Ecuador - Livestock Development Projects

CIRCULATING COPY CONFIDENTIAL TO BE RETURNED TO REPORTS DESK Report No 892 INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT INTERNATIONAL DEVELOPMENT ASSOCIATION Project Performance Audit Report ECUADOR FIRST AND SECOND LIVESTOCK DEVELOPMENT PROJECTS (Loan 501-EC and Credit 173-EC) October 21, 1975 Operations Evaluation Department  PROJECT DATA Loan 501-EC Loan Amount US$4.0 million Amount Disbursed US$3.7 million Amount Cancelled US$0.3 million Date of Agreement June 19, 1967 Date of Effectiveness December 4, 1967 Original Closing Date December 1972 Final Closing Date April 1974 First Bank Mission October 1965 Credit 173-EC Credit Amount US$1.5 million Amount Disbursed US$1.5 million Date of Agreement January 20, 1970 Date of Effectiveness June 29, 1970 Original Closing Date December 1973 Final Closing Date February 1974 Repeater Project Credit 222-EC Credit Amount US$10.0 million Date of Agreement December 10, 1970 Currency Equivalents (1967-1974) US$1 = sucres 18.18 until September 1970 US$1 = sucres 25.25 thereafter  PREFACE This report presents the results of an audit of the two projects supported by Loan 501-EC, signed in June 1967 and closed in April 1974, and Credit 173-EC, signed in January 1970 and closed in February 1974. The audit was based on a review of the projects' Appraisal Reports, supervision reports and other material in Bank files; on consultation with Bank personnel; on discussions in Guayaquil and Quito with officials of the National Development Bank, COFIEC and the Ministry of Agriculture; on conversations with personnel of the National Agricultural Research Institute at Pichilinque; and on visits to a number of participating ranches. The excellent assistance provided by Ecuadorian institutions in the preparation of this report, including the special farm inter- views conducted by the Project Staff in Ecuador, is gratefully acknowledged.  TABLE OF CONTENTS Page Summary I. Introduction 1 II. Background of the Two Projects 1 III. Implementation of the Lending Program 5 IV. Objectives and Impact of the Program 13 A. Objectives 13 B. Summary of Results 16 C. Data 16 D. Investment Costs 17 E. Investment Productivity 22 F. Technical Assistance 24 G. Prices and Marketing Arrangements 26 H. Impact on Aggregate Beef Production 29 I. Rate of Return Analysis 29 i. Financial Rate of Return 29 ii. Economic Rate of Return 31 V. Other Issues 33 A. Seed Multiplication and Development at Pichilinque 33 B. Production Research, Training Centers, and 34 Monitoring Systems C. Agricultural Survey 34 D. Substitution of Bank Funds for Other 35 Available Funds E. Distributive Equity 36 VI. Bank Performance 38 A. Appraisal 38 B. Supervision 39 VII. Conclusions 40 Appendix Tables 42 Map LIST OF TABLES Text Page 1. Contributions to Project Finance 3 2. Degree of Progress on Project Farms Relative to Targets of 23 Development Plans 3. Rates of Return to First and Second Ecuador Livestock Projects 32 Appendix Al Chronological Record of Subloans Approved for Projects 501-EC aid 173-EC. 43 A2 Size Distribution of Subloans for Projects 501-EC and 173-EC. 44 A3 Distribution of Subloans, by Province and Percent of Cattle Herd Affected(1968 Survey) Projects 501-EC and 173-EC. 45 A4 Temporal Distribution of Subloans Approved, by Province, Projects 501-EC and 173-EC 46 A5 Role of Participating Banks, Projects 501-EC and 173-EC. 47 A6 Comparison of Unit Investment Costs as Forecast and as Reported in Farm Development Plans, Projects 501-EC and 173-EC. 48 A7 Comparison of Physical Investments Per Farm as Forecast and As Reported in Farm Development Plans, Projects 501-EC and 173-EC 49 A8 Comparison of Average Farm Size as Forecast and as Reported in Farm Development Plans, Projects 501-EC and 173-EC. 49 A9 Data from Egas' Study of Project Farms in Santo Domingo - Quevedo Area, Referring to Investment and Costs, Projects 501-EC andl73-EC 50 A10 Data on Beef Prices, Investment Costs, Operating Costs and Inflation in Ecuador. 51 SUMMARY In 1965, the Bank began to plan with the Government of Ecuador a livestock development program in Ecuador's Coastal region. Preparation of the project went smoothly, and much of the design of the project was carried out by a special commission in the National Planning Office, assisted by two Bank missions. The first livestock loan (501-EC) was signed in June 1967, providing US$4.0 million, to help finance on-ranch investments such as land clearance, pasture renovation, fencing, water supplies, yards, dips, machinery, and improved breeding stock, both domestic and imported. The program was administered by a specially created autonomous Project Com- mission, composed of representatives from both the public and private sectors, and operated under the technical authority of a foreign Program Director acceptable to the Bank. He selected and directly supervised the loan technicians, who were however employed by the participating banks. The participating banks included the National Development Bank, a public institution and the largest single source of agricultural credit in Ecuador, and COFIEC, a private finance development corporation. The Coastal zone had emerged as Ecuador's principal cattle pro- ducing region only since World War II. Much unutilized land remained which was considered particularly suitable for cattle ranching, and on the great majority of existing ranches, operating conditions were still primitive and their technical efficiency was low. Given the favorable natural conditions, it was believed that new investments, combined with management improvements, would be highly profitable. The first livestock project was designed to operate on a pilot basis and was expected to provide loans to roughly 240 farmers over a period of two years. The sub-loans were expected to average US$27,000, financing 80 percent of farm development costs. Seventy-five percent of the sub-loans were to come from the Bank's funds, and 25 percent from the funds of the participating banks. The farmer was expected to contribute 20 percent of the original investment costs, plus additional amounts as he retained heifers to further build up the breeding herd. Implementation of the project went smoothly. The sub-loans were made more slowly than predicted, but were larger in size, so much so that the funds available were fully committed by July 30, 1969, some six months ahead of schedule. Upon request of the Government, the Bank sent an appraisal mission for a second investment project. This mission concluded that the first project had achieved its major goals and recommended a second interim loan to maintain the momentum of the project until a third, more comprehensive project could be designed and approved. The Bank agreed, and an IDA credit (173-EC) of US$1.5 million was signed in January 1970. A third tranche, an IDA Credit for US$10 million (222-EC), was signed in April 1971. The two loans provided a total of 232 sub-loans to 199 separate borrowers over a four-year period. The average value of the sub-loans, which were made throughout the Coastal zone, was nearly 80 percent more than expected, principally because of increases in unit investment costs. The ranches in- volved, in terms of pasture area and herd size, both before and after develop- ment, were slightly smaller than predicted. - ii - The Project Director and Staff appear to have performed their jobs well. The ranches selected were generally progressive and receptive to the new methods being introduced. The farm development plans were designed to fit the individual farmers' needs. The composition of farm investments, in the aggregate, corresponded closely with those predicted in the Appraisal Reports - approximately 33 percent for pasture creation and renovation, 32 percent for breeding animals, 25 percent for ranch infrastructure, and 8 percent for machinery and equipment. The projects have achieved a notable improvement in the technical parameters on participating farms, increasing, for example, the effective weaning rate from a pre-development level of approximately 45 percent to approximately 62 percent. This is somewhat shy of appraisal report pre- dictions, and considerable gains must still be achieved to meet the ultimate goal of 75 percent, but progress has been unusually good. These gains seem to be due to efforts by the Project Staff to provide continuing technical assistance to borrowers. The first Appraisal Report indicated that the loan technicians would screen borrowers' design, farm development plans, and supervise actual loan investments, but that technical assistance would be provided by Ministry of Agriculture extensionists. The latter have, until very recently, been less available than expected and, facilitated by the relatively small number of sub-borrowers, the Project Staff has filled the gap. The importance of having adequate technical assistance is made clear in this project. The principal deviations in the project from the Appraisal Report predictions are: 1) higher unit investment costs, 2) lower beef prices relative to input costs, and 3) the commencement of rapid inflation. As mentioned, the amount of government-provided technical assistance has also been less than reported, and the beef marketing institutions, particularly slaughterhouse facilities, have been less adequate. Unit investment costs were higher because the Appraisal Report underestimated the cost of pasture creation, partially caused by unforeseen increases in the cost of labor, and did not sufficiently anticipate the effect which the project itself would have on the prices of breeding animals. Relative beef prices have been lower because the Government has not, as previously agreed, allowed cattle and beef prices to fully adjust to export market levels, and because agri- cultural costs have increased markedly in response to shifting sectoral demands. Jointly the two projects have resulted in a 6 percent increase in the Ecuadorian beef cattle herd and also caused a significant improvement in the operating efficiency of a larger proportion of the herd, approximately 17 percent. These effects, when combined, produced an increase in total beef production over pre-development levels of roughly 10 percent, worth approxi- mately US$4.0 million per year. Although it was expected that one-half the predicted increase in output would be exported, providing needed foreign exchange, it appears that a much lower percentage was reached. The discovery of petroleum, which reduced the need for beef exports, was a factor in divert- ing beef production to domestic consumers. The economic rate of return, - iii - which is estimated to be substantially lower than predicted, about 12 percent versus 27 percent, is still acceptable. The financial rate of return would have been lower than predicted as well were it not for the rapid inflation which commenced two years ago, eroding the real value of borrowers' repay- ments and increasing their financial return. The financial return is now approximately equal to the 27 percent predicted in the Appraisal Report. The project was designed to assist commercial beef producers, ex- pecting that this approach would provide the maximum impact on production. Income distribution and employment were not objectives of the project and it is improper to give undue attention to criteria which the project was never expected to satisfy. Nonetheless, we believe several observations are appropriate. The project has had a regressive impact on income distribution by providing development funds only to very large producers, who, after the commencement of rapid inflation, have received a substantial subsidy. The project seems to have been neutral in terms of providing employment, and with respect to the implementation of land reform in Ecuador. The technology proposed to farmers was capital intensive and limited project participation to large producers. Had the project placed greater emphasis on research, it appears that less capital intensive technologies might have been developed. These technologies could have placed even greater emphasis on management improvements. They would have in- creased rates of return to large producers and also been accessible to the more numerous medium-sized producers. Research on the production and marketing of milk from dual purpose herds in the Coastal region was also desirable. The project had a positive impact on the National Development Bank, introducing more effective procedures for providing long-term supervised credit, stimulating other management improvements, and, in the third loan, assisting the modernization of its internal accounting operations. The project also provided on-the-job training for about 25 loan technicians, a number of whom have now assumed more responsible positions within the NDB. The project provided for a seed multiplication and development program at the National Institute of Agricultural Research's (INIAP) station at Pichilinque, which, although getting underway more slowly than planned, is now progressing well, and the third project increased this support to allow the contracting of two foreign technical advisers in pasture production and animal management, the commencing of extended research programs in both areas, and the building of a training cluster for the diffusion of new manage- ment techniques to farmers, farm administrators and foremen. These develop- ments have been positive, and provide the basis for continuing improvements in the beef cattle sector. It would have been wise, however, to implement an expanded research program at Pichilinque, and to institute closer links between the research station and participating farms, at the commencement of the first loan, in order to utilize farmer experience to suggest continu- ing improvements in the project's design. No farm monitoring system has yet been devised, although the first livestock project was explicitly designated as a pilot scheme. Given the overall efficiency of the work of the Project Director and Staff, this gap seems all the more striking. No funds were provided in the project, however, specifically to support such a system. - iv - The Bank has performed its functions reasonably well. The first Appraisal Report designed a project whose basic character has been pre- served, largely because it recommended the establishment of a project unit which worked independently, and successfully, to implement the project The first Appraisal Report mentioned most of the factors which turned out to cause the project to fall short of appraisal predictions, but under- estimated their full and combined effect. The second Appraisal Report failed to confront the problems encountered in the implementation of the first project, in particular, the higher investment costs and lower beef prices. It did this largely, it appears, because their implications were not fully understood - and again the importance of a monitoring system is evident - and because the second loan was viewed as an interim project to the more carefully designed third project. The third Appraisal Report, which expanded assistance to include dairy production in the Sierra region as well as cattle production in the Coastal zone, also placed too little emphasis on costs and prices, and it appears that, were it not for inflation, individual sub-borrowers might be in a difficult financial position. Partici- pating bank collections on both interest and principal have been acceptable, although there is some variation in the performance of different branch banks in this respect, but until recently there were reports that some farmers had been selling breeding animals in order to make payments due. Bank supervision has been competent, but has focused principally on disbursements and collections and did not push as hard as it should have to encourage the collection of better data on technical and economic develop- ment at the farm level, for both continuing and new sub-borrowers. The Bank encouraged substantial Ecuadorian participation in the planning of the livestock projects from the very start and this participation matured rapidly. The foreign Project Director, who was much respected in Ecuador, was replaced by an Ecuadorian shortly after the third project was initiated, and this project has been successfully administered since. The Project Staff continues to enjoy an unusually high morale and sense of pur- pose although it appears that if a fourth livestock loan is not made the Government may allow the project to lapse. The audit suggests that livestock projects can have a significant impact on farm technical parameters, but emphasizes the need for on-ranch management and intensive technical assistance. Questions about replicability and priority arise, however, since the favorable combination of producer commitment and intensive service will not obtain in programs directed at large numbers of ranchers, and since lower cost technologies offering comparable rates of return can apparently be developed for large farmers as well as smaller ones. The increased emphasis on research in the third project is therefore thoroughly consistent with the audit findings. Project Performance Audit Report ECUADOR FIRST AND SECOND LIVESTOCK DEVELOPMENT PROJECTS (Loan 501-EC and Credit 173-EC) I. INTRODUCTION 1.01 The Bank Group has granted three loans to Ecuador for livestock development totalling US$15.5 million equivalent. The first of these loans, for US$4.0 million equivalent (Loan 501-EC, June 1967), helped finance a pilot project for beef cattle development in Ecuador's Coastal region. This loan was followed by an IDA credit, for US$1.5 million equivalent (Credit 173-EC, January 1970), after funds from the first project were exhausted more rapidly than expected. The credit sought to extend the financial assistance for an interim period until a more comprehensive third project could be designed and approved. The third project, also financed by an IDA credit, for US$10 million equivalent (Credit 222-EC, April 1971), continued the beef cattle project in the Coastal zone and added support for: 1) dairy development in the Sierra zone, 2) the expan- sion of project-oriented research and training in animal production at the National Institute of Agricultural Research (INIAP), and 3) the establish- ment of a grass and legume seed multiplication and certification program. This audit covers only the first two projects (501-EC and 173-EC), but the analysis is extended to include the third project (222-EC) whenever necessary to shed light on the development or impact of the first two projects. II. BACKGROUND OF THE TWO PROJECTS 2.01 In mid-1965 the Junta Nacional de Planificacion y Coordinacion Economica (National Planning Office) requested that the Bank review the livestock production and slaughtering industries to determine if a project suitable for Bank financing could be identified. The first Bank missions visited Ecuador in October 1965 and a project was identified in January 1966. The Bank urged the National Planning Office to prepare a detailed proposal, and several subsequent Bank missions assisted a special commis- sion at the Planning Office in elaborating this proposal. 2.02 It is worth pointing out that although the livestock development project initiated in Ecuador was similar in concept and execution to projects initiated during the 1960s in a number of other Latin American countries, the Ecuador project was one of the first in which the Bank encouraged consid- erable local involvement in its formulation and preparation. The Bank hoped that by making an Ecuadorian group responsible for project preparation, despite their previous lack of experience and the consequent need for tech- nical and management assistance from the Bank, the Ecuadorian staff would make a larger contribution to the project during its implementation and also have greater capacity for planning similar future projects. (This approach - 2 - seems to have been successful: three members of the Ecuadorian project commission subsequently became members of the Project Staff, and two of these have served competently as Project Directors during the third live- stock project. The Ecuadorian staff also played a major role in the pre- paration of the third livestock project.) 2.03 The Government of Ecuador submitted its detailed proposal to the Bank for consideration in July 1966. The initial project was to be limited to the Coastal zone. A Bank appraisal mission visited Ecuador soon after, in September 1966, and, following discussions with the Government of Ecuador on several important issues, the Appraisal Report was completed in April 1967. The loan agreement was signed in June 1967, it became effective in December 1967, and the first loans were approved and disbursed in January 1968. It does not appear that the time period required to carry the project from the original request to actual implementation could have been significantly reduced. 2.04 The principal issues considered by the Bank were: 1) the capacity of the cattle sector to absorb the amount of supervised credit requested, 2) the financing plan and on-lending terms, 3) the inclusion of participating banks for on-lending to ranchers, 4) the choice of administrative arrangements for the project, and 5) the existence of restrictions on the marketing of beef in Ecuador. The five issues cited above were resolved as follows: Item 1) 2.05 Capacity. The Bank was uncertain that sufficient qualified farmers could be found, and concerned that insufficient female breeding animals were available to permit a rapid buildup in project borrower's herds without an excessive increase in the price of such animals. As a result, although Ecuador had requested assistance of $6 million to finance the development of 320 farms over a three-year period, the credit approved provided $4 million to finance 240 farms over a two-year period. The Bank opted for the same rate of financing as requested, but preferred to postpone a larger program until greater experience had been obtained. Ecuador also agreed to undertake a sample survey of the livestock industry to provide more detailed information on the livestock subsector before any further Bank group financing could be considered. Funds to assist in making this survey were provided in the loan. The first Appraisal Report stated explicitly that the project would be under- taken as a pilot project, thereby accepting the possibility of some error in project design, as a necessary cost of proceeding more rapidly and of obtaining more information which would be useful in the design of future projects, if any. Item 2) 2.06 Credit terms. The total project was designed to cost US$6.8 million equivalent, of which US$6.4 million would be for farm development and US$0.4 million for technical services. The latter component included funds for the Project Director's salary and a small amount for INIAP's research station at Pichilinque. The cost of the project was to be distributed among the Bank, participating banks, and prospective lenders as follows: Table 1 Contributions to Project Finance IBRD National Farmers' Loan Banks Contribution Total - US$'000 Equivalent - Farm Development 3,840 1,280 1,280 6,400 (Distribution) (60%) (20%) (20%) (100%) Technical Services 160 240 - 400 Total 4,000 1,520 1,280 6,800 (Distribution) (59%) (22%) (19%) (100%) 2.07 The Bank loan was expected to cover 59 percent of the total project costs, with the balance provided by participating banks (22 percent) and the farmer loan recipients (19 percent). For farm development the owner would provide in cash, labor, and materials about 20 percent of the estimated costs. The other 80 percent of the cost would be advanced to the farmer as a loan from participating banks at 10 percent interest per annum for a term of 8 to 12 years, including a grace period of 3 to 5 years. The participating banks would obtain 75 percent of their relevant funds (60 percent of.the total farm development cost) from the proposed Bank loan, rediscounted through the Central Bank, and would contribute the balance of 25 percent (20 percent of total cost) from their own resources. Farm loans were to meet the credit and security policies of the participating banks, and these banks also agreed to provide any short-term credit required by project borrowers for the suc- cessful implementation of the long-term development project. 2.08 The credit terms established were intended to provide sufficient income to the participating banks to permit them to employ qualified loan technicians who, while employees of the participating banks, would work under ithe direction of the Project Director in planning and supervising project loans. 2.09 The Bank loan of US$4.0 million was made to the Government of Ecuador, which carried the exchange risk, and proceeds of the loan for expenditures in sucres were made available to the participating banks through the Central Bank, acting as the Government's fiscal agent. The participating banks made repayments to the Central Bank in sucres and the Central Bank in turn serviced, on behalf of the Government, the debt to the Bank in foreign currency. To allow for a partial rollover of funds which would permit additional lending to the livestock subsector, the terms of the Bank loan to the Government were more lenient than those to individual farmers; the loan was repayable in 18 years, including 6 years of grace. -4- Item 3) 2.10 Participating Banks. It was originally hoped that a number of private banks would join the project. Although the Bank established credit terms which were calculated to provide a return to participating banks of approximately 22 percent on invested capital, no private bank participated in the first two projects (501-EC and 173-EC), principally because their alternative short-term lending opportunities in the commercial sector offered even higher returns. (Central Bank pressure, operating through requirements that private banks expand the proportion of their loan portfolio allocated to agricultural credit, induced a number of private banks to participate in the third livestock loan, but their proportion of total loan activity is still small.) The participating banks during the first two projects were restricted to the National Development Bank (NDB), a public development bank, and COFIEC, a private finance development corporation. The NDB was already the largest single source of credit to the agricultural sector, but COFIEC essentially initiated its livestock lending activities with the Bank project. Item 4) 2.11 Organization of the Livestock Development Project. The Bank per- suaded the Government of Ecuador to create an independent Project Commission to administer the project. The Project Commission was composed of repre- sentatives from the National Planning Office, the Ministry of Agriculture, the Central Bank, the participating banks, and the Coastal Cattlemen's Association. In establishing the Project Commission, the Bank and the Gov- ernment sought to incorporate the assistance and support of leading individuals in both the private and the public sectors, and also to insulate the project from political pressures which the Bank feared might arise if the project were placed directly within an existing government institution. A foreign Project Director, satisfactory to the Bank, was to be appointed and be respon- sible to the Project Commission, but he was to have the technical responsi- bility for the implementation of the project. This responsibility included the approval of the technical staff, consisting of the loan technicians who would work for the participating banks under his direct supervision, the public promotion of the project, the preparation of farmers' development plans, the recommending of these plans to the participating banks for lending, and the supervision of the farm investment projects ultimately approved. Item 5) 2.12 Market Restrictions. The Bank was concerned that existing restric- tions on the marketing of beef in Ecuador would reduce farmers' incentives to develop livestock production along the lines proposed in the project. Accordingly, the Bank sought to obtain a guarantee from the Government of Ecuador that these restrictions would be removed. The Bank wanted to ensure the free movement of cattle and beef to domestic and export markets, which would allow the prices for these commodities to adjust to export market - 5 - levels. The Government refused to be bound in this way, but agreed to provide a letter, separate from the loan agreement itself, stating its intent to remove these restrictions. The letter, which did not give the Bank the right to suspend disbursements in the event of non-compliance, was accepted by the Bank as a second-best solution in order to proceed with the loan project. III. INPLEMENTATION OF THE LENDING PROGRAM 3.01 After approval of the loan in June 1967, several months were re- quired to put the project into operation. One of the members of the first appraisal mission, an Australian experienced in tropical livestock production, was appointed Project Director and he took residence in Ecuador in November 1967. He began his work by selecting and training a staff, consisting of four loan technicians, three to work with the NDB and one with COFIEC. The individuals selected for the Project Staff were university graduates having considerable experience with livestock and all had worked in government or private institutions in the agricultural sector. They were among the most highly qualified Ecuadorians available. The Project Director also rapidly established a set of operational procedures for processing loan applications, developed good working relations with the Project Commission and with the manager of the Guayaquil branch of the NDB (where a significant proportion of the early loans were channeled), and began to familiarize himself with the community of commercial cattle ranchers in the Coastal zone (see map). The first loans were made in January 1968. 3.02 During the first months of the project's operation, Project Staff openly solicited rancher participation by visits to individual farms through- out the Coastal zone and through use of the news media. Following an ex- pression of interest in a loan by a farmer, one of the loan technicans, and during the early stages of the project the Project Director himself, visited the farm to determine the feasibility of a loan. Individual attention of this sort was possible because the number of planned participants - 240 -- was small. An attempt was made to elaborate development plans which, although following a generally similar pattern, were carefully structured to the specific needs of the individual farm. The Project's records indicate that individual projects vary considerably from one farm to another, and the specific invest- ments are generally carefully identified and justified. It is surprising, given the variation in individual farm plans, that the composition of on-ranch investment turned out to be so close to appraisal estimates when taken in the aggregate. This data is given in table a6. The evidence does suggest, however, that farmers capitalized their operations too heavily and, further, that the unit cost of investments exceeded expectations by a considerable margin. More will be said about this issue shortly. - 6 - 3.03 In this early period, farm development plans were sometimes elaborated by the technicians and recommended by the Project Director to the banks for approval before a detailed check on the rancher's credit history, or on the validity of his land title, had been made. This process was changed after the NDB declined to make a number of the loans which had been recommended, having determined either that the farmer's credit position was unacceptable or that his title papers were invalid. The Project Director then required that ranchers obtain a credit and land title clearance before a technical evaluation would be undertaken. Valid title papers could not be obtained for some farms, and it accordingly proved impossible to make a number of loans which were expected to be technically and economically attractive. The Project did cause the clarification and updating of some land titles, which may be considered a small but positive effect of the Project. 3.04 Before a loan was recommended, prospective borrowers also were required to provide assurances that their farms would be adequately managed, either by agreeing to move to the farm themselves or by employing a qualified administrator. We believe this requirement was important. Absentee ownership of farms was very common in the Coastal zone and it was necessary to ensure continuous competent supervision of the new production techniques which the Project was introducing if these were to be successful. Some ranchers ulti- mately employed administrators who were not as qualified as the Project Staff wanted, but the overall reaction to this requirement seems to have been positive. To encourage farmer residence, funds were sometimes included in the loans to build a house on the farm for the owner and his family. This was most frequently done for smaller project borrowers, and care was taken to ensure that this provision was not abused. As a percentage of total costs, this category did not exceed 2 percent. 3.05 The project encountered some early difficulty in working with the NDB. The NDB had been plagued with management difficulties throughout most of its history, and these difficulties were one of the reasons the first livestock project was established as an autonomous institution operating under its own Project Commission. Until 1964, for example, the NDB pro- vincial offices were practically autonomous, each having its own board of directors and separate capital, and following its own operating policies. Political influence was common, both as to the appointment of management and in lending operations. This situation began to change in 1964, when the Military Junta then in power put into effect a Decree centralizing bank policy and administrative supervision under one board of directors, consolidating the system's financial resources and seeking to force the payment of all debts in arrears. Only gradully, however, did these and continuing similar efforts to improve management take effect. At the time the first livestock project was initiated, the NDB also had little significant experience with supervised credits in which loans were approved on the basis of recommendations from bank technicians, as opposed to the personal judgment of the branch manager. Many branch managers saw the Bank's livestock project, which required ap- praisal by a loan technician and then formal recommendation by the Project Director, as threatening their prerogatives. Accordingly, these managers did not actively assist the project to get underway. - 7 - 3.06 The manager of the Guayaquil branch was much more helpful. He was sympathetic to the overall goals of the livestock project, to the specific supervised credit system being implemented, and he had a personal reputation for getting borrowers to make interest and principal payments promptly. Fortunately, the project had its offices in this branch's building and soon established good working relationships with its personnel.. As a result of these advantages, a high proportion of the project's first loans were channeled through the Guayaquil branch. The formal justification for placing these loans in this branch was that the sub-borrower's principal residence was in Guayaquil, even though his farm was located in another area. This circumventing of other branches allowed the project to proceed more rapidly, and placed pressure on managers in other branches to work more actively with the project. As these other branches became more helpful, the Project Director yielded to requests from NDB headquarters in Quito to channel loans through the provincial branch closest to the farm where the actual development was to take place. The final tabulation of loans, by placement in branch office, seems closely to reflect actual geographical distribution of the loans. 3.07 The first Appraisal Report predicted that the project would make loans to approximately 240 farmers over a period of two years, disbursing a total of US$5 million. In fact, Project 501-EC made loans to a considerably smaller number of borrowers - 128, but, as the loans made were larger than expected, the available funds were disbursed more rapidly than predicted and were fully committed by June 30, 1969, some six months ahead of schedule. The chronological evolution of the loans made, and their average value in each period is shown in table al. 3.08 After it became clear that funds from Project 501-EC were being exhausted, the Government of Ecuador requested an additional loan. In response to this request, the Bank sent a second appraisal mission to Ecuador in March/April 1969. The mission concluded that the project seemed to be successful in meeting its principal goals and agreed that a supplemental credit was needed to maintain the momentum of the project until the results of the 1968 agricultural survey, being carried out by the National Planning Office, could be interpreted and utilized in the design of a more comprehen- sive third livestock project. In view of these considerations, the Bank agreed to provide funds to continue the livestock project, on an interim basis, along the same operating guidelines as before. Because Ecuador's general financial situation seemed increasingly precarious, more liberal credit terms were offered. In January 1970 the Bank approved an IDA credit (173-EC) for US$1.5 million equivalent payable in 50 years, with 10 years grace and a service charge of 3/4 of one percent. All loan funds were to be used for ranch development as no additional funds were thought to be needed for technical services. Approximately 60 farm loans were to be made, over a period of approximately 1 year. - 8 - 3.09 Although the rate at which loans were made declined after the first year, there was no major interruption between the two projects except for a short delay while Project 173-EC was being approved. Eighty-four loans were approved during 1968, and 47, 44 and 47 loans were made in 1969, 1970 and 1971, respectively. (Loans to the beef cattle sector from Credit 222-EC which were expected to total 350 over a three-year period, have also been made more slowly, averaging about 45 per year. These loans, in individual amounts, have been larger than predicted because unit investment costs will again be higher than predicted. And the third project also was suspended for one year following a unilateral reduction by the Government in the on-lending interest rate.) 3.10 Project 501-EC financed a total of 132 loans, 4 of which were repeater loans, i.e., an amplification of a previous loan made under the same project. Credit 173-EC, financed 90 loans. Of these 90, one was a repeater loan from Project 173-EC and 18 were repeater loans from 501-EC, so 71 new borrowers were involved. The two livestock projects together made a total of 232 loans, to 199 separate borrowers. Four years, rather than two years, and two loans rather than one, were required to achieve approximately the number of loans which had been predicted for Project 501-EC alone. 3.11 The size distribution of the sub-loans under the two projects, for each project individually and for the two projects combined, is presented in the appendix (table a2). The range, median and mode of the loans in each project are also given. The figures indicate that the loans made in Project 501-EC were much larger than had been predicted in the first appraisal report, averaging 672,000 sucres instead of the planned 384,000, and also larger than those occurring in Project 173-EC. The appraisal estimates are known to be only approximate, but the increase in loan size in this case is serious because, as will be discussed, it arises largely from an unanticipated increase in unit investment costs. 3.12 The increase in the investment costs per ranch, of actual over fore- cast, did not result from having underestimated the size of ranches which would be applying for loans, and therefore of their loan requirements. The average size of the participating ranches was smaller than had been expected, whether size is measured by pasture area or by the number of breeding animals in the herd, either pre- or post-development. 3.13 The average loan in Project 173-EC is somewhat smaller than those in Project 501-EC, averaging 547,000 sucres, but is still larger than the appraisal estimates. The decline in loan size is chiefly due to the fact that the largest producers, requesting relatively larger loans, were early borrowers. Project 173-EC also included a larger number of "repeater" loans (loan amplifications), which were generally smaller than the original loan. 3.14 The first and second livestock projects were not designed to have a favorable impact on the distribution of income in Ecuador, but the ern- centration of funds lent is high. The number of loans in each of the size - 9 - categories utilized is rather similar, but the larger loans utilized a much higher proportion of the total funds available. Indeed, although the Project Director rejected a number of requests for particularly large loans on the grounds that the project was intended to promote widespread development of the cattle sector, rather than the development of only a few ranches, the 16 largest loans, equal to 8 percent of the total number of ranches, accounted for 26.6 percent of all loan funds disbursed, and the largest 39 ranches, 20 percent of the total, accounted for fully 48.5 percent. 3.15 Producers faced higher investment costs in both projects than anticipated, and made larger total investments than expected, but they also contributed a higher than predicted proportion of the cost of their develop- ment plans from their own resources. The first Appraisal Report predicted that producers would provide approximately 20 percent of ranch development costs while the individual farm plans instead indicate that producers planned to contribute approximately 24 percent. It is difficult to ascertain the total amount of funds which producers have since invested in their farms, because there simply are no data on investments made outside of and in addition to the investments listed in the farm development plans. None- theless, it appears from a partial sampling of producers that a number of farmers contributed even greater amounts to their farm development than that which was planned, and that the proportional contribution may have varied positively with the size of the loan. 3.16 The data indicate that the project was successful in affecting a considerable proportion of the Ecuadorian cattle herd. And it had an effect on producers throughout the Coastal zone, rather than simply in one zone. The last row in table a3 show the number of animals which were maintained on project farms at the time the individual sub-loans were made. As can be seen, approximately 8.3 percent of the Ecuador cattle herd was maintained on project farms prior to their development (94,000 of 1,130,000 animals in the Coastal zone), ranging from 5 percent in Manabi province to 13 percent in Los Rios province. 3.17 The Appraisal Report indicated that the first livestock project was intended to concentrate in the Guayas area of the Coastal zone, with some loans also to be made in the Santo Domingo area. In fact, the project spread throughout the Coastal zone almost immediately and, as shown in table a4, the distribution of loans across the different coastal provinces was remarkably equal throughout the whole period of 501-EC and 173-EC. 3.18 There were several reasons for the expansion of loan activity to other provinces. It appears that farmer interest was somewhat lower in the Guayas area than anticipated. The rate of return predicted seems to have been too optimistic, and the returns in alternative agricultural activities were probably higher. A drought in the Guayas area in 1968-69 also reduced the carrying capacity of farms and temporarily deflated interest in,intensive farm development. At the same time, there was growing interest in the project in other areas, especially in Manabi and in Santo Domingo. Manabi, - 10 - a traditional cattle raising area on the coast had suffered from drought throughout much of the early 1960s, and the project was able to assist ranchers in the less arid regions of this province to develop irrigation and water control systems by which drought could be resisted, and to begin restocking. In the Santo Domingo area, the complete collapse of the banana industry in 1968 resulted in a demand from banana producers for assistance in diversify- ing into cattle production. The Government offered credits for clearing banana land for conversion to other crops, but these credits did not provide sufficient funds for investments in other agricultural activities after clearing had been done. The project filled the gap. 3.19 Another factor which may have affected the demand for project credits in the Guayas was land reform. The distribution of agricultural land in Ecuador has traditionally been one of the most unequal in Latin America. Only in the late 1950s did land reform become a political issue, and even then the civilian government in power failed to deal with it. This failure led to open discontent among some sectors of the population, and was one of the reasons given by the military for their ouster of President Arosemena in July 1963. The Military Junta which succeeded Arosemena gave considerable support to land reform, establishing an Ecuadorian Agrarian Reform and Colonization Institute (IERAC) to administer a new law decreed in mid-1964. The new law proposed: 1) the establishment of minimum wages in the agri- cultural sector, 2) dramatic changes in the land tenure system, 3) actual land redistribution, 4) the provision of agricultural extension services, and 5) the incorporation of agricultural workers into the Social Security System. However, although IERAC enjoyed some early successes, it was not given the financial resources with which to fully carry out its legal mandate. The Government established by the Military Junta did not have Government revenues adequate to the task and, in any event, it was ousted in 1966. The succeeding civilian governments again paid only lip service to land reform. 3.20 Land invasions, most of which were promoted by political leaders associated with leftist groups, assumed increasing importance in the latter half of the 1960s. The invasions began in 1964, but ceased during the several years when campesinos felt that they might actually receive land through legal channels. When these expectations went unfulfilled, land invasions resumed, reaching a peak in 1968 and 1969, precisely at the time when project loans were first available. The fear of land reform remained strong among agri- cultural producers at this time, particularly in the Guayas region. 3.21 The land reform-land invasion issue seems to have had two effects on the rate of loan disbursements. On the one hand, the threat of land reform and land invasions made the livestock development plan offered by the project more attractive because it seemed to lower the probability of expropriation. Land reform legislation provided for the expropriation of lands which were underutilized, but producers owning underutilized lands were granted a period of delay during which they could demonstrate that their - 11 - lands were being improved and utilized productively. The existence of a cattle ranching development plan implied a long-term commitment to continuing improvements, and the physical investments which were made, such as fencing, corrals, other structures and cattle, served as an easily observable indicator that improvements were occurring. Cattle ranching also was more attractive to many producers than was cropping as it did not require as heavy an invest- ment in land clearance, leveling and water works and, more importantly, was much less intensive in the use of labor. Land owners were concerned that unoccupied lands would be invaded and/or expropriated, but were much more nervous that any tenants employed in cropping activities would suddenly band together and demand to IERAC that the land be given to them on the basis that the land owner was absentee. This problem almost never arose in cattle ranching because the vaquero possessed a quite different economic and cultural mentality from the campesino, and was also utilized in fewer numbers. 3.22 Although land reform and land invasion may have encouraged loan requests, it may also have discouraged ranchers from participating in the project by threatening with the loss of investments made. IERAC provided for reasonable compensation in the event of expropriation, but considerable un- certainty existed regarding the future and it is understandable that some landowners would be reluctant to make any long-term investments. The first Project Director noted that some farmers, after having agreed to borrow funds, decided to proceed with the development more slowly than planned after be- coming involved in land conflicts. 3.23 Given the importance of the land reform issue in Ecuador, both as a social and political issue and as a potentially disruptive factor in the implementation of the first livestock loan, it seems appropriate that more attention should have been paid to it in the first Appraisal Report. It does not seem, however, that the net impact of the land reform issue has been particularly large. Insofar as land reform-land invasion was most intense in the Guayas area, the dispersion of the project to Manabi and Santo Domingo resulted in less potential conflict with this issue. Most importantly however, the Project Staff exercised care to ensure that the project was not used simply as a means of avoiding land reform. Loan applicants were se- lected on the basis of their potential to perform well as ranchers, and the Project Staff attempted to remain unbiased whenever land reform disputes arose. These disputes seem to have been relatively infrequent, involving only approximately 10 of the 200 participating farms. About half of the incidents have resulted in the expropriation or sale of land to campesinos. 3.24 The Project Director made a considerable effort to ensure that borrowers made prompt payments whenever interest or principal was due. A system was established to remind borrowers well in advance that payments were coming due, and' pressure on-borrowers' overdue payments was maintained. The relatively small number of borrowers facilitated the collection process. A number of loans were foreclosed early in the program because of serious divergences from the development plan, or because of a failure to meet re- payment obligations. Theproject was more strict in this respect than have - 12 - been most other credit lines in Ecuador, but the foreclosures seem to have established useful precedents. At least one rancher attempted to use poli- tical pressure to resist foreclosure, but without success. 3.25 The repayment results seem significantly better than other similar projects in Ecuador, and also seem to be improving with time, although there is still cause for concern in some project sub-regions. As of May 1975, eight of the 132 farms in 501-EC were involved in litigation in which the participating banks were seeking to force cancellation and repayment of the loan, and eight more sub-loans had been declared in default, but without yet proceeding to judicial action. Approximately five other loans already have been paid off in response to demands from the banks. Thus, 21 of the 132 sub-loans, or 16 percent, have required bank action to obtain payment. The number and percentage of sub-loans in 173-EC in default or in litigation is much smaller, about 6 percent. 3.26 The proportion of interest which is in arrears at the NDB is still significant, but it has been declining with time, amounting to 26.5 percent in December 1971, 19 percent in December 1972, and 12.5 percent in December 1973. The comparable data for 1974 are not yet available. The amount of interest in arrears also varies significantly by branch bank, ranging from 1.6 percent in Chone to 33 percent in Babahoyo, as of December 1973. The magnitude of the differences among NDB branches suggests that the policies maintained by the branch manager are quite important in determining loan performance. The amount of principal in arrears is generally greater than the amount of interest, but part of this difference simply reflects the fact that loans which have been declared in default are shown as being totally due. It does not appear that the repayment record on principal differs, with respect as to arrears, from that on interest. 3.27 Table a5 indicates that the NDB made roughly 90 percent of all loans in each of the projects, 501-EC and 173-EC, but, as its loans averaged only half the size of those made by COFIEC, the NDB accounted for only 80 percent of the total funds disbursed. It is worth pointing out that, when loan amplifications are included with the original loan, the NDB is responsible for five of the seven largest loans made in the project. COFIEC has a smaller proportion of its loans in arrears than does the NDB, but among the NDB borrowers, there is no evident difference in repayment record by size of loan. 3.28 Several members of the original Project Staff have asserted that they felt continual pressure during the first months of the project to make more loans. This pressure was felt most strongly by the loan technicians working within the NDB. -These staff members had been hired at salaries above those prevailing for other persons in similar positions, and they felt that they had to justify their salaries by making loans. They were fearful that rancher interest was less than anticipated, and that the project would not develop as planned, and it is also clear that they - 13 - genuinely wished that the, project would bring about a rapid and strong impact on the livestock sector. By their own analysis, their fears and their zeal inclined them to recommend some loans which they would not have recommended at a subsequent date. The early loans were slightly less suc- cessful than later loans, but no major problems developed. 3.29 There was some complaint among project borrowers, particularly in 1970, that the short-term credits which had been promised by the NDB were not being offered in sufficient quantity, and that accordingly ranchers were unable to purchase the animals for fattening which were needed to profitably utilize the pasture which had been created. The shortage of short-term credit does not seem to have been a major problem, however. The NDB did suffer from a credit squeeze provoked by a tightening of monetary conditions in Ecuador during this period, but made efforts to supply short- term credit to project borrowers, particularly when requested to do so by project loan technicians. The small amount of short-term credit requested by ranchers is more noteworthy than their failure to obtain such credits. The lack of demand is apparently associated with the shortage of animals for fattening at a price which made their fattening profitable. The level of technical coefficients in ranches in the Coastal zone was so low that most ranches had been growing at extremely slow rates, and had no excess animals for sale. This situation produced a difficult situation for the project. Increases in carrying capacity brought about by large investments in project ranches could not be fully utilized because animals could not be obtained to stock this capacity. Many producers thus ended up with excess carrying capacity, and lower sales than they could have obtained had more animals been available. IV. OBJECTIVES AND IMPACT OF THE PROGRAM A. Objectives 4.01 The Appraisal Report for the first livestock project indicates that the Bank sought to achieve multiple objectives. The foremost objective was to: 1) Increase beef production by increasing productivity, particularly in cow-calf breeding, and by expanding the total area devoted to beef cattle. The increase in pro- duction was expected to go almost equally for increased domestic consumption and for increased exports. In addition, the Bank and Ecuador planned to achieve a number of subsidiary goals. The project was expected to: - 14 - 2) Provide additional information about the potential for, and the best means to further develop the beef cattle sector, 3) demonstrate to the agricultural sector as a whole the possibilities of increasing production through investments and the application of farm technologies, 4) stimulate the public and private banking systems to provide greater amounts of long-term credit to agriculture, and to improve their operating procedures, 5) encourage the Ministry of Agriculture to improve its extension services, and 6) assist INIAP in improving its technical operations as related to pasture seed and livestock production. 4.02 Additional perspective on these objectives may be obtained through the following background information. The economy of Ecuador, one of the poorer countries in South America in terms of per capita income, has long been predominantly agricultural. In 1967, when the first livestock project was approved, 65 percent of Ecuador's population lived in rural areas. Bananas, coffee and cacao, the principal commercial crops, provided 80 percent of total exports and were grown primarily in the Coastal zone. However, the ravages of Panama disease had recently devastated banana production in some areas, and the expectation of slowly growing and unstable export markets for all of these commodities suggested the need for diversification in agricul- tural production. Livestock production accounted for only 8 percent of GNP in 1967 and cattle for about half of this, but beef was believed to have a potential for rapid output expansion and for sharply increased exports. 4.03 The Coastal zone had emerged as the principal area of beef cattle production only since World War II. Approximately one-third of the land considered usable for agriculture in the Coastal zone was still unutilized in 1967, and approximately half of this land was thought suitable for improved pastures in a beef cattle development program. A few ranches in the area had achieved high levels of technical efficiency, but the overwhelming majority operated very inefficiently. The fact that ranching was a new activity in this area, operating inefficiently, was thought to imply a potential high return to new investments combined with the simultaneous introduction of improved managerial techniques. 4.04 The project designed by the Bank proposed to: 1) increase and upgrade the available pastures, 2) increase the size of the herd through the purchase of new breeding animals and the retention of heifers, 3) improve animal sanitary treatment, breeding procedures, and pasture management tech- niques, and 4) introduce improved blood lines into the herd to develop animals which would respond more rapidly, and more economically, to the improvements in infrastructure and management being adopted. - 15 - 4.05 The Appraisal Report for the first livestock project anticipated a significant impact at the individual farm level. During the first 12 years, the number of animals in a representative herd was expected to increase by 120 percent, sales of animals would increase by 325 percent, and net farm profits would increase by 700 percent (assuming an increase in the price of beef relative to farm inputs as well, as will be discussed below). 4.06 The Appraisal Report presented a model thought to be representative of the investments to be made and their effects. This model, while only indicative, suggested the conditions under which the project would prove to be profitable. The representative ranch contained, before development, a herd of 500 animals, 200 of which were cows, maintained on 650 hectares of pasture. The development plan which was proposed was expected to take three years, during which 150 hectares of new pasture would be converted from banana production or from forests, and 300 of the existing 650 hectares would be renovated. Seventy breeding heifers were to be purchased during the first three years as well, and an additional (net) 30 heifers were to be retained during the subsequent four years to increase the breeding herd a total of 50 percent to 300 cows. 4.07 As a result of these investments, and improved management tech- niques, calving rates were expected to rise from 52 to 80 percent and calf mortality rates would fall from 13 percent to 6 percent so that calf weaning rates would rise from 45 to 75 percent. Mortality rates for the rest of the herd were to fall from 6 percent to 2.4 percent. The stocking rate was expected to increase gradually from 0.8 animals per hectare to 1.3 animals per hectare, in addition to the increase in the total carrying capacity provided by additions to and improvements in existing pastures. Steers were also expected to gain faster, reaching slaughter weights some 20 to 40 percent higher than the previous average of 400 kilograms, and achieve this weight in approximately 3.5 years instead of the 4.5 years required on most farms before project development. 4.08 Finally, as the Bank hoped it had persuaded the Government of Ecuador to permit free trade in beef and beef cattle, it was thought that the domestic price of beef would rise to international levels. The magnitude of the increase in beef prices was not indicated in the Appraisal Report except insofar as the unit value of animals sold to slaughter was shown in the Appraisal Report model to increase by 50 percent over the life ,of the project. Much of this increase in the value of animals was due to the increase expected in slaughter weights, a result of improved breeds, pastures and better management. However, the increase in slaughter weights was 6nly approximately 30 percent, so that.an implicit price increase per pound of beef of 20 percent was also assumed. The increase in beef price relative to input costs was expected to commence in the fourth year of the project and to continue at a constant rate of increase through the twelfth year. - 16 - 4.09 Given the assumptions above, a calculation was made in theAp- praisal Report to ensure that the expected financial return to project borrowers would be sufficiently high that individual borrowers would wish and be well advised to enter the program, and to ensure that the expected rate of return to Ecuador would be sufficiently high to guarantee the attractiveness of the project among alternatives. The Appraisal Report suggested that the financial and economic rates of return would be similar to each other, and approximately 27 percent. B. Summary of Results 4.10 Although the project on the whole has evolved reasonably well, it appears that the financial and economic goals established were too optimistic. And too little attention was paid to the effect which several unanticipated factors were already having on the predicted rate of return. There is a complex interdependence among several factors in determining the rate of return achieved on any investment project, the most important factors being the cost of the investment, the physical productivity of the investment, and the value of this physical output. In the livestock projects in Ecuador, it appears that the costs of the planned investments were higher than estimated, the physical productivity of the investments was slightly less than predicted, and the prices received for the product produced were also less than expected. The financial rate of return for project borrowers would have fallen significantly short of the predicted level had not rapid inflation, which was unexpected, eroded the real value of loan repayments. C. Data 4.11 These conclusions are based on information taken from a variety of sources. Unfortunately, no detailed statistical system has yet been estab- lished to monitor the farm level impact of the livestock projects in Ecuador. An agricultural economist was employed on the Project Staff at the beginning of the third livestock project (222-EC), but his time has been utilized for the analysis of new loans for 222-EC and the design of a fourth livestock project proposal, rather than the analysis of the effect of past loans. Similarly, Project Staff have filled regular formal reports on the progress of each project ranch, indicating the degree to which producers are meeting predicted levels of success, but while these reports have been helpful for the administration of the project they do not provide a sufficiently precise, or comprehensive statistical basis to permit reliable judgments regarding the overall financial success of the project. 4.12 As a result of the lack of data, an attempt was made immediately prior to this audit to undertake a statistical survey of ranches. The most severe rainy season in 40 years in Ecuador's Coastal zone prevented this - 17 - from being completed. The statistical information on which this information is based is therefore taken from several different sources: project records, discussions with Project Staff, a small and partially completed sample survey of project borrowers carried out in March-May 1975, INIAP personnel, and from an academic research project carried out by Jaime Egas in 1972 analyzing a sub-set of project borrowers.1 I D. Investment Costs 4.13 The project's; files, which contain the development plans worked out for each farm, indicate that unit investment costs exceeded the amounts predicted in the first Appraisal Report by a considerable degree. See table a6. For example, the cost of each unit of pasture created or improved, exceeded expectations by more than 100 percent for Project 501-EC (S/770 vs. S/333) and by 150 percent for Project 173-EC (S/825 vs. S/333). The unit cost of each breeding animal purchased exceeded Appraisal Report pre- dictions by similar proportions, 110 and 120 percent respectively, and the 1 / A thesis written by Jaime Egas for the Department of Economics, Central University of Ecuador, Quito, and subsequently published in abbreviated form by INIAP, discusses developments among project borrowers in the area around Santo Domingo de los Colorados and Quevedo. This important sub- region contained 52 loans from 501-EC and 173-EC, of which 10 were re- peater loans. Of the 42 separate borrowers, all but 4, or 38, were inter- viewed in detail with respect to the situation on their ranches both at the time of the loan and at the time of Egas' evaluation, which took place in early 1972. Egas was encouraged by the then Project Director in his research, given access to project files, and accompanied in his visits to project borrowers by two NDB loan technicians responsible for the area in which the evaluation was carried out. Egas divided his sample into three strata to determine whether there were statistical differences in the performance among farms of different size, where size was measured by the number of animals on the ranch at the time of the evaluation. The strata were: I, with a range from 140-300 animals and an average of 221; II, with a range from 380-640 animals and an aver- age of 483; III, with a range from 700-1,200 and an average of 1,115. There were 15 farms in category I, 13 in category II, and 10 in category III. Unfortunately, because Egas did not stratify farms according to the date they commenced ranch development, nor the date the repeater loan was granted, it is impossible to determine the speed with which investments had an impact. The 38 loans were made at different points in time, and some farms had been participating in the program for a relatively short period before the evaluation, so Egas' results must be treated carefully. It appears however, that the average farm in his sample had been partici- pating in the project for about 2 years at the time of Egas' evaluation. - 18 - unit cost of each kilometer of fence installed exceeded Appraisal Report predictions by 20 and 60 percent, respectively. These costs can be determined by dividing the total amount of funds allocated in farm development plans for each category of investment, by the units which the farmers were committed to purchase. This calculation assumes that farm development plans were carefully drawn up to reflect the costs prevailing at the time of investment; the majority of evidence seems to support this assumption. 4.14 The unit costs were so much higher than expected that even though the total amount invested per farm was greater than predicted, the amount of additional pasture and the number of additional breeding animals purchased were far below the amounts predicted for the representative farm in the Appraisal Reports . See table a7. More importantly, these higher costs implied that without correspondingly higher productivity than predicted, or a much greater increase in the price of beef relative to input costs than was predicted, the individual farm investments were certain to be less profitable than expected. 4.15 These data in tablea7 may be combined with the data in table a8 regarding the pre-development situation on project ranches to obtain a more precise indication of the degree to which higher investment costs, cet. par., reduced the expected rate of return. First, we note that farms on which development took place were slightly smaller in their pre-development situation than expected, both when measured with respect to pasture avail- ability and with respect to the total number of cows and heifers .in their herds. Similarly, if we compare the size of the farms in the post-development situation, taking the fourth year for comparison (all investments were to have been complete at this time), we find that the amount of pasture and the number of breeding animals on farms in both projects are still smaller than was predicted for the representative farm. Second, we note again that the average size of investment in both projects was substantially greater than predicted, S/870,000 for Project 501-EC and S/700,000 for Project 173-EC versus an Appraisal Report prediction in both projects of only S/480,000. Third, we note that the planned marginal investment per breeding animal in the herd, post-development, was S/1,580, while that which actually occurred for Projects 501-EC and 173-EC is S/3,175 and S/2,475, respectively. The latter figures, which are 2.0 and 1.6 times predicted levels, can be obtained by simply dividing the investments made by the number of breeding animals in the herd in year 4. They are the most important data. 4.16 Although the actual project farms and the representative farm in the Appraisal Report are not the same size, the marginal investment per breed- ing animal in the herd in the post-development situation will correlate highly with the rate of return achieved on the investment in each situation, assuming that the level of technical efficiency and all prices are given. This assumption allows us to substitute our investment costs per breeding animal, about twice and 1.6 times those originally predicted, in the Ap- praisal report model. For these investment costs, the rate of return falls - 19 - from 27 percent to approximately 13 percent and 17 percent, respectively, for the two projects. These are respectable rates of return, but they leave little margin for error elsewhere in the project if producers are to be able to repay their loans profitably, particularly in the case of farms in Project 501-EC. 4.17 Because of the importance of this issue, we have attempted to determine why costs were so much higher than expected. A major factor seems to have been the impact of the project on the demand for breeding animals, and this fact does not appear to have been adequately recognized in either the supervision reports or in the second Appraisal Report. Other input costs, including labor, also rose more than expected during this period, but for reasons largely unassociated with the Project. 4.18 The second Appraisal Report indicated that in Project 501-EC unit investment costs had been higher than predicted for both pasture improvement and for cattle purchase, but it was not deeply concerned with this diver- gence. It attributed the difference in pasture cost to a shift in area from the relatively cleared Guayas zone to the Santo Domingo zone where forest clearing, with higher costs, was required for pasture establishment. This Appraisal Report justified the costs in the belief that "once pastures are established in the Santo Domingo zone, even at a higher unit cost, the carrying capacity is much greater than that obtained in the main Guayas zone." It also mentioned the fact that the cost of breeding animals had increased by at least 30 percent, but went on to predict that better beef prices and physical responses would make it possible to meet estimated returns. 4.19 The second Appraisal Report seems to have treated the issue of higher costs too casually. First, the proportion of farms located in the Santo Domingo-Esmeraldas area in Project 501-EC was greater than predicted, some 38 percent instead of the expected 17 percent, and the cost of pasture establishment in this area does seem to have been higher than elsewhere. However, it does not appear that these costs are higher by an amount which would fully explain the discrepancy between the costs predicted for pasture creation and those which were occurring in the project taken as an aggregate. This conclusion can be supported by examining the data contained in Egas' study, which specifically evaluates the development of farms in the Santo Domingo area. Manipulation of Egas' data shown in table a9 produces cost figures for pasture creation which are at a maximum, S/1,200 per hectare on average, as opposed to a predicted cost for the project of S/333 per - 20 - hectare. 1/ Assuming all farms in the Santo Domingo area paid S/1,200 per hectare for pasture creation, the cost of pasture creation on other farms would still have to be 50 percent more costly than predicted in order to reconcile the amounts spent on pasture with the pasture increase achieved. In short, the costs of pasture creation were higher than expected across all zones, largely because of rising labor costs. 4.20 It is also important to point out that producers in the Santo Domingo area were not stocking their pastures at higher rates than elsewhere, at least not during the early years. Egas indicates that pasture capacity expanded more rapidly than did herd size for farms in the smallest size categories, so that at the time of his evaluation stocking rates had actually fallen from a pre-development level of 1.2 and 1.1 to 0.75 and 1.0, respec- tively, for categories I and II, and rose only slightly on the farms in category III, from 1.0 to 1.1. This finding is consistent with the fact that the largest farms seem to have spent the least on pasture creation and the most on animal purchase. 4.21 The problem of initial stocking rate declines leads us to discuss the second major development on the side of input costs, that of unexpected increases in the costs of animals for breeding and fattening. These costs rose in response to the demand created by the project itself. The Bank had been concerned in early discussions regarding the project that there might 1/ Egas' three-farm categories evidence cost differences among themselves which apparently result from statistical discrepancies. Only the data regarding small and medium producers appear reasonable, and they have been used here. The following two assumptions were used to obtain cost estimates: (a) if all original pasture is renovated at a cost of S/325 per hectare, 50 percent above appraisal predictions for this area, the cost of creation of new pasture per hectare in his sample is: I II III 821 1,120 2,328 (b) if all funds were spent solely on pasture creation, the cost of pasture creation per hectare is: I II III 1,210 1,810 5,550 It appears that the cost of pasture creation is between S/800-1,200 per hectare, and that large producers either had not completed their investments at the time Egas made hiyv-m-luation, or were diverting funds originally planned for pasture creation to animal purchase. Given the fact that animal prices were much higher than expected, the latter path seems the most reasonable. - 21 - be too few breeding animals in Ecuador, but both the first and second Ap- praisal Reports concluded that the local supply of breeding animals would be adequate to permit the proposed livestock development projects to evolve as planned. This was not the case. The level of technical efficiency in Ecuadorian herds was so low prior to the initiation of the first livestock project that there simply were too few animals which could be retained to permit the herd expansion which was sought. Because breeding animals were in short supply, their prices rose significantly when many producers attempted to purchase them simultaneously. Producers were making large investments in fixed infrastructure, and they had to try to increase their herds if they were to achieve the higher sales by which to amortize these investments. Their efforts only increased the prices of breeding animals, reducing the rate of return achievable. 4.22 It may be that in appraising the project the Bank was misled by an unusual climatic situation occurring in Ecuador at the time. A drought had been occurring in Manabi province since the early 1960s and became parti- cularly severe between 1965 and 1968. Herds, including breeding animals, were significantly reduced on many farms as a result. However, by the end of 1968 the climatic situation began to improve in Manabi, and this sub- region attempted to rebuild its herds. Similarly, the drought in the Guayas zones in 1968/69 restricted herd development and forced the sale of animals to other sub-regions. Both the drought in Manabi and in the Guayas zone reduced the price of breeding animals below what it would otherwise have been, making investments appear more attractive, but the termination of these droughts brought renewed demand for herd rebuilding in these areas, and a resurgence in the price of breeding animals. Although the livestock project assisted in the salvaging of animals in the case of both droughts, it is questionable whether it was wise to induce such significant capacity expansion at a time when the existing pasture was temporarily underutilized. And in any event, the Bank ought to have recognized that the price of breeding animals in Ecuador was going to rise significantly above Appraisal Report levels once the droughts were ended and the Project's own impact begun. 4.23 The shortage of breeding animals might have been partially solved by the importation of animals from abroad had it not been the case that the price of beef in international markets began rising rapidly in 1969, inducing producers in other countries to retain their breeding animals as well. Thus, although the Project Director attempted to arrange for the importation of breeding animals from abroad, few animals were available, and then only at high prices. An agreement was finally made to import 50,000 heifers from Colombia, where producers wished to sell-breeding animals because of artifi- cially restrained market conditions, but this agreement was then vetoed by the Government of Colombia because it wanted to promote the development of the Colombian livestock herd. - 22 - E. Investment Productivity 4.24 The livestock projects in Ecuador have achieved their greatest success in the area of technical improvements at the farm level. This con- clusion can be drawn by comparing the current level of the technical co- efficients on project farms with those predicted by the first appraisal report. The first livestock project was initiated slightly more than seven years ago and the "average" farm included in Projects 501-EC and 173-EC has now participated for 5.8 years. The Appraisal Reports for both projects predicted a weaning rate of 65 percent in the fifth year and 75 percent in the sixth and subsequent years. In comparison, it appears that the calving rate achieved on project ranches currently averages approximately 68 percent, with calf mortality between 5 and 10 percent, implying a current weaning rate of 62-65 percent. These figures show marked improvement over the estimated pre-development weaning rate of 45 percent, and suggest that the Ecuadorian livestock projects have achieved a significantly greater increase in weaning rates than have most of the other livestock projects in Latin America which have been evaluated to date. 1/ Substantial additional effort, particularly as regards management improvements, will nonetheless be required if the average weaning rates are to be raised from their current levels to the steady state level of 75 percent predicted in the Appraisal Report. 4.25 We have almost no information regarding mortality rates save the Egas study and the March-May 1975 survey. The Egas study suggests that mortality rates on medium and larger farms declined from their pre-develop- ment levels at the rate planned, while mortality rates had remained constant on the smaller farms. The recent survey, which seems to have elicited responses chiefly from better as opposed to worse farmers, and is probably somewhat biased upwards, suggests that adult mortality rates were ap- proximately those which were predicted. 1/ The information in the Egas study confirms that these conclusions held at an earlier period for the Santo Domingo sub-region. The weaning rate on farms in his sample, evaluated at the end of their second year in the project, was 52 percent, up from 43 percent pre-development. This improvement was greater than that predicted in the Appraisal Report, which assumed a 45 percent pre-development weaning rate, and an increase at the end of two years to 50 percent. Interestingly, the larger farms in the Egas sample achieved higher levels of technical efficiency than did the smaller ranches, and also achieved greater absolute improvements in these measures over pre-development levels. No evidence exists to determine whether this pattern has been maintained during the three years since the Egas study. - 23 - 4.26 We have no good data on the rate of weight gain by animals, but it appears that the infusion of better breeds, and better pastures and their improved management have permitted farmers to approach the goals established regarding heavier slaughter weights occurring at younger animal age. 4.27 Little information on stocking rates is available; that which exists suggests that farms have achieved stocking rates approximately equal to those predicted. Originally, stocking rates among project bor- rowers seemed to fall somewhat short of the programmed, but the recent survey suggests that at least some farms are now exceeding the predicted levels by a slight amount. 4.28 In March 1975, the Project Staff attempted to evaluate, in a qualitative sense, the technical progress of each of the loans which had been made. These evaluations, which are shown in table 2, were made without the benefit of a systematic survey of farm developments, but reflect the Project Staff's judgements as based on their experience with the farmer. We believe the Project Staff is competent, realistic and well-informed, and that the evaluations provided another useful impression of the projects' results at the farm level. Table 2 Degree of Progress on Project Farm (Relative to Targets of Development Plan) As Marginally Well Superior Planned Below Plan Below Plan Failure Total 501-EC 9 43 41 9 30 132 (%) (7) (33) (30) (7) (23) (100) 173-EC 3 28 38 16 6 90 (%) (3) (32) (42) (16) (7) (100) Total 12 71 79 25 36 222 (%) (5) (32) (36) (11) (16) (100) As these figures show, approximately 73 percent of project ranches are performing technically in a satisfactory way. Eleven percent are perform- ing badly, and 16 percent have quit the program, most because of failure in one respect or another. The results for the two projects when compared with each other are highly similar, though Project 173-EC may be marginally better. On balance, the results suggest that the Appraisal Report pre- dictions were optimistic, but that substantial progress has indeed been made. - 24 - F. Technical Assistance 4.29 Because of the importance of the technical achievements in the Ecuador livestock projects, this audit has considered the specific reasons such achievements occurred. One of the principal problems faced by the project was the lack of technically qualified producers. Approximately 80 percent of the commercial livestock producers in the Coastal zone were absentee most of the year, and they entrusted the day-to-Aay management of their establishments to paid administrators or to semi-skilled foremen. Most large farmers also had other occupations and sources of income, and were not knowledgeable about the technologies available. Technical assist- ance to improve management competence was an essential component of the livestock project. 4.30 The first Appraisal Report indicated that the Coastal zone was adequately supplied with extensionists from the Ministry of Agriculture. Accordingly, the technical assistance included in the project design was confined to the identification and screening of projects, the planning of specific investments, and the supervision of the investments themselves. The Project Staff was not expected to provide ongoing technical assistance to project borrowers, which was to be the responsibility of the Ministry of Agriculture extensionists. 4.31 In fact, the Ministry's extensionists have been much fewer and technically weaker than was predicted, and only recently has its extension service shown signs that it may significantly improve. The Ministry of Agriculture did not have the funds to provide the extension services re- quired at the time the project was initiated, and feuding between INIAP, responsible for agricultural research, and the Ministry of Agriculture, further reduced the effectiveness of the technical assistance potentially available from these sources. 4.32 Since the initiation of the project, the Project Staff, including the loan technicians working for the NDB and COFIEC, have provided more ongoing technical assistance than was envisioned originally, or than has occurred in most of the Bank's other livestock projects in Latin America. The first Project Director was effective in his work and is held in high regard in Ecuador, as have been the subsequent Project Directors and the rest of the Staff. The technical assistance they have provided, together with the requirement that borrowers either reside on their ranches or hire qualified administrators, and a strong interest on the part of many bor- rowers to make suggested operating improvements seem to all informants to be important to the unusual technical achievements which have resulted. - 25 - 4.33 Project Staff numbered approximately 8, including the Director who also worked in the field much of the time. Given a total of 199 bor- rowers, each technician was responsible for approximately 30 loans. The Project Staff was responsible for screening many loan requests which never arrived to the planning stage, and approximately 70 development plans were elaborated for loans which ultimately were not made. Assuming one week for the elaboration of each development plan, time remained for an average of 4 additional technical visits per year to each ranch. These are OED estimates. 4.34 Project Staff prepared and distributed informational pamphlets regarding the construction of corrals and dips, proper sanitary treatment, and better pasture management techniques. Project Staff have also made regular visits to project farms, obtaining substantial understanding of the problems and achievements made on individual farms, and establishing a productive working relationship with many of the farmers. Where the resident management on a farm is better, visits have been less frequent, and have served principally to check on progress and to help in resolving specific problems where specialized knowledge is required. In the case of less knowledgeable farmers, visits have served to provide more general assistance'.l No meetings or training sessions have been organized specifi- cally to bring project borrowers together, but Project Staff have maintained active contact with various agricultural groups, especially the Federation of Coastal Livestock Producers, and have worked to promote livestock develop- ment through these groups. 4.35 The discovery of petroleum and the commencement of its export in 1972 has resulted in a significant increase in Government revenues, and permitted the Government recently to increase its rate of expenditures in the agricultural sector, including those for agricultural research and extension services. Feuding between the Ministry of Agriculture and INIAP has diminished, and INIAP has agreed to provide training courses for the Ministry of Agriculture's personnel, in addition to courses which INIAP is designing for individual farm owners, farm administrators, and foremen. These courses are now being prepared and will be initiated soon. (Funds to assist the development of these courses were provided in the third live- stock project, but progress was delayed by difficulties encountered in locating suitable advisors to help design the courses, and then by a construction contract dispute which prevented the necessary school facilities from being completed at Pichilinque.) We believe that better official ex- tension service and the availability of management training courses will significantly increase the likelihood of continued technical advancement in livestock production in the Coastal zone. 1/In the March-May 1975 survey, the project's loan technicians found that on the 49 farms sampled, 80 percent of the owners are now regularly resident, and the management, whether owner or administrator, was judged technically competent on 80 percent as well. The data did not indicate that administrators were more or less likely to be competent than the owners. - 26 - G. Prices and Marketing Arrangements 4.36 At the time of the first Appraisal Report, the Bank recognized that the cattle and beef marketing system in Ecuador presented several obstacles to more profitable beef production. The principal problem was the controlled suppression of beef prices below their free market levels, both through the fixing of retail beef prices in the major municipal markets, and through the restricting of export trade with Peru. Other obstacles included the fact that animals were purchased live, on the ranch, at an estimated carcass weight basis and without the use of grading which would have offered higher unit prices for better quality cattle; and the existence of inefficient and unsanitary slaughtering facilities which failed to offer producers maximum value for their production. 4.37 The Bank attempted to obtain a guarantee from the Government of Ecuador that the free movement of cattle and beef to domestic and export markets would be permitted. This guarantee would have ensured that domestic prices for these commodities would fully adjust to export market levels. The Government of Ecuador refused to bind itself by such a guarantee, but did provide a letter, apart from the loan agreement itself, indicating this intent. Despite this letter, prices and exports have remained under control, and producers have received less than the potential export price for their animals during much of the period since the first livestock project was initiated. 4.38 By maintaining export and retail price controls, the Government of Ecuador places an effective ceiling on prices to producers and thereby reduces the incentives for higher production. This disincentive results in lower beef production, and works particularly strongly against the introduction of capital intensive production techniques such as those which the Bank's live- stock projects were designed to promote. Equally importantly, the first and second Appraisal Reports assumed that freeing beef prices would signifi- cantly increase the price of beef relative to input costs, and this increase was important to the rate of return which was predicted. It appears that too little consideration was given in the first Appraisal Report to the potential political impact which freeing beef prices might have, and there- fore to the likelihood that the predicted increase would actually occur. 4.39 Ecuador's natural export market for beef is Peru. Peru does not produce enough beef to satisfy its domestic requirements, and is able to obtain animals best suited to its needs from Ecuador. Ecuador enjoys immediate proximity to Peru, reducing transportation costs, but most importantly, it exports live animals which Peru is able to fatten further before slaughter, being able to make use of by-products from agricultural activities which would otherwise be wasted. This was also especially attractive to Ecuador which, until recently, did not have slaughtering facilities which met international hygienic standards, and so could not gain access to other markets. Ecuador also has not had sufficient quantities of beef available on a regular basis to develop other external markets. - 27 - 4.40 Cattle have moved from Ecuador to Peru for many years, but exports to Peru were first legalized in 1963, subject to licensing and taxes. The required licenses and taxes have effectively restricted exports below the levels they would otherwise have reached, and the price of cattle in Peru has.been consistently above the price in Ecuador, by as much as 50 percent. As a result, in addition to legalized traffic, smuggling has been significant. In 1970 Bank staff estimated that approximately 20,000 animals were exported legally to Peru each year, and that an additional 30,000 animals were smuggled across the border. Many Peruvians come to Ecuador with trucks, buying animals directly from producers at the farm gate. 4.41 As a proportion of beef produced in Ecuador, the part exported to Peru has frequently been quite high, despite the restrictions imposed. This may be seen through the following calculations. The Ecuadorian cattle herd contains approximately 2 million animals, and only about one-half of these, or 1 million, are beef cattle produced in the Coastal zone. With an ex- traction rate approximating 15 percent, Ecuador has no more than 300,000 animals for slaughter in any year, and at most half of these are steers as opposed to cull animals. In comparison, the Peruvian demand has been principally for young steers which could be fattened additionally before slaughter, and thus is for the better beef animals produced. This is also the type which was expected to be produced by project farmers. 4.42 The amount of beef exported to Peru is thought to have increased during the first years of the project, especially in 1969 and 1970, but was then reduced by the imposition of tighter restrictions in Ecuador. The Government of Ecuador acted to increase the beef available for domestic consumption, as higher exports were creating domestic shortages and would eventually have forced an increase in fixed retail prices to clear markets. Higher beef prices are unpopular with the consumers of beef, who in Ecuador are almost exclusively the upper-income groups in urban areas, and the Government was hesitant to irritate these groups. 4.43 General inflationary pressures in Ecuador became more pronounced in the early 1970s and, after having been restrained at nearly constant levels since 1967 while other prices were rising, the retail price of beef was allowed to increase significantly in 1972. It remained constant again until late 1974, when it was increased sharply. Inflation, an increase in domestic demand, and pressures caused by the enormous increase internationally in the price of beef were factors in this decision. As a result of the large increases in 1972 and 1974, the price of beef has risen significantly more than the general price level over the last three years, and now shows a greater total increase than the general price level over the life of the whole project. But, there has been a dramatic increase in the price of almost all agricultural commodities in Ecuador in the last several years, reflecting increases in international markets, the impact of the petroleum boom domestically, and stagnating agricultural production in the face of rapid population growth. And agricultural input prices, including the cost of labor, have increased at an equal rate with beef prices, so that producers have still not obtained the increase in price of beef relative to input costs which was foreseen. - 28 - 4.44 The Government of Ecuador has not complied with its letter of intent regarding beef and cattle marketing policy, but Project Staff feel that this letter, which became known to the Federation of Coastal Livestock Producers, was useful to the latter in applying pressure on the Government to permit price increases once inflation became severe. The officers of the livestock federation have expressed a similar belief. It is also possible that as cattle ranching becomes more modern and more economically significant, and as ranchers become more organized as a group, they will be able to exert more political pressure in their own interests, thereby partially countering the pressures from consumer groups to hold prices at lower levels. 4.45 The situation of the slaughterhouse facilities in Ecuador is not so quantitatively important in the short run as are price controls, but it remains unsatisfactory and it will be necessary for it to be significantly improved if Ecuador is ever to export slaughtered beef. The older municipal slaughterhouses, including those in Quito and Guayaquil, are economically inefficient and unsanitary. They make no use of animal by-products, and therefore cannot offer producers full market value for their animals, they do not offer price differentials for animals of better quality, and they cannot export carcass beef as they do not meet normal sanitary regulations. 4.46 The Bank felt that the problem of the slaughterhouses was being resolved at the time the first project was initiated. New, modern and technically efficient slaughterhouses had been constructed in Guayaquil, Santo Domingo de los Colorados, and Loja;another was planned for construction in Quito, and a law establishing sanitary regulations had been passed in 1965 and should have forced the closure of the older municipal plants. However, the sanitary laws have not been enforced and the older municipal plants have not been closed. They also have been able to compete economically with the new slaughterhouses by virtue of the fact that they have an established purchasing and distribution network, effectively operated as a cartel. The new slaughterhouses have not been able to achieve the volume necessary to operate economically, largely due to the cartel, but also because they lack working capital and until recently suffered from labor contracts which imposed inefficient practices. The new packing plant in Guayaquil has recently changed its management and renegotiated improved labor contracts, but Government action will probably still be required, enforcing the existing sanitary laws or, at the least, actively reducing the monopoly power of the old municipal slaughterhouses, before the packing plant situation is favor- ably resolved from the farmers' viewpoint. 4.47 The livestock projects have encouraged producers to purchase scales to ensure full compensation for heavier animals and the introduction of scales has been widespread. This seems to have been justified financially from the producers' viewpoint. The lack of price differentials for higher quality beef, however, which is enforced by the uniform ceiling on retail prices, continues to discourage the production of better grade beef. - 29 - H. Impact on Aggregate Beef Production 4.48 Forty-nine farms were sampled in March-May 1975. These farms show a 74 percent increase in herd size and a 60 percent increase in pasture area over their pre-development situation. Farms in this sample which are in Project 501-EC show an 84 percent increase in herd size, while those from Project 173-EC show a 61 percent increase. These figures compare with Appraisal Report predicted increases of 91 percent and 55 percent respectively, which for the individual farms is close to that achieved. Farms in Project 173-EC compare somewhat better than those in 501-EC. The data from this sample survey are consistent with indications from other sources regarding the increase in farm herds. However, because the indi- vidual farms included in Projects 501-EC and 173-EC were slightly smaller than predicted and, as investment costs were higher, significantly fewer loans were made than predicted. This means that fewer total cattle were affected than expected. Extrapolating from the sample, we conclude that the project resulted in an increase in the Ecuadorian herd by at most 65,000 animals instead of the increase of approximately 120,000 animals which was expected at this date. Though below predictions, the results are significant, implying that the project increased Ecuador's beef cattle herd by 6 percent and improved the operating efficiency of about 17 percent of the total herd (all progress concentrated in the Coastal zone). These effects, when combined, imply an increase in total beef production of roughly 10 percent, worth approximately US$4 million per year. To the impact of the first two projects must be added that of Project 222-EC, which has provided an approximately equal amount of funds in the Coastal zone. It must be concluded that the series of livestock projects has had a widespread impact on cattle ranchers of commercial size. I. Rate of Return Analysis Financial Rate of Return 4.49 A number of adjustments have been made to the Appraisal Report rates of return in order to reflect project developments as they actually occurred. In section a of table 3, we estimate the financial rate of return to borrowers and in section b, the economic rate of return to Ecuador. In section a we have disaggregated three major adjustments, showing the individual effect which each would have had in isolation assuming that the project had otherwise developed exactly as planned. These effects are then combined in line 4) to indicate their joint impact. The effect of inflation is then considered as well. 1/ 1/ Each of these calculations has been-made by applying the indicated adjustements to the Appraisal Report model. The backup tables,showing the evolution of herd development, sales, operating costs, debt service, farmer investment, and the farmer's net cash position are available from OED. - 30 - 4.50 On line 1, we assume that unit investment costs were 100 percent and 60 percent higher than expected for Projects 501-EC and 173-EC, respec- tively. We also assume, as reported in the second Appraisal Report, that individual farm development projects were carried out more rapidly than expected, 70 percent of the investment being completed in the first year and 30 percent in the second year. These adjustments reduce the rate of return from the 27 percent forecast to 13 percent and 17 percent on the two projects, respectively. 4.51 On line 2, we assume that weaning rates have risen slightly more slowly than predicted, reaching a constant level of 75 percent in year 10 rather than in year 6. No adjustment is made to the predicted adult mortality rates or to animal slaughter weights,which are assumed to be valid, although less information is available for these variables. Our assumptions may be optimistic, but we have tried to err on the high side in order to reflect the impressive technical changes which have been intro- duced to Ecuador. The adjustment reduces the rate of return from 27 percent to 25 percent for both projects. 4.52 On line 3, we assume that producer prices declined relative to input costs during the first five years of the project, subsequently recovering to original levels, but without ever reaching the Appraisal Report prediction of a 20 percent increase in the price of beef relative to input costs. This assumption reduces the rate of return from 27 percent to 17 percent for both projects. Project 173-EC suffered less favorable relative prices at its initiation, but also enjoyed a more rapid recovery to the assumed price/cost ratio. The data supporting these conclusions are presented in table alO. 4.53 On line 4, we show the combined effect of adjustments 1, 2 and 3. Together they reduce the rate of return on the two projects from the predicted level of 27 percent to 10 percent and 13 percent, respectively. 4.54 On line 5, we make an adjustment to reflect the effect of rapid inflation in Ecuador which has eroded the real value of repayments on interest and principal by producers. In making this adjustment, both output and input prices have been increased to obtain a current value of net operating profit, and the net cash balances to producers remaining after debt service then have been deflated by the increase in the general price level in Ecuador to obtain a financial rate of return in constant purchasing power. The adjustment for inflation is combined with those already made in line 4, and hence ought to yield a good estimate of the financial rate of return currently being achieved by producers. Inflation brings about an increase in the financial rate of return to 25 percent and 30 percent, respectively, on the two projects, close to Appraisal Report predictions. - 31 - 4.55 The adjustments we have made in the calculations described above reflect the best data available and are thought to be reasonably indicative of the effects which the most important project developments have had on the financial rate of return. We do not pretend that these_adjustments, or the resulting estimates, are precise. We have provided only orders of magnitude. Economic Rate of Return 4.56 A proper estimate of the economic rate of return to Ecuador should include adjustments for differences between market prices and shadow prices, the existence of taxes and subsidies, the effects of inflation on costs fixed in nominal terms, and for other criteria, like income distribution, which might be thought to merit special weighting. Without suggesting that the adjustment given in line 6 is precise, we believe it provides an acceptable approximation to the economic return: 11 percent and 12 percent, respectively, for the two projects. To obtain these returns, we have cal- culated the increased value of output resulting from the original investment, and have assumed that the shadow prices of beef and farm inputs both have risen at the rate specified in the Appraisal Reports, even though the actual market price of beef has been constrained below this level. We have forgone the inclusion of the effects of taxes and subsidies, believing that these are of second order importance in this particular case. We have also ignored income distributional and employment effects in making these calculations, although the former may be important. Note, finally, that the rate of return to the Government of Ecuador would be somewhat higher than shown in table 3 to the extent its payments to the Bank are fixed in currencies which, because of inflation, are themselves being reduced in real terms as time passes. - 32 - Table 3 RATES OF RETURN TO FIRST AND SECOND ECUADOR LIVESTOCK PROJECTS Assumptions 501-EC 173-EC Appraisal Forcast 27% 27% a. Financial Rate of Return 1. Adjustment to reflect unit investment costs which were 100 percent and 60 percent higher than predicted, and that farm development 13% 17% was carried out more rapidly than expected, 70 percent in year 1 and 30 percent in year 2. 2. Adjustment to reflect slightly slower increase in weaning rates, to same ultimate level, with no change in assumptions regarding adult 25% 25% mortality rates or animal slaughterweights. 3. Adjustment to reflect lower producer prices of beef relative to farm input costs, as 17% 17% these occurred over the life of the project. 4. Combined effect of 1) + 2) +3) 10% 13% 5. Adjustment to reflect effect of inflation which has eroded the real value of repayments 25% 30% by producers, combined with the effect of 1) +2) +3). b. Economic Rates of Return 6. Adjustment reflects principally the higher 11% 12% unit investment costs and the slightly lower weaning rates. - 33 - 4.57 This is not the place for a lengthy digression on the calculation and use of rates of return for project evaluation, but two comments seem in order. First, producers faced higher input costs and lower beef prices than expected, but have benefitted from unexpected inflation. However, while inflation has assisted producers, this result does not alter the fact that several of the original assumptions crucial to the financial success of the project have been violated. The Bank attempted to exercise control over investment costs and beef prices, and both departed from their projected levels. So did inflation. It is only by fortuitous circumstances that individual borrowers are coming out, on balance, in satisfactory condition. It might have been reasonable to assume at the time the first project was planned that rapid inflation would not be a problem for there was almost no inflation in Ecuador during the latter half of the 1950s and the rate was quite moderate until 1967. However, inflation began to rise just: as the first livestock project was initiated, reaching 12 percent in 1972, and 25 percent in 1974, and producers are now being subsidized by negative real rates of interest. It was not possible for the Bank to have exercised control over inflation, per se, but it might have sought to introduce in- dexing on its loans. 4.58 Second, despite the emphasis placed on estimating the rate of return to the project, we feel that these rates underestimate the contribu- tion of the project to Ecuador's economic development. It appears that the two projects have gone some distance toward sensitizing non-project farmers to the economic advantages that can be achieved by new investments and improved management techniques and, to the extent that non-project farmers have adopted these techniques as a result of the,project, the true economic rate of return on the project is higher than that shown by the effect on borrower farms alone. Similarly, there is no doubt that the project has had a positive impact on the participating banks, especially the National Development Bank, on INIAP, and possibly on the Ministry of Agriculture. The strengthening of these institutions were important secondary goals specified by the project, and they should be given weight in the final evaluation. V. OTHER ISSUES A. Seed Multiplication and Development at Pichilinque 5.01 Loan 501-EC provided funds to permit the establishment of a seed multiplication and certification program at INIAP's research station at Pichilinque. These funds were available only for the introduction and studies of new species of pasture. This program developed more slowly than originally expected due to difficulties in obtaining mother seed, problems encountered in establishing facilities for the seed experiments, and delays in the - 34 - contracting of personnel to carry out the program, but it is now under way. Trial plots have been established to evaluate pasture grass and legume seeds which have been imported from other tropical areas and which have been gathered from plants growing naturally in Ecuador's Coastal zone. These experiments are expected to require a number of years before sufficient information is available to permit definitive recommendations to producers regarding the optimal types of grasses and legumes for use under different conditions, and before sufficient seeds can be produced for commercial sale. Experimentation with commercial seed production has recently begun and the cost of this process has been quite high. Pichilinque turned out to be unsuitable for producing seed because of climatic and soil conditions. The seed production project was then shifted to an area closer to the coast which, although biologically better suited, suffered from higher labor costs. INIAP has recently imported mechanical harvesters which will sharply reduce total labor inputs and, hopefully, make profitable operations possible. The third livestock project, 222-EC, provided funds for resident foreign advisors at Pichilinque in both the fields of pasture seed research and in animal production. These positions have been filled. B. Production Research, Training Centers and Monitoring System 5.02 Although the Bank provided assistance for a seed multiplication program in the first livestock project, not until the third project, four years later, were funds included to support research on improved investment and management technologies, and to provide courses for the diffusion of this information to producers and farm workers. Similarly, the Project did not develop a farm monitoring system to provide systematic feedback by which experience could yield operating improvements. Research undertaken at Pichilinque in recent years has identified a number of technical bottlenecks preventing the increase of calving rates above their current levels in Ecuador, and may permit a reasonably inexpensive approach to the removal of these constraints, assuming herd management can be further improved. These discoveries, plus the high investment costs encountered by the first and second projects, suggest that the Bank would have been wise to initiate a larger scale research program on cattle production at INIAP concurrently with the commencement of the first project, monitoring developments on a smaller number of project farms for several years before embarking on a larger lending program. C. Agricultural Survey 5.03 Funds were provided in the first livestock project to assist the taking of an agricultural survey in 1968. Previous information was limited to the 1954 CIDA survey and the A96LUgtq!ultural census, which was known to be internally inconsistent. In the 1968 survey, undertaken - 35 - by the National Planning Board, information was collected on 18,000 farms throughout Ecuador. Although this survey was also statistically imperfect, it provides better data than those previously available and has been the basis for agricultural planning in recent years. A full agricultural census was conducted in September 1974 and the census report is now being elaborated. D. Substitution of Bank Funds for Other Available Funds 5.04 It is difficult to determine the degree to which Bank funds sub- stituted for other funds available to project borrowers. It appears that substitution has not been substantial, although it is most likely to have occurred in loans made to larger and wealthier ranchers. These judgments are based on the following considerations. 5.05 Prior to the Bank's loan, long-term credits for cattle develop- ment in Ecuador were very limited. Agricultural loans made by the NDB from its own funds were generally limited to three years. Some longer-term loans up to seven years were available from IDB and Export-Import Bank funds, but these credits, which were available for cattle purchase only and not for infrastructure development, and which did not include technical assistance, were not widely used. The interest charged on these loans was not above that on the Bank's funds, so the provision of funds for infrastructure and the inclusion of technical assistance seem to have been crucial elements. 5.06 The Bank considered that infrastructure development was essential to the improvement of the cattle industry and it is clear that farmers accepted this judgment in drafting their investment plans. Approximately 65 percent of the farm investments made was for infrastructure or machinery and equipment. Therefore, no more than 65 percent of the funds could have substituted for credit available elsewhere, and the actual substitution seems to have been less. 5.07 As for the use of owned assets, Project Staff have estimated that approximately 70 percent of project borrowers could have financed their entire farm development plan had these borrowers been willing to liquidate other assets which they owned. Although this fact indicates that the Bank was working with a group of farmers who were already wealthy by Ecuadorian standards, we do not believe that one can conclude that the Bank's loans fully substituted for existing funds in these cases. It is not clear how much investment would have taken place in the absence of the Bank's project, nor how effective such investment could have been. 5.08 During the period in which Projects 501-EC and 173-EC were being implemented, there was no improvement in the price structure which made cattle ranching significantly more profitable than it had previously been, - 36 - and we have indicated that we do not believe the threat of land reform was a predominant factor in stimulating the investments which took place. It seems likely that it was the package of credit and technical assistance which brought about the investments which occurred, and this package was available only through the project. 5.09 There is some evidence that larger borrowers had made proportionately greater investments in their ranches than had smaller borrowers prior to the project. Among farms in the Egas sample, the larger ranches were the better capitalized and their owners must have had access to and a desire to use funds in order to achieve this position. Larger ranches also had a scale of operations which would have made the hiring of a capable administrator more feasible, and thereby made extraordinary technical assistance less important. Despite these considerations, the Egas study indicates that the impact of the project on ranch productivity coefficients was the greatest on the larger ranches in his sample, and not on the smaller or medium-sized ranches, and the marginal productivity of funds employed was largest, at least during the early stages of ranch development, on the larger ranches as well. These results, if generalizable, provide further evidence that the Bank's funds had an impact easily differentiable from other funds, even if some substitution, in the simplest sense, was occurring. 5.10 Although it seems unlikely that substitution was serious at the level of the individual farm borrowers, the Bank was not as successful as it had hoped in increasing the supply of long-term credits to agriculture. The amount of agricultural credit increased by 60 percent between 1967 and 1971, but as a percent of total credit, it fell. Agricultural credit composed 14.5 percent of total Bank crVdit in Ecuador in 1966 and has never exceeded this percentage during the succeeding nine years. In 1973, the last year for which data are available, agricultural credit was 13 percent of total Bank credit outstanding. 5.11 More importantly, when the amount of agricultural credit is deflated by the GDP deflator, it is shown to have remained constant, or even declined slightly in real terms. E. Distributive Equity 5.12 Livestock Projects 501-EC and 173-EC were designed to assist com- mercial beef producers, expecting that within this group the impact on production of the credit and technical assistance to be provided would be maximized. The projects have had a significant impact on output, and it is improper to evaluate them on criteria which they were never expected to satisfy. Nonetheless, consistent with the Bank's concern for distributional considerations, we believe a number of observations are in order. - 37 - 5.13 First, cattle are produced on 233,000 farm units, or 37 percent of all Ecuadorian farms. Most of these producers maintain only enough animals (1 or 2) to provide for household milk needs, and were not feasible targets of any beef improvement program. However, the project was designed to embrace only 0.1 percent of cattle holders, these being among the largest and wealthiest individuals in the agricultural sector. The economies of scale inherent in the development approach adopted effectively restricted the program to producers having at least 200 animals, and generally re- quired more than this for profitable operation. Only about 1,300 producers in Ecuador have more than 200 head of cattle. The project made loans to 199 producers, each having about 450 animals at the time the loans were received. 5.14 The Bank might have adopted somewhat less infrastructure-intensive techniques and put more emphasis on achieving the benefits of improved management with less capital. In this way, the project,might have been of greater benefit to middle-sized producers. There are 5000 producers having between 30 and 200 animals each, and in their aggregate they possess an approximately equal number of cattle as do those having 200 animals or more. Thus, by affecting the medium-sized producers, the Bank could have had a significant impact on production as well as on distribution. 5.15 The project also might have placed more emphasis on improving dual-purpose herds, providing both beef and milk. The Bank felt that dual- purpose herds were less economic than specialized herds, but recent INIAP research has questioned this assumption. More importantly, smaller producers must worry about their source of daily income, provided by milking, and only by investigating dual-purpose production would it have been likely that the medium-sized producers could have been significantly aided. 5.16 Second, the project borrowers, already wealthy, have received substantial subsidies as a result of the credit which was offered to them with payments fixed in nominal terms. Higher interest rates and/or indexing of the loans might have avoided this problem. 5.17 Third, the larger ranches have mechanized their operations, where geographically possible, particularly in the weeding of pastures. The principal motivation is the attempt to reduce the labor force and avoid pos- sible labor conflicts. The amount of labor which this mechanization has displaced is not large, and ranch development in previously unutilized lands has somewhat compensated by the provision of new jobs. Still, given the need for employment creation in Ecuador's agricultural sector, the Project must receive low marks in this regard. - 38 - VI. BANK PERFORMANCE A. Appraisal 6.01 The first appraisal mission was sensitive to the principal problems faced by the project and worked out solutions which it felt would permit the project's successful implementation. In most respects, we feel the first Appraisal Report was adequate in its analysis of the situation and the project design selected. The basic character of the project has been pre- served and its impact, although less than predicted, has been significant. 6.02 The establishment of an independent project unit, operating under an autonomous Project Commission and a strong Project Director was foreseen as desirable and seems to have been very important to the success which has been achieved. The members of the Project Commission functioned competently to guide and assist the program, and most importantly, the Project Director selected a capable staff, commanded the respect of the Ecuadorians with whom he worked, and with them jointly was able to promote and implement the project as designed. 6.03 Most of the project's shortcomings with respect to the appraisal goals are related to issues which were identified, but whose effects were underestimated. Investment costs were underestimated. The report failed to allow for the temporarily depressing impact which the drought of the mid-1960s had on the prices of breeding animals, and also underestimated the stimulative effect of the demand generated by the project itself on these prices. It was overconfident that the Government would allow beef prices to increase, was overoptimistic regarding the amount of Government-provided technical assistance available in the Coastal zone, foresaw problems in the marketing of beef, but gave too little attention to these institutional needs, and overlooked the possibly harmful effects of inflation. Each of these problems has reduced the success achieved by the project but none was fatal. 6.04 It does appear that the Bank should have re-examined the situation in Ecuador more carefully before proceeding with the second livestock project. The economic outlook for new farm investments in 1969, at the time the second livestock project was approved, was not as optimistic as was portrayed. The on-farm physical responses to investments in the first project had been good, but there is little evidence to suggest that they were better than had been predicted, as was contended. And it was generally recognized that the as- sumptions which had been made in the first Appraisal Report regarding the expected responses to be achieved over the life of the project were still quite optimistic. Even if early year targets were being Net-,there wis a long way to go. Beef prices also seem to have been lower than predicted relative to input costs, not higher. The Bank ought to have pushed more strenuously to get the Government to increase the domestic price or recal- culated the expected rate of return, using lower prices. And more considera- tion ought to have been given to the possibility of reducing unit investment - 39 - costs through less capital-intensive techniques. These problems are not really confronted in the second Appraisal Report. The report makes arguments for the continuation of the loan, but more concern was warranted for a number of issues which were crucial to the economic results. 6.05 The fact that a number of important issues were overlooked in the second Appraisal Report would have been more acceptable, given its interim nature, had the same issues been more adequately treated in the third Appraisal Report. However, the third Appraisal Report did not deal with any of the issues listed above. It also underestimated the costs of investment, as subsequent experience has shown, though by more for dairy production than for cattle. It gave too little emphasis to the prevailing price structure and marketing institutions. And it failed to consider the harmful effects of inflation, although inflation had already reached 11 percent, a rate nearly equal to the interest rate of 12 percent being charged to sub-borrowers. In this situation it was necessarily going to be difficult to determine whether the demand for credit was a result of production opportunities, or simply a demand for credit subsidies. 6.06 The third Appraisal Report expanded assistance to include dairy production in the Sierra region, but did not mention the need for increased milk production in the Coastal zone. A number of borrowers in the coast, with project approval, have been importing dairy animals to increase milk production in their herds. The third Appraisal Report recommended herds for a pasture and beef cattle research program at Pichilinque, and the establishment of a beef ranch training center as well. These were badly needed, and the research has been expected to include dual-purpose animals. The third project provided assistance to improve the business operations of the NDB, and also planned that Ecuadorians would assume the Project Director- ship in two years. Several Project Staff members were thought competent to fill this position, and this has been proved correct; two have since served successfully as Project Directors. B. Supervision 6.07 A total of 16 staff from Washington have participated in the appraisal and supervisory missions which have occurred during the nine years (1966-1974) since the Bank initiated efforts in Project 501-EC, yet there has been considerable continuity in the use of staff on the project One of the members of the first appraisal mission became the first Project Director and remained on the job four and one-half years, providing close liaison with the Bank. Another member of the first appraisal mission served on each of the sUcteding--tWO-a vr ls asvsi ns- and on two supervision missions. Visits by other Bank staff permitted the familiarization of operational officers with the project and provided staff specialized in non-agricultural issues to assist in working out problems encountered in the implementation of the project. - 40 - 6.08 The supervision mission may be criticized, however, in that they concentrated much more on loan procedures, disbursements, repayments, and disputes with the Ecuadorian Government regarding credit terms, than on physical and economic developments at the farm level. Information on the first set of issues was more readily available, and perhaps these problems seemed more immediately pressing. However, several references are made in the supervision reports to the fact that farmers are eager to get additional loans, or to want lower interest rates, and no systematic information is given whether such positions have any economic foundation. Had the supervision missions pushed for better information on developments at the farm level, and considered the implications of changes in several crucial macro variables, it might have been possible to avoid or alleviate a number of difficulties which did arise. VII. CONCLUSIONS 7.01 The first two Ecuador livestock projects provided large amounts of credit and intensive technical assistance to a small group of progressive ranchers, achieving significant increases in the technical parameters on project ranches, and an expansion in pasture and herd size. The audit suggests that the technical success achieved is due to a number of causal factors, each of which seems to have been important. These factors include: resident management of the ranchers, the availability of a technically feasible and financially profitable development plan, vigorous and competent program management, intensive technical assistance, and at least minimally satisfactory macroeconomic parameters. 7.02 It would be unsafe to assume that the results achieved on a small project such as this can be fully replicated in large programs where program management inputs are likely to be more diffuse. 7.03 Although impressive technical results were achieved, the costs of the improvements were much higher than forecast at appraisal and the estimated rate of return on the project is not significantly greater than those achieved on other livestock projects where the technical achievements and costs per unit have been less. The Ecuador livestock projects also were capital-inten- sive, requiring large-scale ranching for profitable operation, and have not had a progressive distributional impact, either through producer incomes, employment, or nutrition. 7.04 The project faced significant obstacles in accomplishing the goals actually achieved, and the success is that much more impressive. It now appears that increased emphasis on livestock research, and establishment of closer links between the Pichilinque research station and project farms, might have permitted more rapid improvements in farm operation and in project design. Recent research at Pichilinque suggests that less capital-intensive - 41 - technologies can be developed, reducing the costs of further increases in production. This research places less emphasis on pasture improvement, paddock division and tick control, and greater emphasis on the provision of .phosphorus supplements, the establishment of a controlled breeding season, and improved culling of unproductive breeding animals. Equally important, the establishment of training centers for farm owners, farm administrators, and farm foremen, by which available information on new production technologies can be transferred and emphasized more efficiently, also seems desirable. Finally, research on less capital-intensive technologies will make the program available to middle-sized producers, also important in aggregate livestock production, and research on the production of both beef and milk by dual- purpose herds will benefit both smaller producers and consumers in the Coastal zone. - 42 - APPENDIX TABLES Table A-1 CHRONOLOGICAL RECORD OF SUB-LOANS APPROVED FOR PROJECTS 501-EC AND 173-EC Percent of Loan Funds Disbursed Percent of Loans Made to Farmers Number of Total Total Value of Average Amount Total Year Quarter Sub-Loans 501-EC 173-EC By Quarter Cumulative Loans, S/. of Loans, S/. 501-EC 173-EC By Quarter Cumulative 501-EC 1968 I 6 4.5 2.7 2.7 6,714,800 1,119,133.33 7.6 4.9 4.9 II 35 26.5 15.8 18.5 25,341,800 724,051.40 28.5 18.4 23.8 III 22 16.7 9.9 28.4 12,243,000 556,500.00 13.8 8.9 32.2 IV 21 15.9 9.5 37.9 10,572,740 503,463.81 11.9 7.6 39.8 1969 I 10 7.6 4.5 42.4 7,781,600 778,160.00 8.8 5.6 45.4 II 19 14.4 8.5 50.9 11,478,000 604,105.26 12.9 8.3 53.7 III 8 6.1 3.6 54.5 5,033,200 629,150.00 5.7 3.6 52.3 IVL 9 6.8 4.1 58.6 6,603,200 733,688.89 7.4 4.8 62.1 1970 I 2 1.5 0.9 59.5 3,000,000 1,500,000.00 3.4 2.2 64.3 132 100.0 59.5 88,768,340 672,487.42 100.0 64.3 173-EC 1969 IV 1 1.1 0.4 59.9 278,800 278,800.00 0.6 0.2 64.5 1970 III 37 41.1 16.7 76.6 18,200,070 491,893.78 37.0 13.2 77.7 IV 5 5.5 2.2 78.8 3,345,440 669,088.00 6.8 2.4 80.1 1971 I 9 10.0 4.1 82.9 6,333,000 703,666.67 12.9 4.6 84.7 II 14 15.6 6.3 89.2 7,453,600 532,400.00 15.1 5.4 90.1 III 17 18.9 7.7 96.9 10,162,400 597,788.24 20.6 7.4 97.5 IV 7 7.8 3.1 100.0 3,460,000 494,285.71 7.0 2.5 100.0 90 100.0 40.5 49,233,310 547,036.78 100.0 35.7 222 100.0 138,001,650 621,629.05 100.0 - 44 - Table A-2 SIZE DISTRIBUTION OF SUB-LOANS FOR PROJECTS 501-EC AND 173-EC Class Interval: Average Total Value Percent of Amount of Loan No. of Percent of Amount of Loan of Loans Funds (Sucres) Loans Loans Made (Sucres) (Sucres) Disbursed I. Project 501-EC 0- 200,000 15 11.4 161,653 2,424,800 2.7 200,001- 400,000 42 31.8 310,462 13,039,440 14.7 400,001- 600,000 21 15.9 516,000 10,836,000 12.2 600,001- 800,000 21 15.9 657,452 13,806,500 15.5 800,001-1,000,000 7 5.3 906,628 6,346,400 7.2 1,000,001-1,200,000 2 1.5 1,130,000 2,260,000 2.5 1,200,001-1,400,000 6. 4.6 1,272,400 7,634,400 8.6 1,400,001-1,600,000 5 3.8 1,468,400 7,342,000 8.3 1,600,001-1,800,000 2 1.5 1,667,400 3,334,800 3.8 1,800,001-2,000,000 11 8.3 1,976,727 21,744,000 24.5 2,000,001-more --- 132 100.0 672,487 88,768,340 100.0 II. Projecq 173-EC 0- 200,000 6 6.7 177,840 1,067,040 2.2 200,001- 400,000 40 44.4 299,603 11,984,130 24.3 400,001- 600,000 17 18.9 523,988 8,907,800 18.1 600,001- 800,000 13 14.4 697,218 9,063,840 18.4 800,001-1,000,00 4 4.4 937,250 3,749,000 7.6 1,000,001-1,200,000 4 4.4 1,061,675 4,246,700 8.6 1,200,001-1,400,000 2 2.2 1,251,000 2,502,000 5.1 1,400,001-1,600,000 1 1.1 1,500,000 1,500,000 3.0 1,600,001-1,800,000 1 1.1 1,712,800 1,712,800 3.5 1,800,001-2,000,000 1 1.1 2,000,000 2,000,000 4.1 2,000,001-more 1 1.1 2,500,000 2,500,000 5.1 90 100.0 547,036 49,233,310 100.0 III. Projects 501-EC and 173-EC Integrated (Lqan Amplifications Added to Original Loan) 0- 200,000 16 8.1 168,190.00 2,691,040 1.9 200,001- 400,000 68 34.3 306,340W73 20,831,170 15.1 400,001- 600,000 35 17.7 502,542.86 17,589,000 12.7 600,001- 800,000 26 13.1 666,028.46 17,316,740 12.5 800,001-1,000,000 14 7.1 911,457.14 12,760,400 9.3 1,000,001-1,200,000 6 3.0 1,074,450.00 6,446,700 4.7 1,200,001-1,400,000 9 4.6 1,255,200.00 11,296,800 8.2 1,400,001-1,600,000 5 2.5 1,462,400.00 7,312,000 5.3 1,600,001-1,800,000 3 1.5 1,679,200.00 5,037,600 3.7 1,800,001-2,000,000 11 5.6 1,966,472.73 21,631,200 15.7 2,000,001-more 5 2.5 3,017,800.00 15,089,000 10.9 199 100.0 696,978.03 138,001,650 100.0 IV. Range, Average- Md Modal Loans 501-EC and 173-EC 501-EC 173-EC Integrated-/I Range of Loans S/.161,000-1,976,000 S/.177,000-2,500,000 S/.168,000-3,000,000 Average Loan S/.672,000 S/.547,000 S/.696,000 Modal Loan S/.300,000 S/.300,000 S/.300,000 /1 Includes repeater loans. Table A-3 DISTRIBUTION OF SUB-LOANS, BY PROVINCE AND PERCENT OF CATTLE HERD AFFECTED (1968 SURVEYI PROJECTS 501-EC AND 173-EC Province Loans Pichincha Guayas E3meraldas Manabi El Oro Los Rios Cotopaxi Azuay Total 501-EC Amount Loaned in S/. 11,270,640 28,606,800 8,995,600 14,426,900 5,325,800 18,666,600 1,476,000 88,768,340 % of Total Funds Loaned 12.7 32.2 10.1 16.3 6.00 21.0 1.7 100.0 173-EC Amount Loaned in S/. 7,302,480 16,460,200 5,983,530 9,241,100 1,392,000 7,810,000 792,000 252,000 49,233,310 % of Total Funds Loaned 14.8 33.4 12.2 18.8 2.8 15.9 1.6 0.5 100.0 Total 501-EC and 173-EC Amount Loaned in S/. 18,573,120 45,067,000 14,979,130 23,668,000 6,717,800 26,476,600 2,268,000 252,000 138,001,650 % of Total Funds Loaned 13.5 32.7 10.8 17.1 4.9 19.2 1.6 0.2 100.0 Number of Farms- a bcd a bcd a bcd a bcd a bcd a bcd abcd abcd a be d 501-EC 17 1 26 0 23 1 25 0 14 0 0 25 2 2 0 132 4 173-EC 16 1 4 19 0 6 11 0 1 25 0 3 5 0 12 0 4 1 0 0 1 0 90 1 18 Total 501-EC and 173-EC 33 27 45 39 34 32 50 47 19 19 37 31 3 2 1 1 222 199 Number of Cattle Total number of cattle in province, pre-development 249,891 303,111 136,606 406,387 66,519 109,383 132,886 156,564 1,561,346 Total number of cattle on participating farms, pre-development 9,921 30,978 10,143 23,717 4,889 14,425 1,117 145 95,335 Participant herd as percent of total 3.97 10.22 7.43 5.84 7.35 13.19 0.84 0.09 6.11 /1 Key: a. Total number of loans: per project and with projects combined. b. Number of repeater loans within each project. c. Number of borrowers from 501-EC receiving loan amplification from 173-EC. d. Total number of individual borrowers. Table A-4 TEMPORAL DISTRIBUTION OF SUB-LOANS APPROVED, BY PROVINCE, PROJECTS 501-EC AND 173-EC 501-30 Date Quarter Guayas Los Rios El Oro Pichincha Esmeraldas Manabi Cotopaxi Azuay Total 1968 I 4 2 - - - 6 6 II 9 7 7 2 5 5 - 35 III 3 1 2 6 6 - 22 IV - 3 1 8 3 5 1 21 1969 I 1 3 2 2 - 2 - 10 1r - 4 3 1 6 5 - 19 III 5 1 - - 1 1 - 8 IV 2 1 - 2 2 1 1 9 1970 I 1 1 - - - - - 2 Total 26 25 14 17 23 25 2 0 132 173-EC 1969 1 1 1970 III 10 - - 8 7 12 - 37 IV 1 1 - 2 - 1 - 5 1971 I 3 3 1 - 1 1 - 9 II 3 1 2 1 1 6 - 14 III 5 3 2 3 2 1 1 17 IV - 3 - - 1 3 - 7 Total 22 11 5 15 12 24 0 1 90 Table A-5 ROLE OF PARTICIPATING BANS, PROJECTS 501-EC AND 173-EC No. of Loans Percent of Loans Funds Loaned Percent of Total Average Loan Size (Million Sucres) Funds Loaned 1. 501-EC NDB 117 88.6 69.60 77 595,000 COFIEC 15 11.4 20.88 23 1,392,000 132 100.0 90.48 100.0 2. 173-EC NDB 87 95 44.78 89 515,000 4-1 COFIEC 5 5 5.47 11 1,094,000 92 100.0 50.25 100.0 J. Total (501-EC and 173-EC Combined) NDB 204 91 114.38 81 560,000 COFIEC 20 9 26.35 19 1,317,000 224 100.0 140.63 100.0 - 48 - Table A-6 COMPARISON OF UNIT INVESTMENT COSTS AS FORECAST AND AS REPORTED IN FARM DEVELOPMENT PLANS, PROJECTS 5o-EC AND 173-EC Forecast 01-EC 173-EC Total Investment SZ11,4OO,OOO s,/64,o0o,oo0 Number of Participating Farms - 132 90 Average Investment/Farm SZ480,000 S' 870,000 S,K 700,000 % % (Units Created) % (Units Created) Investment Composition and Number of Units Created Pasture Improvement 31.3 33.6 (49,784 ha) 32.5 (25,290 ha) Cattle Purchase 38.5 31.0 ( 6,916 animals) 34.5 ( 4,053 animals) Fencing 6.9 8.9 ( 2,780 km) 7.3 ( 1,167 km) Water Facilities 5.2 9.1 6.6 Machinery and Equipment 8.9 8.6 5.2 Corrals 4.4 2.5 4.6 Dips 2.3 2.3 1.5 Ranch Buildings - 1.6 3.8 Other 2.5 2.4 4.0 100.0 100.0 100.0 Cost/Unit Created Pasture (ha) S/. 333 /. 770 S/. 825 Breeding Cattle (animals) S/.2,h0 S/.5,100 S/.5,450 Fencing (km) s/.3,660 S/.3,660 S/.h,750 - 49 - Table A-7 COMPARISON OF INVESTMENTS PER FARM AS FORECAST AND AS REPORTED IN FARM DEVELOPMENT PLANS, PROJECTS 5o1-Ec AND 173-EC % % (Units Created) % (Units Created) Total Number of Ha Improved or Created, Per Farm 450 377 281 Total Number of Breeding Animals Purchased, Per Farm 76 52 65 Total Km of Fences Built, Per Farm 11 21 13 Table A-8 COMPARISON OF AVERAGE FARM SIZE AS FORECAST AND AS REPORTED IN FARM DEVELOPMENT PLANS, PROJECTS 501-EC AND 173-EC % (Units Created) % (Units Created) Pre-Development Situation Pasture (ha) 650 433 344 Breeding Animals (Heifers and Cows) 238 222 244 Post-Development Situation (Year 4) Pasture (ha) 850 810/1 625/1 Breeding Animals (Heifers and Cows) 304 274 289 Marginal Investment Per Breeding Animal in Herd S/1,58o S/3,175) S/2,475 /1 The breakdown between hectares of pastures created and renovated is not known for the development plans. The figures shown assume that all funds spent on pastures created new pastures, an unlikely event, but this assumption produces the maximum pasture available post-development. - 50 - Table A-9 DATA FROM EGAS' STUDY OF FARMS IN SANTO DOMINGO - QUEVEDO AREA REFERRING TO INVESTMENTS AND COSTS, PROJECTS 501-EC AND 173-EC 1. Farm Size Categories I II III Total Investment Per Farm (Sucres) s/. 406,oo S/. 692,000 S/. 1,094,300 Hectares of Pasture Before Investment 106 209 612 After Investment 193 309 676 Change 87 100 64 Animals in Herd Before Investment 160 314 799 After Investment 221 483 1,115 Change 61 169 316 Cows in Herd Before Investment 55 99 248 After Investment 96 242 474 Change 41 143 226 2. Percentage Invested in Each Category According to Approved Development Plan Cows and Bulls 31.4 31.7 30.0 Pasture 26.5 25.5 32.4 Fences 10.2 12.1 8.1 Corrals 3.9 2.1 2.8 Dips 1.0 2.1 2.5 Machinery and Equipment 4.4 4.4 10.8 Other 22.6 22.1 13.4 100.0 100.0 100.0 3. Amount Invested Per Category as Implied by Total Investment and Composition of Approved Development Plan I II III Breeding Animals S/. 128,000 S/. 208,000 S/. 330,000 Pasture S/. 106,000 S/. 180,000 S/. 355,000 4. Cost/Unit, as Implied by Amount Invested and Actual Changes Realized on Farms Cows and Bulls S/. 3,100 S/. 1,470 E/. 1,460 Pasture Creation S/. 1,210 S/. 1,810 S/. 5,550 - 51 - Table A-10 DATA ON BEEF PRICES, INVESTMENT COSTS, OPERATING COSTS, AND INFLATION IN ECUADOR % Increase 1967 1969 1974 1967-74 Beef Prices (live, on the ranch, per lb.) S/ 2.22 (S/350-1972) S1 6.75 300 Investment Costs Breeding Animals (per animal) Cows S/ 2,000 S/ 3,000 S/ 8,000 400 Heifers S/ 2,000 S/ 3,000 S/ 8,000 400 Bulls S/ 7,500 S/ 15,000 S/ 30,000 400 Pasture Renovation (per ha) S/ 250 S/ 500 S/ 2,000 400 Fences S/ 3,000 S/ 4,000 S/ 12,000 400 operating Costs Wages, Stockman (per year) S/ 8,000 S/ 8,000 n.a. n.a. Wages, Laborer (per day) S/ 20 S/ 20 S/ 60 300 Pasture Maintenance (per ha) S/ 60 s/ 60 s/ 200 333 Fertilizer S1 250 S/ 300 S1 500 100 Inflation Wholesale Price Index 104.7 111.0 n.a. n.a. GDP Deflator 108.9 120.9 230 130  80ow MAP I *0 - TULCÅN X SÅN-GABRIEL XI e AIBARRA ..-2 OTAV ýLO CAYAMBE cM,NG NUT AoP- p 0 A AIPIDOMA IP PUY BA--IA ROAMBA A AU 0 TNA JIPIJAPN AZOGUESZ CNDA A ___~ ~ RIOSAMBA GECUADO _________ *ALAUJSI lp___ 6 MACAS P _ ZARUMA - - ___} GENERAL LOCATION OF APPROVED LOANS UNDER 501-EC AS OF MARCH 31, 1969 EGUAYAs PROJECT AREA \O JA I. ~ IESMERALDAS AND SANTO DOMINGO ~~ PJECTUAROA E 1/ OCARIAMANGA . A EXPERIMENT STATION, PICHILINIGUE . NATIONAL DEVELOPMENT BANKS P. å PRIVATE BANKS: GUAYAQUIL. ..11 BANKS . IQUE VEDO.. 1 ANK. PERU /MACHALA...1 BANK 4. COFIEC --- PROVINCIAL BOUINDARIES p RANCHN LOCATIONS MAY 1969 80W MBRD-190 0R2