in real terms and borrowers were given the choice of borrowing in foreign currency, or borrowing in local currency at a higher rate of interest with BTM assuming the foreign exchange risk. All loan applications for more than US$100,000 were expected to show an Economic Rate of Return (ERR) of at least 12% and a Financial Rate of Return (FRR) to exceed the cost of borrowing by at least three percent. (b) Institution Building: This component complemented the one initiated under the previous project and was to provide financing for: (i) extension of the ongoing technical assistance to BTM's training division and to the computer division; (ii) use of short-term consultant services to help BTM strengthen its project promotion and appraisal capacities; (iii) continued financing of assistance to audit and institution building from the internationally recruited firm under BTM I; (iv) financing of the incremental costs of the pilot field level credit management project aimed at reviving smallholder credit; (v) consultant services to assist entrepreneurs with further technical preparation of project ideas that were prima facie viable at the identification stage; (vi) consulting services for a study on agricultural credit policy; and (vii) procurement of specialized computer hardware and development of specialized banking operation software that would enable BTM to initiate and monitor portfolio analysis and improve its current banking operations. 4. Project Design and Organization 4.1 The design of the credit component differed radically from the previous project. At the time the project was prepared Madagascar was introducing a more open market economy, where the role of state enterprises was to be progressively reduced. This was also a period when foreign exchange was scarce and the banks in Madagascar had few funds available for medium term development lending. Project lending was provided only for medium/large scale enterprises, with a ceiling of US$500,000 on individual IDA commitments to prevent undue concentration of lending to a small number of enterprises. Technical assistance was provided to strengthen BTM's rural credit department and improve small farmer credit procedures. Given the unsatisfactory experience with traditional lending to smallholders under BTM I, funding for this type of credit was excluded from the second project. However, in light of the major weaknesses apparent in Madagascar's smallholder agricultural credit system, and the failure of the smallholder credit component in the first agricultural 4 Part!: Project Reviewfrom Bank's Perspective credit project, the second project included provision for a credit study designed to develop proposals for a more appropriate national agricultural credit policy and funding for a pilot operation to test novel ways of providing smallholder credit. 4.2 The staff appraisal report cited three main risks: (i) BTM devoting too much of its energies to short-term commercial activities and not taking sufficient interest in development lending; (ii) the banking sector problems intensifying, which would aggravate BTM's financial problems; and (iii) the Government backtracking on its commitment to develop the agricultural sector. As it turned out, the second risk was underestimated and seriously affected the institutional strengthening activities financed under the project. Moreover, it turned out during project implementation that BTM's institutional strengths and weaknesses had not been adequately reviewed during appraisal. 4.3 The project was to be implemented by BTM within the framework of work programs agreed with its Board and IDA. BTM was to be responsible for all aspects of identification and promotion, assistance in preparation, appraisal and supervision. BTM was to make decisions independently and apply selective technical and solvency criteria included in its current appraisal procedures. BTM was also to manage a study by consultants to formulate an agricultural credit policy with special attention to the needs of smallholders. 5. Implementation A. General 5.1 The Macro-Economic and Political Environment. Project implementation started during a period when Madagascar was introducing its structural adjustment program and this helped to create a more favorable economic environment for the project. However, in 1991, when the country was starting to experiment with a more open political system, massive strikes seriously disrupted the economy and paralyzed the political decision making process. Following the political transition, the macro-economic situation worsened with an overvalued exchange rate and serious shortages of foreign exchange. This caused a financial crisis that affected the state owned banking system in general, and BTM in particular. The poor quality of BTM's overall loan portfolio was negatively affected by these events; most of its loans to public sector enterprises and to private companies with ties to the old Government became non-performing, and Government was unable to honor the guarantees it had provided. For several enterprises that had received project funded loans, these events came at a very bad time, for they had just completed the capital development programs funded under the project, and they could not achieve projected sales and cashflows. It therefore became necessary for BTM to delay or reschedule repayments for these loans. In 1991, Government established a steering committee to oversee BTM's operations, and changes were made at the senior management level. In 1992, BTM was suspended from participating in the IDA supported Financial Sector and Private Enterprise Development Project (APEX; Credit 2104-MAG), because of problems related to overall portfolio management highlighted in the 1991 audit results; this suspension remains in effect. Project performance and staff morale were Part 1: Project Reviewfrom Bank's Perspective 5 adversely effected by the uncertainties about BTM's future. A floating exchange rate was introduced in May 1994, that led to a substantial fall in the value of the FMG (about 50 percent); the subsequent inflation led to a reduction in domestic demand, which further exacerbated the problems that some project supported investments faced - particularly, those investments that were not export oriented. Preliminary studies designed to prepare for BTM's privatization have been funded by IDA under the Economic Management and Social Action Project (EMSAP; Cr. 1967-MAG). However, this option should have been more resolutely pursued from the beginning of the project. 5.2 Project Effectiveness and Closing. The Credit became effective on October 8, 1987, less than three months after the Development Credit Agreement (DCA) was signed. It was closed on June 30, 1994, as originally planned. Even though no activities were funded under the Credit after 1992, the Credit was kept open until its scheduled closing date, because it was believed at the time that a decision on the privatization of BTM was forthcoming and that the balance of the funds would be needed to prepare BTM for privatization. 5.3 Project Cost and Financing. The project was implemented along the lines planned, and 88 percent of the Credit was disbursed. Total project costs in local currency were 30 percent higher than estimated at appraisal, although they were 38 percent lower if costs are expressed in US dollar terms. This divergence arose because of the substantial devaluation of the FMG in relation to the US dollar over the project period. By the closing date, the FMG was worth less than one third of its value at the time of appraisal. The proportion of total project costs spent on individual components was similar to the appraisal estimates. However, within the institution building component the cost overrun was higher for consultants and training as compared with computer equipment purchases. 5.4 At appraisal, total project costs were estimated at US$19.3 million, with IDA financing US$10.0 million (52 percent of project cost), BTM US$6.15 million (32 percent), and borrowers US$3.18 million (16 percent). Actual project costs are about US$12.0 million, with IDA providing US$9.5 million (80 percent); BTM providing US$2.4 million (20 percent); and borrowers providing an unknown amount. Details on project costs and financing are given in Table 3, Part III. 5.5 Disbursements. Eighty-eight percent of the Credit was disbursed (see Table 4, Part III), and utilization by category was similar to that planned originally. At most times disbursements were ahead of the appraisal schedule. The Credit was closed on June 30, 1994, as originally planned, at which time the undisbursed balance of SDR 938,000 was canceled. SDR 5.2 million was disbursed for 21 medium term development loans, equivalent to 82 percent of the amount provided for under that category. Disbursements for computer equipment (SDR 912,000) were almost exactly the same as the amount allocated originally (SDR 900,000). Disbursements for consultants' services and training were 17 percent above the original allocation (SDR 701,000 compared with the original allocation of SDR 600,000), primarily as the result of the introduction of the computerized accounting system. Most project activities were completed more than two years before the Credit's closing date. During the latter stages of the project, BTM asked IDA if the undisbursed balance could be 6 Part 1: Project Review from Bank's Perspective used to finance additional investment credits and for further development of BTM's computer system. IDA did not agree to support these additional investments, primarily because of BTM's unsatisfactory financial position at the time and the uncertainty concerning its future. B. Individual Project Components B.1. Medium-Term Credit. 5.6 By the time the project became effective BTM had a significant pipeline of loan applications, and this made it possible to commit most of the funds available for sub-loans at an early stage. Twenty six loan applications were submitted to BTM, 23 were approved and disbursements were made for 21, of which fourteen were Category A and seven Category B loans. Category A included loans for more than the equivalent of US$150,000 and these required prior IDA approval. BTM was not required to seek IDA approval for the smaller loans covered by Category B. For investments in excess of US$100,000, equivalent estimates of economic and financial rates of return were made, with minimum acceptable rates of 12 percent for the Economic Rate of Return (ERR), and 3 percent over the borrowing rate for the Financial Rate of Return (FRR). Most borrowers used project funding to hire local consultants to help prepare these evaluations. These evaluations were not subjected to a thorough post-review by Bank supervision missions and turned out to be seriously flawed. Sensitivity analysis was not performed to study the impact of foreign exchange rate fluctuations and domestic prices on the financial rate of return of these investments; when the floating exchange rate was introduced in April 1994 several of the borrowers of these loans became insolvent. 5.7 Loans were used for development of fisheries, livestock (production and processing), crop processing and agricultural production (see Table 6). Loans were heavily concentrated on fisheries and livestock. The original project design included a ceiling of US$500,000 for IDA participation in individual loans. During project implementation, this ceiling was increased to US$1.0 million, in order to allow for the participation in several large investments with a high likelihood of success (dairy production and prawn fishing). In practice, project lending was concentrated on a small number of large loans. Five loans for fisheries accounted for 54 percent of total disbursements under this category of the Credit. These were used primarily to finance acquisition of trawlers (for prawns), processing and refrigeration equipment and refrigerated trucks. Only three existing fisheries enterprises benefited from these loans: one enterprise obtained two loans, while two other fishing companies later merged. Four loans for the livestock sector accounted for 26 percent of disbursements under this category. These included a large milk processing factory and a large poultry enterprise. The balance of the Credit funds were used for a variety of crop production and processing enterprises (processing wine, litchis, coffee, perfume, maize milling, production of coconuts, sugarcane, etc.). The eight largest borrowers accounted for 82 percent of lending. Part I: Project Reviewfrom Bank's Perspective 7 5.8 This component suffered from inadequate Bank supervision during the later stages of project implementation, and information on the performance of the investments made with these loans is incomplete (para. 6.1). None of the post-1992 Bank supervision missions followed up on problems with loan repayments or financial performance of the borrowers. BTM followed up systematically with borrowers in order to assess their ability to maintain loan repayments. In part due to the deteriorating macro-economic environment, 1991 loan repayments reached only 35 percent (repayments received as a percentage of repayments due in that period). During 1992, repayments exceeded the amount due (116 percent), primarily because of prepayments on loans denominated in foreign exchange. As of December 31, 1992, out of the SDR 5.2 million lent (about US$7.0 million), eight loans totaling SDR 1.95 million (36 percent) had been repaid early, repayments on eight loans totaling SDR 2.5 million (46 percent) were on schedule, repayments on two loans totaling SDR 0.63 million (12 percent) had been rescheduled, and four borrowers had defaulted on loans totaling SDR 0.37 million (7 percent). In 1993, repayments reached 83 percent and in 1994 they dropped again to 54 percent, primarily because of the effects of the sharp devaluation on the economy. As part of project design, borrowers were given the choice of borrowing in foreign currency at a relatively low interest rate, or borrowing in local currency at a higher rate, with BTM assuming the foreign exchange risk. Initially several borrowers chose to borrow in foreign exchange and assume the foreign exchange risk. But subsequently, as the perception grew that a devaluation was imminent, several borrowers, including one of the largest, decided to repay their loans prematurely. This explains the repayment rate of 116 percent achieved in 1992. At the present time, almost all of the remaining outstanding loans are denominated in local currency. B.2. Institution Building 5.9 Organization and Human Resources Development Component. Initial progress was satisfactory. An evaluation of all personnel was carried out and BTM vigorously set about implementing the recommendations from this exercise (staff redeployment and reductions in force). The results were major public confrontations between the unions, who feared loss of employment, and management, who wanted to streamline operations; at one point the Director General of BTM offered his resignation to Government. Following political intervention, the Personnel Director was transferred to the position of Director of the Rural Credit Department, which calmed down critics. The 1990 supervision mission decided to postpone any further activities under this component until Government was willing to endorse the project proposals. Satisfactory progress of this component was subsequently made impossible because of the political transition in 1991, which lead to uncertainty about the future of BTM management and its ownership. 5.10 Computerization of BTM's Operations. Installation of the project financed computer equipment was completed in 1990 both in BTM's headquarters and in 17 branches. Major problems were encountered in making the newly installed computerized systems operational, and BTM's accounting system was seriously disrupted for several years. The computer hardware was generally satisfactory. The main problem involved the software. BTM decided to develop its own computer programs, instead of using standard software. Initially 8 Part 1: Project Reviewfrom Bank's Perspective the software developed was inadequate and was introduced prematurely, which caused serious disruptions in the operations of BTM. Since that time a major effort has been made to rectify the problems and the system is now operational. 5.11 Technical Assistance. The audits of BTM were done by the internationally recruited firm that was already involved with BTM 1. The audits, were in general well done, although not always complete, given the near absence of accounting and audit standards in the country. For instance, staff from the Industry and Energy Division of the Department reported that the 1991 audit departed from international norms in that the audit lacked a discussion on non- accrual policies. The quality of subsequent audits improved. The audits consistently pointed out the weakness of BTM's portfolio management, which ultimately led to the replacement of the management team. 5.12 Training. Improved procedures were implemented for loan appraisals and staff received training in evaluating the benefits and risks of proposed investments. Training was also provided in other areas of financial management. The training was carried out by an expatriate consultant who provided in-house training seminars, as well as specialized training supported by case studies for the agents dealing with project evaluation. Training focused on every aspect of project analysis, including technical, marketing, economic and most importantly, financial analysis (i.e., financial profitability, internal rate of return, etc.). 5.13 Pilot Field Level Credit Management Project. The rural finance team was reinforced in three pilot areas (Vakinankaratra, Lac Alaotra and Miarinarivo). Where the project worked with efficient agricultural extension, the results were encouraging and the decision was taken in 1989/90 to expand the pilot operations to the areas covered by the French financed South-East Agricultural Development Project, the IFAD financed Second Highlands Rice Project and the IDA financed Agricultural Extension Pilot Project (Cr. 2150-MAG). BTM was initially reluctant to expand its lending significantly, and refused to lend to voluntary credit associations, opting to lend to individual members only. The repayment experience was encouraging (95-100 percent). A test was initiated on lending to associations at harvest against paddy secured in a common store. This enabled farmers to avoid selling at low post-harvest prices. The main problem was the high initial investment in storage facilities, therefore the pilot project provided a subsidy of US$100 per farmer to help build these facilities. The pilot field level credit management project showed the difficulties that BTM faced in providing this type of service, because of the high costs in administering these types of loans. 5.14 Rural Finance Study. This study was in the end not funded from this Credit, but with financial help from EMSAP. Completion of the study was delayed until 1990, compared with the original target date of 1988. The quality of the study was good and laid the foundation for the Rural Finance Project Technical Assistance Project supported by Cr. 2459- MAG, which was approved by the Board on February 16, 1993. The study underscored the problems in rural credit provision in the absence of rural savings and loan institutions. Part I: Project Reviewfrom Bank's Perspective 9 6. Project Results 6.1 Initial results indicate that this project has been unsatisfactory overall. The medium- term credit component has benefited only a small number of enterprises, some of which could have obtained the credit from other sources. An in-depth analysis of the impact of the investments financed from the proceeds of the credit cannot be carried out for lack of pertinent information, as borrowers were not required to maintain separate accounts on the investments made with the loans. BTM did continue to track the overall financial performance of these borrowers. Based on a 1994 review, six of the eight largest borrowers reviewed, were financially healthy, and BTM predicted continued growth. BTM reported that these six borrowers were producing at close to capacity. The review was, however, backward looking-based on past performance-and did not analyze the constraints these companies were facing in a deteriorating macro-economic environment. More up to date information on these eight firms is not available, and no information is available on the other 13 firms. 6.2 The institutional development component was only partially implemented and the activities executed showed major initial problems, such as the computerization program and the human resources and organizational activities, which had to be ultimately suspended. The impact of this component is unsatisfactory, due to BTM remaining a weak financial institution overshadowed by a burden of poorly performing loans. In November 1995, a provisional administrator was appointed to manage BTM's day-to-day operations on an independent basis, prior to selection of private financial advisors to facilitate privatization of BTM. The rural credit study, part of this project but funded from other sources, was instrumental in helping prepare the Rural Finance Technical Assistance Project (Credit 2459- MAG). This pilot project, which became effective in November 1993, is intended to develop savings and loan associations for small farmers and the rural poor. Initial implementation experience under this project is satisfactory. 6.3 At appraisal no attempt was made to estimate an overall ERR for this project, although BTM was required to ensure that all loan applications over US$100,000 could demonstrate an ERR of at least 12%, and a FRR of at least 3% above the borrowing rate. This analysis was done for all subloans granted, but as indicated previously, no ex-post analysis of these subloans can be done for lack of detailed information. 7. Sustainability Although a successful outcome to the majority of the physical investments is expected, sustainability of the lending program is unlikely. The level of loan repayments for the medium term credit component has been unsatisfactory and this will prejudice future operations unless repayment performance can be improved. In fact, BTM would have been technically insolvent, at the start of the project, were it not for the massive equity infusion under the Credit. However, this assistance did not correct the many problems that plagued BTM, and it is once again insolvent: BTM's 1994 accounts show it has a negative net worth 10 Part 1: Project Review.from Bank's Perspective and it does not meet its reserve requirements. This experience has confirmed the experience in other countries, that equity infusions in a poorly managed institution will not work unless they are linked to appropriate institutional and administrative safeguards. This lesson has been assimilated in the Financial Institutions Development Technical Assistance Project (Cr. 2497-MAG) that supports improved banking supervision and the privatization of state-owned banks as a priority of future adjustment lending. 8. IDA Performance 8.1 Eight supervision missions reviewed progress under the project between May 1987 and June 1994 (Table 9). Initially supervision was quite intensive, with three missions during the first year. But thereafter missions were often more than 12 months apart and mission compositions less balanced. During the initial years, when project implementation was in full swing, most supervision missions included a financial analyst and frequently a computer specialist. Later supervision missions generally included only an agricultural economist, who had no experience in the sub- sector, and inadequately addressed the problems that surfaced during project implementation (macro-economic crisis, problems with repayments of sub-loans, poor progress on institutional development component, poor compliance with several covenants of the DCA). Moreover, supervision missions should have placed greater emphasis on BTM's overall development as a financial institution, including more intensive contacts with other Bank staff, who were responsible for supervising APEX. Project ratings did not reflect the development impact of the project. The project was rated a 2 throughout its life, even after 1992, when it was apparent that BTM as an institution faced serious problems and that repayments were lagging. In 1992, IDA decided that BTM should no longer be allowed to participate in the APEX Credit. In retrospect, this decision may have been too modest and the Credit supporting BTM II should have been closed prematurely, rather than maintaining it while hoping for a quick political decision. 8.2 Lessons for future IDA lending include the following: (a) It is difficult to bring about dramatic improvement in the performance of a Government-owned bank with weak management capabilities, political interference in the decision making process and poor oversight. Equity infusions in a poorly managed institution will not work unless they are linked to ownership and appropriate institutional and administrative safeguards. The privatization option should have been far more vigorously pursued before the end of the project. (b) It is inadvisable to undertake such a credit operation without an overall clear assessment and understanding regarding the evolution of the financial and banking sector. Part L: Project Reviewfrom Bank's Perspective 11 (c) Timely remedial actions need to be taken to address weaknesses in project design and operations. For example, the Credit was not closed in 1991 at the start of the macro-economic crisis, nor in 1993 when it was apparent that all planned activities had been implemented and that no scope existed under the prevailing circumstances to continue existing activities or execute new ones. 9. Borrower Performance 9.1 During the project period BTM's overall performance as a bank has been uneven, often unsatisfactory. At one stage (1991) BTM's problems were so severe that Government established a special committee to oversee its operations and institute management changes. The Borrower was slow in taking a decision on BTM's future, even when it became apparent that its future was in serious peril. A provisional administrator was not appointed until November 1995. 9.2 BTM's performance in approving and supervising sub-loans under the project was generally inadequate. Several smaller sub-loans were approved based on flawed analyses where it turned out later that the borrowers were overextended financially. Supervision of the performance of the investments made with the proceeds of these loans was also inadequate. as they were backwards lookings and lacked a review of the continued viability of these investrnents. Moreover, BTM was often slow to take remedial action when borrowers encountered problems. 9.3 Serious problems were experienced during the first few years following introduction of the new computerized accounting systems developed under the project. BTM appears to have contributed to these problems through (a) trying to develop its own software, rather than using standardized software, and (b) through premature adoption of systems which had not been adequately tested. 10. Consulting Services 10.1 Consulting services were provided to help BTM develop several aspects of its operations, including developing new computer systems, improving auditing procedures, strengthening the rural credit department and carrying out appraisals for sub-loans. In general the performance of consultants was satisfactory. There were, however, problems with the development of computer systems. One of these consultants was asked to withdraw prematurely due to technical and administrative problems. However, the most important factor contributing to the initial failure of the new computerized systems was not inadequate technical advice, but BTM's decision to develop its own software and excessive haste in trying to make it operational. 11. Project Documentation and Data 12 Part : Project Review from Bank's Perspective 11.1 During the period up to 1991, when project implementation was proceeding rapidly, progress reports and other data were routinely provided by BTM and these supplied information on the sub-loan and computerization components. More recently progress reports have provided information on sub-loan performance, but not on the current status of the computer systems. The reports lacked information, however, on the impact of the investments made with the proceeds of the sub-loans and no provision had been made for this in the design of the project. Project reporting appears never to have been satisfactory on those aspects of the project which concerned institutional development of BTM as a bank, although some information was available in the audit reports. For example, progress reports have never discussed covenants in the Credit and Project Agreements relating to the need for six monthly reviews of the interest rate structure, measures BTM should take to protect itself against foreign exchange risks, and progress in implementing findings of the rural credit study. Audit reports on the project accounts have also not always been submitted on time. PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL CREDIT PROJECT (CREDIT 1804-MAG) PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE BTM's minor comments on the original draft PCR have been fully incorporated into Parts I and III of the report. No formal Government comments were received. PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL CREDIT PROJECT (CREDIT 1804-MAG) PART III: STATISTICAL INFORMATION Table 1: Related Bank Credits Credit Name Credit Purpose Year Status Comments Number Approved l Agricultural Credit 1064-MAG Extend credit to smallholders and 1981 Closed Closed in 1986 strengthen BTM. Agricultural Sector 1691-MAG Adjustment operation in support of 1986 Closed Closed in 1988 Adjustment Credit rationalizing public investments and liberalizing producer price controls in the agricultural sector Second Agricultural 1709-MAG Institutional strengthening project in 1986 Closed Closed in 1992 Institutions support of implementing Cr. 1691 - Development Project MAG Economic 1967-MAG Social action program and 1988 Ongoing Closing date Management & Social strengthening institutional capacity extended by 2 years Action Project for structural adjustment to June 1996. Agricultural Research 2042-MAG Support national agricultural research 1989 Ongoing I_ program l Financial Sector and 2104-MAG Medium term Credit through 1990 Ongoing Private Enterprise commercial banks. Available for all Development Project sectors. Agricultural Extension 2150-MAG Agricultural extension pilot project 1990 Closed Closed on June 30, Pilot Project 1995 Livestock Sector 2243-MAG Support national livestock 1991 Ongoing Project development programn Rural Finance 2459-MAG This is a new pilot project developing 1993 Ongoing Technical Assistance small farmers savings & loans Project associations Financial Institutions 2497-MAG Strengthening & reforming banking 1993 Ongoing Development Project system Part III: Statistical Data 15 Table 2: Project Timetable Item Date Planned Actual Date Identification 1984 Preparation 1984 Appraisal September 1985 Negotiations March 1987 Board Approval May 21, 1987 Credit Signature July 23, 1987 Credit Effectiveness October 23, 1987 October 8, 1987 Project Completion December 31, 1993 December 31, 1992 Credit Closing June 30, 1994 June 30, 1994 Last Disbursement September 1, 1993 16 Part M.: Statistical Data Table 3: Project Costs & Financing A. In FMG millions Component Appraisal Estimate Actual Actual as % of Appraisal PROJECT COST (1) Sub-Loans 11,966 15,568 130 (2) Institution Building 2,577 3,360 130 Total Project Cost 14,543 18,928 130 FINANCING: IDA 7,527 15,042 200 BTM 4,623 3,846 83 Borrowers 2,393 not available not available Total Project Cost 14,543 18,888 130 B. In US$'000 Component Appraisal Estimate Actual Actual as % of Appraisal Project Costs (1) Sub-Loans 15,900 9,843 62 (2) Institution Building 3,430 2,124 62 Total 19,330 11,967 62 Project Financing IDA 10,000 9,510 95 BTM 6,150 2,432 40 Borrowers 3,180 not available not available Total 19,330 11,967 62 Part Ill: Statistical Data 17 Table 4: Disbursements (SDRs 'OOOs) Date Appraisal Estimate Actual Cumulative Actual as % of l__________________ __________________ Disbursements Appraisal Estimate June 1988 448 356 79 June 1989 1,632 2,818 173 June 1990 4,160 4,230 102 June 1991 5,504 6,554 119 June 1992 6,592 7,218 109 June 1993 7,552 7,218 96 June 1994 8,000 7,062 88 * * Disbursements were negative in the last year because the balance in the special account was reimbursed to IDA. 18 Part III: Statistical Data Table 5: Size of Sub-Loans (SDRs 'OOOs) Size of Loan Number of Loans Average Loan Total Amount Lent Amount Less than 100 10* 47 473 100-249 3 156 469 250-499 5 326 1,628 500 or more 3 885 2,655 Total or Average 21 249 5,225 * These figures are based on IDA disbursement data which is not fully accurate. Two small loans show zero disbursement in IDA records, presumably because the actual disbursements have been shown against different loans or categories. Part III: Statistical Data 19 Table 6: Use of Sub-Loans (SDRs 'OOOs) Purpose Number of Loans* Average Loan Total Amount Lent Amount Fisheries 5 561 2,806 Livestock 4 344 1,377 Crop Processing 6 118 706 Agriculture 6 56 336 Total or Average 21 249 5,225 *Footnote to Table 5 applies here also. 20 Part 111: Statistical Data Table 7: Sub-Loan Repayments Year Repayments Due (FMG Amount Percent millions) Repaid Repaid (FMG % millions) 1988 5 - - 1989 453 338 75 1990 1,979 945 48 1991 1,371 474 35 1992 1,912 2,217 116 1993 1,881 1,556 83 1994 1,182 639 54 Part 111: Statistical Data 21 Table 8: Compliance with Covenants A. Development Credit Agreement Covenant Subject Status Section 2.02 Open Special Account Complied with Section 3.01 (a) BTM to implement project Complied with according to Project Agreement Section 3.01 (b) Proceeds of Credit to be made Complied with available to BTM as equity Section 3.02 Computer equipment to be Complied with procured as provided in Project Agreement Section 3.04 BTM not to pay dividends Complied with Section 3.05 By 1988 BTM to implement Study completed two years recommendations of project late. Recommendation not financed credit study implemented Section 3.06 Improved rules for supervising Complied with the banking system to be introduced by December 1988 22 Part Ill: Statistical Data B. Project Agreement Covenant Subject Status Section 2.01 Project to be adequately staffed Complied with Section 2.02 Sub-loans to be made in Complied with accordance with Schedule to Project Agreement Section 2.03 Satisfactory procurement Complied with procedures to be used Section 2.04 BTM to submit annual work Complied with during programs to IDA majority of project period when lending program active. Section 2.04 Satisfactory project reporting Progress reports have been received regularly on sub- loan performance but not on other aspects of the project. Section 2.07 Interest rate structure to be Not complied with reviewed every 6 months Section 2.08 Credit study to be completed by The study was completed 1988 two years later than expected. Section 3.01 Audit Audit of the project accounts for 1992 was received 6 months late Section 3.02 BTM to make adequate Not complied with provision for bad debts Section 3.04 BTM to maintain specified Not complied with minimum loan recovery rates Section 3.05 BTM not to make loan Complied with repayments in advance of normal loan maturity Section 3.06 BTM to protect itself against Not complied with foreign exchange risks 23 Part 111: Statistical Data Table 9: Use of Bank Resources A. Staff Inputs Staff Weeks 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 Total Preparation 0.6 . Appraisal 31.1 7.7 Negotiation =_=_2.7 = Supervision 2.0 9.7 15.5 7.3 6.3 5.0 2.6 0.1 PCR _ _ __ _ _ _ __ _ _ _ __ _6.5 _ _ Total 31.7 12.4 9.7 15.5 7.3 6.3 5.0 2.6 6.6 * Data not available for 1984-5 B. Mission Data Mission Date Number of Days in Specialization Project Type of ___________ _______ People Field Rating Problem Appraisal Sept. 85 2 N/A. F Supervision 1 May 87 2 _ 6 _ _ F, C 2 M. Supervision 2 Nov. 87 2 7 F, IS 2 M. Supervision 3 June 89 3 N/A F, C, IS 2 M Supervision 4 May 90 2 N/A. F, C 2 M, F, Tr, TA Supervision 5 April 91 1 N/A. F 2 M. F Supervision 6 May 92 1 N/A. AE 2 M, F Supervision 7 June 93 1 N/A. AE 2 M, F, Env. Supervision 8 June 94 1 N/A. AE 2 M, F Specializations: F=Finance, C=Computers, IS=Institutional Specialist, AE=Agric. Economics Problems: M=Management, F=Financial, Tr-Training, TA=Technical Assistance, Env=Environment Note: All of the tables with statistical data on BTM's financial position, balance sheets, etc., which were suggested in the first draft of the PCR, have been left out. IBRD 21799 12 44° 46' Ambre Ecological Complex (L-i) 50' 12°- MADAGASCAR Lokobe ntsianana Nosy Tanikmely Nosy Mitsio LOCATION OF PROTECTED AREAS (L- 2) LEVELS 1, 2, AND 3 (Ll, 2, 3) Lac Sohaka (L-3) * National Parks Hell-Vile * Proposed National Parks Tsaratananoa Vahiarina * Integral Natural Reserves Monongorivo (L-21 Amoj 14' 0 Special Reserves * 1 * Proposed Special Reserves Sarniava A Private Reserves Bcro Andcap C Ecological Complexes Antsohihy (L1 2) Maroey (L-1) 0 Proposed Ecological Complexes Baly )L * SAntaIaha iEEm P.A. Level 3 Katsepy [L-31 Befandriono Paved Roads Mahajanga a-- Masoala I M S S Rir- Ankaralantssko . Tampoketsso Nosy Mongoae '---.Railroads IL I) d'Anolomainso IL-1il 1 6 * National Capitol ILl) 16' Rivers Tsingyde Lac KIkony Monanaro IL-1) Namoroka Tsiombikibo Manmorivos(11 IL 21 Ecological Complex Point Banangozo L2| *(L-2) v \Point Larit Monangozo (L-21 [-21S*-Marie L2 Kas,iy Zks,n I1 KodS'iY I f < \ ~~~~~~~Zahomeno (1. 1 ) "I Ambatovaky Cousse de Kelirely .Abtvy ,Mczcar;' v: . IL2) Ambohtantely Marotandrano (L-2) /"(,-Z(I, 1, ..). Am6oh,tantely ~~~~~~~~~~~Fenoariyo fh M s deL 21 M Atsinanana 18' ? Maintirano D*cl D Betampa Ambolasocr IL-3i Toamasina 18'- Manambolo~~~~~~~~~~~~~~~IL2 msn Bemaraha / Anjozorobe Mangerivola 1L-2) Tsimembo Nosy Barren 3 Tsiroanomandidy R*VO Ecological Complex Basse TsiribihimsMiandrivazo Soavinandriona r a-- na Basso Tsiribihino 2/*--Andosibe Ecological Complex IL-l Aontoirbcdia (Le 20' \ts4r 20' - Morondava O Mahabo Belo Sur Mer Ecological Complex 1L--3) M Mananjory ~~This mop onproducd MangoLy Fianarantsoa Monaniary ly the Maop Design Unei Mangol,y Finratoao Tle Word Bunk. The Moromr e (L)/ boundsr3es,cslo rs soifi,.d soy - 22, (B *h4 *.P fo -shon L2c2 ho try zombitsy Andriro th lis, mop h p0t 22' Morombe Voh,basic (L 2) Pic divoaibe Monakaro The World Book Group, Ecological * solo (L 2) IL-II s) ofr a ny teir-ioi)§ Comple (L-21 Su r y endosenos Sakaraha lsolo Sudor cceptonce o such 11 (L-3) Kolambotritra Farafangana boondories. Recihs Coralliens 5 F L-2)1 Manambo (L-2) 50' de Toliora T U Mido*gy V a Ecological Complex ( i 21 ngindrano 1L-2) -24' Bezo MahaFaly - 24.~~~~~~~11 4 Plateau MaaaAly f AFRIC Ecologicol Complex 1L121 Chainse A Anosyenne 1L-3) Andohaaoelo FL-I) 0 50 100 150 Kilometers ATohnaro O 50 100 MiiesLi > 1 Karimbola Coo Saints Mane 4MADAGASCAR 01 50 100 Miles Ecological C om plex 1LA2) 4 8' AUGUST 1996