Colombia - Irrigation Rehabilitation Project

Documnt Of The World Bank FOR OFFICIAL USE ONLY Rep"t No.8542 PROJECT COMPLETION REPORT COLOMBIA IRRIGATION REHABILITATION I PROJECT (LOAN 1996-CO) APRIL 17, 1990 Agriculture Operations Division Country Department III Latin America and the Caribbean Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their offical duties. Its contents may not otherwise be disclosed without World Bank authorization.  . > ~~~~~~~J -. CURRENCY EQUIVALENT Currene Unit Colombian Peso (ColS) Appraisal 1980 US$1 - ColS 50.0 January 1989 US$1 _ Col$ 339.7 WEIGHTS AND MEASURES Metric System GLOSSARY OF ABBREVIATIONS CAJA - Caja de Credito Agrario, Industrial y Minero (Agricultural, Industrial and Mining Credit Bank) FEDEARROZ - Federacion Nacional de Arroceros (National Federation of Rice Growers) HIMAT - Instituto Colombiano de Hidrologia, Meteorologia y e^decuacion de Tierras (Colombian Institute for Hydrology, Meteorology and Land Improvement) ICA - Instituto Colombiano Agropecuario (Colombian Agricultural Institute) INCORA - Instituto Colombiario de la Reforma Agraria (Colombian Institute for Agrarian Reform) SENA - Servicio Nacional de Aprendizaje (National Apprenticeship Service) GOVERMENT OF COLOMBIA FISCAL YEAR January 1 - December 31  FOM MIAL USK ONLY THE WORLD SANK Washpon. D.C. 20433 UIA. Olkfie o f 0Mgc.Coww April 17, 1990 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECTs Project Completion Report on Colombia Irritation Rehabilitation I Pro_ect (Loan 1996-CO) Attached, for information, is a copy of a report entitled "Project Completion Report on Colombia - Irrigation Rehabilitation I Project (Loan 1996-CO), prepared by the Latin America and the Caribbean Regional Office. No audit of this project has been made by the Operations Evaluation Department at this tzime. Attachment I This document has restrvicted disibuton my be uwsd bY t etus oly in t pefmC of dtit McWdutias.Its cawats may not othewise be disclosd witmou WOWl Sank autbouinlion  FOR OMCLAL USE ONLY COLOMBIA IRRIGATION REHABILITATION I PROJECT (LOAN 1996-CO) PROJECT COMPLETION REPORT TABLE OF CONTENS Paze No. Preface ....................... Evaluation Summary .................i... Part I - PROJECT REVIEW FROM BANK'S PERSPECTIVE . ...............1 A. Project Identity . . . .............................................. 1 B. Project Background ........... . ................................... 1 C. Project Objectives and Description . .. 1 D. Project Design and Organization . . . 2 E. Project Implementation .. . ........................................ 3 F. Project Results .................................................. 4 G. Project Sustainability ... . . ................................. . 7 H. Bank Peformance .................................................. 7 I. Borrower Performance ................................... . .. ....... 8 J. Project Relationship ................. . . .......... 8 K. Consulting Services ....................... 8 L. Project Documentation and Data ................ .................. . 9 Part II - PROJECT REVIEW FROM BORROVER'S PERSPECTIVE ................... 10 0, Part III - STATISTICAL INFiRs&TION ................................ 15 - Table 1 - Related Bank Loans............................. 15 - Table 2 - Project Timetable ............................ 16 - Table 3 - Loan Disbursement ............................ 20 - Table 4 -Project Costs .......*................ 21 - Table 5 - Direct Benefits ........................2 - Table 6 - Status of Covenants ........... .........,... 28 - Table 7 -Staff Inputs ............. 30 Map No. - IBRD 15418, February 1981 This document has a rstricted distribution and may be used by cipients only in the performance of their offcial duties. Its contents may not otherwise be disclosed without Wodd Bnk authodzation.  COLOMBIA IRRIGATION REHABILITATION I PROJECT (LOAN 1996-CO) PROJECT COMPLETION REPORT PREFACE This is the Project Completion Report (PCR) for the Irrigation Rehabilitation I Project in Colombia, for which Loan 1996-CO in the amount of US$37.0 million was approved on April 16, 1982. The loan was closed on December 31, 1988, six months behind schedule. The loan has been fully disbursed (US$34.4 million), taking into account the US$2.6 million which was cancelled, due to savings, on September 8, 1987. Final disbursement was on June 30, 1989. The PCR was jointly prepared by the Agriculture Operations Division of the Latin America and the Caribbean Regional Office (Preface, Evaluation Su-mary, Parts I and III) and the Borrower (Part IZ). Preparation of this PCR was started during the Bank's final supervision mission of the project in July 1988, and is based, inter alia, on the Staff Appraisal Report: the Loan &ad Guarantee Agreements; supervision reports, correspondence between the Bank and the Borrower; and internal Bank memoranda.  - ii - COLOMBIA IRRIGATION REHABILITATION I PROJECT (LOAN 1996-CO' PROJECT COMPLETION REPORT EVALUATION SUMMARY Obiectives The main objectives of the project were to: (a) halt deterioration of existing irrigation infrastructure (canals, dikes, ttructures and roads); (b) construct new irrigation infrastructure to improve ase of available water and land resources; (c) improve the operation and maintenance of the irrigation infrastructure; (d) improve livestock farming methods and increase yields of irrigated crops, including the introduction of rice cultivation in basins; (e) provide technical and seasonal short- and medium-term financial assistance to allow beneficiary farmers to improve their production capability; (f) gradual transfer responsibilities for operating and maintaining the irrigation and drainage works to the user associations; and (g) strengthening the Borrower's staff capability in the areas of water management, appraisal and supervision of irrigation works (para. 4). Implementation Experience Apart from a loan closing date extension of six months, the project was implemented as planned. At times, the project suffered from insufficient counterpart funds. However, Bank actions, such as retroactive financing, use of a Special Account, introduction of uniform international competitive bidding procedureis of Bank loans in Colombia, and modification to Loan Agreement to allow H&MT to contract civil works up to project closing date, contributed favourably to successful project implementation (para. 10). Of the potential risks identified by the appraisal mission, only the farmers' reluctance to invest in land levelling was of significant consequence (para. 11). Final project costs were _out US$80.5 million, 93Z of appraisal estimate of US$86.3 million (para. 14, Table 4). Results The objectives for the major civil works, comprising rehabilitation of eight irrigation districts were fully met (para. 16). But the appraisal target for land levelling of 12,400 ha was not achieved, with only 5,700 ha levelled (462 of appraisal target). A contributing factor was the lack of suitab.' land levelling equipment available in the private sector, and HIMAT's limited experience with on-farm development works. Consequently, there is still a need for more videspread use of land levelling techniques (para. 17). The objective of financing feasibility  - iii - studies for a second phase project was reached (pare. 18). The project training program was very successful. Activities included training for technicians and farmers through informal courses at national and district levels, and training in administrative and managerial aspects for the water users' associations (para. 19). Project objectives for extension services were not fully met because of inadequate emphasis by ICA to on-farm water management (para. 20). Likewise tha objective of preparing districts for financial independence from the Government and for their transition into independent water management authorities was not fully reached. Improvements were slow in the level and structure of water tariffs to recover O&H costs (para. 21). The actual economic rate of return of the project is estimated to be 342 compared with the appraisal estimate of 50Z. The projett has contributed significantly to the growth of the agricultural seceor, with about 77,000 ha being cropped annually in the eight districta, which is 362 more than the pre-project situation and compares with the SA2 fucremental area at full development of 412 (para. 24, Table 5). SustainabiLity The project significantly contributed to strengthening HVAT*'s management. Because of HIMAT's satisfactory perfor&ance, it has received two subsequent Bank loans (Loan 26e7-CO, signed on May 27, 1986 and Loan 3113-CO, approved in August 1989 and ready to be signed). In particular sustainability has been enhanced by HIMAT's and Government's recent decision to be more active in improving the level and structure of water tariffs, with the view to having irrigation districts attain financial independence (para. 25). Lessons Learned Some useful lessons have been learned from this experience for both the Bank and the Borrower: (a) annual rather than semi-annual progress reports from HIMAT proved to be adequate and, to encourage more focus on productivity aspects, semi-annual evaluation reports on the agricultural impact of the project proved to be useful; (b) contracting of demonstration services (for land levelling) to a private producers'organization (PEDEARROZ) proved to be very successful; (c) provision for the automatic suspension of disbursements on a district basis, when the Borrower failed to meet billing and collection of water charges, proved easy to apply; (d) the establisbment of the Project Special Account was crucial and ensured HIMAT's liquidity during project implementation; (e) the introduction of standard bidding documents under Bank loans in Colombia greatly accelerated ICB and LCB activities;  - iv - (f) Inclusion of project in the Bank's Special Action Program helped to overcome severe counterpart funding problemst (8) procurement efficiencies, which have been applied in the follow-on project (Loan 2667-CO), were determined by carrying out equipment purchases in a few consolidated ICBs rather than several processes each year. Also, the winning bidders have been required to train HIMAT's mechanics and equipment operators; (h) decentralization of administrative and accounting responsibilities to the Irrigation Distric, led to improved accounting procedures, which raised HIMAT's awareness as to the real administrative and O&M costs of providing irrigation. HIMAT is now conducting studies in each district to more accurately determine costs (parse. 27 and 29); and Mi) in irrigation/drainage development projects in which the main executing agency is an engineering agency, it is essential to ensure that adequate emphasis is given to the non-civil works and on-farm development and services components from the outset.  COLOMBIA IRRIGATION REHABILISATION I PROJECT (LOAN 1996-gO) PROJECT COMPLETION REPORT Part I - Proiect Review from Bank's Persgective A. Proiect Identity - Project Name: Irrigation Rehabilitation I Project - Loan No. s 1996-CO - RVP Unit : Latin America and the Caribbean Region Country Department III - Country s Colombia - Sector s Agriculture - Subsector s Irrigation B. Proisect Background 1. Sector Develoment Obiectives. The objectives put forward at the time of the project were tot increase resource productivity in the agricultural sector; improve efficiency and competitiveness of the marketing system: develop agroindustries to stabilize demand and increase employment and rural incomes; and increase foreign exchange earnings through exports of products of the agricultural sector. 2. Policy Context. The main sectoral policy, which was the subject of considerable discussion between the Bank and the Government, wasOthat, before embarking on new construction, priority should be given to the rehabilitation of existing irrigation and drainage schemes. The second key policy area was to be the reduction of subsidies for irrigation development through improved cost recovery measures. 3. Linkates Between Project. Sector and Macro-nolicy Obtectives. The project was important to the agricultural sector and in turn to the economy through productivity gains made possible through the Increased availability of water. Moreover, the project, together with subsequent Bank support to the agriculture sector, has explicitly promoted Intensive rather than extensive agriculture, given Colombia's need to greatly increase its agricultural productivity, while at the same time protecting its natural resource base from encroachment. C. Proiect Obiectives and Description 4. Proiect Obiectives. The main objectives of the project were tot (a) halt deterioration of existing irrigation infrastructure (canals,  dikes, structures and roads); (b) construct new irrigation infrastructure to improve use of available water and land resources; tc) improve the operation and maintenance of the irrigation infrastructuret (d) increase yields of irrigated crops, including the introduction of rice cultivation in basins and improving of livestock farming methods; (e) provide technical and seasonal short- and medium-term financial assistance to allow beneficiary farmers to improve their production capability; 'f) gradual transference of responsibilities for operating and maintaining the i4rigation and drainage works to the user associations; and (g) strengthening the Borrower's staff capability in the areas of water management, appraisal and supervision of irrigation works. 5. Proiect Components. The project includeds (a) construction of irrigation infrastructure in eight districts; (b) acquisition of equipment for operation and maintenance (O&M) to replace obsolete equipment and vehicles in the project districts; (c) provision of short- and medium-term credit to farmers in the districts for financing irrigation works and promoting agricultural production at the farm level; (d) provision of technical assistance to small-scale farmers, by means of preparation and supervision of irrigation and investment plans at the farm level and establishment of about 90 levelling and pasture management demonstration plots; te) feasibility studies and engineering designs for prepariag a second phase project; (f) training of the Borrower's staff and Technical Assistance Agencies' staff in the fields of project prepara*'.on, management, economic analysis, and accounting and management for representatives of user associations; and (g) monitoring and evaluating the project. D. Project Design and Organization 6. The conceptual foundation for the project was appropriate and clear. The design, however, required that the project rehabilitate eight separate irrigation/drainage districts, scattered all over the country, each requiring its own preparation work. For this reason, the project required extensive preparation by FAO/CP and the Bank and above-average Bank staff appraisal time. 0. 7. Preparation Phase. The FAO/CP mission visited Colombia in June 1979, September 1979, January 1980 and June 1980. During this time there were changes in project definition and scope. The initial fifteen irrigation schemes proposed by the Government were divided into two projects, with eight schemes being included in the first phase, i.e., the subject of this PCR. A second phase project now under execution involves the other seven schemes. A total of 169 staff weeks was required for the preparation stage up to appraisal, more than double the usual time taken for preparation (Table 7). However, in retrospect, the staff input was well spent in designing a successful project with a new Borrower. Subsequent Bank-financed projects with the same Borrower have required far less time to prepare. Indeed the most recently approved project required considerably less than the average. 8. Appraisal. This was the first Bank loan with NIMNT as executing agency, and, therefore, the Bank proposed that each of the eight districts should be evaluated. This posed logistical problems, which were resolved  - 3 - by sending a 10 staff appraisal mission (five Bank staff and five consultants) requiring one month in the field. A total of 101 staff weeks was required for appraisal (Tabl 7). There was much confusion over the establiubtent of a revolving fund. Unnecessary discussions could have been avoided if there had been a preappraisal issues/decision procedure during which the revolving fund issue could have been resolved. However, at the time of appraisal the establishment of Special Accounts was not a comon practice in the Bank. The appraisal mission felt that the establishment of a 'revolving fund' would be required, because Colombian budgetary procedures only permitted HIMAT to commit itself to expenditures within its counterpart budget allocation. Therefore, the appraisal mission was aware of the chicken and egg situation, but at that time, establishment of Special Accounts were not common practice. 9. HIMAT was res onsible for the implementation of the project. HIMAT's role and responsibilities were clearly defined and understood, and this contributed significantly to the project's success. E. Proiect ImDlementation 10. Apart from a loan closing date extension of six months, the project was implemented as planned. At tLmes, the project suffered from insufficient counterpart -ds. However, the following Bank actions favorably contributed to successful project implementations (a) the Bank retroactively financed US$0.4 million; (b) tte Bank opened a Special Account (revolving fund); (c) uniform International Competitive Bidding (ICB) procedures for Dlik loans were introduced in Colombia on September 25, 1984; (d) the Bank incluied the project in its Special Action Program on February 11, 1985. This allowed the increase of Bank financing, to the project, from 692 to 802 and ensured HIMAT's liquidity during project implementation (paras. 14 and 27); (e) the Loan Agreement was modified on December 20, 1985 to allow HIMAT to contract civil works up to the project closing date; and (f) the Bank extended the project closing date on May 24, 1988 for six months from June 30, 1988 to December 31, 1988. In addition, the project contributed to the smooth follow-on of a subsequent project, and this was facilitated by the Bank's authorization on September 1, 1985 to allow HIMAT to use project funds to contract the final designs for civil works to be carried out during the first year of implementation of Phase II, in addition to feasibility studies for Phase II. The Bank cancelled US$2.6 million of the Bank loan on September 8, 1987 due to savings in consultant and extension services; the latter was associated with slow start-up and only partial implementation. The Borrower had requested that the savings be used to finance additional civil works. However, because such works were beyond the scope of the project and were insufficiently prepared technically, the Bank declined the request. 11. Proiect Risks. Of the potential risks identified by the appraisal mission (such as credit availability and fanrers' reluctance to invest in land levelling), only the farmers' reluctance to Invest in land levelling was of significant consequence. With the exception of Zulia district, which reached the goals on land levelling, the other irrigation districts did not reach project targets estimated at appraisal.  -4- 12. Unforeseen Factors Affecting Proiect ImDlementation. There was a delay in starting-up the extension service to farmers, and a lack of specific extension programs for on-farm water management. This contributed to the reluctance of farmers to invest in both land levelling and improved water control. Also, the lack of a monitoring system for the credit service did not permit an evaluation of the adequacy and quality of the service. 13. Procurement. Procurement of equipment and vehicles was slow in the beginning due tos (a) HIMAT's lack of familiarity with Bank procedures; (b) Government's cumbersome administrative regulatio:.s for procurement; and (c) several separate ICBs rather than consolidated ICBs were carried out during a short period of time, which contributed to the administrative burden and delays. After ICB and Local Competitive Bidding (LCB) procedures were revised and approved on September 25, 1984, no major problems were encountered. 14. Proiect Costs. The final cost of the project was about US$80.5 million, 932 of appraisal estimates (US$86.3 million). In spite of cancellation of US$2.6 million of the Bank loan and also the increase in Bank financing (from 692 to 802 effective February 11, 1985), the Governmert financed about US$0.8 million more than anticipated in equipment mainly due to higher local expenditures in clearing the equipment and vehicles acquired under ICB from customs. There were also higher incremental O&M costs (Table 4). 15. Disbursements. It was estimated at appraisal that the Bank loan would be disbursed in six and one-half years. Actual disbursements took place over approximately seven years (Table 3). F. Proiect Results. (See also Part III for Project Profile at Completion) 16. Proiect Obiectives. The objectives for the major civil works, comprising rehabilitation of eight irrigation districts were fully met. The execution of the civil works was very satisfactory in spite of YDMAT's limited technical contracting :apacity during the first year of project implementation. (SIMAT had to prepare contracts for a large number of contracting firms to work all over the country.) No serious problems were found during execution of the contracts and the performance of contractors was satisfactory. 17. The target estimated at appraisal for land levelling of 12,400 ha (of which 9,600 ha were in the rice districts) was not achieved. At project completion only 5,700 ha were leveled (462 of appraisal). However, in the 2ulia district, 5,000 ha were leveled (125Z of appraisal estimates) mainly because agricultural equipment was readily available and rice was a very profitable crop in this district. In other project districts the lack of equipment, HIMAT's limited experience with land levelling, poor promotion, and comparative advantage in some competing crops did not sufficiently motivate farmers to invest in land levelling fir puddled rice. At appraisal, the Bank had contracted a consultant to visit the project rice growing areas to recommend the appropriate type and amount of equipment that was to be financed with the Bank loan for the land levelling program. The recommendation was based on dry soil levelling techniques.  During project execution, HIMAT decided not to acquire the equipment because it felt that the dry soil land levelling methodology was oni, suitable for large holdings on relatively flat land, and not for small holdings wich sloping terrain. It therefore requested Bsank approval to procure equipment and tractors to work in wet soils. With some delay the equipment was procured and delivered to FEDEARROZ, with whom HIMAT had signed a contract for establiebment of demonstration plots on land levelling in the rice districts. This program was successful in achieving its target of 450 ha (90 plots of 5 ha each). The performance of FEDEARROZ was also satisfactory. However, HIMAT did not put any effort into promoting land levelling because its main concern was the construction of works to rehabilitate the irrigation/drainage districts. Consequently, the need persists today for more widespread use of land levelling techniques. FEDEARROZ has particularly become aware of this and is taking steps to promote the technology. 18. The project also reached the objective of financing feasibility studies for a second phase project. HIMAT also contracted local firms for the preparation of final civil works designs for the first year of implementation. The cost of these studies was US$1.0 mi1lion, as estimated at appraisal. The performance of the consultants was satisfactory, and the studies greatly facilitated the preparation of a follow-on project which has continued the rehabilitation of irrigation districts. 19. The project training program was very successful. HIMAT signed a contract with the National Apprenticeship Service (SENA) to carry out this component. SENA concentrated its activities in three specific fields: (a) community development, organizing and reactivating producers' associations; (b) training for technicians and farmers through informal courses at national and district levels; and (c) training in administrative and managerial aspects for the water users' associations. The performance of SENA was satisfactory. 20. NIHAT contracted the Colombian Agricultural Institute (ICA) to carry out extension services in the project areas. This component took more time than anticipated to get under way. It was not until 1984 *(two years after effectiveness), that ICA began field activities. Initially, progress was slow due to lack of vehicles and the fact that most of the extension agents were not living within close proximity to field activities. In addition, field demonstrations were too small and were conducted in research areas of ICA, thereby reducing their effectiveness. Sta-ting in 1986, the scope of the technical assistance activities was changed by ICA. The extension agents were required to live in the project areas, mew vehicles acquired with project funds were assigned to them, and field deamonstrations were conducted on the farmers' own plots at a field scale. While this wes a great improvement, ICA did not fully achieve the project objectives because it gave inadequate emphasis to on-farm water management. HINAT also must share responsibility for not training the extension agents in on-farm water management. 21. In this first phase project, HIMAT did not reach the objective of preparing all districts for financial independence from the Government, and for their transition into independent water management authorities.  - 6 - However, two districts (RUT and Rio Recio) were transferred to the users' associations after the project closing date. Improvements were slow in the level and structure of water tariffs to recover O&M costs. HIMAT was unable to establish water tariffs for the district of Sibundoy (an area with a large Amerindian population who has refused on legal grounds to pay any tariffs), and the Bank subsequently suspended disbursements for civil works in Sibundoy on January 1, 1986. No action has, as yet, been taken by HIMAT to charge tariffs in Sibundoy. HIMAT's record keeping on cost recovery was also deficient. HDHAT did not provide the BP ik with an analysis of proposed water tariffs in accordance with O&. costs, and it was only in the latter years that HIMAT began to introduce improved accounting procedures as it decentralized responsibility for such to the irrigation districts. On the other hand, HIMAT did introduce a computerized central billing procedures for all its districts during the execution of the project. In the second phase project, HIMAT is conducting studies to identify the main weaknesses in administration and accounting with the purpose of improving its management and procedures. 22. Impact of Proiect. Project impact has been measured through both financial analysis (at farm level) and economic analysis. 23. Financial Analysis. Based on similar illustrative farm model sizes that were envisaged at appraisal for representative enterprises in each district, actual figures of available pre-tax income for project beneficiaries and financial rates of return were estimated (using January 1989 prices) and compared with appraisal estimates. Details are presented in Part III, Table 5, Page 4. The actual pre-tax income at full development for project beneficiaries increased from 1992 to 952? in four districts. This compares very favorably to appraisal estimates whereby incomes were projected to increase from 1002 to 185?. However, in three districts, income only increased from 182 to 45S and in one district decreased to -272. This was because of higher costs of production (labor and transport) in San Alfonso, (equipment and operation) in Roldanillo, Toro and La Union (RUT) and a very low level of development in Sibundoy where there is a large Indian population. The calculated financial rate of return (FRR) was higher than 502 in all project districts, comparedtwith appraisal in which it was estimated higher than 502 in five districts, and lower in three (from 322 to 44?). In part the higher FRR's were because the full cost of water was not charged. 24. Economic Analysis. The actual economic rate of return in the project computed over 20 years, is estimated to be 34? compared with the appraisal estimate of 50. The project has contributed significantly to the growth of the agricultural sector. About 77,000 ha are being cropped annually in the eight districts, which is 36? more than the pre-project situation and compares with the SAR incremental area at full development of 412. About 55? of the area is cultivated with rice which produces about 127,000 MT of paddy per year. In general, yields have increased. The technique of puddled rice cultivation has been introduced with about 202 higher yields than anticipated at appraisal, but the area of such was only about 50? of the appraisal target because of the slow adoption of land levelling techniques discussed in para 17. Other crops have increased their yields by about 15t (average), over appraisal estimates. Table 5 in Part III, shows the evolution of crops, yields and production per year in the project districts.  -7- G. Prolect Susta