Colombia - Third Atlantico Irrigation Project

Report No. 3959 L COPY Impact Evaluatilon Report Colombia: Atlantico Irrigation Project June 21, 1982 Operations Evaluation Department FOR OFFICIAL USE ONL'V Document of the Wbrld Bank 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 authoi-ization.  FOR OFFICIAL USE ONLY IM_ACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) TABLE OF CONTENTS Page No. Preface .................. ........................................ i Summary and Conclusions ....... ............ ......... ...... ... ... .. ii I. o INTRODUCTION ........................................ .............. 1 A. Historical Background . ....... ... .. 0.. . -. . .. . . .. . . . 1 B. The Project ............. 3 II. MAIN DEVELOPMENTS SINCE PROJECT COMPLETION ............. 6 A. Drainage System ............. .............. ...... ...-.. 6 B. Irrigation Scheme * . .. o..o.... .... ..... ...... ....... .. 7 C. Connection with the Main Electric System ........... 7 III. SOCIAL IMPACT .................................. 8 A. Agrarian Reform. . ......... ....... .... ....... .... ... 8 B. Failure of the Cooperative System ............... o. 9 IV. AGRICULTURAL IMPACI ... o.. .. ..... ... o ........ . ............ 12 Ao Cropped Area ooo .................................. 12 B. Net Value of Agricultural and Livestock Production ....* .......... o.o.o.o.o .................. o o ... ... 13 V. ECONOMIC IMPACT .. .. ........ .. ..... . o ..... .... o. 16 VI. IMPACT ON INSTITUTIONS . ...... ......... ...... . e...o. 17 VII. FUTURE PROSPECTS .. .. ..... .o..o ........ . ....... .o. . 18 VIII. CONCLUSIONS o.............o .oo..o..ooo .......o.....................o 19 Appendix 1: Borrower's Comments ..... e...*........................ 23 Annex 1: Table 1 - Appraisal Cost Estimates . . ...................... ...... 31 2 - Size Distribution of Rural Property in Southern Part of Atlantico Province in 1966 .. o.... 33 3 - Comparison of Estimated Costs and Actual Expenditures o...... o.oo* o... ........................... .oo.... 34 4 - Allocation of the Proceeds of the Loan ............. 36 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. TABLE OF CONTENTS (cont'd.) Page No. Annex 2: Table 1 - Distribution of Farming Families in the Project Area (Phase I and II) ...................... 37 2 - Land Acquisition by INCORA .......................... 38 3 - Distribution by Year of Land Acquisition by INCORA .. 39 4 - Evolution of Cash Crops in the Settlement Area ...... 40 5 - Distribution of Land Use by District in Phase I and II Areas (1980) ............................... 41 6 - Land Use by Settlers in Phase I and II Areas (1980) . 42 7 - Land Use by Private Farmers Under Rainfed Conditions. 43 8 - Key Indicators: Summary of Actual and Projected Cultivated Area, Net Value of Production and Net Income Per Family *s .......................... 44 Maps: - IBRD - 10163R (Map 1) IBRD - 3245R2 (Map 2)  IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) PREFACE This is an Impact Evaluation of the Atlantico Irrigation Project in Colombia, for which Loan 502-CO was approved in June 1967 in the sum of US$9.0 million and closed- eight and a half years later in December 1975 after cancellation of US$0.3 million. The Impact Evaluation Report is based on a review of the Appraisal Report (No. TO-589-b) dated June 13, 1967, the President's Report (No. P-552) of June 15, 1967, the Loan Agreement dated July 31, 1968, and the Project Performance Audit Report (PPAR, OED Report No. 1841) of December 28, 1977; relevant Bank files have also been consulted, and Bank staff associated with the project have been interviewed. OED missions visited Colombia in April and October 1981. Farm surveys and data gathering, which were carried out in August-September 1981 by the Colombian Institute for Agrarian Reform (INCORA) and the Hydrology, Meteorology and Land Development Institute (HIMAT), yielded many valuable insights regarding the social and economic impact of the project. OED mis- sions held discussions with officials of INCORA and HIMAT. A field trip to interview a number of part:icipating farmers was undertaken. The information obtained during these missions was used to test the validity of some conclu- sions of the 1977 audit report and permitted discussions of the project's impact on the lives of the project's beneficiaries and on agricultural produc- tion. A copy of the draft report was sent to the Borrower on March 16, 1982. However, no comments were received. This project was selected by OED for impact evaluation for the following reasons: (i) it was the first Bank financed irrigation and land settlement project in Colombia, (ii) its implementation time overrun was about 57%, mostly because of technical problems; and (iii) its economic and social impact was still uncertain at time of the audit. The valuable assistance provided by INCORA and HIMAT and their staff, as well as the farmers interviewed during the preparation of this report, is gratefully acknowledged.  - ii - IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) SUMMARY AND CONCLUSIONS The Project 1. Loan 502-CO (US$9.0 million), approved in June 1967, was to finance the construction of irrigation and drainage systems on 3,300 ha of previously undeveloped land (Part I); flood protection and drainage on 6,000 ha of rainfed land (Part II); cgrarian reform and land settlement on both parts; construction of roads and marketing facilities; purchase of agricultural equipment; and the feasibility study for a Stage II project (Loan 849-CO, approved by the Bank in Juae 1972). 2. It took more than eight years to disburse the loan. Technical problems were the main factor slowing down project implementation: most of the soils to be irrigated proved to be saline and/or alkaline. Procurement difficulties in constructing pumping stations were another reason for delays in implementation. The project envisaged developing fruit and vegetable production for export under a sophisticated fixed sprinkler irrigation system. However, the intensive cropping pattern could not be implemented due to the poor quality of soils and had to be replaced by a traditional cropping pat- tern, with output for the internal market. The land reform program was only partly completed; land acquisition and redistribution to settlers were delayed by difficulties encountered in completing negotiations with landowners and insufficient funding to purchase the land. Settlers were grouped in produc- tion cooperatives based on collectivized production and centrally planned farming operations, without: the right to farm individual plots. 3. At completion, it: was found that the net value of production at full development would be only about 25% of that originally estimated and that full agricultural development would not be reached before 1981. The project's overall ERR was estimated at 2.2% compared with 11% for Part I and 25% for Part II estimated at appraisal. 4. The PPAR, issued in December 1977, concluded that (i) because of improper project preparation and appraisal and delays in project implemen- tation, the economic results of the project were unsatisfactory but that (ii) the project had a positive impact on family incomes and living standards of project beneficiaries. Project Impact 5. Since completion, the project was adversely affected by (i) the poor efficiency of the drainage system resulting in severe floods and (ii) by the failure to complete the irrigation scheme and connect the pumping stations with the main electric grid. However, land acquisition and redistribution were completed in 1981. Land occupation by squatters, which started during project implementation, continued during the recent years. 6. In 1981, the social impact of the project turned out to be somewhat mixed but more favorable, on balance, than at project completion. As a result of the land reform and land occupancy by squatters, the population of the area (Stage I and II) increased by 650%, and the average farm size decreased by about 85% compared to the pre-project situation. The production cooperative system failed, and the main cooperative, which in the early 70s grouped about 1,000 members, went bankrupt. In 1975, settlers were given an identifiable plot which they started to work individually. All settlers interviewed during the mission expressed their disappointment with the collectivist experience and satisfaction to be settled finally on their own 'individual holdings. 7. The collapse of the cooperative system and allocation of individual plots to settlers, combined with the deficiencies of the irrigation and drainage systems, led to fundamental changes in the distribution of crops grown in the project area. The area of cash crops was reduced by about 85%, while food crops and livestock production increased considerably. Marketing of products, supply of inputs and access to credit remain major constraints to agricultural development. The net value of production in 1981 was about the same as in 1975 and only 48% of that projected at completion for 1980. The project ERR is now estimated at 1.7%. 8. Although most of the farmers can be considered subsistence farmers, their net family income is significantly higher today than in the pre-project conditions but only slightly better than the minimum wage in agriculture. The satisfaction generally expressed by settlers reflects more a contentment to be landowners for the first time rather than beneficiaries of higher living standards. 9. In conclusion it can be said that, in economic terms, the Atlantico Project was a failure. Nevertheless, its social impact is important since an orderly agrarian reform program and an acceptable land tenure system have been introduced and living standards of about 22,000 people have been improved. The project's positive social impact remains precarious, however, because of the drainage system's deficiencies and the subsequent increased risk of floods if no prompt remedial action is taken. 10. In addition to the main findings of the PPAR, the following points which were not apparent at the time of audit may be of particular interest:  - iv - - settlers, who were initially considered the poorest segment of the population, tturned out to be a privileged group compared to other farmers (parts. 3.02, 3.05 and 4.03); - failure of the production cooperative severely dampened the cooperative spirit of settlers and demonstrated the risk of (i) Introducing systems modeled after those of other countries and not adapted ito local socio-economic conditions and (ii) developing a set,tlement scheme which left little initiative to settlers and was not based on a self-help concept (paras. 3.06-3.10); - the major flaw of the project was that no attempt was made to fundamentally change the project design or drop the irrigation component when unfavorable soil conditions became apparent (paras. 5.02-5.03); - the Bank requirement that consultants be responsible for project management proved ill-founded (paras. 6.01-6.03); - data provided by consultants were not reviewed critically by the Bank for accuracy, completeness and consistency (paras. 6.01 and 8601) .  IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) I. INTRODUCTION A. Historical Background 1.01 In 1961, the Co:Lombian Government adopted a land reform policy to correct what it considered to be an inequitable land tenure situation. The 1960 census had shown that 765,000 small farms with an average size of 1.7 ha held 55% of the land while 14,600 large estates with an average size of 750 ha owned 45% of the land, much of it under-utilized. The Colombian Institute for Agrarian Reform (INCORA) was established in December 1961 not only as a land reform organization but also as an important agency for agri- cultural development. Its main objectives were to: (i) redistribute large land holdings; (ii) develop unused lands and those not being used to their fullest potential; (iii) increase both agricultural productivity and total agricultural production; (iv) improve security of land tenure for small farmers and make it easier for them to acquire land; (v) improve farmers' living standards by providing them with technical assistance, credit and various social services; and (vi) ensure better conservation and use of natural resources. 1.02 The Land Reform Law required INCORA to establish "family farm units" wherever possible in its la-nd settlement work. A "unit" was defined as the area of land sufficient to provide a normal family with an income adequate for its maintenance, for paying off debts incurred in acquiring or improving the land and purchasing equipment and for progressively improving housing and general living standards. It was accepted that the size of settlement/ development units would vary in relation to the soil, climate, location, relief and type of crops to be produced; subdivisions of land smaller than 3 ha were forbidden by law. 1.03 To allow INCORA to achieve its objectives the Land Reform Law permitted it to acquire fanrs larger than 100 ha in irrigation projects by compulsory purchase at valuntion. INCORA, however, had no legal right to force owners of farms smaller than 100 ha to sell their land, nor was it able to acquire by compulsory purchase all the land owned in holdings larger than 100 ha, since under the law owners of such holdings can retain 100 ha for their own use. 1.04 As an agency for agricultural development, INCORA was to operate a supervised credit program in association with the Caja Agraria. The program aimed at providing annual production loans as well as medium- and long-term loans to farmers. INCORA was to find the funds, decide on credit allocation and supervise its use. The Caja Agraria handled the financial transactions, kept records and took care of repayment collection.  - 2 - 1.05 In the INCORA project areas, all private farmers had to conform to INCORA's irrigation and cropping programs. The Land Reform Law permitted INCORA to enforce its rules by applying sanctions to farmers who did not cooperate. INCORA could exercise its legal right to withhold water deliveries from them and could exclude them from its programs of technical assistance and credit. INCORA was to recover costs of the irrigation investment through water charges. Private owners who decided to remain in the project area were also forced to adopt the official cropping plan and follow agricultural practices established for the project. 1.06 INCORA commenced its work by granting land titles to occupants of existing holdings and later began to redistribute lands. These operations were soon followed by agricultural development programs. At the end of 1965, INCORA implemented irrigation and agricultural development projects in an area of about 200,000 ha, and a further 140,000 ha was under investigation (map 1). In 1965 INCORA requested a Bank loan to finance its settlement and development program in the Department of Atlantico in northern Colombia, where the Guajaro Reservoir was under construction and was expected to provide water to irrigate about 35,500 ha. 1.07 The proposed project was located in the southeast of the Department of Atlantico, bordered by the Magdalena River to the east and by the Canal del Dique to the southwest (see map 2). The climate of the region is tropical and semi-arid with an average annual rainfall of 1,000 mm but with irregular dis- tribution throughout the season. The project area is an undulating plain, divided into a series of low ridges by a number of small lakes or swamps. First project investigations indicated that most land was under natural forest or pastures, carrying beef and dairy cattle under extensive systems of manage- ment. The area was considered one of the most economically depressed in the country due to the decreasing productivity of the fishing industry, which for several decades had been the primary source of income to the local popula- tion. A limited amount of subsistence farming was practiced in the area, but crops were subject to damage by annual flooding from the Rio Magdalena and local runoffs and the usually unfavorable rainfall distribution. 1.08 As initially conceived, the Atlantico Project envisaged agricultural and rural development in an area covering about 66,000 ha in which three sectors to be developed separately had been identified: the Repelon Sector (4,700 ha) to the west, which was to be developed by INCORA with its own resources; the Molinero Sector (11,700 ha) to the north, being developed by private interests, and the South Sector (35,000 ha) to be developed with the Bank's assistance (see maps). A feasibility study on agricultural development of the South Sector had been carried out by consultants, covering also aspects of irrigation, drainage and flood control. A preliminary soil survey had also been carried out by a local firm. 1.09 A Bank mission visited the project in June 1966. Its main preoc- cupation was to evaluate the managerial capacity of INCORA to undertake a 35,000 ha project and the suitability of some soils for irrigation. The mission proposed, and the Government accepted, to reduce the risk by phasing  - 3- the project in two stages. The project under Loan 502-CO, appraised in September 1966, was designated as Stage I, including irrigation and drainage of 3,300 ha and flood protection and drainage of 6,000 ha to be farmed under rainfed conditions. Loan 849-CO, appraised in October 1970, was designated as Stage II for drainage and land preparation on 17,000 ha for rainfed farming. B. The Project Project Description 1.10 As approved by the Board, the project which is the subject of this Impact Evaluation Study included two parts. Part I entailed the construction of a complete sprinkler irrigation system and drainage network and on-farm development needed for intensive cropping, mostly fruit and vegetables for export, on about 3,300 ha; purchasing agricultural equipment; setting up marketing facilities including an orange packing plant; establishing trial farms; providing agricultural management including technical assistance to farmers accompanied by credit facilities; acquiring most of the land and establishing settlers; preparing feasibility studies for future stages of the project and training personnel abroad. Part II included construction of an interceptor channel to protect an area of 6,000 ha of cultivable land from flooding, access roads and the introduction of improved cropping practices for traditional rainfed agriculture. The project cost (Annex 1, Table 1) was estimated at US$15.4 million (US$53 million in 1981 terms) and the Bank loan (US$9.0 million or US$31 million in 1981 terms) was expected to be disbursed over a five-year period. 1.11 At the time of appraisal, all land of the Atlantico South Sector (35,000 ha) was in the hands of about 850 owners (Annex 1, Table 2); approx- imately 4% was in holdings of less than 10 ha, 39% in holdings of 10 to 100 ha and 57% in holdings larger than 100 ha. INCORA was expected to complete the acquisition of farms larger than 100 ha before the end of 1967 and to select settlers from the local farmers according to their "farming and business ability, physical fitness, willingness to work, adaptability, intelligence, discipline and mechanical knowledge".-' The land acquired by INCORA was to be divided into family farm units of about 3 ha in Part I and 5 ha in Part II. Project management was to direct settlers as to the types of crops to be grown and cultivation practices to be applied and would provide techni- cal assistance, machinery services, supervised credit and marketing services. 1.12 INCORA was to employ a qualified consulting firm which would be responsible for engineering, agricultural development, management and market- ing during the construction period of the project and for at least five years thereafter. As the projeci: became well established and INCORA gained experi- ence, its own staff would. progressively take over full responsibility for project management. 1/ Appraisal Report, para.. 4.18.  - 4 - 1.13 At full development, family incomes were expected to increase 1200% for Part I and 400% for Part II beneficiaries. Based on the assumption that about 75% of the fruit, vegetable and tobacco production would be exported, investments in Part I of the project were expected to yield an ERR of 11% while double cropping of traditional crops in Part II would generate an ERR of 25%. Project Implementation 1.14 The loan was declared effective on November 2, 1969 when a contract with a foreign consulting firm was signed. Project start-up was rapid and effective. However, reconnaissance soil surveys and groundwater testing over 8,000 ha in the South Sector by the consulting firm established that (i) salinity and alkalinity could be found to varying degrees throughout the Stage I and Stage II areas, (ii) groundwater was also saline and (iii) about about 1,250 ha (Block II) in Part I area (see map) were unsuitable for field crop cultivation. An area in the eastern part of Part I (Block III) was selected to replace Block II, but the total irrigable area was reduced from 3,300 ha to 2,100 ha. Because of delays in replanning the project, the irrigation and drainage works were not completed in Block I until in 1971, or two years behind schedule, while irrigation works in Block III were severely delayed and were not complete at project closing in June 1976. Other serious delays in project implementation were due to technical and procurement diffi- culties in constructing the San Pedrito pumping station, which was to supply irrigation water for Part I. Construction of San Pedrito was finally com- pleted in April 1977, five years behind schedule and two years after the Closing Date. At project completion a large part of the area to be irrigated still remained to be cleared, leveled and prepared for cultivation. 1.15 During the eight implementation years, there were several changes in the original plan: (a) Loan 502-CO funds were used to provide flood protection for both Stage I and II areas and not only for the 6,000 ha dry- farming area as originally planned; (b) several roads were constructed under Loan 502-CO in the Stage II area; (c) the Boquitas drainage pumping plant and the main canal in the Stage II area were constructed to a capacity sufficient to accommodate all the estimated drainage for both Stage I and II areas; (d) because of the unsuitable soils, the plan to produce high value export crops had to be abandoned and was replaced by a traditional annual crops program; consequently there was no need to construct facilities for storage, grading, packing and processing of export crop products; (e) the Manati Experimental Farm and the electric interconnection of pumping stations with the main regional grid were not financed under Loan 502-CO, but incorporated under Loan 849-CO; and (f) the purchase of operation and maintenance equipment for both Stage I and II was made under Loan 502-CO. Total project expenditures (Annex 1, Table 3) exceeded the originally estimated cost by about 9%. This relatively low increase in total cost over a period of eight years may be explained by the fact that the major works were completed or contracted in 1973, before the period of high inflation began, and that changes in project components eliminated or reduced the cost of certain items. The original  - 5- allocation of loan proceeds was amended twice (Annex 1, Table 4) during the course of project implementation to conform more closely with actual expenditures. 1.16 INCORA's land reform program was to acquire and redistribute to settlers about 14,000 ha out of the 22,000 ha cultivable (and 26,300 ha total area) in Stages I and II. However, the program encountered delays in complet- ing negotiations with many landowners, and insufficient funds were provided by Government to purchase the land. At project completion, INCORA had acquired about 8,000 ha and settled about 840 families in both Stage I and II compared to the 2,000 originally projected. Settlers were grouped into production cooperatives, the most important of which was "Cooperativa Agropecuaria del Caribe, Ltda." supervised by INCORA and its consultants. Smaller livestock cooperatives were also established. For the Stage I project area alone (9,300 ha) about 5,500 ha were acquired and redistributed to about 600 new settlers, while the remaining 3,800 ha were retained by their 310 original owners. The land reform program continled under Loan 849-CO, both in Stage I and II areas. 1.17 The loan was closed on December 31, 1975, three years later than originally planned. At coimpletion it was found that the net value of produc- tion at full development would be only about 25% of that originally estimated mostly because of shifting from high value export crops to traditional crops. In Part I (3,300 ha) of Staige I, out of 2,150 ha of irrigable land only 900 ha (650 ha crops--mostly maize, sesame, cotton and sorghum--and 250 ha pasture) were irrigated, while the balance was used for rainfed crops and natural pasture. In Part II (6,000 ha) crops covered about 2,200 ha and natural pasture about 3,800 ha. It was estimated that full agricultural development and production would not be reached before 1981 because of delays in complet- ing the San Pedrito pumping station, as well as land preparation and drainage works. The project's overall ERR was estimated at 2.2% compared with 11% for Part I and 25% for Part II calculated at appraisal. Increases in per capita income of settlers were about 4.6 times above pre-project incomes in Part I and 1.9 times in Part II, compared with 12 and 4 times respectively projected at appraisal. Main Findings of the PPAR 1.18 The PPAR, issued in December 1977 (OED Report No. 1841) concluded that, because of improper review of soil studies at appraisal and the slow pace of project implementation, the economic results of the project were unsatisfactory. The PPAR pointed out that (i) the project design was overly optimistic and did not properly take into account the regional constraints, and (ii) the rate of return computation was not given enough attention since dry-farming development was shown in the Appraisal Report to yield higher ERR than irrigation. The project, however, had a positive impact on family incomes and living standards; of project beneficiaries and made a satisfactory contribution to institution-building.  - 6 - II. MAIN DEVELOPMENTS SINCE PROJECT COMPLETION 2.01 The Stage II Project (Loan 849-CO) was approved by the Bank in June 1972, three and a half years before the closing of Stage I, and the two projects soon became interdependent. Part of the Stage I components (mainly flood protection works and roads) benefitted landowners in both areas. Simi- larly part of the Stage II project works (land preparation, enlargement of operating headquarters, drainage works and equipment) also benefitted both project areas. Since 1975, INCORA has not distinguished between Part II of Stage I and Stage II, and separate figures indicating agricultural development under rainfed conditions are not available for the two stages. Therefore, the impact of Stage I must be evaluated in conjunction with Stage II since both pro ects aimed at improving the economy of the overall Atlantico South Sector.1 2.02 Since the completion of the Stage I project at the end of 1975, three important factors have had a negative effect on the development of the Atlantico South Sector: (i) deficiencies of the drainage system; (ii) the continuing failure to complete the irrigation scheme; and (iii) delays in connecting with the main electric network, resulting in adverse effects for both the drainage and irrigation systems. A. Drainage System 2.03 The project area is a depression, enclosed on the north by hills and on the other three sides by bodies of water which are at elevations above those of the fields. Thus, any internal drainage water has to be elevated by pumps before being discharged. The drainage system, which was started under Stage I and completed in 1979 under Stage II, consists of 37 lateral canals and 6 main canals which discharge into a main collector and thence to the Boquitas pumping station, where the water is raised into the Canal del Dique (see map). In 1981, the drainage system was found to be performing poorly for the three following reasons: (a) Although originally constructed to discharge water from both Stage I and II, the design and capacity of the Boquitas pumping station proved inadequate during the rainy season (October-November); in 1971, a consultant's report recommended that the discharge capacity be increased from 5 to 10 or more m3 /sec when drainage works were implemented in the Stage II area, but the recommendation was not taken into consideration by INCORA and the Bank; in addition, recent 1/ As Loan 849-CO (Stage II) was not yet closed at the time of this Impact Evaluation, the economic rate of return was calculated for Loan 502-CO (Stage I) in order to compare the project economic viability both at completion (1975) and at impact evaluation time (1981) (see paras. 5.01-5.03).  - 7- inspection of the Boquitas station found that, because of wear of pumps, discharge capacity is now reduced to about 3.5 m3 /sec; finally, during the rainy season, the level of water in the Canal del Dique is above the Boquitas station outlet, thus hampering water discharge from the main collector. (b) Because of poor maintenance and lack of equipment, most of the main and lateral canals are silted up and do not function properly. (c) The Boquitas swamp, which was originally planned to serve as a retention reservoir for peak floods, has now been taken over by squatters and therefore cannot be used for this purpose. 2.04 As a result of the inefficient drainage system, severe floods adversely affect every year an area of about 2,000 to 2,500 ha of cultivated land, mainly in the Manati sector (see map), where main canals converge and merge into the main collector. Flooding of various degrees and duration as well as a rise of the groundwater table cause destruction of crops or serious yield reductions in a large number of other sectors in both Stage I and Stage II areas. HIMAT commissioned a study to a consulting firm to investigate the shortcomings and to recommend possible remedies of the existing drainage sys- tem. The conclusions of the study are expected to be known by May-June 1982. B. Irrigation Scheme 2.05 Irrigation started on about 1,000 ha in Block I in 1972 with tempo- rary pumps. In 1977 the San Pedrito station and five sub-stations located along the main irrigation canal were completed, and the temporary pumps were removed. However, due to a lack of funding, irrigation works were not com- pleted by 1981: one pumping sub-station in Block I as well as all pumping stations and secondary and tertiary canals in Block III remain to be con- structed. As a result, out of 2,100 ha of irrigable land, only 940 ha can actually be irrigated. C. Connection with the main electric system 2.06 The San Pedrito pumping station, the 5 irrigation substations and the Boquitas drainage pumping station were to be temporarily operated by a diesel-electric 2,600 kVa generating plant located close to the San Pedrito station. Loan 849-CO provided for a link-up with the main electric system. However, although a power line was constructed it was still not connected in November 1981. The whole pumping system has been, and still is, expensive to operate and hampered by inadequate supplies of diesel fuel and lack of spare parts for the generator. Consequently, both the irrigation and the drainage systems have remained unreliable so far.  III. SOCIAL IMPACT A. Agrarian Reform 3.01 Redistribution of land was one of the main project objectives. INCORA started acquisition of land in 1965 and virtually completed redistribu- tion in 1981 in the Stage I and Stage II areas. In 1981, land distribution to settlers (Annex 2, Table 1) reached about 13,400 ha, of which about 6,700 ha (50%) were acquired from 93 landowners, 2,700 ha (20%) were expropriated from 11 landowners and the remaining 4,000 ha (30%) were recovered from unoccupied land and drained swamps (cienagas). Evolution of land acquisition and redis- tribution in the two project areas (Annex 2, Tables 2 and 3) shows that: (i) about 57% of the land was redistributed during the Stage I (Loan 502-CO) project and the remaining 43% during Stage II (Loan 849-CO); (ii) at the end of 1981, 1,252 farming families were settled in Stage I and Stage II areas, compared with 840 at completion of Loan 502-CO (1975) and 2,000 originally planned; (iii) about 250 settlers (20%) have abandoned their holdings but have been subsequently replaced by others; and (iv) the average farm size is about 11 ha per settler family. 3.02 The large majority of settlers are former fishermen, laborers of previous landowners or ex-sharecroppers. During the impact evaluation mission (October-November 1981) all settlers interviewed indicated that they are better off now than before the project and expressed their satisfaction with being landowners despite shortcomings in the project design and implementa- tion. Records show that repayments of land costs are about 30% in arrears. Since the amount of the original debt has not changed despite inflation, repayments have gradually become less and less of a burden. (Land was sold at 1,000 to 2,300 pesos/ha while its value today is about 60,000 pesos/ha.) About 80 settlers have repaid their outstanding debts in advance to be able to sell their land or part of it. 3.03 In addition to settlers, about 2,400 farming families own about 5,900 ha in Stage I and Stage II areas, on average 2.50 ha per family (Annex 2, Table 1). These farmers are either former owners whose land was only partly expropriated, or not expropriated at all because of the small size of their holdings, or squatters who settled mainly in the drained swamps. A few owners of the first group elected to retain part of their land (to the limit of 50 ha irrigable and 100 ha rainfed land) in accordance with the provisions of the Agrarian Reform Law. Land occupation by squatters started during project implementation and has never been prevented nor legalized by INCORA. Squatters are mostly very small farmers (1 or 2 ha per family) who cultivate swamps or land considered unsuitable for distribution to settlers. As a result, squatters are generally more affected by floods and soil salinity and constitute the poorest group of the project area population. 3.04 In conclusion, the agrarian reform program of the project was imple- mented in a completely different way from what was anticipated at appraisal. As originally planned, about 4,800 family farm units of 5 ha each in rainfed  - 9 - and 3 ha under irrigated conditions were to be established in both Stage I and II project areas. Such a land distribution was based on the assumption that the quality of soil as well as irrigation and drainage works would permit cultivation of high value crops for export in irrigated land and double cropping of traditional crops under rainfed conditions. When salinity and alkalinity problems were discovered, the settlement targets were reduced to about 2,000 families and an average of about 10 ha cultivable land per family. In 1981, overall land occupation actually covered about 19,300 ha divided between 3,651 farming families, i.e. an average farm size of 5.25 ha per family, well below the project re-estimate. These figures have to be compared with the pre-project sit:uation where the same area of about 20,000 ha was distributed between about 560 landowners, with an average of 35.70 ha per family and with 57% of the land in holdings of more than 100 ha. As a result of the land reform and subsequent squatter immigration, the population of the area increased by 650% while the average farm size decreased by about 85% during the 15-year period. 3.05 Settlers who were to benefit from the agrarian reform policy were initially considered the poorest segment of the population. However, it turned out that settlers, and former small landowners, who because of the small size of their holdings escaped the agrarian reform law, are today a privileged group compared with squatters. It can be said that the Govern- ment's first objective tc redistribute large land holdings was fully achieved in the Atlantico South Sector but that a new group of underprivileged has emerged, i.e. the squatters. The other objective to prevent excessive sub- division of land was not achieved since the family farm unit of a large number of smallholders remains below the 3 ha minimum limit permitted by the Agrarian Reform Law. As in a number of other Bank settlement projects appraised in the 1960s, the Atlantico Project illustrates that: (i) the settlement objectives which at that time were considered reformist and conceptually advanced now appear conservative and somewhat outdated and (ii) prevention of illegal popu- lation in-flow and enforcement of minimum land occupancy laws are extremely difficult unless preventive measures are planned and taken in time. B. Failure of the Cooperative System 3.06 The most imporl:ant development of the project in recent years has been the collapse of the cooperative system in 1975 and the ensuing disinteg- ration of the Cooperativa. Agropecuaria del Caribe (CAC) which was established in 1965 under Project Management. The basic principles of CAC were the following: (i) compulsory membership for all settlers farming in the project areal/, (ii) collectivized production without individual plots and (iii) centrally planned farming operations. This production cooperative concept was based on the perception that new settlers, who previously were mostly fisher- men or laborers with a low degree of literacy, would lack basic knowledge and 1/ Stockbreeders were grouped into separate smaller livestock cooperatives.  - 10 - skill to develop their land adequately and adapt to improved cropping tech- niques. Although the objective of producing high value crops was abandoned soon after project start-up, the Project Management thought that settlers would be unable to adopt a modern cropping system for traditional crops unless production was collectivized and strong technical assistance provided to cooperatives. 3.07 The CAC's operating rules and farmers' obligations were the follow- ing: (a) determination of land use was the sole responsibility of Project Management; (b) the project area was sub-divided into large blocks of 100 ha or more, in which only one crop would be grown; (c) at the beginning of each cropping season, farmers had to decide on which crop they preferred to support. When shifting from one crop to another in following years, farmers were unable to identify themselves with their "own" land, since the new crop was farmed in a different block. (d) farming was mechanized using collectively owned agricultural equipment; (e) agricultural credit was granted only to the cooperative, but credit risk had to be carried by each member proportionately to his selected crop and land size; and (f) shares in the profits were proportionate to the number of hours worked and land rights. CAC started its operation in 1966, when only small areas of land were redis- tributed, and expanded its operating area proportionally as land redistribu- tion to settlers progressed. It reached a peak of about 7,000 ha cultivated land in 1970. The main crops produced by CAC were cotton, maize, sorghum, groundnuts, tomatoes and peppers. 3.08 The cooperative system, however, proved to carry the seeds of its own destruction, and signs of deterioration appeared soon after CAC start-up. First, for a number of crops there had been no previous experience in the project area, which soon became more than anything a risky experimental center; yields and profits proved low for some crops, and other crops were quickly abandoned because of technical failures or marketing difficulties. Second, CAC was unable to face up to rapid increases in input and equipment requirements, particularly harvesting equipment, resulting in further depres- sing already poor yields. Third, although settlers were represented -on the CAC Board by elected members, the latter were unable to influence decisions made by Project Management; the system of annual plot selection also meant  - 11 - that settlers were not associated with a permanent plot but were given culti- vation rights in some part of a large cropping block, the location of which changed from year to year. Finally, possibilities for their labor input, and thereby increased earnings, were poor since most of the tasks were mechanized and undertaken by CAC wordkers; the accounting system was poorly designed and kept up, and consequently farmers were not properly informed on their profit or indebtedness situation at the end of the cropping year. As a result of these shortcomings, settLers regarded themselves more as CAC employees than smallholders, yet they were expected to continue bearing the risks for agri- cultural credit granted to them by the Caja Agraria through the cooperative. 3.09 CAC was able to pursue its operations as long as credit, which was plentiful at that time, was granted regardless of profits or losses. In 1974, however, losses of CAC as a whole, and of its members in particular, were such that agricultural credit was suspended, thus triggering the collapse of CAC and, by the same token, of most of the livestock cooperatives. In 1975 settlers were finally given identifiable plots which they started working individually, and many who previously were living in villages started con- structing their houses on their own land. All settlers interviewed during the Impact Evaluation Mission expressed their bitterness and disillusionment about this unfortunate collective experience and their satisfaction to be settled on individual plots. However, since liquidation of CAC was not completed by 1981, a number of previous CAC members are still today not eligible for agricultural, credit because of the debts they contracted and did not reimburse to the Caja Agraria during the CAC experience. 3.10 As in several projects in a number of countries during the 1960-70 period, failure of the p:roduction cooperative system demonstrates the risk of (i) introducing systems modeled on those of other countries and not adapted to local, national, social, and traditional conditionsl/ and (ii) developing a settlement scheme which left little initiative to settlers and was not based on a self-help concept.2/ The failure of CAC was the main reason why about 250 settler families had to leave the project area and opted to emigrate to Venezuela as permanent cr casual laborers. Another negative effect of this unfortunate experience has been and still is an extreme resistance by the farmers towards any kind of association or cooperation. Although the remote- ness of the project area would normally lead to close cooperation between farmers for the purchase of inputs and marketing of products, this is not done, and strong individualism prevails. By going too far and too fast, the cooperative system established under the project not only failed but also destroyed any cooperatLve spirit among project beneficiaries. With the 1/ See PPAM of Senegal - Second Agricultural Credit Project (OED Report No. 3514), dated June 25, 1981. 2/ PPAM of Mauritius - Tea Development Authority Project (OED Report No. 3761), dated January 13, 1982.  - 12 - benefit of hindsight, it can be said that a better approach would have been to settle beneficiaries on individual plots first and then to help them to organize service cooperatives for input supply, marketing and agricultural equipment and other supporting services which are nonexistent in the project area today. 3.11 The failure of CAC also had a long-term adverse effect on relations between INCORA and project beneficiaries. Although INCORA's consultants were responsible for project management in general, and CAC management in particular, in accordance with the terms of the Loan Agreement (see para. 1.12) INCORA had to take the blame for the failure of the cooperative system when the consultants left and INCORA took over full project responsibilities in 1975. It is worthwhile to note that INCORA managed over time to progres- sively and successfully regain farmers' confidence through modest but effici- ent agricultural extension and training programs, benefitting both men and women in the project area, and through functional literacy programs for children (para. 6.02). During its field visit, the Impact Evaluation Mission found that the technical and social assistance of INCORA's officials is highly appreciated by farmers. IV. AGRICULTURAL IMPACT A. Cropped Area 4.01 The collapse of CAC and allocation of individual plots to settlers triggered a fundamental change in the project area's cropping system. The cash crop area, which reached a peak of 7,000 ha in 1970 and still covered about 5,000 ha in 1974, fell to about 1,000 ha in 1976 and has remained approximately constant since (Annex 2, Table 4). Most of the settlers are now convinced that poor soil quality, unresolved drainage problems, unreliable irrigation water availability and marketing and input supply difficulties make cultivation of cash crops particularly risky and therefore have shifted to what they consider safer food crops and livestock production. Although some cash crops (peppers or tomatoes for processing) are still considered more profitable than livestock and food crops, only a limited number of settlers are taking the risk to venture into these crops, the majority preferring lower but more secure incomes. 4.02 A census carried out in 1981 (Annex 2, Table 5) shows that: (i) of the 940 ha irrigable in Part I, only about 460 ha (49%) were cultivated and irrigated while the remaining 480 ha (51%) were used for pastures only partly irrigated or not irrigated at all. In the rainfed sector (about 18,400 ha in both Stage I and Stage II for which data are available), about 5,400 ha (30%) were cultivated (with cash crops and food crops accounting for 15% and 85% respectively of the total cropped area), and (ii) the remaining 13,000 ha were used for pasture, of which about 45% were improved, the rest natural or undeveloped land.  - 13 - 4.03 Different cropping patterns are used by settlers and the remaining private farmers according to farm sizes and past experience in the project area. As stated above (para. 3.05), the size of settlers' farms is generally above average, and settlErs' past experience in cash crops has been particu- larly disastrous for those who were grouped into CAC. Therefore, settlers shifted more deliberately than other farmers to livestock production, and used to limit their cultivated area to about 1 ha per family, growing mainly food crops like maize, beans, cassava and bananas (details in Annex 2, Table 6). The relatively large size of their farms also permitted them to borrow more money for building up their herds and to own a sufficient number of cattle to make a living. Although livestock production remains generally traditional and extensive (about 1 head per ha on average), some settlers succeeded in intensifying their production with improved cattle and pastures (allowing for up to 6 head per ha) and by developing milk and cheese production. On the other hand, small farmers and squatters are not able to derive sufficient incomes from raising cattle on the small 2.5 ha average plots and are there- fore compelled to use their land for both food and cash crops. For "private" farmers (Annex 2, Table 7), crops account for 37% of the land used under rainfed conditions while for settlers, only 26%. Since a large proportion of the small private farmers is concentrated in the Manati region, which is more severely affected by floods, incomes of these small farmers remain low and uncertain. Finally, a Eew of the former landowners, who escaped agrarian reform to a certain degree and who are generally experienced farmers, are suc- cessfully maintaining a mixed system of cash crops and livestock production. 4.04 In addition to the low agricultural potential of the project area and the shortcomings of the irrigation and drainage systems, a number of other constraints hamper agricultural development. The remoteness of the project area (about 100 km from Barranquilla) means that marketing of products, supply of inputs and access to credit facilities are difficult, contributing to high production costs. Transport is also hampered by floods and poor road maintenance. Only strong and efficient service cooperatives could help to break these bottlenecks, but as stated above (para. 3.10) the cooperative spirit no longer exists. Following the bankruptcy of CAC and the widespread failure of CAC members to repay their debts, about 50% of settlers have no access to credit. In addition, the project is also adversely affected by some nationwide agricultural sector problems such as (i) a steeper rise in costs of inputs than in farmgate prices in 1980-1981 and (ii) an increase in rural wages consistently higher in recent years than the increase of agricul- tural product prices. B. Net Value of Agriculrural and Livestock Production 4.05 Although it was difficult to disaggregate crop and livestock produc- tion and the net value of" Stage I from Stage II in 1975 (no separate statis- tics existed for the rainfed part of the two projects), the PPAR made an attempt to separately ca:lculate the incremental output of Stage I (9,300 ha out of a total of 18,400 ha cultivated land) including Part I (irrigation sector) and Part II (rainfed sector). In order to compare the net value of  - 14 - production in 1981 with that prevailing in 1975, the calculation of crop and livestock production was based on (i) the estimated Stage I area under crops and used for livestock production in 1981 and (ii) the production value and costs per hectare as established by INCORA and reviewed by the Impact Evalua- tion Mission. 4.06 In 1975, actual and projected cultivated area and value of produc- tion (Annex 2, Table 8) showed that: (a) for Part I (3,300 ha), the irrigated area was limited to 650 ha but was expected to increase up to 4,400 ha in 1980 through comple- tion of irrigation works and introduction of a double cropping system allowing for a 146% cropping intensity; (b) for Part II (6,000 ha), field crops, which then covered about 2,200 ha (compared with 3,400 ha in 1972) were expected to reach 3,000 ha in 1980, mostly through double cropping, while pastures were expected to increase from 3,800 to 4,000 ha; cropping intensity was to reach 116% at full development; and (c) the actual net value of production was estimated at US$1.01 million (in 1980 terms) in 1975 (compared with US$0.7 million in 1972) and was expected to reach US$2.20 million in 1980. 4.07 In 1981, the actual cultivated area and value of production (Annex 2, Table 8) were found to be as follows: (a) in Part I, about 460 ha were under irrigated crops, 480 ha under partly irrigated pasture and the remaining 2,360 ha were natural pasture and not irrigated; (b) in Part II, about 1,800 ha (a decrease of about 20% and 36% compared with 1975 and 1972 respectively) were cultivated and mostly under food crops, while natural and improved pastures had increased by 10% since 1975 and now cover about 4,200 ha; and (c) the actual net value of production is about US$1.01 million, i.e. the same as that of 1975 (in 1980 terms) and only 48% of that projected at completion for 1980. 4.08 All indications are that cropped areas will decline and pasture area and livestock production will increase unless the drainage problems are resolved, thus eliminating part of the risk confronting agricultural production. Improved pastures tend progressively to substitute for natural pastures and undeveloped land; a trend towards better breeding techniques and introduction of improved cattle can also be noted. Poor drainage remains, however, the most serious limiting factor to wider introduction of improved pastures in a number of districts.  - 15 - 4.09 The project target at the time of appraisal in 1967 was to raise the minimum net family incomes from an average of US$500 to the equivalent of about US$2,000 (in 1980 terms) by 1980. Based on the predominance of high value crops planned at appraisal, net family incomes were expected to reach the equivalent of US$4,300-6,000 in Part I and slightly more than US$2,000 in Part II at full development. At completion, in 1975, it was found that family incomes in Part I and Part II were about US$2,300 and US$960 (in 1980 terms) respectively and were projected to reach about US$4,300 and $1,650 at full development in 1980, i.e. an increase in pre-project family incomes of more than eight times in Part I and 3 times in Part II. 4.10 By 1981, however, net family income of a typical project settler in the irrigated area of Part I is estimated to have reached the equivalent of US$2,000, reflecting the low cropping intensity and the underutilization of irrigation water. In Part II, under rainfed conditions, an eleven-ha family farm with 30% of the area under crops and 70% under pasture now yields a net family income of about US$1,100. These figures show an increase of 4 and 2.2 times over pre-project family incomes. When compared with minimum wages prevailing in agriculture (equivalent to about US$1,000 per annum in 1981), average net family incomes of settlers are only 2 and 1.1 times above the earnings of farm laborers. This explains why settlers with farm sizes below average are inclined to seek part-time, or even full-time, jobs outside the project area, particularly in Venezuela, an OPEC country, where wages are three times higher than in Colombia. The satisfaction generally expressed by settlers to the Impact: Evaluation Mission appears to reflect more their contentment to be landowners for the first time and to have security of land tenure, rather than appreciation of higher living standards. Other additional benefits, not to be underestimated, are the social services provided, includ- ing new schools, health centers, rural electrification and communications facilities which were constructed during project implementation. 4.11 Squatters and small farmers!/ with a 2 or 3 ha farm constitute the poorest segment of the population. By intensifying their cropping system, they tend to reach a higher productivity per hectare than the settlers, but they remain more subjected to floods, the vagaries of the climate and the uncertainties of marketing. Their net family incomes are likely to reach only US$500-600, significantly below the minimum wage in agriculture. Their future appears gloomy since land in the Southern Sector has now been entirely appropriated and developed. 1/ Small farmers who used to live and farm in the area prior to the project.  - 16 - V. ECONOMIC IMPACT 5.01 An attempt was made to evaluate the project economic rate of return of Stage I alone (Part I and Part II totalling 9,300 ha), and on the basis of the same costs and benefits allocation used at completion. At appraisal, it was estimated that the rate of return on Part I would be 11% and on Part II, 25%. The high rate for Part II was justified in the Appraisal Report by the low investments required for flood protection. For both Parts I and II, the rate of return was based on assumed benefits from double-cropping of high value crops. In view of the unfavorable soil conditions and the high risks of double-cropping under rainfed conditions, appraisal projections were revised at completion when it was found that (i) the net value of production at full development would reach only 25% of the original projections and (ii) full agricultural development would not be reached before 1980-81. On the basis of these projections and of information on agricultural production available at completion, the project ERR was re-estimated at 2.2%. The analysis covered both Parts I and II and did not distinguish between the two, since part of the investments in drainage, roads and flood protection benefitted both areas and even Stage II. 5.02 In 1981, the 1975 projections proved to be over-optimistic since net value of production only reached 48% of that which was expected for 1980 and 12% of that anticipated at appraisal. The project's ERR is now estimated to reach 1.7%. High investment cost of the irrigation infrastructure (74% of the total cost), combined with high operation cost and under-utilization of irrigation water, is the main reason for the low economic return of the project. The question was already raised in the PPAR as to whether irrigation should have been included in the first phase, before the overall area of 35,000 ha was fully developed under rainfed conditions and the farmers had experience in modern agricultural practices and area-specific problems. Five years later, it can be said that the high investment cost of irrigation was not justified economically, and only a limited number of families benefitted from this part of the project; in addition, the problem was exacerbated by the project relying on a sophisticated and costly fixed sprinkler system. 5.03 In retrospect, the major flaw of the project was that no attempt was ever made to fundamentally change the project design and to eliminate the irrigation component when reconnaissance soil surveys and groundwater testing indicated that production of fruits and vegetables for international markets was impossible. At that time, the project should have been re-designed limiting it to (i) concentrate on drainage designed to adequately drain both Phase I and II areas; and (ii) develop the whole project area under rainfed conditions with emphasis on livestock production. Irrigation was certainly not an effective way to develop the region; improved dry-farming and develop- ment of new lands through properly designed flood protection and drainage works could have produced acceptable results and a reasonable rate of return at a fraction of the cost.  - 17 - VI. IMPACT ON INSTITUTIONS 6.01 It was customary during project implementation to blame INCORA for poor preparation and design of the project and for delays in implementation. However, INCORA was at that time a young organization, and the role played by local and foreign consultants turned out to be of paramount importance during both preparation and implementation of the project. Unfortunately, a prelim- inary soil survey was badly handled by a local consulting firm, and project preparation by a foreign consulting firm proved inadequate. In addition, although INCORA was nominally responsible for project implementation, the Bank requested that consultants be recruited to manage the project during the construction period and for at least five years thereafter. The consul- tants were to be responsible for "all technical aspects of agricultural development, including selection of cropping patterns and cultural practices, maintenance of quality standards, processing and marketing, and would be given wide latitude for independent decision and action in these matters."l/ 6.02 In retrospect, the Bank requirement appears ill-founded in a country as developed as Colombia in 1967 and proved to be damaging to the project by unnecessarily dividing project responsibilities between INCORA and the consul- tants. Consultants were undoubtedly needed for the detailed technical design but not necessarily for project management. Consultants played a major role in designing, organizing and managing the CAC, which turned out to be the most dismal failure. The Bank passively witnessed the creation, deterioration and disintegration of the cooperative system despite INCORA's reluctance to pursue what soon appeared to be an unfortunate experience. Following CAC's collapse, the consultant's contract was terminated in 1975 instead of 1977 as initially planned, and INCORA was left alone to shoulder the blame for the project's failure. 6.03 In contrast to this, it is surprising that the project, which was the first of its type ever to be undertaken by INCORA, had a strong and beneficial influence on the development of this organization. Many of the procedures and the methodology which became standard for other projects emerged from the operations and problems experienced by the Atlantico project. Many of the INCORA staff members who later assumed more responsible positions received their practical training through the project. INCORA initiated agrarian reform and land development in 35 specific areas throughout Colombia and became the most important institution in the country responsible for executing land development projects and providing assistance to small farmers. After project completion, Government interest in agrarian reform diminished and this led to a reorganization of INCORA and a reduction of its activities. However, INCORA remains responsible for most of the agricultural extension services in the whole Atlantico area, with a staff of 96 both in Barranquilla 1/ Appraisal Report, para. 5.03.  - 18 - and at field level. As noted above (para. 3.11) INCORA also established successful training programs for men and women and literacy programs for children. Technical assistance to settlers is reasonably good, with one extension agent dealing with 128 farmers and 1,213 ha, but INCORA's effec- tiveness is hampered by the small number of livestock specialists. 6.04 After project completion, responsibility for operation and main- tenance (O&M) of the irrigation and drainage system was transferred to the Hydrology, Meteorology and Land Development Institute (HIMAT), which is now responsible for the implementation and the O&M of main irrigation and drainage systems in Colombia. HIMAT is not adequately equipped to maintain the drain- age system of the Atlantico South Sector, and its US$645,000 annual budget for O&M of the project works (Phase I and II) is estimated to cover only 60% of actual requirements. HIMAT is also responsible for collecting water charges, including a fixed levy of US$20/ha/year and a volumetric charge of 4 cents/m 3 , as well as a drainage fee of US$6/ha/year in the rainfed sector. Cost recov- ery rates are low (about 10%) as farmers are reluctant to pay for irrigation and drainage services they consider inefficient. The cost recovery rate is not expected to increase in the near future unless rehabilitation of the drainage system is implemented. 6.05 In conclusion, it can be said that the project's impact on institu- tion building has been strong during implementation and profitable to the country as a whole. However, this impact had little long-term effect, mostly due to a change in government policy regarding agrarian reform and agricul- tural development as well as to the national budget constraints of the recent years. VII. FUTURE PROSPECTS 7.01 Deficiencies of the drainage system now constitute the main con- straint to crop and livestock development of the Atlantico Southern Sector. The problems are particularly severe in the Manati area and were notably aggravated by the completion of the drainage network under Phase II (para. 2.04). Reinforcing the pumping capacity of the Boquitas station, changing the drainage layout, and improving maintenance of the drains are among the solu- tions now being reviewed by HIMAT's consultants. There is no doubt that in the absence of remedial action the current situation is doomed to deteriorate because of increasing siltation of canals and obsolescence of the pumps. During the Impact Evaluation mission, a large part of the Manati subzone was flooded and local farmers expressed their anger and their intention to dam canals upstream from Manati in order to protect themselves against water flowing in from neighboring districts. A situation of this kind may lead to conflicts between villages and to social unrest. In the mission's view, the important social benefit of the project and existing crop and livestock production can only be preserved by additional investments in rehabilitating the drainage system.  - 19 - 7.02 Expanding irrigation should not be considered until the overall area is fully developed under rainfed conditions and farmers have gained experience in modern agricultural practices. In addition, the risk of increasing soil salinity and alkalinity is higher under irrigation than under rainfed condi- tions, and the results of irrigated crops have not been conclusive so far in Part I. 7.03 Helping farmers to organize themselves and overcome their reluctance to mutual cooperation appears of utmost importance. Because of the remoteness and difficult living conditions in the project area, only strong and efficient service cooperatives would permit sound solutions to problems of marketing, input supply and mechanization, which today hamper agricultural development and compel settlers to remain subsistence farmers. Livestock production proved to be a suitable alternative and a partial solution to soil deficien- cies and drainage problems. However, milk and cheese production, which proved profitable for some individuals, cannot be developed on a large scale unless an adequate marketing system is established. Similarly, the need for each farmer to go to Barranquilla to purchase inputs and agricultural equipment now constitutes a serious problem for most of the small farmers. VIII. CONCLUSIONS 8.01 With the benefit of hindsight it can be said that the Government's decision to transform the economy of the depressed Atlantico area through agrarian reform and agricultural development was correct. It is unfor- tunate that the project was poorly prepared and appraised, that the feasibil- ity study and the first soil survey were superficial and that information and data provided by the consultants were not sufficiently critically reviewed by the Bank and checked for accuracy, completeness and consistency. The proj- ect, as approved by the Bank, was overly optimistic and complex and did not properly take into account some of the regional constraints such as poor soil quality, difficult drainage conditions and limited availability of people with any farming experience. The case illustrates the difficulties of developing new land: if any land has not been developed by the end of the 20th century, it is likely that severe technical problems exist, and these have to be carefully reviewed before implementation of any project; if the project is designed to settle unemployed and unskilled people, it is unrealistic to assume that agricultural development can be quickly achieved through sophis- ticated irrigation methods and modern cropping techniques. 8.02 When it became evident that the project area would never be trans- formed into a garden producing large amounts of citrus, other fruits and vegetables for international markets, hope remained that a collectivized production system of annual crops for internal markets was still feasible, and about 1,000 settlers were grouped into the Cooperativa Agropecuaria del Caribe. The bankruptcy of CAC and the end of the cooperative system in the project area show that a. settlement project which is not based on a self-help  - 20 - concept and presupposes relinquishing of rights to farm individual plots is likely to fail. Moreover, this unfortunate experience, based on systems modeled after those in other countries and not adapted to local conditions, had and still has an adverse effect on the whole region. 8.03 Compared with appraisal expectations as well as with the new plans which were worked out two years later, and even with the situation prevailing at completion, the project appears a failure in many respects today. Expecta- tions at appraisal were too optimistic. The question is whether or not the area's real potential could have been recognized at that stage and projections made which would have estimated attainable results with a higher degree of accuracy.!' The area is of low potential, its remoteness accentuates marketing and input supply problems, farming is risky, credit is scarce and most of the settlers were and still are unskilled farmers. By turning to livestock production and food crops to sustain their basic needs, settlers proved to be smarter and more flexible than the Bank or project authorities. Had the project limited its objectives to settling farmers, properly resolving the drainage problems and helping farmers develop livestock production and foodcrops and supported them in building an efficient service cooperative system, the project would have been less costly and probably economically more viable. The major flaw of the project was that overall feasibility and project targets were never reconsidered and an entirely different and more modest approach to developing the area was never sought. 8.04 As an investment project and in economic terms, the Atlantico Project must be termed a failure. However, its social impact is high. About 1,250 landless families have been settled in both Phase I and II areas. An orderly agrarian reform program has been carried out, and an acceptable and stable condition of land tenure has been achieved. Agricultural development, though much lower than anticipated, has improved the economy of the area, halted the former complete dependence on a declining fishing industry and provided more secure and adequate incomes for the local population. In 1/ CPS notes that "During the 1960s, adequate field and laboratory tech- niques were available to enable the consultants, the borrower's agency or the Bank to recognize the widescale presence of saline and alkaline soils. Furthermore, engineering knowledge at that time should have ensured the recognition that drainage design criteria were or were not adequate to meet the needs of local soil and rainfall characteristics. It should be noted also that poor drainage, salinity and alkalinity are each correctable deficiencies. The need is for those involved in land suitability classification to recognize their presence and to make realistic estimates of the cost of remedial measures in evaluating land suitability and project economic feasibility. Almost certainly, if proper procedures had been followed and alternatives evaluated, the more appropriate project for land drainage and rainfed production would have been arrived at."  - 21 - addition, about 2,420 small private farmers and squatters have benefitted from project investments and social infrastructure established during project implementation. Despite its shortcomings, the project has had a positive impact on the incomes and living standards of about 26,000 people in the project area. The project fulfilled, in an unforeseen way, the Government's original objectives of improving social and economic conditions for the poor and correcting the inequities of the old land tenure system. However, the project's positive social impact appears jeopardized in the future by the deficiencies of the drainage system and the subsequent risks of worsening floods, which may lead to large scale abandonment of farms. 8.05 The Atlantico project also contributed some important, although not measurable, benefits 1:hrough institution building. A considerable number of professionals were trained through the project, including many who were to become INCORA executives. One major contribution of the project was the build-up of economic and technical capacity in INCORA, which permitted it to handle different kinds of development projects, both in rainfed and irrigated agriculture. - 22 -  - 23 - APPENDIX 1 Page 1 BORROWER'S COMMENTS Translation of cable received from INCORA - Atlantico Regional Director. "I share findings of Impact Evaluation Report Atlantico, Credit 502-CO. Government will send comments through University of Agriculture. Regards, Manuel Linero." - 24 -  - 25 - APPENDIX 1 Page 2 Dr. Fernando Royos P. Private Secretary to Minister of Agriculture Bogota Bogota, May 24, 1982 Rot DT Loan 502/CO Dear Dr. Hoyos: Regarding the final ealuation of the Atlintico ProJect partially .nanced by the loan undler reference, I wish to inform you that this-report accurately reflects vhalc happened during the project's execution and subsequent history, acknowledging 'he mistakes made both by INCORA and by the Bank and the consultants who prepare& the project and advised INCORA on its implementation. Notvithstanding the foregoing, it is satisfactory that now that the evaluation has been completed, it has been concluded that the implementation of the project furthered in various vays the development of the area and the improving of the living standards of the campesino, something that had never been considered becausat execution of the project was alvays compared against the targets set in the initial project, which was very much out of line with the actual conditions prevailing in the area. In this connection, the report submitted by the Bank brings out clearly and unquestionably the situation experienced in that project. It is also quite accurate in its references to institutional circumstances during its execution and to technician training In this field. However, while the conclusions reflect a situation now past concerning wbich no action can bit taken, the chapter relating to future expectations does not provide a detailedl analysis of certain problems that this office considers highly important and which I feel varrant emphasist  - 26 - APPENDIX 1 Page 3 1. For all projects involving partial financing of works, appropriate steps should be taken to ensure availability of counterpart resources at the time of execution and therefore to identify the sources of the resources needed for the operation and maintenance of the works and not, as was the case this time, once it has become necessary to virtually reconstruct a district due to an almost total lack of maintenance resources. 2. The development policy for the area Vas not the most effective in that no thought was given to the interrelationship between the various zones of activity of the Atlantico project (Santa Lucia, Repelon and Manatpi), nor was advantage taken of the experience gained by the agencies during the execution of these three projects, so that in many cases the same mistakes were repeated. 3. It would be desirable for future loan agreements to include a clause that vould facilitate the transferring of resources from one category to another so that the project executing agencies, acting in agreement with Bank supervision missions, can effect such transfers without the cumbersome and time-consuming procedures currently required which in practice make it impossible to amend the original project even if the conditions which were the reason for the loan have changed. I. Finally, I would like to note that the social situation in the area in question is becoming daily more critical, and serious social problems are expected in the short term unless the Government steps in to provide a final solution to the recurrent flooding problems suffered by the area which have reduced the farmers to a subsistence existence, swollen, INCORA's portfolio of bad debts and made it impossible to recover the investment made in these works. Sincerely, /s/ Pedro Jose Ramirez Ramirez General Manager  - 27 - APPENDIX 1 Page 4 National Planning Department UEA - 101 Bogota, May 26, 1982 Mr. Shiv S. Kapur Director OED Dear Mr. Kapur: We have received your "Impact Evaluation Report, Atlantico Irripa3tion Project", which we have carefully reviewed. The report is indubitably a careful analysis of the impact of the project on the region and its inhabitants. We bave nonetheless taken the opportunity to make a few brief comments which we hope will be useful in preparing the final version of the report. Yours, etc. /a/ Alvaro Silva Chief, Agricultural Studies Unit  - 28 - APPENDIX 1 Page 5 COM4ENTS ON "IMPACT EVALUATION REPORT, ATLANTICO IRRIGATION PROJECT" (Credit 502-CO) The evaluation report is quite extensive and includes the background, description, implementation, effects and prospects of the project. On each point some questions are raised which it would be useful to clarify in the report. Firstly, there is in our view excessive emphasis on the historical background, while the economic background is largely ignored. TLere is no mention of the pre-feasibility study, if one was made, or the feasibility study that preceded the granting of the loan. A question immediately arises as to the Bank's reasons and criteria in granting this first loan, especially when it is recalled that, as is shown later, both the feasibility study and the soils study proved to be unreliable and to a large extent were responsible for the failure of the project. The Bank basically limited itself to assessing INCORA's administrative capacity and to proposing a sub-division of the Atlantico project, without bothering to make a critical review of the actual design of the project. As regards the description and implementation of the project, the report is somewhat confusing, though this can be explained by the changes that were made in project definition. According to the report, of the ten or more project objectives (part 1 and part 2) only some were partially achieved. It would be useful to compare these objectives and the results in turn in order to evaluate and explain the causes of the delays and failures to attain targets. In places the report gives considerable space to what it considers as one of the successes of the project, namely agrarian reform. But as regards the drainage system, for example, it merely lists the reasons for deficient performance but does not explain the causes; although it is clear that this was a case of poor design, there is no explanation of why remedial measures were never taken. Likewise, inadequate maintenasce and the lack of equipment were not in themselves an explanation of the  - 29 - APPENDIX 1 Page 6 shortcomings of the draim&ge system, especially if it is remembered that these tvo points (maintenance sa.d equipment) should have been planned as part of project design. Even more surprising is the fact that no explanation is given as to why the pumping stations have not yet been connected to the central electricity network. Other objectives, such as the introduction of marketing facilities and trial farms, are not evaluated at all. As regards agrariEan reform, which is regarded as one of the successes of the project, it is undeniable, as the report indicates, that from a redistribution point of view its implementation was important; however, from an economic point of view -- and the report does not dwell on this -- it would seem to have been a failure, insofar ags the investment did not produce a substantial improvement in family income, and also led to the establishment of a group of settlers on lands that are not very productive, as a result of their small size and poor conditions. Surely an increase of 650% in the population and a reduction of 85% in the average size of farms cannot be regarded as successes if they are not accompanied by the kind of improvement in iamily income and in the living standards of the population that would justify the substantial investment made. The biggest question raised by the entire document is why the project was never redesigned when from the first years it was realized--so it would appear- that construction of a sprinkler irrigation system was not really viable because of poor soil conditions. As regards project prospects, it seems very unlikely that a cooperative can be organized again and it is strange that, after an analysis of the complete failure. of the coopereative system, it is proposed to revitalize it using new features which nonetheless are not specified.  - 30 - APPENDIX 1 Page 7 Republic of Colombia Ministry of Agriculture Bogota, May 24, 1982 Mr. Shiv S. Kapur Director, Operations Evaluation Department World Bank Washington, D.C. Ref: 000670 Subject: Impact Evaluation Report -- Colombia Atlantico Irrigation Project (Loan 502-CO) Dear Mr. Kapur: In reply to your letter of March 17 on the above-mentioned report, I am pleased to inform you that we regard as satisfactory the Bank's evaluation of the activities carried out by HIMAT. As for the work carried out by INCORA, the Bank clearly and unequivocably reflects the situation experienced with the project. It is also quite accurate in its references to institutional circumstances during project implementation and the training of technicians in this area. It would be desirable for future loan agreements to include a clause facilitating the transfer of funds from one category to another so that the project executing agencies, acting in agreement with Bank supervision missions, can effect such transfers without the time-consuming formalities required at present, which in practice make it impossible to amend the original project even if the conditions that were the basis for the loan have changed. Cordially, Isl Lufs Fernando Londono Capurro Minister of Agriculture  -31- ANNEX 1 Table 1 Page 1 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) COST ESTIMATE AT APPRAISAL/a Part I - Irrigation Local Foreign Total (US$'000) 1. Engineering Works a. Main Pumping Stations 150 440 590 b. Canals and Structures 115 230 345 c. Sprinkler System 100 1,510 1,610 d. Drainage and Levy Strengthening 540 350 890 e. Roads and Structures 115 230 345 f. Electrical Generation and Distribution 60 660 720 g. Land Preparation 70 130 200 h. Pilot Farms 275 465 740 i. Engineering and Supervision (including 25% contingency) 1,075 285 1,360 Sub-total 2,500 4,300 6,800 2. Operation and Maintenance Equipment and Spare Parts -- 210 210 3. Spare Parts for Permanent Equipment -- 160 160 Sub-total -- 370 370 Agricultural Investment and Supporting Services 4. Agricultural Machinery and Spare Parts -- 820 820 5. Establishment of Perennial Crops and Tobacco Drying Facilities 220 -- 220 6. Orange Packing-hous, and Equipment (including 25% contingency) 170 200 370 7. Establishment of Field Trial Farms and Nurseries 90 220 310 Sub-total 480 1,240 1,720 Common Services and General Administration 8. Consulting Services 120 1,520 1,640 9. Vehicles -- 150 150 10. Buildings and Utilities 220 30 250 Sub-total 340 1,700 2,040 TOTAL Part I 3,320 7,610 10,930 /a Contingencies of 15% are included in all items except where otherwise specified.  -32 - ANNEX 1 Table 1 Page 2 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) Part II - Flood Protection Local Foreign Total (US$'000) 11. Engineering Works a. Interceptor Channel and Structures 390 310 700 b. Access Roads to Farms and Drainage 370 250 620 c. Contingency 15% 115 85 200 d. Engineering and Supervision (including 25% contingency) 115 65 180 Sub-total 990 710 1,700 12. Operation and Maintenance Equipment -- 50 50 13. Buildings and Utilities 30 10 40 Sub-total 30 60 90 TOTAL Part II 1,020 770 1,790 Other Expenditures 14. Buildings and Utilities Allocable to Future Stages 110 20 130 15. Feasibility Studies for Future Stages of Irrigation (nil Contingency) 200 100 300 16. Training Abroad (nil Contingency) a. Atlantico Project Personnel -- 160 160 b. General Strengthening of INCORA Staff -- 340 340 17. Interest during Construction (6% per annum on IBRD loan) 1,290 -- 1,290 18. Future Additions to Orange Packing-house Equipment to complete the Project (including 25% Contingency) 300 -- 300 19. Land Purchases by INCORA 500 -- 500 Sub-total Other Expenditures 2,400 630 3,020 TOTAL Project Costs 6,740 9,000 15,740  ANNEX 1 Table 2 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) SIZE DISTRIBUTION OF RURAL PROPERTY IN SOUTHERN PART OF ATLANTICO PROVINCE IN 1966 Size of Property Properties Owners Area No. of Properties % No. of Owners % Hectares % Less than 10 ha 267 31.3 290 35.8 1,150 3.8 10-100 ha 462 54.3 426 52.6 11,773 39.0 Larger than 100 ha 123 14.4 94 11.6 17,253 57.2 Total 852 100.0 810 100.0 30,176 100.0 AVERAGE FARM SIZE (in hectares) Size Class Area Per Property Area Per Owner Less than 10 ha 4.3 3.9 10-100 ha 25.5 27.6 Larger than 100 ha 140.2 183.5 Source: Feasibility report on Atlantico No. 3 Project (dated November 1965) prepared for INCORA by Alfonso Olarte C. Ingenieria y construccion (Bogota); and Development and Resources Corporation (New York).  ANNEX 1 Table 3 Page 1 of 2 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) COMPARISON OF ESTIMATED COSTS AND ACTUAL EXPENDITURES Expenditures as of Estimated Costs 06/30/76 Items Loan Local Total Loan Local Total …-----------------------US$ 000-----… Part I - Irrigation 1. Engineering Works a. Main Pumping Stations 440 150 590 415 283 698 b. Canals and Structures 230 115 345 400 529 929 c. Sprinkler Systems 1,510 100 1,610 1,097 - 1,097 d. Drainage and Dikes 350 540 890 921 1,222 2,143 e. Roads and Structures 230 115 345 130 170 300 f. Power Generation and Distribution 660 60 720 672 51 723 g. Land Preparation 130 70 200 64 206 270 h. Pilot Farms 465 275 740 231 549 780 i. Engineering and Supervision 285 1,075 1,360 - 600 600 2. Operation and Maintenance Equipment 210 - 210 1,038 - 1,038 3. Spare Parts for Equipment 160 - 160 - - - 4. Agricultural Machinery 820 - 820 717 - 717 5. Establishing Perennial Crops - 220 220 - 220 220 6. Orange Packing-House and Equipment 200 170 370 - - - 7. Field Trial Farms and Nurseries 220 90 310 - 230 230 8. Consulting Services 1,520 120 1,640 1,692 624 2,316 9. Vehicles 150 - 150 241 - 241 10. Buildings and Utilities 30 220 250 214 284 498 Sub-total, Part I 7,610 3,320 10,930 7,832 4,968 12,800  ANNEX 1 Table 3 Page 2 of 2 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) COMPARISON OF ESTIMATED COSTS AND ACTUAL EXPENDITURES Expenditures as of Estimated Costs 06/30/76 Items Loan Local Total Loa. Local Total _ _____-_ …_____---- US$ 000… - … Part II - Flood Protection 11. Engineering Works a. Interceptor Canals 310 390 700 300 397 697 b. Drainage and Access Roads 250 370 620 230 305 535 c. Contingencies 85 115 200 - - - d. Engineering and Supervision 65 115 180 - 190 190 12. Operation and Maintenance Equipment 50 - 50 33 - 33 13. Building and Utilities 10 30 40 - - - Sub-total, Part II 770 1,020 1,790 563 892 1,455 Other Costs 14. Future Buildi.igs and Utilities 20 110 130 - - - 15. Feasibility Study for Stage II Project 100 200 300 85 215 300 16. Training 500 - 500 178 25 203 17. Interest During Construction - 1,290 1,290 - 2,093 2,093 18. Future Expansion of Orange Packing-House - 300 300 - - 19. Land Purchases - 500 500 - 360 360 Sub-total Other Costs 620 2,400 3,020 263 2,693 2,956 Total Cost of Project 9,000 6,740 15,740 8,658 8,553 17,211  ANNEX 1 Table 4 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) ALLOCATION OF THE PROCEEDS OF THE LOAN (US$ million) Original As Revised As Revised Actual Disbursements Category June 29, 1967 April 18, 1973 Sept. 12, 1974 August 31, 1976 A. Civil Works I. Construction and Installation 2.5 2.5 2.6 2.7 II. Generators, pumps, line equipment, valves and accessories, sprinkler equipment, asbestos-cement pipes, orange packing-house equipment, and other equipment, supplies and materials to be incorporated into civil works 2.2 2.2 2.0 2.1 B. Machinery, Agricultural Items, Consulting Services and Training I. Agricultural machinery and equip- ment, imported planting material, operation and maintenance equip- ment including motor vehicles and spare parts 1.3 2.4 2.4 1.9 II. Consulting services and training 1.9 1.9 2.0 2.0 C. Unallocated 1.1 _ _ _ Total 9.0 9.0 9.0 8.7  - 37- ANNEX 2 Table 1 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) DISTRIBUTION OF FARM FAMILIES IN THE PROJECT AREA (1981) (Stage I and II Areas) Area Settlers Private Farmers* Total Area (ha), Families Area (ha) Families Area (ha) Families Manati 7,667 735 1,680 1,156 9,347 1,891 Candeleria 1,462 147 407 507 1,869 654 Campo de la Cruz 2,528 190 3,513 714 6,041 904 Santa Lucia 1,738 180 344 22 2,082 202 Total 13,395 1,252 5,944 2,399 19,339 3,651 * Including squatters.  - 38 - ANNEX 2 Table 2 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) LAND ACQUISITION BY INCORA (1965-1981) (Stage I and II Areas) Area Purchase Expropriation Permutation Total Units Ha Units Ha Units Ha Units Ha Campo de la Cruz 28 841 1 175 - - 29 1,016 Candeleria 14 669 1 800 - - 15 1,469 Manati 29 3,695 7 1,521 - - 36 5,216 Juan 4 660 2 127 - - 6 787 Santa Lucia 18 822 - - 1 76 19 898 Total 93 6,687 11 2,623 1 76 105 9,386  - 39 - ANNEX 2 Table 3 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) DISTRIBUTION BY YEAR OF LAND ACQUISITION BY INCORA (in hectares) Campo de Santa Year la Cruz Candeleria Manati Juan Lucia Total 1965 646 - 285 - - 931 1966 - 159 591 106 96 952 1967 - 36 1,107 - 65 1,208 1968 - 40 459 554 - 1,053 1969 17 - 374 - 161 552 1970 - 114 762 - 372 1,248 1971 100 129 340 - 166 735 1972 - 872 - - - 872 1973 - - 24 - - 24 1974 - - - - 38 38 1975 - - 591 - - 591 1976 78 - 244 - - 322 1977 - - 182 - - 182 1978 175 80 - 41 - 296 1979 - 39 129 - - 168 1980 - 21 86 - 107 1981 - - 107 - - 107 Total 1,016 1,469 5,216 787 898 9,386  ANNEX 2 COLOMBIA: ATLANTICO IRRIGATION PROJECT Table 4 (Loan 502-CO) IMPACT EVALUATION REPORT Evolution of Cash Crops in the Settlement Area 7.000 6.00 0 5.OOC 4.000 0 2.000 1.000 0- 1970 1971 1972 1973 1974 1975 1976 1977 1978 1.979 YEARS  -A - ANNEX 2 Table 5 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) DISTRIBUTION OF LAND USE BY DISTRICT IN STAGE I AND II AREAS (1980) (Including double-cropping - in hectares) Campo de Santa Land Use/District Manati la Cruz Candeleria Lucia Total Percent Irrigation Crops - - - 460 460 (49) Pasture - - - 480 480 (51) Total - - - 940 940 100 Rainfed Crops 3,379 1,789 111 185 5,464 (30) Pasture 5,968 4,252 1,759 944 12,923 (70) Total 9,347 6,041 1,870 1,129 18,387 100  - 42 - ANNEX 2 Table 6 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) LAND USE BY SETTLERS IN STAGE I AND II AREAS (1980) (In hectares) Campo de Santa Land Use/District Manati la Cruz Candeleria Lucia Total Crops Irrigated Area Food Crops 99 99 Cash Crops - - - 293 293 Subtotal - - - 392 392 Rainfed Area Food Crops 2,133 379 54 141 2,707 Cash Crops 619 - - 44 663 Subtotal 2,752 379 54 185 3,370 Pastures Irrigated Area Natural Pasture - - - 47 47 Improved Pasture - - - 170 170 Subtotal - - - 217 217 Rainfed Natural Pasture 2,172 617 1,003 505 4,297 Improved Pasture 2,385 1,507 367 405 4,664 Undeveloped Land 359 24 38 34 455 Subtotal 4,916 2,148 1,408 944 9,416 Total 7,668 2,527 1,462 1,738 13,395  -43- ANNEX 2 Table 7 IIIPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) LAND USE BY PRIVATE FARMERS UNDER RAINFED CONDITIONS IN STAGE I AND II AREAS (1980) 1975 1976 1977 1978 1979 1980 Land Use Rainfed crops (ha) 1,455 1,963 1,771 2,087 2,586 2,094 Pasture (ha) 4,145 3,637 3,829 3,513 3,014 3,506 Total (ha) 5,600 5,600 5,600 5,600 5,600 5,600  - 44,- - 44 ANNEX 2 Table 8 IMPACT EVALUATION REPORT COLOMBIA: ATLANTICO IRRIGATION PROJECT (LOAN 502-CO) KEY INDICATORS SUMMARY OF ACTUAL AND PROJECTED CULTIVATED AREA, NET VALUE OF PRODUCTION, AND NET INCOME PER FAMILY IN STAGE I AREA Area Cultivated (Including Double-cropped) (ha) 1982 1972 1975 1980 1980 (Projected at (Projected at Com- Appraisal) (Actual) (Actual) pletion in 1975 (Actual) Part I Field Crops 5,695 649 650 4,000 460 Livestock - 183 2,600 400 2,840/-a Subtotal 5,695 832 3,250 4,400 3,300 Part II Field Crops 6,400 3,393 2,188 3,000 1,800 Livestock - 3,700 3,812 4,000 4,200 Subtotal 6,400 7,093 6,000 7,000 6,000 Net Value of Production (US$ 1980) Part I 6,090 114 680 1,300 388 Part II 2,810 661 334 900 623 Total 8,900 775 1,014 2,200 1,011 Net Income - Per Family Part I 4,300-6,000 - 2,300 4,300 2,000 Part II 2,050 - 960 1,650 1,100 /a Of which about 480 ha is irrigated pasture.  1i8,L 1016jR 7d8 7t6i 74 72 JULY 1979 e on eo SCo OUTH AMIC 2 rlIbb St. M.o t / |., MAGDALENA _ S I g* - t ,# ~~~M.M i. UC i 0 X R D O a}b CORDO A S 2 > r E t - - = V e E N ~~~~~~~~E Z U E L A C.. QABO 5AGOLN - CV'~ (J 8 C H O C ° _ B / oUP \B OYASA C)I z RISARALDAl u RA CALDAS t qX olcnlZles CUNDINAMARCA) > CYoSAN REAIh f LQ 0 C / 4 <°UINOI_ VI C H ADAc CMI OLI/AMACA1A0 V A LIL E I T O LI M A ' (UEJA 6 / ) D - PO IGTI 0HA r) 0 / O { / (SUCPAPAZ/ LJL.~~~~~~~~~- RI)l' Dr CA CA rt V r~ A L/ UAI _ I N 0 - IA ANDINA _ '7 r COSrA j<A CAUCb-: .HUILA G U A I N IA o I P .p , NA RIN0 1AMAOAA I. .,r , C_',<_ _=/0ee. _ L I. _ _ 0~~~~~~~~~~~~~~~~~~~0N --/ e-^ .. D~~~-- , / _ncI C' -. A MA C, N  IBRD 3245R3 C CDLCD VAB IA 75. 8. T. Carrogen. 75- %oSsobnolargo To4rrrii 5 ~JNO1 SECOND ATLANTI CO DEVELOPMENT PROJECT 50 -0132' ~EXISTING FEATURES 10'32' Roads so tr embankmsen ,oiet L-1Guajro darm and reservoir - <aood de PabO. Coldel Dique -- Lre6 Streoms/ PROJECT FEATURES ~.l• 6es FlGaao Main roads Secondary roads A'eA,rvo, Roads -tn embankm,ent Auxiliary embankrn,ents OULT0NT,`PNAPL1 - Main drainage canaisSRxU EEVIRdl,o Lateral drainage canoaisNT ILPOET rR I -a--Main irrigation canals 01 Pumspingplants Fi.an,ced under laxe 502-CODa Project bocndary ForIoan 847-CO -10-24' L- ,i-- - -1%4- Coxxbbeanr Sea 3~~~~~~~~~~I- / .I- Aren.1 ~~~~~~~~~Q~~. ~ orrxrr 5 ~ j' ~-- ~ irisrroru,aitv'sdura ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~Conjo- \. VENEZUELA a ~~~~~~~~~~~~~~LOAN 502C0 U -. ~~~~~~~~~~~~~~~~~~~Part I Irrigated Area O ~~~~~~~~~~~~~~~~~~~Drain I U ~~Bugutd .~~~~ Boquitas P.moinp,g Plant " Son Pedriso Pumping PlantStL Irigation systemPat) COLOMBIA ~~~~~~~~~~~~~~~Irrrigated area Part II Dry Farming Area Flood Interceptor Works Drains M and I Drains IVand V auth laterals Roads K ~~~~~~~LOAN B49-CO ECUADOR r~Dry Farming A rea4 ~~~~~Balance of drainage system -~~~~~~~~~~~~~ ir,''supD9' K, no ~ FRU ~ /Project B R A Z I L ~~~~~Ba.lance of roads lands rr PIN PLNT,0a- sie- 45GENERATOR 0 1 2 3 4 5 0 0~ 200 300 400 500 & m~n., t -,an''Oedrit s0o 300*0~.,,,,-- 200 ~ M,Ie,I ~ ____________________~ ~ ~ ~ ~ ~~758'75 L~~~~~~~~25 To Mredj. -i - :0Pent 74-52