Colombia - Second Integrated Rural Development Project

Document of The World Bank FOR OMCIAL USE ONLY Report No. 9821 PROJECT COMPLETION REPORT COLOMBIA INTEGRATED RURAL DEVELOPMENT PROJECT PHASE II (LOAN 2174-CO) AUGUST 19, 1991 Agriculture Operations Division Country Department III Latin America and Caribbean Regional Office This document bas a restricted distribution and may be used bY recipients only in the performance of. their offcial duties. Its contents may not otberwise be disclosed wFithout World Banl; autborization.  CURRENCY EOUIVALENT Currency Unit Colombian Peso /ColS) Appraisal 1981 US$1 = Col$ 61.0 June 1990 US$1 = Col$ 497.0 WEIGHTS AND MEASURES Metric System GOVERNMENT OF COLOMBIA FISCAL YEAR January 1 - December 31  FOR OFFCIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Olke DotfOvttGerwl Opwatton lvatkm August 19, 1991 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Colombia Integrated Rural Development Pro1ect (Loan 2174-CO) Attached, for information, is a copy of a report entitled "Project Completion Report on Colombia Integrated Rural Development Project (Loan 2174- CO)" prepared by the Latin America and the Caribbean Regional Office. No audit of this project has been made by the Operations Evaluation Depa:tment at this time. Attachment 2~~ 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.  GLOSSARY OF ABBREVIATIONS CAJA - Caja de Cr6dito Agraria, Industrial y Minero (Agricultural, Industrial and Mining Credit Bank) CAVECINALES - Fondo Nacional de Caminos Vecinales (National Fund for Rural Roads) DNP - Departamento Nacional de Planeaci6n (National Planning Department) DRI - Programa de Desarrollo Rural Integrado del Fondo DRI (Integrated Rural Development Program, which became DRI Fund) EDO - Programa de Extensi6n Dirigido a Objetivos (Extension Program Directed to Objectives) FEDECAFE - Federaci6n Nacional de Cafeteros de Colombia (National Federation of Colombian Coffee Growers) FERS - Fondos Educativos Regionales (Regional Educatior Funds) FINANCIACOOP - Instituto de Financiamiento y Desarrollo Cooporativo de Colombia - (Colombia Institute for Financing and Development of Cooperatives) FONDO/MEN - Fondo del Ministerio de Educaci6n Nacional (National Fund of the Minsitry of Education) ICA - Instituto Colombiano Agropecuario (Colombian Agricultural Institute) ICCE - Instituto Colombiano de Construcciones Escolares (Colombian Institute for School Building) ICEL - Instituto Colombiano de Energia El6ctrica (Colombian Institute for Electrical Power) INDERENA - Instituto de Recursos Naturales Renovables del Medio Ambiente (Institute for Renewable Natural Resources and the Environment) INS - Instituto Nacional de Salud (National Institute of Health) IVA - Impuesto al Valor Agregado (Value Added Tax) MINISALUD - Ministerio de Salud (Ministry of Public Health) PAN - Plan Nacional de Alimentaci6n y Nutrici6n (National Food and Nutrition Plan) PDIC - Progran,a do Desarrollo Integral Compesino (Integrated Rural Development Program for Peasants) PNR - Plan Nacional de Rehabilitaci6n (National Rehabilitation Plan) SENA - Servicio Nacional de Aprendizaje (National Apprenticeship Service)  FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT COLOMBIA INTEGRATED RURAL DEVELOPMENT PROJECT PHASE II (LOAN 2174-CO) Table of Contents Pare No. Preface . ........ sees ... ...... s .**............. s ........ i Evaluation Suimmary sees ................. .*00 ................ ........ .0.. ii Part I - PROJECT REVIEW FROM BANK'S PERSPECTIVE 1. Project Identity .............s..... .. ... ....... 1 2. Project Background so...... ...... .......... 1 3. Project Objectives and Description ........................ 1 4. Project Design and Organization ........................... 1 5. Project Implementation ... s ............ se........ .. *.- 2 6. Project Results ......... .................................. 4 7. Sustainability ........ ............ O0 ........8 8. Bank and Borrower Performance ........................ 9 9. Project Relationships.... ......... .... .. ... ......... o 11 10. Consulting Services ...... ................ oo.......... 11 11. Project Documentation and Data ................O.............s. 11 Part II - PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ........... 12 Part III - STATISTICAL INFORMATION ........................... 20 Table 1 - Related Bank Loans .oso.......o*. ....*.... 20 Table 2 - Project Timetable **..... *.. *....... ............ 22 Table 3 - Loan Disbursement ....... . . . . .. . . . . . *...... ..... 25 Table 4 - Project Costs .............*................. *...... 26 Table 5 - Direct Benefits *.O..* ........ ****.0*t** * ............. 29 Table 6 - Status of Covenants ....................... ......... 37 Table 7 - Staff Inputs .. o ...... o .... .s. s ............. ....... . ... 39 Map IBRD No. 15802 This document has a restricted distribution and may be used by recipients only in the performance oftheir official duties. its contents may not otherwise be disclosed without World Bank authorization.  - 1 - PROJECT COMPLETION REPORT COLOMBIA INTEGRATED RURAL DEVELOPMENT PROJECT PHASE II (LOAN 2174-CO) PREFACE 1. This is the Project Completion Report (PCR) for the Integrated Rural Development Project Phase II in Colombia, for which Loan 2174-CO in the amount of US$53.0 million was approved on June 8, 1982. The project w-.s a follow-on to a successful Bank-supported first phase project (Loan 1352-CO) which closed in 1984. The loan was closed on June 30, 1990. Only US$40.8 million was disbursed because decentralization reforms that took place in Colombia in 1990 changed the role of the implementing agency to such a degree that it was felt better to appraise a new project, rather than try to modify the project. The loan balance of US$12.2 million was canceled effective December 31, 1990. A follow-on Bank/IDB-supported Integrated Rural Development Program (DRI) (Loan 3250-CO) was approved on July 31, 1990 which incorporates the new role of DRI under the decentralization reforms mentioned above. 2. The PCR was jointly prepared by the Agriculture Sector Operations Division of the Latin America and the Caribbean Regional Office (Preface, Evaluation Summary, Parts I and III) and the Borrower (Part II)V. 3. Preparation of this PCR was started during the Bank's final supervision mission of the project in April 1989, and is based on a completion mission during April-May 1990, the Staff Appraisal Report, the Loan Agreement, supervision reports, correspondence between the Bank and the Borrower, interviews with relevant project staff, and internal Bank *aemoranda. Sample impact evaluations werc also done, and provided valuable information on the project for this PCR, as well as for the design of the recently approved follow-on loan (para. 6.14). 1/ Part II includes unedited but selected sections of the Borrower's Btatement. A complete text is available in LAC Files.  l ~~~~~~~~~~~~- 'Iii - PROJECT COMPLETION REPORT COLOMBIA INTEGRATED RURAL DEVELOPMENT PROJECT PHASE II (LOAN 2174-CO) EVALUATION SUMMARY Objectivee 1. The primary objectives of the project were tot (a) improve rural living conditions through the construction of infrastructure (road., electric power connections, and piped water supply) and through the increase of primary health and education services and facilities; and (b) increase food production of small farmers through increased yields and intensified land use, coupled with programs of supervised credit, technical assietance, training, and marketing (para. 3.01). Implementation Experience 2. The loan closing date was extended on two occasions totalling 18 months, from December 31, 1988 to June 30, 1990. Fiscal constraints between 1981 and 1984, and again between 1986 and 1988, delayed project execution. However, the overall project did succeed in fully meeting many of its targets and it has contributed to the continued development of rural Colombia (para. 5.01). 3. Other problems that contributed to project delays were: (a) institutional changes which took place after 1981 when the Integrated Rural Development Program (DRI), was moved from the National Planning Department (DNP) to the Ministry of Agriculture. This was done because it became clear that the task of administering projects was too large for a policy making authority such as the DNP. DRI became a fully autonomous agency in 1985. However, it did not become fully operational until January 1988, one year before the original loan closing date. These institutional changes, while conceptually sound, did result in some transitional problems of duplication in some administrative activities, coordination difficulties with the executing agencies, and poor monitoring of project activities; (b) processing of disbursement requests by Government and subsequent movement of disbursed funds back through Government to the executing agencies was very slow due to complicated Government procedures; and (c) DRI's difficulties in getting the various executing agencies to submit timely audit reports; in 1986 this led to the Bank's non-acceptance of disbursements against Statement of Zxpenditures (SOEs) for almost a year. In addition, poor progress reporting (due primarily to the lack of an effective monitoring and evaluation system)  - iv - and frequent changes of DRI management, hampered smooth implementation (paras. 5.02, 5.07, 5.08, 5.09, and 6.11). 4. The final project cost was about US$739.2 million, 80% of aippraisal estimate (US$173.2 ..Illion). The lower cost is due to the higher than expected devaluation of the local currency (para. 5.06). Only US$40.8 million (77%) of the loan was disbursed (para. 5.07). Results 5. In spite of the difficulties during implementation, the majority of the physical targets were met or exceeded and the project continued to sustain the achievements of the previous (first phase) projectV. The Bank and Government objective of poverty alleviation was largely achieved through the project, as evidenced by the improved standard of living of project beneficiaries (para. 6.14). About 21,000 (105%) beneficiaries of this project, and about 5,100 (94%) beneficiaries of the first phase project received production credit (para. 6.02). Technical assistance to farmers was less successful because the hiring of extension agents was prevented by a freeze on Govevnmen' hiring during several years in which Colombia faced severe fiscal constraints (para. 6.03). Nevertheless, training for farmers exceeded targets, although the technical staff of the executing agencies did not receive the projected level of training (para. 6.04). By contrast, targets were exceeded for the construction of farm fish ponds, roads, water supply systems, health centers, school classrooms, and implementation of the electrification program (paras. 6.05 - 6.10). Institutionally, DRI has developed into a very capable agency. 6. In 1989, in line with the Government's Decentralization Reform, the project was modified to incorporate a pilot activity involving 38 additional municipelities. The pilot was done to see how DRI's operations could be brought in line with the decentralization reforms of the Government, and yet ensure the sustainability of the rural development effort. Under the pilot the role of DRI was changed such that instead of contracting public implementing agercies, DRI cofinanced investments with municipalities. These, in turn, contracted private and public entities to carry out works. The pilot did contribute to developing effective mechanisms for cofinancing between DRI and the municipalities. This in turn has become a key feature of the recently approved follow-on project, the Rural Development Investment Program (RDIP) (Loan 3250-CO) (para. 6.12). Subsequently in 1989, the project was modified to support the further strengthening of DRI in line with the revised moduf ooerandi, and in preparation for the follow-on phase of the program. This 2/1 The PCR for the first phase project (Loan 1352-CO) concluded that the primary strategy for improving farmers' income through increasing yields, reducing idle land, and introducing a managerial approach to farming and marketing was successful. Implementing agencies which failed to perform satisfactorily during Phase I were not included in the second phase of the project. And the issue of slow disbursement processing in Colombia (about 18 months) for the first phase project has been reduced to 3-4 months for this project.  - v - involved a modified institutional structure of DRI at the regional level, along with additional computers, equipment, and vehicles (para. 6.13). 7. The economic rate of return (ERR) at project completion is estimated to be 39% compared with the appraisal estimate of 23%. The economic analysis was done for the agricultural component only and is based on the illustrative farm model. used for the financial analysis. The higher ERR at project completion results from the increased yielde and output prices for the agricultural products (para. 6.16). Bustainabilitv 8. Prospects for sustainability of benefits appear to be good since: ,a) the Government continues to recognize and be committed to supporting DRI's key role in building self-reliance among poorer rural populationsl (b) the project has significantly strengthened DRI and, indeed, transformed its functions and procedures in line with changing circumstances in Colombia, notably the decentralization reforms; and (c) the DRI program has now been in operation for some fourteen years and, while it has changed in form considerably over the years, it has developed a good track record of sustained support. Thus, it can be expected that project beneficiaries will continue to receive production support and improved services. Indeed, the follow-on DRI project places even greater emphasis on local participation in both the planning and implementation of subprojects, and provides for the contracting of both public and private sector agencies to implement works, depending on who is best placed to do so. Thiw should further improve the effectiveness of Colombia's rural development efforts (paras. 6.13 and 7.01). 9. As noted previously, the DRI program has been relatively successful in improving the standard of living of the rural poor and it has been sustained throughout several Colombian administrations. The program incorporates a number of the more successful features of rural development efforts identified in the Bank's Report on Experience with Rural Development (Report No. 6883), in that it: (a) uses existing institutions/structures to implement projects; (b) coordinates its assistance with other sectoral programs; and (c) backs its agricultural production projects with extension services and technology. Indeed, a key to DRI success has been its ability to change with changing circumstances. As an implementing agency it h%s never been directly involved in doing the physical works under the project. Rather it has contracted other line agencies of Government to do such works and thereby avoided the duplicative, uncoordinated and often unsustainable nature of mabny of the earlier integrated rural development projects (para. 7.02). Loesons of the Proiect 10. The main lessons stemming from the project and the manner in which they have been addressed in the follow-on project are as follows: (a) Counteroart Fundina: The untimely availability of counterpart funds proved to be an intractable problem throughout much of the project implementation. The importance of timely and adequate  - vi - couwlterpart funding in well documented and it is hoped the combined efforts made under the follow-on project will alleviate the problem. In the follow-on project special efforts have boen made to address this problem by establishing a one year advance planning and budgeting program cycle. Coupled with this, Government introduced a new national budget system requiring that specific provisions be made to earmark counterpart funds for all externally financed projects in the budget. More importantly, the new function of DRI requires that, rather than fully fund rural development subprojects, it would now cofinance these with municipalities who would be =equired to earmark a share of the IVA taxes they receive for this purpose. (b) Mozitorino and Evaluations This was not adequately done under the prrject. By getting key actions taken at the outset of the follow-on project:, it is hoped that better monitoring will be possible. Therefore, in the follow-on project a conditlon of effectiveness was that DRI establish and put into operation a monitorlng system to track the processing of projects through the entire project planning, execution, and supervision cycle, Also, as a condition of effectiveness, it was required that consultants be hired to undertake a baseline survey. During the project, provision has also been made for a review of how well administrative procedures and management is functioning, after the second year of the project. This would be followed up with a mid- term evaluation of the project by independent consultants, to be followed by the preparation of an action plan to allow for mid- course corrections. (c) Institutional CAoacitv And Authority& The project encountered difficulties in obtaining disbursements in a timely manner, since Government procedures typically took over 100 days to process disbursement requests. Also, DRI had difficulty coordinati.lg the activities of the 11 cooperating agencies at the national level (and numerous agencies at the regional level) implementing the project, and had no authority to ensure the adequacy of the accounting procedures followed or the accuracy and timeliness of submission of agency audit or progress reports. In the follow-on project, these difficulties have been addressed by having DRI function as a cofinancing agency, whereby ORI cofinances agreed activities with municipalities and a small number of other agencies. Contracts for implementing the various programs are, therefore, no longer the responsibility of DRI, thereby reducing 3 DRI's administrative burden. To improve the administrative processes, a condition of Board presentation was that Government introduce new procedures reducing the time taken to process j disbursement requests from 103 to 40 days. This has been done. Further improvements in the management of DRI's program include strengthening bottom-itp planning by the municipalities and by local communities. Also, the scope of the DRI program has been narrowed to focus primarily on improving agricultural productivity and directly associated infrastructure. Aspects dealing with  - vii - health, education and power would no longer be supported by DRI. Inote-ad, such activities would be financed and managed by the municipalitios in cooperation with the reopective eectoral Ministries.  PROJECT COMPLETION REPORT COLOMBIA INTEGRATED RURAL DEVELOPMENT PROJECT PHASE II (LOAN 2174-CO) Part I - Proiect Review from Bank's Perspective 1. Proiect Identity Project Name: Integrated Rural Development Project Phase II Loan No.: 2174-CO RVP Unit: Latin America and the Caribbean Country: Colombia Sector: Agriculture Subsector: Rural Development 2. Proiect Backaround 2.01 The major sector objectives were to: (a) increase resource productivity in the agricultural sector; (b) improve efficiency and competitiveness of the marketing system; (c) develop agroindustries to stabilize demand and increase employment and rural incomes; and (d) increase foreign exchange earnings through exports of agricultural products. The small farmer was to be assisted by specific integrated rural development programs, accompanied by technical assistance and credit services, to consolidate the agrarian reform process. 3. Proiect ObJectives and DescriRtion 3.01 The primary objectives of tLi project were to: (a) improve rural living conditions through the construction of infrastructure (roads, electric power connections, and piped water supply) and through the increase of primary health and education services and facilities; and (b) increase food production of small farmers through increased yields and iti'tnsified land use, coupled with programs of supervised credit, technical assistance, training, and marketing. 4. Proiect Design and oraanization 4.01 The project was complex, with numerous components and executing agencies. Nevertheless, it was carefully defined in scope and understood by all relevant parties, primarily because of the experience gained in the implementation of the first phase (1976-1984). The project was prepared by the DRI/PAN Directorate in the National Planning Department (DNP). The appraisal mission, comprising fourteen persons, remained one month in the country; two follow-up missions were required to finalize project appraisal. In the Bank, the loan approval process required the participation of many of the Bank's technical units. This high level of staff input was well spent in  designing what has become a successful approach for achieving sustainable rural development in Colombia. 5. Pro_ect Imolementation 5.01 The loan closing date was extended for 18 months in two steps, from December 31, 1988 to June 30, 1990 (Table 2, page 1 of 3). Fiscal constraints between 1981 and 1984, and between 1986 and 1988, delayed project execution. However, the overall project did succeed in fully meeting many of its targets and has contributed to the continued development of rural Colombia. There were delays of about seven months in declaring the loan effective (May 18, 1983) (Table 2, page 1 of 3); and the first Bank disbursement took place 20 months after loan signing. Delays in meeting the conditions of disbursement and the signing of subsidiary agreements between DNP and each of the 11 national executing agencies contributed to slow project start-up. In addition, during project implementation, several more executing agencies were added, totalling 27 agencies at one point in time. 5.02 Other problems that contributed to project delays weret (a) institutional changes taking place after 1981 when the Integrated Rural Development Program, the DRI Fund (DRI), was moved from DNP to the Ministry of Agriculture. This was done because it became clear that the task of administering projects was too large for a policymaking authority such as the DNP. DRI became a fully autonomous agency in 1985. However, it did not become fully operational until January 1988, one year before the original loan closing date. These institutional changes, while conceptually sound, did result in transitional problems of duplication in some administrative activities with the Agricultural, Industrial, and Mining Credit Bank (CAJA) (such as coordination with the Bank and the executing agencies, and monitoring of project activities). There was also poor progress reporting and frequent changes of DRI management (in eight years, there were six general managers); (b) processing of disbursement requests by Government and subsequent movement of disbursed funds back through Government to the executing agencies was very slow due to complicated Government procedures. The process, from expenditure to receipt of funds, on average required three to four months (para. 5.07); and (c) DRI's difficulties in getting the various executing agencies to submit timely audit reports, led to the Bank's non-acceptance of disbursements against Statement of Expenditures (SOEs) for almost a year in 1986 (para. 5.09). 5.03 Proiect Risks. Of the potential risks identified by the appraisal mission (functioning of the marketing system and DRI's administrative and management capacity), only the institutional changes taking place during project implementation were of significant consequence.  - 3 - 5.04 Unforeseen Factors Affecting Proiect Im2lementation. The Government'"s insufficient budget allocations since project inception hampered project implementation. Starting in 1981, the country experienced a deteriorating financial situation because the coffee bonanza of the late 1970e had come to an end. Also, insufficient number and high turnover of extension agents did not facilitat3 implementation of the project (para. 6.03). 5.05 Procurement. Procurement of equipment and vehicles was slow in the beginning due to the Government's cumbersome administrative regulations for procurement. After the bidding procedures were revised and approved by the Bank on September 25, 1984, no significant problems were encountered. 5.06 Prolect Costs. The final project cost was about US$139.2 million, 80% of appraisal estimate (US$173.2 million). The lower cost is due to the higher than expected devaluation of the local currency (Part III, Table 4). 5.07 Disbursements. Several factors contributed to the elow rate Gf disbursement. Only US$40.8 million (77%) of the loan was diebursed. There were delays in the Government processing of disbursement requests. Also budget releases were often only made towards the end of the Colombian fiecal year, when it was too late to contract and pay for works within that period. Government regulations require that funds not committed before the end of the fiscal year have to be returned to the Treasury. As a result, budgetary funds could often not be utilized for the project. However, in spite of slow disbursements, physical implementation did proceed reasonably well (para. 5.01). 5.08 Reportino. Progress reporting was poor because DRI management was unable to organize its own monitoring and evaluation unit for the project. Since 1976, project administration, control and monitoring activities had been under CAJA. This had worked effectively during the first phase project, but DRI management failed to clarify with CAJA how this was to be done under the subsequent project. Therefore, some project administration and monitoring activities were carried out by the two institutions, other activities were omitted. The agreement with CAJA was revised only in December 1984. But the performance of CAJA was not satisfactory, in part because of its passive attitude and the poor collaboration of the executing agencies. Both institutions reported conflicting and unreliable project progress data, in part due to methodological differences and generally poor coordination. 5.09 Auditing. The audited financial statements of the executing agencies were always submitted late to the Bank, and the auditors' recommendations were often not carried out by DRI and the executing agencies. DRE and 13 of the executing agencies failed to include an explicit opinion in the 1985 audit reports on the legitimacy of expenses against which the Government had requested disbursements using the Statement of Expenditures (SOEs) procedure. For this reason, on October 28, 1986, the Bank ceased to accept disbursements by SOE procedures. This situation lasted almost a year for some of the agencies and led to project delays. 5.10 ComDliance with Loan Conditions. The Borrower did not fully comply with the following covenants of the Loan Agreement: 3.01(a), 3.02, 3.04(b),  - 4 - 5.10 gomoliance with Loan Conditions. The Borrower did not fully comply with the following covenants of the Loan Agreements 3.01(a), 3.02, 3.04(b), 3.06(d)(e), 3.20(b)(iii)(c) and 4.02(e)(d). Such covenants required sufficient counterpart funds be available; hiring consultants for project control and monitoring; maintaining files for registering project activities; preparing, and on-time presentation of, progress reports and the preparation by the Ministry of Public Health (MINSALUD) of a program to improve the provision of primary health care services (Part III, Table 6). 6. Proiect Results 6.01 In spite of the difficulties during implementation, the majority of the physical targets were met or exceeded and the project continued to sustain the achievements of the previous (first phase) project./ The Bank and Government ob4ective of poverty alleviation was largely achieved through the project, as evidenced by the improved standard of living of project beneficiaries (para. 6.14). 6.02 Agricultural Credit. The project provided seasonal, medium- and long-term subloans to about 25,000 beneficiary farm families (including about 5,400 of the first phase project) financed entirely by CAJA. Credit to cooperatives and farmer groups was provided by the Colombian Institute for Financing and Development of Cooperatives (FINANCIACOOP), for marketing and processing plants. Project families received credit for home improvement. Except for working capital for marketing and processing for which subloans financed up to 100% of requirements, sub borrowers contributed at least 15% of production costs financed by the subloans. Interest rates varied from 18% to 24%. With regard to the appraisal targets, about 21,000 (105%) beneficiaries of this project, and about 5,100 (94%) of the first phase project received production credit. By contrast marketing subloans reached 36 (59% of target), while subloans for agro-processing reached only 13 (24% of target). A contributing factor to the lack of demand is that the farmer groups were too small to avail themselves of this type of creditY. 6.03 Technical Assistance and Trainino. The Colombian Agricultural Institute's (ICA) extension agents were to prepare farm plans, make projections of small farmer credit needs for four to five years, provide 1/ The PCR for the first phase project (Loan 1352-CO) concluded that the primary strategy for improving farmers' income through increasing yields, reducing idle land, and introducing a managerial approach to farming and marketing was successful. Implementing agencies which failed to perform satisfactorily during Phase I were not included in the second phase of the project. And the issue of slow disbursement I processing in Colombia (about 18 months) for the first phase project has been reduced to 3-4 months for this project. ai No information is available on cost recovery; there is no statistical information per year of the DRI loan portfolio, and the DRI subloan portfolio does not separate information for phase I and phase II beneficiaries.  implementation, because it had to assign its extension agents to a complementary program, the National Rehabilitation Program (PNR). At the time, PNR had somewhat higher priority than the DRI program because it dealt with providing public assistance to areas of the country where security problems were an issue. ICA also had difficulties in hiring now extension agents due to a freeze on Government personnel hiring during several years in which Colombia faced severe fiscal constraints. To resolve this problem, DRI signed an agreement with the National Federation of Coffee Growers (FEDECAFl), whereby the latter provided agricultural extension services in the Departments of Caldas, Huila and Tolima. However, there was a high turnover of extension agents hired by the local coffee federations, because their oontracts lasted only one year. DRI further tried to strengthen the extension service by signing another agreement with CAJA, which provided additional extension agents under its extension program. This did improve the services to DRI beneficiaries. The adaptation trials and demonstration plots were designed and carried out for a few marketable commodities and livestock activities (in each district). However, it took more time than anticipated to get results. Also the technological recommendations to farmers have emphasized productivity without adequately taking into account the marketing problems which arose from time to time. Furthermore, there was poor coordination between the executing agencies in charge of the agricultural production, credit and marketing. Nevertheless, these problems do not appear to have seriously affected the overall project. 6.04 The National Apprenticeship Services (SENA) was to provide technical courses to farmers and technical personnel of the executing agencies involved in the production activities. SENA provided 1,543 program induction courses to about 26,000 farmers (237% of the target); 5,070 technical courses to about 68,000 farmers (166% of the target); and 1,659 marketing courses to about 40,000 farmers (243% of the target). The marketing courses were reported to have been very helpful. However, SENA did not reach the target set for technical courses to staff of the executing agencies, i.e, only 91 courses to about 1,700 participants (54% of appraisal estimate). The shortfall is primarily due to ICA staff hiring problems and lack of trip expenses for participants. 6.05 Natural Resources. The Institute for Renewable Natural Resources and the Environment (INDERENA) was successful in providing technical assistance for the construction of 1,600 farm fish ponds (225% of target). However, the target to plant 1,100 ha of trees was not reached and there is no reliable data as to how many hectares were planted. The same conclusions apply to the protection of microwatersheds. Infrastructure 6.06 Roads. The roads program, executed by the National Fund for Rural Roads (CAVECINALES), completed 723 km of new roads (268% of target) and improved 424 km of existing roads (157% of appraisal estimate). The main problem with the road component is the lack of maintenance. This is being addressed in the follow-on project through a requirement that municipalLties set aside a certain amount of theLr budget for road maintenance. Maintaining  - 6 - a satisfactory road maintenance record is now a requirement for receiving additional DRI funding. 6.07 Water Suonl. This component was executed by the National Institute of Health (INS). INS installed, or completed, 210 systems as compared to the target of 176 (119%). The main problem was that INS had severe financial and administrative problems. It was replaced in the beginning of 1988 by organizations at the departmental level. 6.08 Electrification. The electrification program was executed by the departmental power companies and supervised by the Colombian Institute for Electrical Power (ICEL), which provided the design and supplied the construction materials. The number of house connections was 6,423 (134% of the appraisal target). Farmers contributed 25% of the power installation charge. The problems encountered with this component were: higher than expected costs; initial financial problems, because procedures for cost recovery had not been established; some of the villages included in the project did not meet the criteria established, e.g., in terms of the allowed cost per house connection; and the lack of agreement between ICEL and the Electrificadora of Caldas. Despite these problems, the component was successful. A major improvement resulting from the electrification program was the move from firewood to electric cooking, which has taken the pressure off the land for wood fuel, and contributed to natural resource protection. social Services 6.09 Health. The Ministry of Public Health (MINSALUD) was responsible for constructing and equipping 41 new health centers (228% of revised target); remodelling 135 existing centers (113% of revised target); and remodelling 34 hospitals (189% of revised target). Special training was provided to 2,149 staff, including health promoters (442% of the revised target). Many of the newly constructed health centers had limited use or were not used at all because MINSALUD's operational budget and pay scale did not allow hiring of the necessary staff. In 1987, targets were revised to give priority to renovation rather than building new hospitals and health centers. However, in the Tolima Department, many existing centere had to be rebuilt. The major operational problem with this component has been the poor coordination between MINSALUD and Servicios Seccionales (SSS). 6.10 Education. The National Fund of the Ministry of Education (FONDO/MEN) was responsible, at the national level, for the execution of this component. Several regional and local agencies, such as the Colombian Institute for School Building (ICCE) and Regional Education Funds (FERs), among others, had the local responsibility. The number of school classrooms constructed was 1,296 (185% of the target) and the number of schools that received furniture, purchased with project funds, was 1,081 (108% of the target). As in the health component, the main problem has been poor coordination between FONDO/MEN and the regional and local agencies.  -7- other Comoonentg 6.11 Management. Monitorina and Evaluation. The poor project monitoring has been explained in para. 5.08. The evaluation unit, originally part of DNP's project management, never developed as an analytical tool to continuously evaluate the project's impact. This was exacerbated by the fact that project management was overwhelmed by the task of day-to-day control of the activities of the participating agencies. To get some more detailed evaluation, DRI management, as in the first phase, subcontracted for an impact evaluation to be done through two surveys in 1984 and 1989. These were done in approximately one third of the municipalities under the project and, within the limits of the data, provide a good evaluation which has been helpful in preparing this PCR. 6.12 Pilot Activity. In 1989, in line with the Government's Decentralization Reform, the project was modified to incorporate a pilot activity involving 38 additional municipalities to test new procedures. The pilot was done to see how DRI's operations could be brought in line with the decentralization reforms of the Government and yet ensure the sustainability of the rural development efforts. DRI, instead of contracting public implementing agencies, would in future cofinance municipalities. The municipalities in turn, would contract private and public entities to carry out works. The pilot did contribute to developing effective mechanisms for cofinancing between DRI and the municipalities, a key feature of the recently approved follow-on project, the Rural Development Investment Program (RDIP) (Loan 3250-Co). 6.13 The Strengthening of DRI. In 1989, the project was modified to support the further strengthening of DRI in line with the revieed modus operandi following from the decentralization reforms. This involved a modified institutional structure of DRI at the regional level, along with additional computers, equipment, and vehicles. 6.14 Proiect Impact. While the data are not available for the overall project area, the impact evaluations done in 19 of the 55 municipalities, i.e., in the districts of Tolima (eight municipalities) and Huila (11 municipalities), do provide useful data/. The Tolima District studies (1984 and 1989) concluded that 56% of the population improved their houses, 64% and 55% had access to electricity and potable water, respectively. However, education and health services remained at the same level in spite of the project actions. The participation of the rural families in community organizations increased from 39% in 1984 to 84% in 1989. During the same period (1984-89), average agricultural productivity showed considerable increases in some crops such as maize (71%), sorghum (38%) and sesame (128%). For the agricultural production component, the evaluation impact study estimated an economic rate of return of 37% which confirmed the economic efficiency of the DRI investments. The Huila Sur studies of evaluation impact (1984 and 1989) concluded that 58% of the households surveyed improved their i/ The impact studies are available in LAC Files.  - 8 - houses, 25% purchased now houses, 74% gained access to electricity and 58% gained access to potable water. Education and health services appear to have improved somewhat. Average agricultural productivity (1984-89) showed increases in beans (26%), banana (62%), and cassava (400%). In addition, the area planted to coffee increased by 100%. This together with the coffee boom of the mid 19801, certainly contributed to the project's good results. 6.15 Financial Analysis. Based on the same illustrative farm models that were envisaged at appraisal for representative small farms in different altitudes, actual data was used and compared with appraisal estimates. The analyses indicate a financial rate of return (FRR) ranging from 37% - >50% at project completion, which compares favorably with an FRR of >S0% estimated at appraisal. There are indications that the project is still attractive to small farmers because the relative increase in incomes of 52% - 208% at project completion compares favorably with the relative increase in incomes of 64% - 185% estimated at appraisal (details are presented in Part III, Table 5, page 5 of 8). 6.16 Economic Analysio. The economil' rate of return (ERR) at project completion, computed over 20 years, is estimated to be 39% compared with the appraisal estimate of 23%. The economic analysis was done for the agricultural component only and is based on the illustrative farm models used for the financial analysis. Details for farm model No. 1 (1.5 ha) are given in Part III, Table 5, page 7 as a sample. The higher ERR at project completion results from the increased yields and output prices for the agricultural products (assumptions in Part III, Table 5, page 4). 7. Sustainabilitv 7.01 Prospects for sustainability of benefits appear to be good since: (a) the Government continues to recognize and be committed to supporting DRI's key role in building self-reliance among poorer rural populations; (b) the project has significantly strengthened DRI and, indeed, transformed its functions and procedures in line with changing circumstances in Colombia, notably the decentralization reformsl and (c) the DRI program has now been in operation for some fourteen years and, while it has changed in form considerably over the years, it has developed a good track record of sustained support. Thus, it can be expected that project beneficiaries wlll continue to receive production support and improved services. Indeed, the follow-on DRI project places even greater emphasis on local participation in both the planning and implementation of subprojects, and provides for the contracting of both public and private sector agencies to implement works, depending on who is best placed to do so. This should further improve the effectiveness of Colombia's rural development efforts. 7.02 As noted previously, the DRI program has been relatively successful in improving the standard of living of the rural poor and it has been sustained throughout several Colombian administrations. The program incorporates a number of the more successful features of rural development efforts identified in the Bank's Report on Experience with Rural Development (Report No. 6883), in that its (a) uses existing institutions/structures to implement projects; (b) coordinates its assistance with other sectoral  - 9 - programsl and (c) backs its agricultural production projects with extension i services and technology. Indeed, a key to DRI success has been its ability to change with changing circumstances. As an implementing agency it has never been directly involved in doing the physical works under the project. Rather it has contracted other line agencies of Government to do such works and thereby avoided the duplicative, uncoordinated and often unsustainable nature of many of the earlier integrated rural development projects. 8. Bank and Borrower Performance 8.01 Bank Performance. The Bank's performance was satisfactory throughout project implementation. Relations between the Bank and DRI were cordial at all times and the Bank acted supportively and flexibly to resolve project problems or to respond to changing circumstance. in Colombia. The following Bank actions contributed to successful project implementation: (a) introduction of uniform bidding procedures for Bank loans in Colombia, on September 25, 1984; (b) inclusion of the project in the Bank's Special Action Program, on February 11, 1985. This allowed Bank financing for the project to increase from 50% to 80%, and helped ensure DRI's liquidity during project implementation; (c) opening of a Special Account, on Nlovember 5, 1985, to increase the pace of disbursements and, in April 1989, increasing the amount of this account from US$4.0 million to US$6.0 million; (d) agreement to modify the description of a small farmer in the Loan Agreement in order to include farms up to S0 ha; (a) flexibility by the Bank for extending deadlines for DRI submission to the Bank of the field surveys for project impact evaluation; (f) agreement to include two new project components (the pilot scheme and the strengthening of DRI) which has contributed to the smooth follow-on of the subsequent project; and (g) flexibility by the Bank in extending the project closing date on two occasions. 8.02 Borrower Performance. In spite of the implementation problems stated in paras. 5.01 and 5.02, DRI's effective performance in project coordination may be due to the experience gained in the first phase. However, DRI performed less than satisfactory in monitoring project activities. DRI is, in fact, a young institution, created in 1976. Since 1982, when the Bank loan was approved, DRI's annual budget has increased 300%, but DRI has been able to maintain staff growth to just 56%. The successful execution of this, and the previous project, led to Bank approval of a subsequent operation (para. 6.12). The main weaknesses of DRI have been: (a) failure to clarify the project administration and monitoring responsibilities with CAJA, causing duplication of some of these activities and a weak management information system in DRI; and (b) lack of a personnel policy providing staff members with adequate skills training for their work. 8.03 Lessons of the Proiect. The main lessons stemming from the project and the manner in which they have been addressed in the follcow-on project are as follows: (a) Counteroart Fundina: The untimely availability of counterpart funds proved to be an intractable problem throughout much of the project. The importance of timely and adequate counterpart funding is well documented and it is hoped the combined efforts made under  - 10 - the follow-on project will alleviate the problem. In the follow-on project, special efforts have been made to address this problem by establishing a one year advance planning and budgeting program cycle. Coupled with this, Government introduced a new national budget system requiring that specific provisions be made to earmark counterpart funds for all externally financed projects in the budget. More importantly, the new function of DRI requires that, rather than fully fund rural development subprojects, it would now cofinance these with municipalities who would be required to earmark a share of the IVA taxes they receive for this purpose. (b) MonLtorino And Evaluation: This was not adequately done under the project. By getting key actions taken at the outset of the follow- on project, it is hoped that better monitoring will be possible. Therefore, in the follow-on project a condition of effectiveness was that DRI establish and put into operation a monitoring system to track the processing of projects through the entire project planning, execution, and supervision cycle. Also, as a condition of effectiveness, it was required that consultants be hirod to undertake a baseline survey. During the project, provision has also been made for a review of how well administrative procedures and management is functioning, after the second year of the project. This would be followed-up with a mid-term evaluation of the project by independent consultants, to be followed by the preparation of an action plan to allow for mid-course corrections. (c) Institutional Capacitv and Authority: The project encountered difficulties in obtaining disbursements in a timely manner, since Government procedures typically took over 100 days to process disbursement requests. Also, DRI had difficulty coordinating the activities of the 11 cooperating agencies at the national level (and numerous agencies at the regional level) implementing the project, and had no authority to ensure the adequacy of the accounting procedures followed or the accuracy and timeliness of submission of agency audit or progress reports. In the follow-on project, theme difficulties have been addressed by having DRI function as a cofinancing agency, whereby DRI cofinances agreed activities with municipalities and a small number of other agencies. Contracts for implementing the various programs are, therefore, no longer the responsibility of DRI, thereby reducing DRI's administrative burden. To improve the administrative processes, a condition of Board presentation was that Government introduce new procedures reducing the time taken to process disbursement requests from 103 to 40 days. This has been done. Further improvements in the management of DRIs program include strengthening bottom-up planning by the municipalities and local communities. Also, the scope of the DRI program has been narrowed to focus primarily on improving agricultural productivity and directly associated infrastructure. Aspects dealing with health, education and power would no longer be supported by DRI. Instead, such activities would be financed and managed by the municipalities in cooperation with the respective sectoral Ministries.  - 11 - 9. Proiact Relationships 9.01 Imoact of Relationshis on Proiect Imolementation. The success of this project has been assisted by the good working relationship between the Bank and the Borrower, and between the various institutions involved. In addition, DRI has been able to maintain effective coordination with other sectoral programs in the country, a key feature which has not existed in rural development programs elsewhere. 10. Consultino Services 10.01 The project hired local consultants to: (a) prepare designs of the infrastructure component; (b) carry out field studies to evaluate the project impact; and (c) prepare specific studies to help DRI and the municipalities implement the pilot scheme. However, in the early year. of project implementation, DRI failed to contract consultants for specific technical assistance for some of the executing agencies. In general, the local consultants' inputs during project execution were important, and their performance was satisfactory. 11. proiect Documentation and Data 11.01 The Staff Appraisal Report was useful during project execution. Overall, DRI monitoring and evaluating capacity was weak. The PCR has been prepared based on the information existing in DRI offices and in the field surveys of evaluation impact conducted in 1984 and 1989.  - 12 - Part II. Project Review from Borrower's Perspective The Borrower has completed a Full Project Completion Report (PCR). Attached is the unedited text of Burrower's PCR from which we have deleted the descriptive portions of the report. The Full PCR is available in LAC Files.  - 13 - Selected sections from the Pgoiect Completion Report for the Second Integrated Rural Dezelopment Proiect iLoan 2174-COt 2.06 Prior to 1986, the Second Integrated Rural Development Project was implemented during a time of economic crisis and stagnation in agricultural production due to the dislocations caused by the coffee boom at the end of the 1970s and the fiscal deficit. In late 1982 the first corrective measures were introduced in the form of import controls, faster devaluation, limitations on government spending and higher custom duties and taxes. As these measures were not sufficient to put an end to the economic recession, the Government established the Macroeconomic Adjustment Plan in 1984. The economy as a whole and the agriculture sector in particular began to recover significantly in 1986, stimulated by the coffee boom in that year and the ensuing rise in external aggregate demand. In addition, the Social Economic Plan (1987-90) sought to restore the dynamism of agricultural production through price mechanisms and customs protection. That policy of sectoral protection was strengthened in 1988 with the Selective Supply Plan, which sought to concentrate all of the sector's technical and financial resources on eight products (maize, rice, cassava, beans, soybeans, cinnamon, sesame and sunflower). 2.07 The Government's political base for the rural sector now consists of the National Rehabilitation Plan (Plan Nacional de Rehabilltacl6n -- PNR), Integrated Rural Development Program for Peasants (Programa de Desarrollo Integral Campesino -- PDIC) and the new Agrarian Reform Law, which are the regional and local instruments that seek to improve the living conditions of the population as a whole. 4.03 The Government also decided that domestic funds would finance the continuation of the program in the areas covered by the first project, with a view to the gradual establishment of an ongoing program in those areas, independent of external financing. (When the loan was negotiated with the Bank, the Government agreed to provide all funds for credit and farm and marketing advisory services necessary to ensure the direct beneficiaries of the first project that the development plans on their farms would be completed.) 5.2(e) Because of the importance accorded to integrated rural development and its high priority within the agriculture sector, the Government established the DRI Fund. However, DRI could operate as such only as of January 1, 1988 because the regulations governing its operation had not been finalized. Decree 77 of January 1987 established the bases of the administrative decentralization reforms in which DRI would support the local governments in the preparation and cofinancing of the rural development plans,  - 14 - based on grass-roots participation and diversification of ecor.omic alternatives for small farmers. Decree 77 also recognized DRI as a public entity of nationwide scope with administrative and financial autonomy, responsible for setting policy guidelines for rural development, especially as regards peasant, minifundlo and settlement areas. Decree 2428 of October 24, 1989 org&nized DRI, giving its 21 regional offices greater authority and responsibility. 6.02 In general, project objectives to improve the living standards in the rural areas specifically selected as well as to increase the availability of basic foodstuffs for urban dwellers, were satisfactorily achieved, as evidenced by the DRI impact evaluation studies made in 1989 in two of the four departments where the project was carried out, and where a baseline diagnostic study had previously been done in 1984. Project Stability 7.01 It is not anticipated that there will be problems with project self-sustainability because the first project is still stable six years following the closing of the Bank loan and because both projects continue to receive credit and technical assistance services from the Government (in the case of the second project, production credit has always been granted with funds from FFAP and CAJA AGRARIA). 7.02 In the wake of the satisfactory execution of the first and second Integrated Rural Development Projects, the Bank has approved the financing of a third project (May 1990), entitled the Integrated Rural Development Program for Peasants (PDIC), which is nationwide in scope. The Program, reflecting the policy changes of the Government in support of administrative decentralization, will essentially provide institutional building for DRI and support for local municipal management. Nevertheless, implementation will differ from the two preceding projects, but will help ensure their sustained stability principally because the project has the same primary objective of raising rural income, and because the activities will be carried out preferably in regions with a high concentration of small farms where there are already DRI beneficiaries. In addition, activities are concentrated in the production component, facilitating access to the factors of production and modern techniques that tend to preserve the productive potential of the natural resources, which should help ensure that the increases in productivity on the small farms can be maintained over the longer run. Bank Performance 8.01 Bank performance was satisfactory throughout project implementation, and Bank relations with DRI and other executing agencies were cordial. The complexity of the project required an extraordinary effort, first during preparation and subsequently during appraisal and supervision. The technical and financial problems that were resolved during execution show that when the project was prepared allowance was made for possible and unforeseen problems. The Bank likewise achieved good results when it acted with flexibility and, in conjunction with DRI, took corrective measures to step up loan disbursements.  - 15 - 8.03 Nevertheless, the Bank was very inflexible in requiring compliance wlth Article 4.02 Ic and d) of the Loan Agreement, with regard to the submission of audit reports, especially in connection with statements of expenditures for each project executing agency. Submission to the Bank of the audit reports on the statements of expenditure for 1985 was substantially delayed precisely because in previous years the Bank did not enforce that requLrement. In addition, the Government and DRI did not accord those covenants their special importance when the loan was being negotiated. When, in 1986, DRI was requLred to comply fully with the covenant, it was almost a year before it could have the corrective measurea taken within the executing agencils, which posed obstacles to obtaining Bank funds. 8.04 The main lessons learned, which could be important for other projects, are as followst (a) The Bank placed great importance on the monitoring and evaluation of project activities. Throughout project implementation, the General Directorate first and DRI subsequently could not organize their own monitoring and evaluation eystems, primarily because of the successive changes within the organization and their institutional etructure. Moreover, the uncertainty as to whether CAJA AGRARIA was to resume monitoring and evaluation of project activities, as under the first project, also delayed the decision that DRI itself would set up its own monitoring and evaluation unit. Nevertheless, beginning in 1988, DRI has sought to organize its own monitoring and evaluation system without succeeding (despite Bank support and the fact that a procedure implemented in 1984 was tried in the field, because DRIVs objectives were changed in view of the new concept of cofinancing between it and the municipalities). The lesson for the Bank is that the monitoring and evaluation system must be located within the same project unit that reports directly to management, and that it must be established and operating before project implementation begins. (b) Durlng execution of the first project, the Bank required DRI and the executing agencies to submit their audited financial statements and accounts for each fiscal year, no later than four months after they ended. The Bank learned that four months was not realistic because each year the provisions of the Loan Agreement were not carried out. In this second project, the Bank required the audit reports to be submitted no later than six months after the end of each fiscal year. This did not resolve the problem of timely submission of audit reports either. In projects such as this, in which many national, regional and local executing agencies are involved, it is difficult to have them all submit their financial statements to the Office of the Controller on time. The Bank and the Controller should rate each entity before it becomes an executing agency to determine whether it qualifies on the basis of its administrative and accounting  - 16 - structures. In addition, if for justified grounds the favorable ruling of the Controller cannot be acquired within the set time- frame, the Bank should approve an extension. Such extensions would be done only once or twice at a maximum during project implementation. The Bank should also determine whether noncompliance with the deadline for submission for audit reports is due to delays emanating from the Office of the Controller itself, in which case the executing agency should not carry the entire burden of the Bank's charges. Another lesson that the Bank has learned on this aspect of audit reports is that each year the auditors make the same recommendations to certain executing agencies that never correct th6 deficiencies encountered. These should be resolved within a reasonable period so that the recommendations of the auditors are always carried out withln the next fiscal year. The Bank should suspend disbursements or exclude from the project, depending on the seriousness of the nonperformance, agencies that do not implement the recommendations of the auditors. Moreover, on the basis of its experience, the Bank has recommended that the entities with a negative ruling or abstention from the auditors should not be executing agencies under the new project. In this respect, the Dank should give DRI the responsibility through its central office and regional departments to ensure compliance with this recommendation, as regards both the deadlines for audit reports and the recommendations of the auditors. (e) Furthermore, Bank experience in Colombia showed a lack of sufficient and timely counterpart budget funds. In order to resolve this problem in part, the Bank included this project in its special Action Program that increased the disbursement percantage of the main categories of the loan, which facliltated and accelerated implementation. (g) One of the most importan'.t lessons learned by the Bank is to have finaiced a pilot plan (in 1989) with unutilized funds from the project, involving 38 municipalities in an experiment that would be usef4l to the implementation of the Integrated Rural Develor ant Program for Peasants (PDIC), partially financed by the Bank. 9.01 DRI performance was uneven. In general, general project coordination was satisfactory, owing primarily to the experience DRI had acquired in the implementation of the first project. However, performance was unsatisfactory as regards the monitoring and evaluation of project activities. 9.04 A weakness of the DRI-PAN program and DRI itself was that the situation with CAJA had not been defined in time as regards the co- administration of the project, which impeded overall administration and coordination of the project from 1982 until 1987, causing duplication in both management with the executing agencies and with the Bank, and duplication in the handling of information because the original 1976 DNP/CAJA contract was  - 17 - still in force and, as noted above, the new December 1984 contract, covering CAJA involvement in certain activities, was not working as expected. 9.05 As also noted above, another DRI weakness during project * implementation was that it did not have adequate monitoring and evaluation or management infLrmation systems, the result of which was poor preparation of progress reports. Other institutional weaknesses -- also important -- were that the recommendations in the audit reports prepared by the Controller of the Republic were not considered important and much of the personnel was inexperienced in project formulation and evaluation techniques, especially in the agriculture sector. 9.10 With regard to the production credit component, despite its financial problems during certain periods, such as in 1988, CAJA performed satisfactorily and, for all intents and purposes, achieved its targets, although this cannot be shown with actual figures because the available data are not reliable. In addition to FFAP funds, CAJA used its own resources and when it became evident that the number of credit agents would have to be increased, CAJA hired them. Lastly, when technical assistance to farmers suffered because of ICA limitations in recruiting extension agents and because the Coffee-Grower Committees could not retain their agents, CAJA also worked with DRI through its Extension Directed to Objectives Program (Programa de Extensidn Dirigida a Objetivos -- EDO). 9.11 In the component covering credit for marketing and collection and processing plants, the performance of FINANCIACOOP was satisfactory as regards the granting of marketing credit and technical assistance. It was much more difficult to grant credit and technical assistance to processing and collection plants. 9.12 With regard to the component covering technical assistance for production, the performance of ICA was poor as regards farm development planning for the granting of credit and technical assistance direct to individuals and to groups of farmers. This function was carried out by other entities such as FEDECAFE and CAJA. In general, these other entities performed satisfactorily despite the problems of personnel and logistics that delayed the attaintment of the targets. 9.14 The main lessons learned by DRI that can be important to the future of the institution are as follows: (a) The subsidiary agreements with other entitiee should be revised or terminated in due course, such as occurred with CAJA, the Ministry of Health, FONDO/MEN, ICEL and INDERENA. The situation with CAJA, regarding project coordination and administration, resulted in duplication of functions, and in the case of other entities such as the Ministry of Health, ICCE and ICEL, DRI learned somewhat late that it should have contracted direct with the departmental and not national offices. In the case of INDERENA, the auditors indicated the problems faced by that entity under the first project, which worsened when the project began.  - 18 - (b) DRI should set up its project unit (external credlt unit) at the highest administrative level, with appropriate technical personnel at headquarters and in the regional offices. The personnel should be suitably trained so that they can collect, process and publish data frr Management, DRI itself and the Bank. Because such a unit was not set up from the outset, all project information was dispersed throughout the various departments, making it difficult to prepare periodic reports for the Bank and consequently this final report. During the years in which the External Credit Division existed within DRI's basic structure, the problem was temporarily resolved, but it reappeared when DRI was reorganized and the division eliminated. communications between DRI and the Bank began to suffer at that time. Even when most communications are signed by the General Manager, they originate in various offices. Because of uneven coordination, it is difficult to maintain an institutional memory with uniform criteria and centralized information. (C) In the light of DRI's new obligations and activities under the PDIC, personnel policies must ensure incentives and career stability, along with ongoing training, both in-service and at training centers, for all personnel. (d) DRI should give greater importance to the audit reports and implement the recommendations of the auditors. Executing entities that do not satisfy the contractual stipulations or the recommendations of the auditors should be terminated in due course. (e) A review of the Special Account since its establishment (1985) shows that there were long periods during which it had no resources available, not because requests for repayment (reposloi6n) were pending, but because of advances to the executing agencies. The improper use by DRI of the funds in the Special Account, authorizing advances that took up to a year to be justified, impeded the proper recycling of funds and consequently hampered implementation. In order to avoid a repetition of this and because counterpart funds are not available at the beginning of the year, the Bank should authorize a high disbursement percentage during the first six months (for example, 80%), compensating par! passu that agreed in the loan document. (f) Even though the Bank organized seminars on the preparation and transmittal of disbursement requests for both DRI and the executing agencies, the heads of the department in question and not those in charge of that activity are occasionally trained. Moreover, because of the extremely rapid turnover of personnel, the training should be given more frequently. In the case of the DRI Central Office, only one person, who has already left, was sent to Bank headquarters.  - 19 - (g) When it appears that some of the loan funds will not be utilized within the authorized period, they should be cancelled in order to avoid unnecessary commitment fees. 10.03 The relations between DRI and leading government agencies -- DNP, the Ministry of Finance and the Ministry of Agriculture -- were cordial and helped the project produce a positive outcome. DRI always had political support to carry out the Integrated Rural Development Program. After the implementation of this second project ends, a task that the Government has accorded the highest priority should be carried out, namely, the Integrated Rural Development Program for Peasants (PDIC), for which DRI is having substantial changes made to its legal and organlzational structure.  - 20 - Part I Table 1 Pago 1 of 2 OJECT COHPLETION REPORT COLONBIA INTEGRATED RURAL DEVELOPNENT PROJECT PEASE It (LgAN 2174-CO) Related Bank Loans gJ Year of Loan Title Purpose Approval Status Comments Second Atlantico Irrigation 1972 Closed At closing, the project Donelopment internal rate of return was (Loan 849-CO) minus five. The conclusion was that the project could have been implemented as a flood control project instead of a sprinkler irrigation project. caqueta Rural Rural Dev. 1975 Closed Works execution was adequate Settlement II but very slow due to the Project permanent INCORA liquidity (Loan 1118-CO) problems. More counterpart funds were needed due to dulays in execution and high inflation rates. Cordoba 2 Rural Dev. 1975 Closed Land use did not change and Agricultural extensive livestock production Development Project continued in spite of the fact (Loan 1163-CO) that improvements were introduced by the project. INCORA had a precarious financial situation. Lack of counterpart funds and experienced technical staff. a/ Agricultural loans before the Integrated Rural Development Project Phase II.  -21 - PA=-= Table 1 Page 2 of 2 Year of Loan Title Purpose Approval Status Comments Integrated Rural Rural Dev. 1976 Closed Project executlon was Development Project extremely complex wlth 12 (Loan 1352-CO) national executing agencLes but results were posLtLve since farmers Lncreased thelr incomes. Bank disbursement process and Bank funds dellvery to the executLng agencLes were exceedingly slow due to complex administrative procedures. Second Agricultural Agricultural 1976 Closed The FFAP credlt system was Credit Project Credlt inadequate for small farmer.' (Loan 1357-CO) needs. Lack of counterpart funds especlally reserved for small farmers. Third Agricultural Agricultural 1979 Closed The Government budget pollcy Credlt Project Credit limLted the brldge flnancing (Loan 1737-CO) responsLbliLty of the executing agency. AdmLnistratLve and technlcal problems due to INCORA weaknes. IrrLgation Irrigation 1981 Closed Inadequate O&M cost Rehabliltatlon I recuperatlon in the irrlgation Project districts. (Loan 1996-CO) The rehablIltated districts did not achieve thelr admlnistrative lndependence nor did they organLio their own irrigation associatLons. Accountlng procedures at HIMAT's headquarters did not Lmprove. Upper Magdalena Watershed 1981 Closed INDERENA, the executlng Pilot Watershed ProtectLon agency, had serlous (Loan 2069-CO) admLnLstratLve problems, weak management and lack of counterpart funds.  - 22 - zSL I.. Table 2 Page 1 of 3 PROJECT COMPLETION REPORT COLOMBIA INTEGRATED RURAL DEVELOPMENT PROJECT PHASE II (LOAN 2174-CO) Proiect Timetable Date Date Date Planned Revised Actual Identification Bank 2/80 2/80 Preparation Bank A/ Mar-Apr/80 Mar-Apr/80 Follow-up Preparation Bank 9/80 9/80 Appraisal Feb/Mar/81 Feb-Mar/81 Post-Appraisal I 6/81 6/81 Post-Appraisal 1I 9/81 9/81 Loan Negotiations 3/4-9/82 3/4-9/82 Follow-up Negotiations 4/19-23/82 4/19-23/82 Board Approval 6/8/82 6/8/82 Loan Signature 7/26/82 7/26/82 Loan Effectiveness 10/26/82 12/16/82 3/16/83 5/20/83 5/18/83 Loan Closing 12/31/88 12/31/89 6/30/90 6/30/90 Project Completion Report 3/31/91 3/31/91 a/ The Executive Project summary was issued on June 30, 1980. The issues raised at each stage of the project cycle were: Identification - control and monitoring of project activities had not been clarified between DNP and CAJA; - CECORA, the marketing institution which did not perform satisfactorily under the previous project, had to be replaced; - overlapping with sector projects (infrastructure components) under preparation; and - - the inclusion in the project of the health and education components would depend on Governments formulation of adequate sectoral policies.  -23 - Part III Table 2 Page 2 of 3 Pr-2aration - The institutional problem between DNP and CAJA remained au one of the project's problems; - the continuation of activities in the areas of Phase I; - the amount of Bank loan; - overlapping with other Bank sector projects; - interest rates to farmers, subloans since CAJA had two different rates of small farmers; and - under Phase I the main problems were related to delays in processing Bank disbursements and procurement of equipment, vehicles and services. Aporainal - the continuation of program activities in the areas of Phase 1; - the financing of agricultural credit operations; - the Bank percentage financing of the project and the Bank retroactive financing; - the social components of the project; - changes in responsibility for administration and control of the project; - changes in responsibilities for the marketing component; - Bank disbursements; and - procurement of goods and services. Neaotiations - the definition of small farmer was adjusted; - project total cost and Bank loan were also adjusted; - the Government would finance the credit and technical assistance activities to Phase I beneficiaries; and - CAJA would not increase the interest rates of farmers' subloans and would use FFAP funds for credit operations (the Bank loan would finance only marketing and agroindustry subloans; the agricultural production subloans would be financed by FlAP funds).  - 24 - PrII Table 2 Page 3 of 3 Ln Cloina - Project supervision and monitoring of project activities were mome of the failures of DRI. As a result, the preparation of project progress reports was poor; - tho presentation of auditing reports to the Bank took more time than anticipated and DRI did not give importance to compliance with auditors' recommendations; - DRI technical staff suffered from mobility during project implementation. The new staff lack experience and need training in project preparation, appraisal and supervision; and - DRI staff do not have adequate incentives to remain in the institution.  -25 - Part II Table 3 PROJECT COMPLETION REPORT COLOMBIA INTEGRATED RURAL DEVELOPMENT PROJECT PHASE II (LOAN 2174-COl Cumulative Estimated and Actual Disburegments (US$ millions) 1983 1984 1985 1986 1987 1988 1989 1990 Appraisal estimate 5.0 12.5 22.0 32.5 43.0 50.5 52.2 53.0 Actual 1.3 6.4 9.8 11.6 17.3 22.8 29.7 40.8 Actupl as % of estimate 26 51 44 35 40 45 56 77 Date of last disbursement: January 22, 1991 Amount cancelled: US$12.2 million  - 26 - ENCIII Table 4 PROJECT COMPLET10N REPOU P 1 of 3 COLOMI INTEGRATED RURAL DEVEMNT PROJQCT PRASE 11 (LOAN 2174-CO) A. Proiet Cost (US$0'O) Appraisal Actual as t of Cat-gory EstLmate A1 Actual Appraisal Estimato 1. Marketlng 8ubloans 7,970 5,220 65 2. Civil Works - Rural Roads 20,880 21,370 102 - Water Supply 16,550 4,390 27 - Electrical Energy 5,540 2,760 S0 - Health Services 2,930 1,770 60 - Education Services 7,390 4,350 59 3. Acquisition of Equipment and Vehicles - extension, Natural Resources 1,500 1,190 79 - Training 420 290 69 - water Supply 1,490 2,570 172 - Health 1,300 1,030 79 - Project Administration 1,000 230 23 - Education 2,450 800 33 4. Equipment for Electrical Energy Distribution 1,930 2,030 105 5. operational Costs - Tochnical Assistance 17,870 12,910 72 - Training 8,120 3,540 44 - Control and Monitorlng (Health) 630 290 46 6. Training for Education Svu. 4,820 60 - 7. Consultant Sve, Administration 3,690 1,660 45 8. Uncoimmitted 5,917 - - Subtotal 102L397 A66460 Al Other Local Costs Production Credit 70,040 72,000 103 Financial Cost of Bank Loan 783 783 100 Subtotal 70,823 72,783 Total 1r3a220 139 243 pl A/ InClUdes reallocations of the Bank loan during project implementation.  -27 - Part III Table 4 Page 2 of 3 B. PROJECT FINANCING (USS million) Benefi- Govt. Local Category ciaries (DRI) FFAP a/ Agencies Bank Total I. APPRAISAL ESTIMATES 1. Marketing Subloans 0.82 2.67 1.33 3.15 7.97 2. Civil Works - rural roads 10.98 9.90 20.88 - water supply 1.27 3.43 1.85 6.55 - elect. energy 1.51 2.53 0.60 0.90 5.54 - health services 0.11 2.19 0.63 2.93 - education services 0.69 3.30 3.40 7.39 3. Acquisition of Equipment and Vehicles - extension nat. resour. 1.50 1.50 - training 0.42 0.42 - water supply 1.49 1.49 - health - project administ. 0.10 0.10 - education 2.45 2.45 4. Equip. for Electrical Energy Distribution 1.93 1.93 5. Operational Costs - Technical Assit. 8.53 9.34 17.87 - Training 4.63 3.49 8.12 - Control & Monitoring * (health) 0.63 0.63 6. Training for Education 3.40 1.42 4.82 7. Consultant Sve. Admin. 2.19 0.50 2.69 8. Financial Cost of Bank Loan 0.78 0.78 9. Unallocated _ _ 7.82 7.82 Subtotal 4.40 39.22 6.56 53.00 103.18 other Local Costs Production Credit _ 45.40 24.64 70.04 Total 4.40 39.22 45.40 31.20 53.00 173.22 a/ The Agricultural Financial Fund.  - 28 - Part LI Table 4 Page 3 of 3 Benefi- Govt. Local Cat-gory ciaries (DLI) FFAP A/ AgencLem Bank Total II. ACTUAL 1. Marketing Subloans 0.73 1.67 2.82 5.22 2. Civll Works - rural roads 9.63 11.74 21.37 - water supply 0.84 1.32 2.23 4.39 - elect, energy 0.55 0.82 1.39 2.76 - health services 0.07 0.75 0.95 1.77 - education servLces 0.36 1.96 2.03 4.35 3. AcquiLitLon of Equip. and Vehicles - extension, nat. resour. 1.19 1.19 - training 0.29 0.29 - water supply 2.57 2.57 - health 1.03 1.03 - project administering 0.23 0.23 - education 0.80 0.80 4. Equip. for Electrical Energy Distribution 2.03 2.03 5. Operatlonal Costs - technical arst. 6.04 6.87 12.91 - training 1.51 2.03 3.54 - control & monitoring (health) 0.14 0.15 0.29 6. Trainlng for Zducation 0.03 0.03 0.06 7. Consultant 8ve. Adminiet. 1.66 1.66 8. Flnancial Cost of Bank Loan - 0- .28 0 Subtotal 2.55 22.20 1.67 40.82 67.24 Other Local Financing Production Credlt 46.80 25.20 72.Q0 Total 2.55 22.20 46.80 26.87 40.82 139.24 8/ The Agricultural Financial Fund.  - 29 - Part tI Table 5 * PROJECT COM LETION REPORT 8 COLOBIA INTEGRATD RUMR^ DEVELOPMENT PROJECT PU&SE II (LOAN 2174-CO) A. Direct Benefit. Targets Expected Revised Actual as % at Project Targets of Revised Key Indicators Completion (1985) Actual Targets A. Credit -Production Total Beneficiaries (No.) First Project 7,000 5,400 5,100 94 Second Projwct 20,000 20,000 21,000 105 Total Lending (Col$ m) 6,420 6,420 n/a - Average/beneficiary (Col$) -Marketina Subloans Granted (No.) 61 - 36 59 - Processing Plants Subloans Granted (No.) 55 28 13 46 B. Technical Assistance - Production Adaptation Trials (No.) 465 900 953 106 Demonstration Plots (No.) 2,500 1,700 1,005 59 - Marketina Associative Groups Assisted (No.) First Project 27 27 27 100 Second Project 24 24 24 100 Producers who sell through Assisted Cooperatives (No.) 4,200 4,500 8,750 194 Centros de Acopio (No.) 16 27 n/a -  - 30 - Part ISI Table 5 Page 2 of 8 Targets Expected Revised Actual ao % at Project Targets of Revised Key Indicators Completion (1985) Actual Targets C. Natural Resources - Reforestation Plants (million) 5.6 2.1 n/a - Area (ha) 2,800 1,100 n/a - Microwatersheds (No.) 500 - n/a - - ipheries Fishponds (No.) 300 710 1,600 225 D. Trainina - Courses for Farri.ers (No.) Induction 650 650 1,543 237 Technical 3,050 3,050 5,070 166 - Courses for Staff (No.) Technical 168 - 91 54 - Cooperatives Trained in Marketina Number 22 - 159 723 Courses (No.) 680 1,659 244 E. Rural Infrastructure - Roads New Roads (km) 270 - 723 268 Rehabilitation Roads (km) 270 - 424 157 - Water Su0DD1 Systems Completed (No.) 126 176 210 119 - Electrification Houses Connected (No.) 4,800 - 6,423 134 - Health New Health Center Completed (No.) 15 18 41 228 Remodeling of Health Centers (No.) 68 120 135 113 Remodeling of Hospitals (No.) 10 18 34 189 Training of Health Promoters (No.) 323 486 2,149 442  -31 - Part III Table 5 Page 3 of 8 Targets Expected Revised Actual as % at Project Targets of Revised Key Indicators Completion (1985) Actual Targets - Education School Rooms Completed (No.) 706 700 1,296 185 Sanitary Units Completed (No.) 653 550 750 136 School Furniture Granted (No.) - 1,000 1,081 108  - 32 - Part III Table 5 Page 4 of 8 B. Economic Imgact Appraisal Actual Estimate (at Final Development) Economic Rate of Return (%) 23 39 Underlyina Assumptions 1. The foreign exchange premium was assumed to be 120% of the foreign exchange rate on June 30, 1990. This is the official shadow price for the foreign exchange rate. Other prices were taken on June 30, 1990 constant prices. 2. The maize price was derived from its import parity value because it more accurately reflects its real price. The banana price was derived from its export parity value because It more accurately reflects its real price. 3. The maize/bean cropped jointly was valued as bean price. This price as well as rice and sesame prices were taken as those fixed by IDEMA a/ for the first semester of 1990. Other commodity prices such as potatoes, cassava, and vegetables were taken as its market price. 4. The economic rate of return (ERR) was calculated for a 20-year project period based on illustrative farm models used for the financial analysis. 5. The sensitivity analysis shows that if costs increase by 42%, the ERR is equal to the economic opportunity cost of capital (i.e. 12%). Also, if investments increase by 400%, the ERR reduces to 12%. a/ The Agricultural Marketing Institute.  ~ 33 - Part III Table 5 Page 5 of 8 C. Financial Imoact ------------Illustrative Farm Models------------ 1 2 3 4 5 Climate Hot Temperate Hot Cold Temperate Area Suitable for Production (hectare) 1.5 1.5 5.4 5.4 13.5 Crops (ha) 1.5 1.5 3.5 3.5 8.0 Annual Family Income After Debt Service b/ -C---_____________---col$