Madagascar - Second Agricultural Institutions Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 12610 PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL INSTITUTIONS PROJECT (ATIA II) (CREDIT 1709-MAG) DECEMBER 17, 1993 Agriculture Operations Division South-Central and Indian Ocean Department Africa Regional Office This document has a restricted distribution and may be used bv recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  EXCHANGE RATES Year Malagasy francs Malagasy francs per SDR per USS 1985 698.4 662.5 1986 941.6 676.3 1987 1,751.0 1,069.2 1988 2,054.1 1,407 1 1989 2,085.4 1,603.4 1990 2,012.0 1,494.1 1991 2,673.8 1,835.4 1992 2,626.5 1,910.2 1/ WEIGHTS AND MEASURES i are (a) = 0.0247 acres 1 hectare (ha) = 2.47 acres i kilometer (km) = 0.62 miles I square kilometer (km=) = 0.39 square mile i kilogram (kg) = 2.20 pounds i liter (I) = 0.26 US gallon 0.22 Imperial gallon i ton (t) = 2,204 pounds ACRONYMS ATIA I First Agricultural Institutions Technical Assistance Project ATIA II Second Agricultural Institutions Project ASAC Agricultural Sector Adjustment Credit FOFIFA Center for Agricultural and Rural Development Research IRNT Natural Resources Unit of the Ministry of Research MPAEF Ministry of Livestock, Forestry and Fisheries MPARA Ministry of Agriculture MRSTD Ministry of Research USAID United States Agency for International Development FISCAL YEAR January 1 - December 31 Period averages not yet available; exchange rate is for end of year.  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation December 17, 1993 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Madagascar Second Agricultural Institutions Project (ATIA I) (Credit 1709-MAGj Attached is the Project Completion Report on Second Agricultural Institutions Project (ATIA II) (Credit 1709-MAG) prepared by the Africa Regional Office. Part II was prepared by the Borrower. The operation was not actually a "project" but a complex "umbrella credit" which financed 39 different technical assistance activities executed by three ministries and two agencies. Since many of the most important activities were properly implemented and led to outcomes of good quality (including the basis for two subsequent IDA-supported Credits), the operation is rated as satisfactory and sustainable, and the institutional development impact is rated as substantial. Not all aspects of the operation were successful. Some activities faced important difficulties in implementation, when the technical assistance provided was not clearly related to the work program and goals of the beneficiary units or when volume overwhelmed units' capacity to select consultants, supervise them, and make use of their output. A commendable aspect of the operation was its spare use of covenants, with reliance on annual work programs as a more flexible project implementation instrument. The PCR is complete and informative. No audit is planned. Attachment 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 authorkzation.   FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT MADAGASCAR SECOND AGRICULTURAL INSTrITONS PROJECT (ATIA IT) (CREDIT 1709-MAG) TABLE OF CONTENTS PREFACE .................................................... i EVALUATION SUMMARY ......................................... iii PART I: PROJECT REVIEW FROM IDA'S PERSPECTIVE ................... 1 Project Identity ............................................. I Project Background .......................................... 1 Project Objectives and Description ................................ 2 Project Design and Organization .................................. 4 Project Implementation ........................................ 5 Project Results ......................................... 6 Project Sustainability. ......................................... 7 IDA Performance . ................................... 7 Borrower Performance .................................... 8 Project Relations . ................................... 9 Consulting Services .................................... 9 Project Documentation and Data .................................. 9 PART II: PROJECT REVIEW FROM THE BORROWERS' PERSPECTIVE ... ...... 10 PART m: STATISTICAL INFORMATION ............ .. ................ 16 Table 1: Related IDA Credits .................................. 16 Table 2: Project Timetable .................................... 17 Table 3: Disbursements and Financing ............................. 18 Table 4: Studies .......................................... 20 Table 5: Compliance with the Credit Agreement ...................... 21 Table 6: Use of IDA Resources ................................. 22 MAP 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.   MADAGASCAR SECOND AGRICULTURAL INSTITUTIONS PROJECT (CREDIT 1709-MAG) PROJECT COMIPLETION REPORT PREFACE Attached is the Project Completion Report for the Second Agricultural Institutions Project for which Credit 1709-MAG, in the amount of SDR 8.7 million (US$10.0 million equivalent), was approved on June 10, 1986. The Credit was closed on December 31, 1992, two years behind schedule. 2/ The last disbursement was on MIay 5, 1993, at which time the undisbursed balance in the amount of SDR 1.07 million (US$1.5 million equivalent) was cancelled. The report was prepared by the Agricultural Operations Division of the South-Central and Indian Ocean Department (Preface, Evaluation Summary, Parts I and III). This report is based, inter alia, on the Staff Appraisal Report, the Development Credit Agreement, supervision reports, correspondence, and the mid-term evaluation report. The Borrower's comments are presented in Part II. 2/ The credit closed on June 30, 1991 for all categories of disbursement, except for categories l.a and 3.a through c. The closing date for these categories was extended to December 31, 1992.   iii MADAGASCAR SECOND AGRICULTURAL INSTITUTIONS PROJECT (CREDIT 1709-NIAG) PROJECT CONIPLETION REPORT EVALUATION SUMMARY Project Background 1. At the time the project was appraised, in 1985, Madagascar was emerging from a decade of ill-conceived economic dirigism and mismanagement. An adjustment program had begun the necessary macroeconomic reforms, which initially concentrated on demand management. By 1985 the focus was shifting to supply side adjustment with a program supporting economic liberalization and improved public sector management. In the agriculture sector, the government's program was to promote private, smallholder production within a market economy and to support this production with improved public investment in infrastructure and essential public services, notably research and extension. 2. The credit for the First Agricultural Institutions Project (ATIA I; Cr. 1249-MAG) became effective in 1982 and was substantially completed by 1985. Its objective was to support the nascent reform program in agriculture. Project activities concentrated on strengthening sectoral planning, improving parastatal management, and reorganizing and strengthening producer services. The Project Completion Report found that ATIA I had been instrumental in preparing the liberalization of the agricultural economy and in creating tools for public resource management. Project implementation was, however, often difficult as it relied on a process of change that had both political and technical dimensions, and which depended on astute deployment of technical assistance. Objectives 3. The objectives of ATIA II, framed within the overall sector strategy, were: (a) to improve the management of the most important agricultural institutions - the Ministry of Agriculture (MPARA), and the Ministry of Livestock, Forestry and Fisheries (MPAEF); (b) to help these two Ministries use scant resources efficiently; (c) to build up planning capacity and help develop specific operational strategies for important public services (agricultural extension and research); and (d) to develop capacity in policy and economic analysis. These objectives were specifically linked to the reform program undertaken by the Government, which was supported by the Agricultural Sector Adjustment Credit (Cr. 1691-MAG), which was approved on May 8, 1986. Implementation 4. Most activities in MPARA were implemented as planned. Activities in support of MPAEF started in 1985, but came to a halt in 1987 when MPAEF entered a period of internal crisis where goals were not clear and management deteriorated. A consultant report following an internal audit in 1990/91 was ignored. Activities in support of extension and research were efficiently carried out, except for a natural resources inventory, which was delayed because of managerial problems. The inventory was carried out with the help of primarily local consultants.  iv 5. Actual project costs are estimated at US$11.8 million versus US$12.8 million planned. Financing was provided by the Government of Madagascar (US$1.2 million actual versus US$2.3 million planned), the United States Agency for International Development (USAID; US$0.5 million actual, same as planned) and IDA (US$10.1 million actual versus US$10.0 million planned). Results 6. In MPARA, the project achieved many of its objectives, notably in improving systems for investment programming, managing financial resources, personnel management and information management, and in developing a pilot for a national extension program (supported by the Agricultural Extension Pilot Project, Cr. 2150-MAG). Support to parastatal reform, with financing from the United States Agency for International Development, resulted in some privatisations and restructuring. 7. Less impact was obtained in project monitoring and economic analysis as the units responsible had difficulties in clearly defining objectives and fulfilling their complex task due to lack of qualified expertise and adequate incentives. The training program was unsatisfactory due to lack of a training plan with clearly-defined objectives. It consisted mainly of overseas training courses, which were proposed individually by each service, in lieu of a coordination body empowered to decide on priority needs for the Ministry, which would have made training more relevant to its development goals. 8. MPARA staff have mastered some of the computer-based systems introduced by consultants (notably those related to personnel and financial management), but are not manipulating others with much ease (primarily those related to project monitoring and economic analysis). However, the gains are fragile. Lack of hands-on computer experience makes it vital that systems be simple and easily mastered. In retrospect, it would have been preferable to have labored longer on the development of simpler systems while simultaneously trying to improve the quality of field data. 9. Within MPAEF, internal problems prevented implementation of the project for three years and when a part was implemented, not much time was left for execution. However, the project was seen as helpful by officials who took office in 1990. 10. Activities in support of research resulted in a major reorganization of the Center for Agricultural and Development Research (FOFIFA) and multi-donor support for the research program (supported by the National Agricultural Research Project, Cr. 2042-MAG, approved on June 15, 1989). 11. The natural resource inventory activities achieved their objectives and contributed substantially to the strengthening of local expertise. Sustainability 12. The results of three important project activities are likely to be sustained: the research and extension programs, and the natural resource inventory and capability building program (a second phase will probably be financed by another donor). For the management systems and procedures, their sustainability will only be known in the future. The chances of sustainability are highest where local staff were sufficiently associated in their development, and feel they have mastered the techniques and can modify them as needed, such as the financial management and investment programming techniques. Less likely to be sustainable are the procedures introduced for personnel management, economic study methods and project monitoring techniques. However, the overall  V picture regarding sustainability within MPARA will only be known once decisions have been made on its priorities, strategy and future organization, which will determine the implementation of the various systems and processes that the project has helped to develop. Lessons Learned 13. The lessons from this project can be summarized as follows: - The project was too complex. Design should be kept as simple as possible and project cost should be reasonable in relation with local agencies' budgets; - In order for changes in procedures to be acceptable, they must be simple, adaptable and well understood by local staff; The use of technical assistants should be limited to the ability of agencies to supervise them (para 1 1. 1) and where possible, local staff or consultants should be used. More focused use of technical assistants saves substantial resources and enables institution building and the transfer of expertise; Although there were few covenants in the Development Credit Agreement for this project, this has not harmed its performance. In fact, project control and monitoring through a boilerplate covenant for annual work programs has shown to be a more flexible and efficient instrument for project implementation than the use of specific covenants (para. 4.5); The success of a wide ranging TA project depends crucially on whether the services receiving it have clear overall priorities and a commitment to those priorities. In agencies with such commitment, the project was successful with sustained results; where the commitment did not exist or disappeared, the project failed; The project should have been supported through adequate documentation. There was no preparation report, appraisal report or technical annexes, only a four-page description in the President's Report. As time went on and staff changed, it became very difficult to understand either the rationale or even the intended content of the project; Objectives of the project were too ambitious. Proposals for restructuring a Ministry should have been made in line with the general policy of the Government. It appeared unrealistic to propose, for example, implementing a management- decentralized structure and/or a sector retrenchment plan, when the Government was still unclear on these issues; and Since the project focussed on institutional issues, it was essential that a strong commitment had been made by the Government on the main orientations of the reform, and that a strategy statement seen satisfactory to both parties had been included in the Development Credit Agreement.   I MADAGASCAR SECOND AGRICULTURAL INSTITrTIONS PROJECT (CREDIT 1709-MAG) PROJECT COMPLETION REPORT PART I: PROJECT REVIEW FROM IDA'S PERSPECTIVE 1. Project Identity Name: Second Agricultural Institutions Project Credit Number: Cr. 1709-MAG RVP Unit: Africa Department: South-Central and Indian Ocean Department Country: Madagascar Sector: Agriculture 2. Project Background 2.1 Macroeconomic and Sectoral Background. At the time the project was appraised in 1985, Madagascar was emerging from a decade of ill-conceived economic dirigism and mismanagement. An adjustment program had begun the necessary macroeconomic reforms, which initially concentrated on demand management. By 1985 the focus was shifting to supply side adjustment with a program supporting economic liberalization and improved public sector management. In the agriculture sector, the government's program was to promote private, smallholder production within a market economy and to support this production with improved public investment in infrastructure and essential public services, notably research and extension. First Agricultural Institutions Technical Assistance Project (ATIA I; Cr. 1249-MAG) 2.2 The credit for ATIA I became effective in 1982 and the project was substantially completed by 1985. Its objective was to support the nascent reform program in agriculture by providing analytic tools to decision makers, and to begin the process of rehabilitating run-down agricultural institutions. Project activities concentrated on strengthening sectoral planning, improving parastatal management, and reorganizing and strengthening producer services. The Project Completion Report found that ATIA I had been catalytic in beginning the liberalization of the agricultural sector and in fashioning instruments for public resource management. Project implementation was, however, often difficult as it relied on delicate management of a process of change that had both political and technical dimensions, and which depended on astute deployment of technical assistance. In addition, potential advances at the sectoral level were menaced by an unfriendly environment, for example lack of resources to create motivation or operatidnal efficiency within the civil service. Overall, ATIA I played an invaluable part in the process of adjustment and had a beneficial impact on key public resource management functions for the sector.  3. Project Objectives and Description 3.1 Project Objectives. The objectives of ATIA II, framed within the overall sector strategy, were: (a) to improve the management of the most important agricultural institutions - the Ministry of Agriculture (MPARA), and the Ministry of Livestock, Forestry and Fisheries (MPAEF); (b) to help these two Ministries use scant resources efficiently; (c) to build up planning capacity and help develop specific operational strategies for important public services (agricultural extension and research); and (d) to develop capacity in policy and economic analysis. These objectives were specifically linked to the reform program undertaken by the Government, supported by the Agricultural Sector Adjustment Credit (Cr. 1691-MAG), which was approved on May 8, 1986. 3.2 Project Components. The project comprised 37 activities to help develop human resources, systems and specific products in five institutions responsible for policy analysis and public services in agriculture. 1. For NIPARA, the overall objectives were to set priorities, improve human and financial resources management, strengthen monitoring and analytical capabilities, and resolve some basic operational issues (notably organization of agricultural extension services on a national basis). The project activities (19) were grouped as follows: (a) Support monitoring and evaluation, public investment programming and parastatal reform; (b) Assist in monitoring financial resource flows; (c) Improve personnel management; (d) Develop computer planning and information management; (e) Carry out a training program for managers; (f) Conduct sector studies, including a fertilizer study; (g) Design a national extension system; and (h) Bolster MPARA's economic and policy analysis capability. 2. For MIPAEF, priorities were to streamline essential services and make them efficient, to bring financial resources into line with operational needs by restructuring personnel, and to strengthen programming and analytical capabilities. The nine project activities were to: (a) Strengthen project preparation; (b) Support project monitoring and evaluation; (c) Restructure and reform parastatals; (d) Improve finance and accounting systems;  3 (e) Strengthen personnel management; (f) Develop a medium-term computer and information plan; (g) Plan and execute training programs for Ministry managers and staff; (h) Develop an improved system for maintaining and managing the vehicle fleet; and (i) Support economic and sector studies in order to improve policy and economic analysis. 3. For the Center for Agricultural and Rural Development Research (FOFIFA) under the Ministry of Research (MRSTD), objectives were to define long-term agricultural research priorities, to design an efficient and sustainable agricultural research system, and to prepare and strengthen FOFIFA for carrying out the agreed national agricultural research strategy. The project activities (7) were grouped as follows: (a) Reinforce financial management and research management; (b) Train FOFIFA staff in research and station management, and in specific skills like test design, sampling methods and statistical analysis; (c) Prepare a complete inventory of past research results; (d) Prepare an international symposium on agricultural research in Madagascar; (e) Prepare a long-term master plan for agricultural research; and (f) Prepare a research project to execute the first five year tranche of the master plan. 4. For the Natural Resources Unit (IRNT) under MRSTD, the objective was to develop an information base on natural resources. Conceived as a first phase of a long-term program, the project was to support: (a) collection and analysis of existing information; (b) preparation of inventories (maps and supporting documentation) for land and water resources and for current land-use patterns; and (c) training of national staff. 5. For the Ministry of Finance, the project was to provide logistical support and technical assistance to processes agreed under the ASAC to improve public sector resource management.  4 4. Project Design and Organization 4.1 The project was born out of ATIA I and was designed to be the natural continuation of the earlier project. However, new institutions participated in ATIA II and many new activities were added. Indeed, at the time of appraisal, ATIA II was seen as an "umbrella project' sheltering a number of independent programs to be implemented by different institutions. At negotiations, Government had produced a program showing the underlying coherence of project activities around broad themes of institutional development, resource management and sectoral analysis and planning. However, negotiations appear to have been the only occasion on which Government ever came near to having a grasp of the whole project. Even within individual institutions, the number of disparate activities made it hard for staff to grasp the objectives. In MPARA and MPAEF only one or two very senior staff even understood the project. The project lacked any underlying unity that could have guided and motivated the people who were supposed to implement it. 4.2 MIany components placed too great a demand on scarce management time and skills. They were designed as processes, with summarily defined objectives and built-in flexibility. This created a need for scrupulously careful management and monitoring. The magnitude of this challenge was not grasped by the institutions concerned. The components that did best were those where decision- makers understood and wanted the product, and where technical staff understood and subscribed to the objective and had the skills and resources to produce it. Examples are MPARA's public investment program, the research master plan and the National Agricultural Research Project (Cr. 2042-MAG), the Agricultural Extension Pilot Project (Cr. 2150-MAG), and the natural resource inventory. 4.3 There was too great a provision for technical assistance - two-thirds of project funds were for consultants. The institutions concerned were poorly-equipped for finding the right consultants and in many cases never worked out a satisfactory method of supervising their work, nor a lasting institution-building impact in the form of transfer of usable systems and expertise. In fact, project design could have been improved by less reliance on consultants and more emphasis on self-help. For example, the project had provided for turn-key execution of the natural resource inventory by consultants. Design was amended during execution to provide for building up local capacity, with minimal use of consultants. As a result, the inventory is being successfully completed, a year later than planned but at a lower cost and with a measurable institution-building impact. 4.4 The project was too big in relation to local financial resources. Planned disbursements from the credit were about equivalent to MPARA's total non-personnel operating budget from domestic resources. This created distortions in motivation - a false priority to project actions and a preference for project financed equipment and trips rather than real commitment to project activities. In some cases, Government was unable to come up with adequate counterpart funding, and the Development Credit Agreement was revised to allow IDA to finance a larger share of project costs. 4.5 Legal covenants were very few. During preparation, more extensive use of covenants had been proposed to assure satisfactory implementation conditions for staffing, organization, budget, etc. In the event it was the annual work program mechanism which was used to verify these conditions, and this was certainly the right approach. The covenants would have been unpalatable, unwieldy and quickly outdated.  5 5. Project Implementation 5.1 The credit became effective on February 3, 1987; a delay of about ten weeks caused by time consum.ng bureaucratic procedures. The credit closed on December 31, 1992, after three extensions. 2I Disbursements slowed down after an initially promising start because of delays in implementing the MPAEF component and because of cumbersome local procurement procedures. The last disbursement took place on May 5, 1993, at which time the undisbursed balance in the amount of SDR 1.07 million (US$1.5 million equivalent) was cancelled. Actual project costs are estimated at US$11.8 million (US$12.8 million planned). Financing was provided by the Government of Madagascar in the amount of US$1.2 million (US$2.3 million planned), by the United States Agency for International Development (USAID) in the amount of US$0.5 million (US$0.5 million planned) and by IDA in the amount of US$10.1 million (US$10.0 million planned). 5.2 Most activities in MPARA were implemented as planned. The exceptions were: a food crops study, which was included after appraisal (The funds were used to finance a rice markets study at the end of the project period); a fertilizer study, which was financed instead by a USAID grant; support to the Food Security Department, where IDA and MPARA did not agree on the activities to be supported; and activities in support of extension, which were revised to support implementation of a pilot project on a bigger scale than foreseen. This change of pace on extension successfully resulted in a free-standing extension project financed by IDA (Agricultural Extension Pilot Project, Cr. 2150- MAG, approved on May 31, 1990). 5.3 MPAEF began implementation of the project with PPF financing in 1985, but subsequently entered a period of internal crisis where structure, objectives and staffing were not clear and management deteriorated. Implementation of ATIA II in MPAEF was virtually suspended until 1990, when a large consultant contract was let for an institutional audit. The consultants concluded that a reorganization was the essential preliminary to institutional strengthening. The consultants' analysis was disputed by MPAEF and the recommendations and the originally planned activities were not implemented. 5.4 Activities in support of research were carried out efficiently and resulted in a major reorganization of FOFIFA and effective multi-donor support (including IDA) for the research program (National Agricultural Research Project, Cr. 2042-MAG, approved on June 15, 1989). 5.5 The natural resource inventory was executed in a different way from that foreseen at appraisal. IRNT took two years to take root as an institution and had persistent management problems, but ended up doing a good job technically. At the end of the project, the foundation had been laid for a pernanent national capacity. 5.6 The planned activities at the Ministry of Finance were carried out, in a different way, under projects supported by other donors. 5.7 Deviations from plan are to be expected in a project of this nature, and the annual work program mechanism was designed to give the necessary flexibility during implementation. This system worked fairly well. The programs proved useful budgeting tools. They were less useful in 3/ The credit closed on June 30, 1991 for all categories of disbursement, except for categories l.a and 3.a through c. The closing date for these categories was extended to December 31, 1992.  6 providing a qualitative evaluation of performance. However, submission of annual work programs and subsequent IDA approval was always behind schedule. The risk of institutional change that prevented the MPAEF activities from being implemented was correctly identified at appraisal. In the case of MPAEF, the annual work program mechanism was used, as foreseen at appraisal, to defer implementation until conditions improved. 5.8 The sustained political and technical commitment required in an institution building project was not always available. This was clearly the case in MPAEF, and for some activities within MPARA. Managers were not able to give time to consultant supervision, so that the product was sometimes below standard or consultants worked in isolation from local staff or without clear objectives. Staff were either too busy or lacked clear work unit goals that they could relate to the assistance being provided. Perhaps most importantly, both MPARA and MPAEF lost some of their strategic vision during the project period. If it was clear where the Ministries were heading - and why in 1985, it was less so in 1990. The project did support a redefinition of roles and priorities in 1989-91 for both MPARA and MPAEF but this process has not yet produced a clear plan and has been disrupted by political events. 6. Project Results 6.1 In MPARA, the project achieved important objectives, notably in improving systems for investment programming, managing financial resources, personnel and information management, and in developing an effective pilot for a national extension program (para. 5. 1). Support to parastatal reform (with parallel financing from USAID) resulted in some privatisations and restructuring. 6.2 Less impact was obtained in project monitoring and economic analysis as the units responsible had difficulties in clearly defining objectives and fulfilling their complex task due to lack of qualified expertise and inadequate incentives. In spite of important and costly efforts provided by national staff and consultants in defining a new structure of organization and working procedures for MPARA, no decision was taken to implement them, due to the adverse political and social environment that has been prevailing since 1991. The training program was unsatisfactory due to the lack of a training plan with clearly-defined objectives. The program consisted mainly of overseas training courses, which were proposed individually by each service, in lieu of a coordination body empowered to decide on priority needs to make training more relevant to its development goals. 6.3 MPARA staff have mastered some of the computer-based systems introduced by consultants (primarily those related to personnel management) but are not manipulating others with much ease (primarily those related to production and project analysis). The gains in the latter cases are fragile. Lack of hands-on computer experience makes it vital that systems be simple and easily mastered. In retrospect, it would have been preferable to have labored longer on the development of simpler systems while simultaneously trying to improve the quality of field data. 6.4 In MPAEF, the project had virtually no impact on strengthening the institution, as the consultants' analysis was not accepted by some key high-level staff members (who were displaced later in 1990). Since then there has been some progress made in implementing the component, but the impact was not important in restructuring MPAEF due to the fact that the Ministry itself was divided into two Ministries for political reasons (the Ministry of Water Resources and Forestry and the Ministry of Livestock and Fisheries). 6.5 The research and natural resource inventory activities achieved their objectives.  7 6.6 Overall, the impact of the project has been satisfactory. Two solid programs within a producer services strategy have been produced - for extension and research - and a prime planning tool prepared - the natural resource inventory. In MPARA there has been a qualitative improvement in resource management. The foundations of a national capacity in photo-interpretation and resource mapping have been laid. 7. Project Sustainability 7.1 Three major project activities are likely to be durable - the research and extension programs 4/, and the natural resource inventory and capability building program (a second phase will likely be financed by another donor). For the management systems and procedures, their sustainability will only be known in the future. The chances of sustainability are highest where local staff were sufficiently associated in development to feel they have mastered the techniques and can modify them as needs change. At MPARA, the work done by consultants has been appropriated by Ministry personnel to varying degrees. The most successful are the financial and personnel management, and investment programming techniques. Staff were fully involved in participating in and/or conducting the related operations, as consultants in most cases played the role of a coordinator or a facilitator. Less successful have been economic study methods and project monitoring techniques. As to the restructuring of MPARA, a decision on implementing the consultants' proposals has been pending. (However, the true picture regarding sustainability within MPARA will only be known once a decision has been made on future priorities, strategy and organization, and on the place of the various systems and processes that ATIA II has helped to put in place). 8. IDA Performance 8.1 At appraisal, the complexity of the project was recognized and a supervision time allocation of 24-30 weeks a year was suggested. However, this proved unrealistic and actual allocations were of the order of 11-12 weeks, no more than for a standard project. Within this constraint, IDA made every effort to respond to project needs. Supervision was delegated to the Resident Mission which allowed regular consultation as issues arose, and permitted in-field supervision of consultant progress. It did, however, not prevent a lack of progress on the development of a training management system. Specialist expertise in research, extension, public administration, finance, computers and teledetection was fielded at key points and contributed materially to shaping the project and helping implementation. Flexibility was shown in the light of changing needs, regarding content, implementation and financing mechanisms and timetables. The weakness of delegating supervision to the Resident Mission was that the project was not well enough known at headquarters and was poorly integrated into IDA sector strategy. 4/ Implementation of the projects in support of these programs (i.e. the National Agricultural Research Project, Cr. 2042-MAG, and the Agricultural Extension Pilot Project, Cr. 2150-MAG) has encountered difficulties, primarily because of inadequate counterpart funding.  8 9. Borrower Performance 9.1 ATIA II attracted strong commitment from a handful of senior officials within MPARA and these officials invested heavily in coordination and contacts with IDA. Commitment at the technical level was variable, dependent on the clarity with which work unit objectives were defined and the perception of ATIA II's relevance to internal priorities. (Commitment at decision-making level was uncertain towards the end of the project; it was not clear that successive new Ministers accepted the improved resource management systems as decision-making tools.) 9.2 Within MPAEF, internal problems prevented implementation of the project for three years and when parts were implemented, insufficient time was left for execution. The project was, however, seen as helpful by officials who took office in 1990. 9.3 FOFIFA fully adopted the project as its own and completely identified with both process and product. IRNT suffered protracted birth pangs and ownership was in doubt. Once the institution building activities began to take shape and technical performance was good, MRSTD fully adopted the project. 9.4 The least satisfactory aspect of borrower's performance was the emphasis on equipment purchase and overseas training that emerged, sometimes independent of project objectives. Particularly where project objectives were vaguely defined, no longer relevant or poorly appropriated by officials, ATIA II was seen simply as a source, in the resource-poor environment of the Malagasy civil service, of funding for acquisitions and trips. 9.5 This distortion of motivation in a resource-poor environment reflects a fundamental, structural problem that the appraisal mission identified - how can an external project help effect lasting institutional change in a poorly structured, underdeveloped, demotivating civil service environment? This problem became more pronounced during implementation as public sector budgets and salaries shrank in real terms. One supervision report summed up the general culture within MPARA: - Prevalent sense of impotence, stemming from the annihilatingly low levels of personal remuneration and operating budget, and from the suffocating mantle of civil service regulations. - Lack of sense of purpose, arising from lack of an overall strategy. Many services no longer fit, or feel they no longer fit, with the current economic situation or with the role of the Ministry, yet they have no other instructions. Hence they lapse into routine and submerge themselves in comforting detail. - Uncertainty created by erratic management style, of which the worst aspects are unpredictability and lack of delegation. Lack of an open culture. Retention of information by units is a defense of their territory, an assertion of power, an attempt to define a place in an unstable and enfeebling environment. Within these constraints, the successful units were those that had an overall strategy that defined and justified their existence and function, and that had within the unit, a clear conception of unit goals and a commitment to those goals. Other features that characterized successful units were interest  9 from line management and, preferably, from outside agencies in the product of the unit; and access to some supplementary resources. 10. Project Relations 10.1 IDA stayed in close touch with project agencies throughout and many problems were resolved through the regular contacts made possible by Resident Mission supervision. This allowed a flexibility invaluable in a project of this nature. The disadvantage was that it led to a large number of ad hoc requests, notably for training. This was not only inefficient but probably impeded the development of the intended training management systems. 10.2 IDA cooperated with USAID in supervision of the USAID financed component of the project (parastatal reform). I1. Consulting Services 11.1 Consultants played a large role in the project and had their greatest impact where they played a supporting role in a well-defined and nationally-managed activity. Agencies had difficulty in choosing the right consultants and in supervising them. MPARA ended up using repeater contracts with those firms that had worked well in the early stages. In two cases, this was relatively successful in assuring continuity and responsibility and in building up good relationships. In a third case, the firm made less effort and work was of poor quality. MPAEF failed to find a means of managing the large consultancy it undertook in 1990 and the consultants proved to lack experience in dealing with institutions like MPAEF. MRSTD, FOFIFA and IRNT used consultants sparingly and only to support strong national teams and thus achieved good results. It is not possible to get the required information on the actual number of man-months of consultant time used under the project. 12. Project Documentation and Data 12.1 The project was conceived as a series of processes and prepared through dialogue. Although this gave great scope for flexibility and creativity, it should have been supported by adequate documentation. In fact there was no preparation report, no appraisal report or technical annexes, only a four-page description in the President's Report, and voluminous cost tables. As time went on and staff changed, it became very difficult to understand either the rationale or even the intended content of the project.  10 MADAGASCAR SECOND AGRICULTURAL INSTITUTIONS PROJECT (ATIA 11) (CREDIT 1709-NIAG) PROJECT COMPLETION REPORT PART II: PROJECT REVIEW FROM THE BORROWERS' PERSPECTIVE COMNIENTS BY THE MINISTRY OF LIVESTOCK AND FISHERIES 5/ ON THE DRAFT PROJECT COMPLETION REPORT OF THE SECOND AGRICULTURAL INSTITUTIONS PROJECT (ATIA II) This follows up on the draft report by the OED audit mission and the project completion report on the ATIA 1I project, whose contents came to my attention, and the very practical discussions that took place with Mr. Tran-Luu. 6/ I would like to thank you again for the interest that you and your staff have shown in the institutional and organizational problems in my department. In this context, and without prejudice to any future decisions by the Bank, let me draw your attention to the following matter, which in my view is of fundamental importance and which in general terms one might assume should have been resolved, namely our management capacity and our readiness to undertake the reorganizations and changes required once the external funding is no longer available. From January 1993 onwards, for the next five years, my department will be required to manage five projects in the livestock sub-sector, for a total amount of more than US$70 million. Each project and each component has its own organization and financing system. In light of the difficulties and complexity surrounding each entity, my department has established, at the Bank's suggestion at the time of appraisal of the Livestock Sector Project (Cr. 2243-MAG), a Livestock Sector Program Coordinating Committee. This Committee, bringing together the project managers and the various donors, has an executive office or unit known as the "Steering Group." This Steering Group will be responsible for overseeing the smooth progress of activities in each project and also flows of funds. It will in addition ensure the harmonization of operations and the consistency of decisions. 5/ Under the present structural arrangement, the Ministry of Livestock, Forestry and Fisheries (MPAEF) is divided into two: the Ministry of Water Resources and Forestry (MEF) and the Ministry of Livestock and Fisheries (MERH). Consultant for IDA in the December supervision mission.  11 The Coordinating Committee and Steering Group are located within this Ministry so as to ensure that each entity does not engage in a multiplicity of operations independent of others. Appropriately, each project or sub-project includes a "training' component designed to enhance the technical skills of the agents working for the program. This is not yet true, however, as regards the ministry itself, which is responsible for coordination; here there is an urgent need for organization, methods, systems and procedures that will generate performance charts to facilitate decisions. Specialized technical assistance is planned for this purpose. Nonetheless, it will not be possible for ministry staff to be involved in the introduction and operation of such a system unless they have been trained beforehand. It is for this reason that I regard it as essential to propose the establishment of a "management training' component to complement the activities of the livestock sector program. This training would enable local personnel to manage coordination activities, particularly in the areas of financial management of projects and the programming of investments in financial supervision, monitoring and evaluation and human resources. In this way they could enhance their contacts with service providers and extend the methods and systems inherited from the consultants on their departure.  12 COMMENTS BY THE MINISTRY OF WATER RESOURCES AND FORESTRY 7/ ON THE DRAFT PROJECT COMPLETION REPORT OF THE SECOND AGRICULTURAL INSTITUTIONS PROJECT (ATIA 11) 1. GENERAL CONSIDERATIONS 1. 1 The basic purpose of the Agricultural Institutions Project was to strengthen the capacity of the operational departments of the NMinistry of Livestock, Forestry and Fisheries (MPAEF) in the areas of planning, economic studies, project preparation and financial and human resource management, and also to develop, in the medium term, an information and data processing system, and to strengthen management resources. 1.2 The credit for ATIA II, which was signed on August 1986, was closed on December 31, 1991, one year behind schedule. 8/ While waiting for the Credit (1709-MAG) to become effective, MPAEF embarked on the program in 1985, with a PPF advance of US$350,000. Implementation of ATIA 11 began in 1987 but did not, in practice, reach full development until 1990. 1.3 The draft Project Completion Report on ATIA 11 sent to the Malagasy Government paints a particularly negative picture of MPAEF's performance, without clearly identifying the grounds for such a conclusion. 1.4 While it is true that implementation of components related to MPAEF was considerably delayed for administrative and institutional reasons, the World Bank's Project Completion Report cannot accurately state, without proper justification, that the project had virtually no impact on MPAEF. It would seem that the author of the report based his analysis on the situation that prevailed mid-way through the project and, more specifically, prior to the period when project activities, as revised in 1989, were implemented. 1.5 Some frank clarification is needed for a better understanding of the project's progress and results. 2. PROBLEMS AND CONSTRAINTS ENCOUNTERED 2.1 Several problems and constraints were encountered during project implementation. There was a considerable delay in the start-up of the second phase because of administrative and institutional reasons. 7/ Under the present structural arrangement, the Ministry of Livestock, Forestry and Fisheries (MPAEF) is divided into two: the Ministry of Water Resources and Forestry (MEF) and the Ministry of Livestock and Fisheries (MERH). 8/ The disbursement categories related to MPAEF were closed effective December 31, 1991. A retro-active extension to June 30, 1992 was given to allow for final disbursements.  13 Nonexistence of preparatory documentation 2.2 The lack of agreed minutes of negotiation or a report made it difficult to evaluate the Annual Work Programs. The Development Credit Agreement does not, in fact, contain any details of the costs and activities entailed in each component of the project. Problems concerning the Mtinistry 2.3 The initial project design, prior to negotiations, was handled by teams who were no longer in place by the time the Development Credit Agreement came into effect. The change-over in personnel was apparently poorly handled and the new teams had difficulty entering into the spirit of the project and devising the annual work programs and the terms of reference for the consultants. 2.4 Unlike the staff of the Technical Directorates (Livestock, Fisheries and Forestry), those in the Directorate of Studies, Programming and Financing, who were asked to take charge of project execution, generally lacked the professional training needed to be able to perform the functions involved. The new staff should therefore first have been given additional training in planning and programming as part of sessions held to bring the staff up to standard. 2.5 There also seems to have been insufficient information given to top Ministry officials, explaining the rationale for and the objectives of the Project, to convince them of the activities required and of the results expected. Problerns concerning the World Bank 2.6 There were frequent changes in World Bank staff responsible for monitoring the project between the time of project preparation and appraisal and the time of the negotiations and project implementation. 2.7 These changes were particularly detrimental to the continuity of the monitoring of project documents, but perhaps also to the spirit in which the monitoring was done. The delays in decision- making related to bidding documents (importation of equipment and supplies, consulting services) and approval of the annual work programs are indicative of this. 2.8 This was why the annual work programs for 1987 and 1988 could not be prepared on time. In 1989 all these programs had to be changed, updated, re-discussed with the World Bank and then finally approved. 2.9 After recommending, in 1988, that the support program for the introduction of the programming and budgeting system be adjusted in light of the findings of the evaluation carried out by MPARA, the World Bank itself then questioned the adoption of this system as the result of an evaluation mission in 1989. Problems attributable to external factors 2.10 Certain factors not related to the Project itself thwarted the Ministry's implementation efforts: regulations governing procurement procedures and public finance management. Compromises had to be found between the legislation in effect in Madagascar and the World Bank's procedures governing the hiring of consultants and the procurement of goods. Difficulties were encountered,  14 in particular, prior to the arrival of the Agricultural Country Officer assigned to the World Bank Resident Mission. 3. PROJECT EXECUTION Training: courses and seminars 3.1 Various types of training were organized, both locally and abroad, for senior and support staff of the Ministry. The training component enabled participants to: - further strengthen their knowledge and practical skills in terms of modern methods and techniques of human and financial resource management, accounting and budgetary supervision, development planning and project analysis; - expand their veterinary skills to include areas relating to epidemic disease; - intensify the strategic management of enterprises under the Ministry's supervision; - gain a better grasp of how to approach and monitor the execution of the public investment program (PIP). Missions 3.2 The senior management of the Department conducted two operational missions abroad in response to invitations from the World Bank and the Indian Government: - a review of the cooperative programs between MPAEF and the World Bank; - field trips to India, with the participation of dairy sector entrepreneurs, as part of a study trip of dairy producer associations. 3.3 The missions resulted in the strengthening of technical, economic and financial cooperation between MPAEF and its international partners. Technical assistance 3.4 The Ministry used the services of a consulting firm to give staff initial training in the methods and techniques of project analysis and in how to conduct and monitor feasibility studies, as part of the activities covered by the PPF advance. 3.5 In order to strengthen the activities of the various work units, the Ministry, with the consultants' assistance, had the current situation analyzed by means of fact-finding missions, as a result of which mission participants were able to identify areas of dysfunction and analyzed strengths and weaknesses within the agencies targeted by the missions. 3.6 The task continues of taking the diagnostic study still further, by extending the analysis to all the branches of the Ministry, at both the central and decentralized level, in keeping with MPAEF's role and responsibilities. --- - ----  15 Goods 3.7 Goods acquired under ATIA 11 made it possible to replenish the resources available to the Ministry and to improve the management of its various work units. 4. CONCLUSIONS 4.1 Despite the difficult first few years of the project, the objectives established for the implementation of ATIA II in MPAEF were fully accomplished, thanks to the extension of the closing date of the credit to December 31, 1992, from 1990, as originally scheduled. 4.2 The challenge accepted when the program of activities was revised in 1990 was met. 4.3 The Project Completion Report sees the institutional change within MPAEF, the replacement of certain project leaders, and the discussions with the consultants concerning the methodology used for performing the actions required, as administrative weaknesses or failings. 4.4 All these measures were aimed at improving the structural organization of the work units as a whole and at making sure that the various actions taken by the consultants yielded acceptable results, which was the case in the end. 4.5 The mechanism for establishing the annual work programs was particularly responsible for delaying implementation of ATIA II, for reasons originating in both the Ministry's work units and the World Bank. 4.6 The action taken in the course of this second phase made it possible to build on and strengthen the gains of ATIA 1, particularly in terms of project planning and analysis, human and financial resource management, and economic studies.  16 MADAGASCAR SECOND AGRICULTURAL INSTITJTIONS PROJECT (ATIA II) (CREDIT 1709-MAG) PROJECT COMIPLETION REPORT PART III: STATISTICAL INFORMATION Table 1: Related IDA Credits Credit Tide Year of USS Status ApprovaJ (million) Agricultural Credit Project 09/1980 11.50 Closed (Cr. 1064-MAG) (12/31/86) Second Mangoro Forestry Project 06/19S1 20.00 Clo(ed (Cr. 1161-MAG) (06/30/87) Second Village Livestock and Rural 0219S2 15.00 C1owd Development Project (Cr. 1211-MAG) (12t31/8S) Agricultural Institutions Technical Assistance Project 05/1982 570 Closed (Cr. 1249-MAG) (06t30/87) LAc AJaotra Rice Intensification Project 03/1983 18.00 Clo(3d (Cr. 1337-MAG) (03t31/91) Cotton Development Project IZV 1983 7.90 Clowed (Cr. 1433-MAG) (06/30/91) Irrigation Rehabilitation Project 04t1985 12.00 Ongoing (Cr. 1589-NtAG) Agricultural Sector Adjustment Credit 05/1986 20.00 Clo(ed (Cr. 1691-MAG) (06/30/90) Second Agricultural Credit Project 05/1987 10.00 Ongoing (Cr. 1804-MAG) Forestry Management and Protection Project 0211988 7.00 Ongoing (Cr. 1878-MAG) National Agricultural Rerewch Project 06/1989 24.00 Ongoing (Cr. 2042-MAG) Environment Project 04/1990 26.00 Ongoing (Cr. 212.5-MAO)2. Agricultural Extension Pilot Project 05/1990 Ongoing (Cr. 2150-MAG)  17 Table 2: Project Timetable Original Actual Preparation 1984/5 Appraisal Mislion June, 1985 Negotiations March, 1986 Board Approval June 10, 1986 Credit Signature August IS, 1986 Credit Effectivencss November 18, 1986 February 3, 1987 Credit Closing December 31, 1990 December 31, 199229 LaSt Disbursement June 30, 1991 May 5, 1993 2/ By telex of July 13, 1992, the closing date was extended to December 31, 1992 for disbursement categories l.a (for specific expenditures only) and 3.a-c.  18 Table 3: Disbursements and Financing A. Estimated and Actual Cumulative Disbursements of Credit 1709-MAG (US$ '000) Fiscal Year Appraisal Estimate Actual 10/ Actual as % of Credit 1987 1,000 1,597 16 1988 3,800 2,870 29 1989 7,000 5,602 56 1990 9,400 7,094 71 1991 10,000 8,627 86 1992 9,580 96 1993 10,101 101 Cancellation 1,529 IO/ In SDR terms, only 88% of the credit was disbursed. Discrepancies are due to US$ depreciation vis a vis its main trading partners.  19 B. Disbursements by Category of Cr. 1709-MAG (SDR '000) Original Revised Actuat Actual as a _1/ % of Revised Estimate 1. MPARA Consultants' Services and training under Part A(1) through (5) & (7) 2,390.0 2,390.0 2,272.4 95% Equipment and vehicles uncder Part A(1) through (5) g (7) 480.0 480.0 751.6 157% ConsuLtants' services and training under Part A(6) 145.0 137.1 95% Equipment and vehicles under Part A(6) 145.0 71.1 49% Incremental operating costs under Part A(6) 130.0 109.6 84% 2. MPAEF Consultants' services and training 1,300.0 1,300.0 985.5 76% Equipment and vehicles 220.0 220.0 235.8 107% 3. IRNT 2/ ConsuLtants' services and training 1140.0 1140.0 585.6 51% Works' equipment and vehicles 560.0 560.0 263.9 47% Incremental operating costs 250.0 250.0 626.5 251% 4. MPFE ConsuLtants' services and training 140.0 140.0 152.6 109% Equipment 30.0 30.0 12.8 43% 5. MRSTD Consultants' services and training 490.0 490.0 191.4 39% Equipment and vehicles 150.0 150.0 121.7 81% Incremental operating costs 32.0 32.0 0.0 0% 6. FOFIFA Consultants' services and training 630.0 630.0 571.8 91% Equipment and vehicles 8.0 8.0 60.9 761% 7. Refunding of Project and Preparation Advance 760.0 421.0 420.8 100% 8. Unallocated 120.0 39.0 57.6 148% TOTAL 8,700.0 8,700.0 7,628-5 88% C. Proiect Financiny Source Original Actual Actual as % of (USS thousands) (USS thousands) Estimated IDA 10,000 10,101 101 USAID 500 500 100 Goverrnent 2,300 1,200 Lt/ 52 12,800 11,800 92 11/ On May 15, 1990, the disbursement percentage for IRNT was increased from 100% of foreign expenditures and 80% of local expenditures to 100% of expenditures. 12/ Revision effective on May 24, 1988. 13/ Estimate based on incomplete information.  20 Table 4: Studies 14/ Study Consultant Bureau Etude pour I'amelioration de la gestion des ressources humaines Societe Conseil Maheu Noiseut du Ministere de la Production agricole et du patrimoine foncier Etude du marche du riz Centre d'etudes et de recherche sur le developpement international (CEDI) Diagnostic approfondi du Ministere de la Production animale Cabinet Mpanazava and Coopers and (e1evage et peche) et des eaux et forets Lybrand Etude de l'amelioration du reseau des pistes cafe robusta et Societe d'6tudes techniques de l'Ocean arabica Indien Evaluation et amelioration des systemes de gestion et de Cabinet d'audit et de conseil Rindra comptabilite du Programme national de vulgarisation agricole (PNVA) Etude du systeme d'assiette et collecte de taxation et des Cabinet Fivoarana revenus par les collectivites decentralisees en vue de leur amelioration Inventaire des ressources naturelles terrestres TAMS Plan directeur des cultures d'exportation SCET Agri-BDPA and EEDR Mamokatra 14/ The list of studies is based on documents available from files and may be incomplete. The French titles of the studies have been maintained, because no formal translation of these studies was undertaken.  21 Table 5: Compliance with the Credit Agreement 3.03 AnnuaL Work Programs to be subnitted each year Condition met with delays up to and including by Septeriber 30. 1990. Prepared by MPAEF and IRNT for 1991. Not prepared by MPARA for 1991. 3.04 (a) FOFIFA staffing, budgets and research program Condition met. to be presented each year. (b) FOFIFA programs to be fully funded. Condition met. 3.05 Research Master Plan to be presented to Condition met. donors. l 4.01 (a) Accounts to be kept ... Condition met. (b) ... and audited. Report to furnished within Condition met with del ays. six months. (c) Statements of Expenditures to be accounted for Condition met. and audited separately. I  22 Table 6: Use of IDA Resources A. Staff Inputs Stage of Project Cycle FY83 FY84 FY85 FY86 FY87 FY88 FY89 FY90 FY91 FY92 FY93 Total Precaration 0.1 0.6 32.3 _ 33.0 Appraisal 15.2 46.5 _ = = = 61.7 Negotiations 12.9 _ 12.9 SuDervision 10 .7 12.4 10.3 10.2 9.4 3.5 56.5 Project Comrletion Repcort _ 3.3 _ 6.1 9.4 Total 0.1 0.6 47.5 59.4 10.7 12.4 13.6 10.2 9.4 3.5 6.1 173.5 B. Supervision Mission Information Supervision Month/ Number of Specialization Performance Mission Year Persons Represented Rating 1 l 12186 1 Economist 2 2 12187 4 Research Specialist Note I Financial Analyst Extension Specialist Tcledetection 3 08/88 3 Financial Analyst 2 Public Administrution Teledetection 4 06/89 2 Financial Analyst 2 Teledetection 5 03/90 5 Financial Analyst 2 Economist Public Administration Computer Specialist Teledetection 6 04/91 3 Financial Analyst 2 Public Administration Teledetection Note 1: Mission concentrated on research and extension components and issued only BTOR, no updated Form 590.  IBRD 20035R 44* A8 I?. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~12 Antsiranana MADAGASCAR P.o-d road, R