Mexico - Sixth Agricultural Credit Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 1968-ME STAFF APPRAISAL REPORT SIXTH AGRICULTURAL CREDIT PROJECT MEXICO May 1, 1978 Regional Projects Department Latin America and the Caribbean Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS US$1 = Mex$23 Mex$1 = US$0.0435 Mex$1 million = US$43,500 WEIGHTS AND MEASURES Metric System 2 1 hectare (ha) = 10,000 m = 2.47 acres 1 kilometer (km) 2 = 0.62 miles I square kilometer (km ) = 0.39 sq. miles = 100 ha 1 kilogram (kg) = 2.20 pounds 1 liter (1) = 0.26 gallons 1,000 kg = I metric ton = 0.98 long ton ABBREVIATIONS ANAGSA - National Crop and Livestock Insurance Agency ANDSA - National Warehouse Agency BANAGRO - National Agricultural Bank BANCO AGRICOLA - National Agricultural Credit Bank BANCO EJIDAL - National Ejido Credit Bank BANXICO - Bank of Mexico BNCR - National Rural Credit Bank CIMMYT - International Center for Maize and Wheat Research CONAFRUT - National Fruit Development Corporation CONASUPO - National Marketing Corporation CNG - National Livestock Producers Federation FEFA - Special Agricultural Credit Trust Fund FEGA - Technical Assistance and Loan Guarantee Trust Fund FINASA - National Sugar Finance Agency FIRA - Agricultural Trust Funds in the Bank of Mexico FONAFE - National Ejido Development Fund IDB - Interamerican Development Bank FONDO - Trust Fund for Crop, Livestock and Poultry Credit IMPA - National Sugarcane Research Institute INIA - National Crop Research Institute INIP - National Livestock Research Institute LIP - Low Income Producer NAFINSA - National Financiera PIDER - Investment Program for Rural Development PRONDAAT - National Program for Agricultural Development in Rainfed Areas SARH - Ministry of Agriculture and Water Resources SRA - Ministry of Agrarian Reform UNPASA - Sugar Producers' Organization GOVERNMENT OF MEXICO FISCAL YEAR January - December  FOR OFFICIAL USE ONLY MEXICO SIXTH AGRICULTURAL CREDIT PRbJECT TABLE OF CONTENTS Page No. I. THE AGRICULTURAL SECTOR A. Agriculture in the Economy ....................... I B. Rural Incomes, Employment and Land Distribution .. 1 C. Output Trends ..... 0.............. ............................ 3 D. Agricultural Research and Extension .. ............ 6 E. Agricultural Development Strategies .............. 7 II. AGRICULTURAL CREDIT A. Agricultural Trust Funds in the Bank of Mexico ... 9 B. Public Sector Agricultural Banks ..... ............ 11 C . Private Banks . . ................. .......... ........... .... 12 D. Performance Under Previous Agricultural Credit Projects ....................................... 12 III. THE PROJECT A. Introduction ...... .0................... ** .................. 15 B. Brief Description .o, ........... .. ........... 15 C. Detailed Features ........ .. .. . . . . . . . ........ ......... . 16 D. Project Cost ........ ................ . . .. ... . . .. . ....... 23 E. Financing ...... .... . .. .... . ................. ..... ....... 25 F. Procurement ......................................... 25 G. Disbursement ..... ................................ 27 IV. PROJECT IMPLEMENTATION A. Project Administration *.......................... 28 B. Sublending Policies and Procedures .... ........... 29 C. Accounts and Auditing ... ................ ......... 31 D. Monitoring and Reporting .................... 32 This report is based on the findings of an appraisal mission which visited Mexico during November 1977. The mission comprised D. Fitchett, J. Intrator, C. Moret, A. Duersten, V. Freeman and C. Wolffelt of the Bank and D. Stoops and C. Percival, consultants. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  TABLE OF CONTENTS (Continued) Page No. V. SPECIFICATIONS OF ENTERPRISE MODELS A. Livestock Development Models .. .................. 33 B. Crop Development Models .......................... 33 VI. PROJECT PRODUCTION, DEMAND AND MARKETING A. Production .................... ................... 36 B. Demand and Marketing ................... . o. . 37 VII. FINANCIAL ANALYSIS A. Farming/Ranching Operations ........ .............. 39 B. Project Cash Flow ................................ 39 VIII. ECONOMIC ANALYSIS A. 'Economic Benefits ...... ................. ......... 41 B. Economic Rate of Return and Sensitivity Analysis . 41 C. Project Risks .................................... 44 D. Environmental Impact ................... ....... 44 IX. SUMMARY OF ASSURANCES OBTAINED AT NEGOTIATIONS ........ 46 TEXT TABLES Table 3.1 Number of Subloans, Average Investment and Average Subloan ............................. 22 Table 3.2 Total Project Costs .24 Table 3.3 Project Financing .26 Table 5.1 Production Data for Livestock Models ..34 Table 5.2 Projected Development of Yields .35 Table 7.1 Producers' Benefits and Financial Rates of Return .......40 Table 7.2 Projected FEFA Receipts and Disbursements 42 Table 8.1 Economic Rate of Return Calculations .45  TABLE OF CONTENTS (Continued) Page No. ANNEXES Annex 1. Supporting Tables T.1 Principal Indicators of FIRA's Operations, 1971-1976 ...... 49 T.2 Credits for Agriculture through the Banking System and FIRA Participation . .............. . 50 T.3 Medium- and Long-Term Loans Rediscounted by FIRA, 1971-1978 ........................ ....................... 51 T.4 Comparative Statement of FEFA's Assets and Liabilities, 1970-1977 ............................................... 52 T.5 Comparative Statement of FEFA's Income and Expenditures, 1970-1977 ........................ ....................... 53 T.6 Consolidated Comparative Statement of FONDO/FEFA's Income and Expenditures, 1970-1977 ........... ........... 54 T.7 Stabled Dairy Production: Cash Flow Projection ........... 55 T.8 Grazing Dairy Production: Cash Flow Projection ........... 56 T.9 Dual Purpose Livestock: Cash Flow Projection ............. 57 T.10 Beef Production: Cash Flow Projection .................... 58 T.11 Pig Production: Cash Flow Projection ..................... 59 T.12 Rainfed Annual Crops: Cash Flow Projection ............... 60 T.13 Tubewell Irrigated Annual Crops: Cash Flow Projection ....... ................................... 61 T.14 Gravity Irrigated Annual Crops: Cash Flow Projection ..... 62 T.15 Walnut Grove: Cash Flow Projection ........... ............. 63 T.16 Cocoa Orchard Rehabilitation: Cash Flow Projection ....... 64 T.17 Sugarcane Production: Cash Flow Projection .. ............. 65 T.18 Cattle Slaughtering Plant: Cash Flow Projection .......... 66 T.19 Milk Processing Plant: Cash Flow Projection .............. 67 T.20 Feed Mill: Cash Flow Projection .... ..... 68 T.21 Fruit Packing Plant: Cash Flow Projection ................ 69 T.22 Logging and Sawmill Operation: Cash Flow Projection ...... 70 T.23 FIRA Training Program, 1978-80 ............ .. .............. 71 T.24 FIRA Demonstration Centers Program, 1973-76 ............. .. 72 T.25 FIRA Demonstration Centers Investment Program, 1978-80 .... 73 T.26 Weights and Prices for Livestock Development Models ....... 74 T.27 Prices Used in Economic and Financial Analysis ............ 75 Annex 2. Banco Nacional de Credito Rural ....................... 76 Annex 3: Selected Documents and Data Available in Project File.. 81 MAP IBRD 11789R - Distribution of FIRA Offices by Region   MEXICO SIXTH AGRICULTURAL CREDIT PROJECT A. Agriculture in the Economy Background 1.01 Agriculture played an active role in Mexico's economic development during the 1945-65 period, growing at a rate of more than 5% per annum; however, since the mid-1960s, the rate has slowed considerably. This reduc- tion is largely due to the poor performance of the crop subsector; livestock production growth, which accounts for about one-third of sectorial output, has been relatively stable around 4% p.a. The sector's share of GDP fell from 16% in 1960 to 9% in 1976. As a consequence of the reduced agricultural produc- tion growth rates, the agricultural foreign trade balance deteriorated during the 1970s, although it recovered sharply in 1976 due mainly to the jump in coffee prices. Since the early 1970s, large imports of maize, wheat and powdered milk have been necessary to offset the shortfalls in domestic pro- duction, and the domestic supply of sorghum and oil seeds also had to be augmented. On the export side, unfavorable world market conditions caused a decline in the export of cotton, tomatoes and beef cattle, while sugar exports fell to zero in 1976 due to the failure of output growth to match the increase in internal demand. Regional Production Patterns 1.02 The wide range of latitudes and altitudes in Mexico results in much climatic and regional variability. Moving North to South, the North is a semi-arid region, in which beef breeding herds produce weaner calves under range pasture conditions. Irrigation is highly developed in Sonora, Sinaloa, and Chihuahua, where cash crops such as wheat, cotton, beans, fruit, vegetables, maize and alfalfa for milk production are grown. Across the Central Altiplano in the states of Jalisco and Michoacan, maize, beans, sorghum and other food crops predominate, as well as milk production, owing to the proximity to the Mexico City and Guadalajara markets. In the states of Mexico, Puebla and Hidalgo, the Altiplano is temperate and dry, producing rainfed maize and beans, while small irrigated areas yield fruit and horticul- tural crops. Along the Gulf Coast, sugarcane, maize and sorghum predominate. The area devoted to cotton has declined because of pest problems, and sorghum has largely replaced the crop. To the south in Tabasco, beef, cocoa, bananas, and sugarcane constitute the principal crops. In the neighboring Chiapas region, maize, beans, sesame, and cotton predominate. B. Rural Incomes, Employment and Land Distribution 1.03 Of the total population of 62 million in 1976, about 23 million were classified as rural. The national labor force in 1976 was about 17  -2- million of which about 40% were engaged in agriculture, forestry and fishing activities. The rural population is made up of three distinct groups: ejido families, 1/ private farmers and landless families; the total of about 3 mil- lion is divided approximately equally among the three. However, the landless group is increasing due to the pressure of a 2% annual rural population growth rate on the limited supply of farm land. Preliminary data for 1975 indicate that about one-third of the rural families had an income of less than US$670. Rural income distribution is skewed, following the pattern of land holdings. It is estimated that in 1970, about 6% of all farms accounted for about 85% of sales of farm products while 91% of the farms contributed only 5% of the total. The upper 10% of the rural population accounts for about half of the rural income, while half the population has about one-sixth of the income. 1.04 Unemployment and underemployment are major problems in rural Mexico. Since 1950, Mexico has experienced remarkable demographic growth with increases in both life expectancy and fertility. Consequently, the population grew by about 3.4% during the 1960s and 3.3% in the early 1970s. From 1965 to 1970, agricultural employment declined at about 0.5% per year, and, although specific data are not yet available, it appears to have con- tinued to decline through 1974, thereby compounding problems of absorption of a growing labor force in the rural sector and accelerating migration into urban areas. 1.05 The tap root of the Mexican land reform program traces back to the 1910 revolution; land reform has been implemented in fits and starts over the intervening years and a wide variety of assessments as to its polit- ical, economic and social impacts and benefits are available. Some 16,000 of the approximately 20,000 ejidos have by now been granted definite titles of usufruct. The Ministry of Agrarian Reform (SRA) is implementing a plan to ensure beneficiaries of land reform the supplementary services, supplies, and credit necessary to bring the redistributed land to full productive use. The Ministry is expected to provide technical, promotional and organizational assistance, while credit and inputs will come from the public agricultural banks and marketing support from the National Marketing Corporation (CONASUPO). About 2,500 ejidos are already involved in this integrated promotional effort. 1.06 In the private farming sector, the Agrarian Reform Law establishes the maximum size of individual holdings for ranches as the area required to carry 500 animal units and forbids growing of commercial crops on ranchland. However, the Government has attempted to encourage output by granting 320,000 certificates of exemption for crop production and 1,300 certificates for livestock, covering 8.2 million ha. The possession of these legal titles is intended to stimulate investment in improving carrying capacity and develop- ing crop production in potentially productive land by ensuring that farmers would not suffer a reduction in land holdings due to such capital improvements. 1/ The term "ejido" is derived from the Spanish equivalent of the village "common." In present Mexican law, the ejido is basically a group of families with joint tenure rights to land. The members of the ejido are called ejidatarios.  This process is being expedited, but until a more permanent legal exemption is instituted, tenure uncertainty may act as a disincentive to increasing produc- tivity on private farms and ranches. C. Output Trends Crop Production 1.07 Of Mexico's total area of nearly 200 million ha, about 35 million ha are considered potential crop land. Until 1940, the total cultivated area was about 7.5 million ha, located mainly in the highlands 1/ surrounding Mexico City and Guadalajara. During the 1945-55 period, substantial increases in both area and yields determined the rapid growth of sectorial output. The aggregate (irrigated and rainfed) cropped area in the then five major crops (maize, beans, wheat, cotton and sugarcane) increased from 7.2 million ha in 1945 to 9.7 million ha in 1955, i.e., 3.1% per annum. Yields per hectare also tended to increase, appreciably for cotton, wheat and beans (at annual rates of 5%, 4.5% and 4%), less for maize (2%) and negligibly for sugarcane. During the 1955-65 period total crop production grew at an annual rate of 4.4%, 40% lower than the preceding period. 1.08 The gross value of crop production has grown moderately -- about 11.5% -- during the 1965-74 period, in spite of a contraction in cropped area of 3.2%, implying a yield increase on the order of 15% during the decade. The major area reduction occurred for grains; both maize and beans decreased appreciably and the expansion of sorghum -- while significant -- did not offset this trend. Oilseeds cultivation rose rapidly, almost doubling, particularly due to increases in soybean and safflower. The area in indus- trial crops, cotton and sugarcane fell by 16%, due to the reduction in cotton acreage. 1.09 Maize and wheat are the major cereals in the Mexican diet, while sorghum is utilized as an animal feed. Maize is the most widely cultivated crop, with some 8 million ha accounting for approximately half of the country's cropped area, but output has been disappointing. Thus, the Government has moved to increase support prices for it and other staple grains to encourage production. Given the increased efforts by Government, and potential for area and yield expansion, an annual overall growth rate for crop output of 4% to 5% from 1977-82 would appear feasible. However, it is probable that expansion of maize, beans, and sugar production will still be less than the expected population growth rate of 3.3%; thus, deficits are anticipated through 1980. 1/ These highlands comprise the Central Plateau or Altiplano incorporating the states of Queretaro, Guanajuato, Hidalgo, Mexico, Morelos, Tlaxcala, Puebla and the Federal District.  -4- Livestock Production 1.10 Beef. Mexico has three major beef cattle producing regions: (a) the arid and semi-arid northern rangelands; (b) the wet tropics along the Gulf Coast; and (c) the dry tropics along the Pacific Coast. The first is used mainly for extensive cattle breeding, with stocking rates ranging from 5 to 50 ha per animal unit and income derived mainly from the sale of feeder cattle and cull cows. The area supplies the US market with feeder r.-Ltle (male weaners) and boneless beef. The dry and wet tropics produce beef almost exclusively for domestic consumption, with the wet tropics devel- oping into a major area for cattle fattening and the main source of supply for the Federal District. Because of the high initial fertility of the soil, abundant and well distributed rainfall and the high quality of the pastures, stocking rates in the wet tropics are one to three animal units per ha. 1.11 The national cattle population grew at a rate of 2.6% between 1965 and 1975 while carcass meat output increased from 2% p.a. between 1960 and 1965 to 5.2% p.a. between 1970 and 1975. Average carcass weights rose from 152 kg in 1960 to 166 kg in 1976. Over this same period, annual beef consump- tion varied from about 9 kg to 11.5 kg per capita, depending upon exports of feeder cattle and boneless meat exports. 1.12 The volume of annual exports of live cattle and boneless beef to the United States has fluctuated, depending upon the export price and the export quota. The latter is regulated by the Government and the National Livestock Confederation (CNG) in order to ensure adequate beef supplies for the Mexican domestic market. The volume of exports of feeder cattle depends on the difference in beef prices between the United States and Mexico. Exports of feeder cattle, which averaged around 700,000 head annually between 1966 to 1975, peaked at nearly I million head in 1972, but the depressed market prospects for Mexican beef in the United States in 1974 resulted in a drastic decline to around 200,000 head in 1975. Exports for 1976 climbed to 500,000 head and, with buoyant market prices, exports for the 1977-82 period are expected to average around 600,000 head per year. Boneless beef exports should average around 27,000 m tons annually. 1.13 Milk Production. Official statistics for 1975 estimate the national dairy herd at 10 million head. Total milk production in 1975 was estimated at 4.8 million tons produced by about 4.2 million cows, averaging 1,150 liters per cow. Nearly 80% of total production is from about 2 million dairy cows located predominately in the central plateau temperate zone. Here, some 0.6 million Holstein cows, mainly in commercial herds, produce 40% of total production. Another 40% of national production in this zone is from a very large number of small producers owning about 1.4 million cows, while the remaining 20% is produced by around 2.2 million dual purpose cows located in the tropics. Domestic production provides about 85% to 90% of the national consumption while the balance is imported. 1.14 Traditionally, commercial herds feed on irrigated alfalfa in the spring and summer and on forage oats and corn silage, generally supplemented with high levels of concentrate, in the autumn and winter. Although average  production under the system is reasonable at 11 liters/head/day, the cost of production is high because of the concentrates. However, studies by tech- nicians of the SARH (Ministry of Agriculture and Water Resources) and FIRA (Agricultural Trust Funds in the Bank of Mexico) in both temperate and tropical areas have shown that costs can be lowered by as much as 30% by grazing herds on permanent legume-based pastures. Further, FIRA has financed a number of pilot ejidos to demonstrate the financial viability of using the new tech- nology, but there remains ample scope for more applied research to determine appropriate pasture mixtures, fertilizer and water requirements, and grazing management methods. 1.15 Pork. The slaughter of pigs grew at an annual rate of 10% during 1970-75, to 6.5 million in 1975. About one-half of the Mexican swine popu- lation is located in the central plateau region, where commercial swine production has been concentrated because of the temperate climate and proximity to the main consumption centers. In recent years, production has spread to the northern states as a result of increased availability of lower cost sorghum and is also increasing in Yucatan in response to a high consumer demand. Large commercial producers usually combine breeding and fattening operations under feedlot management. Hog producers sell either directly to small bacon and canning plants or to dealers who are authorized to slaughter hogs at the municipal slaughterhouses and then sell to butcher shops and supermarkets. There is, however, a need for specialized slaughterhouses for hogs in order to improve hygiene standards and to utilize byproducts more effectively. 1.16 Pork consumption was about 9.6 kg per capita in 1975 with the domes- tic market normally absorbing all production, about 90% in the form of fresh meat. The balance is used for cold meats and by industry. Only when the internal price declines is there some export of pork carcasses. Hog prices have increased steadily since 1970 under the influence of rising grain prices. The average producer price for pork from 1970 to 1973 was 13% less than for beef, but since then the prices have tended to equalize. Expansion into new areas in the northeast and northwest for sorghum production and the growing demand for pork indicate good prospects for increasing production. Agroindustries 1.17 The food processing industry (excluding beverages) in Mexico employed 337,800 persons and had a total output value of Mex$86,600 million in 1975 -- 18% of the total value of industrial production. The growth of output has tended to match that of the industrial sector overall in recent years. During the past 15 years, relatively rapid expansion has taken place for meat plants, grain mills and silos, fruit and vegetable packing, storage and processing plants and vegetable oil processing. The Government has provided numerous financial incentives and financial support for both private operators and groups of private or ejido producers to initiate and/or expand these activities. These enterprises have been important in providing addi- tional employment opportunities in smaller cities and rural areas. In view of growing domestic demand for processed agricultural products and Mexico's access to the growing United States market, there should be continued strong growth of agroindustrial activities.  -6- D. Agricultural Research and Extension Research 1.18 Despite the international reputation for the successful breeding of hybrid maize and wheat strains, most of the benefit of agricultural research in Mexico has been taken by those private farmers and ejidatarios with irrigated land. Constrained by lack of funds and staff, the Mexican agricultural research system has, until relatively recently, had little impact on the rainfed cultivation activities typically carried out by low- income Mexican farmers. The National Institute for Crop Research (INIA) has received substantially increased financial allocations in recent years, rising from Mex$44 million in 1970 to Mex$283 million in 1975 and Mex$350 million in 1977, and thus has been able to expand the research staff and improve its caliber. Functions have been delegated to regional centers so as to make the work more region-specific and problem-oriented, and changes have been made in the distribution of its budget. For example, larger allocations have been made for research in forage and legumes to overcome a major technological bottleneck in Mexico's livestock development. While significant improvements have been made, there is still need to further upgrade academic standards of staff, strengthen the research effort in rainfed areas, integrate individual research findings into technological packages for specific ecological regions, and improve coordination with the extension service. 1.19 Livestock research, headed by the National Institute of Livestock Research (INIP), I/ is still concentrated on genetics, veterinary medicine, preventive vaccine, and tick control, but greater emphasis is now being placed on animal-production-management problems, including pasture management and stocking rate experiments, both of which are crucial for Mexico's livestock industry. Still there is need for more work on the production potential of sown and natural pastures under alternative fertilizer regimes and grazing techniques. Also, it is necessary to remove two basic impediments to an expanded and more production-oriented livestock research program: the rela- tive scarcity of funds allocated for the purpose and the lack of adequate research coordination between INIA and INIP in livestock production. Extension 1.20 Organized extension services provided through SARH, State Govern- ments, FIRA, public agricultural banks and some private organizations have expanded considerably in recent years. SARH staff alone has gone from 367 extensionists in 1970 to 5,200 in 1976, while FIRA's technical staff numbers 716, public agricultural banks employ about 1,800 technicians, and private banks about 425. The number of farmers served by SARH technicians has increased from 202,000 in 1971 to 1.5 million in 1976 and the area covered from 859,000 ha in 1971 to 5 million ha in 1976. This was accompanied by upgrading the service to a general directorate; increasing salaries; and taking measures to make the staff more mobile, to strengthen the link to research, and to work increasingly through farmers' groups. 1/ INIP's budgetary allocation for 1977 is about Mex$120 million.  -7- 1.21 However, even if allowance is made for farmers who are not potential candidates for extension, either because they do not require it or are not likely to benefit from it, there is need to extend the scope of the services, which, in the case of SARH, covers only 20% of the farm area. Further steps have also bec- taken to (a) overcome deficiencies in internal coordination and delays in administrative procedures of the services, (b) reduce staff turnover and improve the professional standards of the extensionists, (c) make greater use of crop demonstration and extension aids and communication techniques, (d) derive recommendations from region-specific production development programs and (e) promote farmer participation. SARH established the National Program for Agricultural Development in Rainfed Areas (PRONDAAT) in 1974 as an alter- native approach to its traditional extension methods. The program is based on experiences with the Plan Puebla and it is supported also from the Graduate College of the Chapingo Agricultural University. In essence, the PRONDAAT method seeks to give the extension agent a broader development role by directly involving him in the organization of farmer level institutions in marketing and in the provision of credit and inputs. The approach is supported by an incentive program of adaptive field trials and evaluation designed to produce technical packages for each localized ecological zone. The program is oriented to small, low-income farmers; initial results have been most promising and SARH is now planning a considerable expansion of the program which could lead eventually to the adoption of the PRONDAAT approach by the whole of the extension service. One of the first objectives of this expansion is to provide PRONDAAT services to all micro-regions of Government's Investment Program for Rural Development (PIDER). E. Agricultural Development Strategies 1.22 In recent years, the Government has sought three main objectives through its agricultural policy: (a) self-sufficiency in basic foodstuffs; (b) increased production of export crops; and (c) higher incomes, improved living standards and greater employment opportunities for the rural poor. With the production shortfalls in basic grains since the mid-1970s, the objective of self-sufficiency appears to be a dominant concern of current policymakers. In view of these priorities and with favorable prospects for increased petroleum production and export earnings, Mexico is in a posi- tion to structure its agricultural development strategy in order to put the sector on a more competitive basis, provide access to modern farming to a higher proportion of its rural population, and return to the higher sec- torial growth rates of earlier periods. The major elements in this strategy would be the opening up of new areas suitable to rainfed agriculture, partic- ularly in the humid-tropical zone, as well as rehabilitation and expansion of the irrigation and drainage infrastructure, and the modernization of farming techniques in existing areas, irrigated as well as rainfed. Investments in education, research and extension would be a cornerstone of this new approach and, where necessary, would be linked to the provision of physical infrastruc- ture, particularly access roads, land and water resource development, and storage and marketing facilities. This approach would permit fuller utiliza- tion of existing and incremental physical infrastructure.  -8- 1.23 One of the most successful contributions by the Bank to Mexican agricultural development in recent years have been loans for agricultural credit to the Bank of Mexico trust funds under FIRA. These projects, compe- tently administered, have expanded their scope from basically livestock development financing to include other activities, and from largely high- and medium-income farmers to low-income farmers as well. Bank strategy for lending in the agricultural sector will continue to build upon the achieve- ments of the FIRA program and the experience gained in the rural development projects presently under execution (PIDER I and II, Loans 1110-ME and 1462-ME, and Papaloapan Rural Development, Loan 1053-ME). In addition, increasing importance will be assigned to projects in the humid tropics and for small- scale irrigation works. Expansion of Cropped Areas 1.24 Humid Zones. Development of the rainfed agricultural potential of the Gulf Coast could make an important contribution to the expansion of sectorial output, farm incomes and rural employment opportunities. In addition, it would put Mexican agriculture on a more competitive footing; maize, sorghum, oilseeds, rice and sugarcane could almost certainly be grown more economically in this zone rather than in irrigated areas of the northwest, permitting the latter to shift scarce land and water resources to higher value crops. Finally, new products like rubber, palm oil and cassava, as well as substantial increases in cocoa and livestock, would become available to meet primarily domestic demand. Rainfed agricultural development along the Pacific Coast, although of more limited scope because of differences in soils, topography, and rain- fall, could also make a significant contribution to improved conditions in the countryside and overall growth of the sector. A program of area expansion within the limits of available technical information, manpower and budgetary allocations should receive high priority in agricultural sector planning. The recently approved Tropical Agricultural Development Project will initiate a significant effort to intensify agricultural production in the tropical lowlands of Mexico, as well as support the improvement and strengthening of the agricultural research and extension services in the tropics. 1.25 Irrigation. There are presently about 5 million ha of irrigated land in Mexico, accounting for about 50% of the value of sectorial output. Some 2.7 million ha are organized into irrigation districts under SARH responsibility, while an additional 1.1 million ha are organized into small-scale irrigation units, also under SARH. The remaining approximately 1.2 million ha of irrigated land are under private operation, although subject to the provisions of the National Water Law (1971). During the period 1972-75, the Mexican Government carried out a National Water Plan (PNH), under a tripartite agreement with the UNDP and the Bank as executing agency. On the basis of the National Water Plan Report, investment in water resource development would continue to play a significant role in the expansion of agricultural output. Rehabilitation of some irrigation districts would be of high priority, accompanied by a continued emphasis on small-scale, less capital intensive irrigation programs and, in addition, further investment in large irrigation, drainage and flood control  -9- projects. Each specific program or project would be considered according to a careful and realistic assessment of the costs and benefits involved. Special attention would continue to be given to beneficiary charges adequate to cover not only all operation and maintenance expenditure, but appreciably higher proportions of investment costs as well. In addition, the adequacy of sup- porting services, particularly technical assistance to farmers, would be assigned a high priority. Education, Research and Extension 1.26 Programs to improve and expand the education, research and extension facilities will be key components for the future sectorial growth strategy. Improvements in these services are essential for increased productivity in existing areas (irrigated and rainfed) as well as for successful implementation of area expansion programs. The Mexican Government is keenly aware of the importance of these services, and further steps are being taken to strengthen them, as pointed out in paragraphs 1.18-1.21. II. AGRICULTURAL CREDIT A. The Agricultural Trust Funds in the Bank of Mexico 2.01 FIRA is the common name for the trust funds established by the Federal Government with the Bank of Mexico: (a) the Fondo de Garantia y Fomento para la Agricultura, Ganaderia y Avicultura (FONDO): (b) Fondo Especial para Financiamientos Agropecuarios (FEFA); and (c) Fondo Especial de Asistencia Tecnica y Garantia para Creditos Agropecuarios (FEGA). 2.02 The establishment of FONDO in 1955 as a trust fund in the Bank of Mexico was a landmark in the development of agricultural credit in Mexico because, by providing a combination of credit and technical assistance to the public and private banks, the FONDO has contributed significantly to the growth in the volume of such credit and to its qualitative improvement in terms of sound appraisal and supervision practices. 2.03 The FONDO was created to encourage the private banking systems to participate more actively in financing the credit needs of the agricultural sector. FONDO discounted subloans made by private banks, allowing the latter to expand their portfolio of loans to the agricultural sector, and also, given the differential margin between the lending rate to farmers and FONDO's discount rate to the participating banks, permitting higher rates of return on the private banks' invested capital. The private banks bore the risk of lending and the responsibility for collection, but their risks were greatly reduced by the development of the FONDO's regional offices and staff, which carried out much of the technical analysis of prospective borrowers and projects.  - 10 - 2.04 Initiated with Government funds, the FONDO originally restricted its activities principally to short-term agricultural production credit, such as was needed to finance the annual crop cycle and the upkeep of live- stock herds. The FONDO's resources were later augmented by loans from AID and the Inter-American Development Bank (IDB), and it also began to discount longer term subloans for investments. The real movement into on-farm devel- opment lending occurred only after 1965, when a second trust fund (FEFA) was established within FIRA. This fund was established to channel foreign loans such as those from the Bank with corresponding counterpart funds in local currency, to medium- and long-term lending, financing investments in agricul- ture, livestock and agroindustries. 2.05 Apart from loans from external sources, both FONDO and FEFA receive financing from the Government and the Bank of Mexico. Government equity in both trust funds is increasing through time, because the principal of external loans is repaid by the Government and the local currency equivalent, deposited with the institution at the time of receipt of the external loans, is accumu- lating in the trust funds as additional equity to permit increased agricul- tural lending. Thus, under its current structure, FIRA has available funds from both domestic and international sources for rediscounting development subloans made by private and public banks to both commercial and low-income producers in the agricultural sector. FIRA also rediscounts to private banks short-term credits for basic field crop and livestock production. 2.06 FIRA has a central office in Mexico City, nine regional offices, 37 state offices and 87 local offices. Of the total FIRA staff of 1,176 in mid-1977, 716 were technical specialists. The majority (557) of these technical specialists are located in the regional, state and local offices. In addition to its own staff, FIRA has trained and/or designated as qualified to evaluate and approve subloans, 1,232 technical/credit specialists in parti- cipating banks. Of this total, 807 technicians work in public banks and 425 technicians work in private financial institutions. For a small propor- tion of the loans approved by private banks, the loan evaluation and technical assistance are carried out directly by FIRA staff. 2.07 The FEGA was created in 1973 as a fund to defray to participating banks part of the costs of loan evaluation and technical assistance associated with lending operations to ejidatarios and other low-income producers (LIPs) (up to 3% of amount of loan, and in special cases up to 6%), and to issue guarantees to private banks for the recovery of loans. Since its creation until September 30, 1977, FEGA issued guarantees amounting to Mex$2,377 million (Mex$1,373 million for short-term credits and Mex$1,004 million for long- and medium-term loans) and paid to banks on account of these guarantees Mex$18.7 million -- 0.8% of the total guarantees issued. This reflects the favorable subloan recuperation record of participating banks under the FIRA program; a survey in mid-1977 of banks accounting for 35% of FIRA's rediscount portfolio on Bank-financed loans shows arrears of only 0.8% of the outstanding subloan balances. During 1976, FEGA reimbursed participating banks Mex$23.5 million to cover technical assistance costs, and in 1977, about Mex$25.0 million will be paid by FEGA to banks for technical assistance. All payments made by FEGA on account of guarantees and for technical assistance as well as FEGA's administrative budget are paid by the Government.  - 11 - B. Public Sector Agricultural Banks Banco Nacional de Credito Rural (BNCR or Banrural) 2.08 The BNCR was created in 1975 through an amalgamation of three public sector banking institutions: the Banco Nacional de Credito Agricola, the Banco Nacional de Credito Ejidal and the Banco Nacional Agropecuario. On September 30, 1977 the fully paid-up capital and reserves of BNCR amounted to Mex$3,202 million. In 1976, total lending operations were Mex$20,400 million, 10% of which was for medium- and long-term development loans. The lending programs for 1977 and 1978 were Mex$24,300 million and Mex$28,000 million, respectively, of which 12% to 13% would constitute medium- and long-term lending. BNCR predominantly services low-income producers; 92% of its clients are ejidatarios and it accounts for 75% of the institutional lending to this group. 2.09 It will take some time for the BNCR to overcome numerous weaknesses and deficiencies inherited from its predecessor institutions. These problems are reviewed in Annex 2, which presents further background information on BNCR. A rehabilitation program has been initiated and the central management is making serious efforts to unify operational procedures and increase coordina- tion and efficiency. The Bank has had a continuous dialogue with the Government and BNCR management regarding the changes and improvements in its operations. In November 1977, the Government and BNCR agreed that all subloans would be classified at time of loan approval into loans which constituted normal bank lending operations and those whose recuperation was doubtful and which were being undertaken for social or other reasons, the latter to be granted from special Government trust funds to be established within BNCR for such purposes. Of BNCR's medium- and long-term loan portfolio at September 30, 1977, about 22% is more than 30 days overdue. BNCR has initiated a program to clear this portfolio of overdue loans through loan rescheduling, legal and other actions; a special department has been established for this task. In addition, im- proved reporting procedures on loan collection have been introduced to enable management to undertake necessary actions on a timely basis, while recently introduced higher interest rates and more aggressive collection of overdue loans should increase BNCR's current income. The BNCR program therefore should result in the gradual strengthening of the operational and financial structure of the institution. Financiera Nacional Azucarera (FINASA) 2.10 FINASA is a wholly Government-owned financial institution, assigned the sole responsibility for meeting all the credit needs of the sugar industry, both short-term and medium-term, for cane producers and sugar mills, including production, marketing and development. FINASA also operates, as an agent of the Government, the Sugar Trust Fund, which lends to sugar mills unable to provide normally required collateral, and the Sugar Price Stabilization Fund. Apart from a paid-up capital of Mex$1,000 million, the principal sources of FINASA's funds are borrowings from the public and from foreign and domestic banks.  - 12 - 2.11 FINASA's loan portfolio consists mainly of loans to sugar mills, for meeting short- and medium-term credit demands for mill operations, and for the farmers supplying cane to the mills, as well as for marketing and the expansion of existing or construction of new mills. Cane production is largely undertaken by ejidatarios and private smallholders; in 1974, 94% of the producers cultivated less than 10 ha of cane. All loans to cane producers, such as those to be provided under the proposed project, are channelled through the mills and, to the extent possible, are in kind (fertil- izer, tractor services, transportation, and the like) on the basis of an appraisal of needs by field staff of the mills, who also supervise these loans, with technical support from the national organization for sugarcane research (IMPA). The loans are recovered when the farmer delivers his cane to the mill, which is the sole purchaser of cane in its geographical jurisdiction. During the 1975-77 period, FIRA rediscounted Mex$146 million of medium- and long-term credit extended by FINASA for 37,500 ha and 13,500 beneficiaries. C. Private Banks 2.12 Until the establishment of the FONDO rediscounting facilities, commercial banks were hesitant to undertake agricultural term lending because of the short-term nature of the bulk of their resources, problems of collateral and their lack of familiarity with agriculture. This situation has appreciably changed during the last decade, largely owing to the support extended by FIRA in terms of funds and services. Many private banks have been establishing and rapidly expanding agricultural credit departments and hiring significant numbers of agricultural technicians; by the end of 1977, some 425 agricultural technicians in private banks had been authorized by FIRA to evaluate and approve subloans. The Bank of Mexico has also encouraged this trend by stipulating that an amount equal to 2% of the total deposits in banks be held in the form of agricultural loans to low-income producers. Another important development has been the reimbursement from FEGA for technical assistance costs and guarantees for small farmer loans (para 2.07). The banks have, as a result, increased their technical staffs and expanded their lending to farmers, including ejidatarios and other low-income producers. Nevertheless, their total loans outstanding for agriculture have increased only from Mex$6,582 million in 1970 to Mex$21,163 million in mid-1977, a modest expansion in real terms if allowance is made for inflation. Commercial banks have been making increased use of FIRA rediscount facilities. Private banks rediscounted Mex$674 million of their medium- and long-term lending in 1973, Mex$1,120 million in 1974 and Mex$1,758 million in 1976. D. Performance under Previous Agricultural Credit Projects 2.13 The proposed loan represents the sixth of a series of loans to Mexico to support lending for crop and livestock development. These first five loans totalled US$400 million, and an additional amount, approximately US$63.4 million, is being channelled through FIRA under two Bank financed  - 13 - rural development projects and the recently approved Tropical Agricultural Development Project. The first loan (430-ME, US$25 million, effective Oct. 1, 1965) provided funds for medium- and long-term on-farm investments to medium- sized and commercial farmers in the tropical zone, central plateau and northern Mexico, financing a variety of activities but largely livestock, annual crops and perennial crops, with a lesser amount for agroindustries. On-farm subloans amounted to some 5,500, with 80% of them having a value of less than US$6,700 equivalent, while 2% of the loans were over US$80,000 equivalent and accounted for 23% of the amount lent. Livestock activities accounted for 60% of the disbursements, annual crops for 19%, perennial crops 5%, and agroindustries 16%. 2.14 The favorable performance of FIRA under the First Project encouraged the Bank to process a second loan to maintain program continuity. The Second Project (Loan 610-ME, US$65 million, effective September 19, 1969) continued the activities initiated under the first loan and added a cattle development component for southeastern Mexico. In this project, 11,651 sub- loans were made, averaging US$17,000 equivalent each; the smallest 85% aver- aged US$6,400 and loans over US$80,000 were only 1% of the total and accounted for about 10% of the funds. Livestock activities absorbed 61% of the funds disbursed, annual crops, 28%; perennial crops, 8%; and agroindustries, 3%. 2.15 The Third Agricultural Credit Project (Loan 747-ME, US$75 million equivalent, effective September 1, 1971) was fully disbursed in April 1974 and closed on September 30, 1975. In this project, 12,326 subloans were made; 81% of these were less than US$20,000 equivalent, averaging US$6,500, while 6% were in excess of US$80,000 equivalent and accounted for 29% of the amount disbursed. Of the amounts lent, 58% were for livestock enterprises (four-fifths of these beef ranches), 24% was for annual crops, 8% for peren- nial crops, 7% for agroindustries, and 3% for mixed farming enterprises. 2.16 Subloans under the first three projects had included an appreciable number of modest size loans, and -- in the Third Project -- an increasing number of ejidos participated. The Fourth Agricultural Credit Project (Loan 910-ME, US$110 million, effective October 24, 1973), while continuing to sup- port the activities financed under the previous projects, included a specific component directed to low-income producers (LIPs), totalling 20% of project subloans. Total disbursements under the Fourth Project were US$303.5 million; disbursement of the Bank loan was completed by December 31, 1975, one and one-half years ahead of schedule. Overall, 10,881 subloans were made, aver- aging US$24,260. Of the total number of subloans, 1,910 went to low-income producers, averaging US$19,140. Individual smallholders received 1,118 sub- loans and groups (ejidos, sociedades and grupos solidarios), 792. The number of direct beneficiaries was 23,552 and the average subloan per beneficiary in the low-income producers' component was US$1,552. Of the total amount lent under the project, 54% was for livestock enterprises (two-thirds of this for beef ranches), 24% for annual crop farms, 7% for perennial crops, 13% for agroindustries, and 2% unclassified. 2.17 The Fifth Agricultural Credit Project (Loan 1217-ME, US$125 million, effective August 29, 1976) is currently being implemented by FIRA. It represents a continuation of the activities financed under the previous Bank  - 14 - loans for livestock, crop and agroindustrial development. It includes a substantially expanded low-income producer component, slightly over 40% of the lending program, and also provides for field demonstrations, farmer and staff training, project monitoring and pilot schemes for group farming enter- prises. The medium-income farmer component was virtually fully disbursed by February 1978. Bank disbursement for the low-income producer component was slow moving initially, as the Bank awaited clarification from the Mexican authorities on the operating policies to be introduced by the Banrural. The Banrural was fully incorporated into the Fifth Project in the latter part of 1977. It is anticipated that the proceeds of Loan 1217-ME will be fully disbursed by October 1978. 2.18 A Project Performance Audit (Report 1573, April 27, 1977) carried out by the Operations Evaluation Department on the Third Project (Loan 747-ME) and the Completion Report on the Fourth Project (Loan 910-ME) presently under preparation point out some achievements as well as some problems encountered in the Bank's agricultural credit program in Mexico. The institutional growth of FIRA has clearly been one of the major accomplishments of the program, with impressive expansion and up-grading of FIRA staff, decentralization of opera- tions, and support to participating banks, which has resulted in an effective use of loan funds and an effective mechanism for the channelling of credit to the agricultural sector. The economic rates of return on subloans finansed under the Third Project were estimated to fall into the 12% to 20% range. For subloans made under this Third Project, only 4% of the private participating banks' portfolio was overdue more than six months; arrears constituted 4.1% for the Banco Agropecuario, the only public bank participating in the project. Preliminary analysis of the subloan portfolio financed under the recently closed Fourth Project indicate that appraisal expectations have been met. Under the Fourth Project, a low income farmers component was initiated, designed to complement FIRA's on-going operations for medium and commercial producers. Commitment of funds to low-income farmers proceeded well and a slightly larger number of relatively smaller subloans were financed than forecast at appraisal. 2.19 However, certain common problems have also been encountered during the course of the implementation of the Third and Fourth Projects. The rates of return for livestock operations have tended to lie in the lower end of the range due, in part, to the tendency of sub-borrowers to pursue development programs with less emphasis on pasture intensification. A relatively high percentage of total loan funds financed livestock development, 58% under the Third Project and 54% under the Fourth Project; within these subloans 68% of investment funds were utilized to purchase breeding stock under the Third Project, and 66-69% under the Fourth. Such a pattern reflects in part the continuing farmers' uncertainty regarding implementation of agrarian re- form in Mexico, resulting in a hesitation to invest in long-term investments in land improvement. In addition, experience gained under both projects indicates that without intensive technical supervision of subloans, deviations from appraisal expectations for on-farm development may occur. Servicing of small farmers in particular has required an intensive degree of technical support and assistance from FIRA staff. The need for intensive technical supervision of subloans is clear, and FIRA has stretched its resources in  - 15 - attempting to prepare technically sound subloans as well as provide on-going technical services to farmers. The Fourth Project initiated the operation of a monitoring unit within FIRA; while some progress was made in estblishing this unit, problems were encountered in achieving an adequate coverage of the full range of fIRA investment activities. In addition, due to the rapid rate of disbursement of funds, data collected reflect a point in time relatively early in the implementation of farm investment plans, and thus, are not fully indicative of the impact at full development of subloans financed under each discrete project. However, progress has been achieved in the evolution of the system under the on-going program. III. THE PROJECT A. Introduction 3.01 The Government of Mexico has requested a Bank loan to help finance an agricultural credit project for medium- and long-term invest- ments in crop, livestock and agroindustrial development in a variety of enterprises located throughout the country, as well as supporting FIRA's activities in training, demonstration and technical assistance. The proposed Sixth Agricultural Credit would be a continuation of the credit programs initiated with the first loan granted in 1965. Under the preceding five projects, US$400 million were loaned to Mexico for crop, livestock and agroindustrial credit. Of the Fifth Loan (1217-ME) for US$125 million, US$83 million were disbursed by March 29, 1978, and it is anticipated that the remainder will be fully disbursed by October 1978. The feasibility study for the proposed project was prepared by FIRA staff and submitted to the Bank in October 1977. A list of studies and working papers relating to the proposed project is given in Annex 3. B. Brief Description 3.02 The proposed project would finance, over a three-year period, subloans for medium- and long-term investments channeled through private and public financial institutions, with the following objectives: (a) to increase the supply of crop and livestock products to meet the needs of a rapidly expanding national market; (b) to enable the agricultural sector to expand its contribution to export earnings and import substitution; (c) to improve the opportunities for employment in rural areas both on farms and in agroindustrial enterprises; and (d) to raise the incomes of farm families.  - 16 - 3.03 The project would finance technically and financially sound invest- ment plans for annual and perennial crop farms, beef, dairy and dual purpose ranches, and small agroindustries. About 13,600 subloans are expected to be committed and disbursed over approximately three years (Table 3.1). The project would also include finance for training, demonstration, technical assistance and monitoring activities carried out by the executing agency (FIRA) and the participating banks. 3.04 The countrywide project would be executed by FIRA, under arrange- ments similar to those which have proven to be successful under earlier projects, with credit channelled to farmers through some 150 participating private and public sector banks. FIRA would rediscount specific proportions of subloans disbursed by these private institutions to sub-borrowers. Respon- sibility for the preparation, evaluation and approval of subloans is largely vested in the technical staff of the participating institutions, who have been pre-qualified by FIRA. FIRA technical staff would provide back-stopping to the loan approval, technical assistance and subloan supervision activities of participating bank technicians. C. Detailed Features Project Area and Scope 3.05 The credit program would provide finance for a wide variety of technically and financially sound on-farm and agroindustrial investments, based on the experience gained from the previous and on-going projects, and the assessment of FIRA and other institutions of the demand for the different types of credit. About 47% of the sublending would be for livestock invest- ment, 43% for annual and perennial crops, and 10% for agroindustries. The farm, ranch and agroindustrial models described in the following paragraphs are illustrative of the types of investment to be financed under the project and reflect FIRA's actual lending program. In practice, there would be considerable variation in the size of the subloans, as well as in the invest- ment package financed, given the wide range of regional conditions and producer organization. The numbers of subloans for each model (Table 3.1) are therefore only indicative of the volume and pattern of the lending program. Livestock Lending Program 3.06 Physical and financial parameters for livestock activities have been obtained by analyzing FIRA's actual 1976 on-farm loan program and by updating costs and prices to June 1977 in order to derive average unit in- vestment costs. The production parameters are based on those employed by FIRA in evaluating the 1976 loan program. 3.07 Dairy - Stabled Feeding. This model (Annex 1, Table 7) represents the traditional system for milk production on commercial dairy farms, mainly concentrated in the temperate central plateau and, to a lesser extent, in the  - 17 - -rid and semi-arid irrigated north. Investments would include alfalfa estab- lishment, extensions to existing dairy buildings, water supplies, irrigation, milking and cooling equipment, agricultural machinery, and the purchase of in-calf Holstein heifers. Feed is harvested and fed to stalled animals. Alfalfa is fed as green forage during the spring and summer and corn silage and forage oats, as well as purchased alfalfa hay, is used during the winter. Concentrate supplements are fed at the rate of about 1 kg per 3 liters of milk. These farms generally have a limited area of land on which to expand feed production so that herd expansion is associated with a greater reliance on purchased feedstuffs. Under the development plan, farm-produced feedstuffs would decrease from a level of 75% before development to 60% at full develop- ment. As a consequence of the proposed investment and accompanying technical assistance, the calving rate is expected to improve from 70% to 75% and average milk production per lactation from 3150 liters to 3600 liters over a five-year period. The model is based on an operating unit of 30 ha; most likely subloans will be for units ranging from 20 to 60 ha. 3.08 Dairy - Grazing System. The model for investment in dairy produc- tion under grazing management (Annex 1, Table 8) represents a composite program projected for 500 farms comprising the conversion of existing stabled- type dairy farms (88%), the establishment of new dairy farms (8%), and the establishment of specialized heifer rearing farms (4%). The system represents a pilot low cost technology developed on FIRA demonstration farms to reduce the use of high cost feed concentrates. Investments would cover establish- ment of permanent pastures, remodeling or building new milking parlors, installation of milking machines and milk cooling equipment, fencing and water supply, purchase of agricultural machinery and purchase of in-calf Holstein heifers and heifer calves for rearing. On existing farms, four-year rotation of alfalfa would be substituted by higher yielding permanent pastures and winter forage oats by annual ryegrass, thereby lowering dependence upon purchased concentrates to 1 kg per 4 liters of milk. New farms would commence with the establishment of permanent pastures. In the model, farm-produced feedstuffs decrease from a level of 76% before development to 69% at full development. It is estimated that the calving rate would be raised from 70% to 75% and milk production per lactation increaed from 3100 liters to 3500 liters. Dairies served by this type of subloan would mostly be in the range of 25 ha to 80 ha; the model represents a 31-ha operation. 3.09 Dual Purpose Cattle. The dual purpose model (Annex 1, Table 9) is representative of the tropical regions and is a composite program for 3500 farms. About 80% of the ranches to be financed would require an investment of about US$17,500 per farm in new and renovated pastures, improvements to farm buildings, fences, water supply, milk cooling equipment and purchase of breeding cattle. The other 20% of ranches of this type require an investment around US$35,000 per unit in order to establish new farms and to upgrade to more specialized dairy production, involving additional investment in tropical pastures, irrigation, milking equipment and crossbred cows with improved milking qualities. The dual purpose farm utilizes natural pastures as the feed source, without using concentrates. In the wet tropics, production is oriented chiefly to the sale of ranch-bred fattened cattle and milk for local consumption. It is estimated that, on average, the stocking rate would  - 18 - increase from 1.3 Au/ha to 1.6 Au/ha by year 2, extraction rate from 18% to 22% by year 5, and milk production per lactation from an initial level of 450 liters to 750 liters by year 5. Subloans would largely be for ranches ranging from 80 ha to 200 ha; this model assumes a 120-ha ranch. 3.10 Beef Production. The beef model (Annex 1, Table 10) is a composite program proportionally representing the three major specialized beef produc- tion regions. Investments would include establishment and renovation of pastures and fences; improvements to farm water supply, dips, corrals and farm buildings; and the purchase of breeding cattle. In the northern and semi-arid region (20% of the subloans), output comprises feeder steers and boneless beef from cull cows, both products for the USA market. Ranches in the dry tropical Pacific (20% of the subloans) produce cull cows and feeder steers for fattening. The major component represented by the model is the Gulf Coast region (60% of the subloans) represented by the fattening component, including ranch-bred and purchased feeder steers. This Gulf Coast region is the major source of quality beef for the domestic market. It is projected that the proposed investment would lead to a 33% increase in stocking rate by year 8, weaning rate would be raised from 60% to 65% by year 4, and adult mortality rate would be reduced from 3% to 2%. These changes would lead to an increase in the annual extraction rate from 18% to 24% by year 7. The model assumes a 700 ha operation; most subloans would be for ranches in the range of 400 ha to 1,000 ha. 3.11 Other Livestock. An important part of the investment in other livestock would be for swine production (Annex 1, Table 11) and the remainder would mainly cover poultry, sheep and goat production. The swine model envisages a combined breeding and fattening operation based on purchased feed require- ments. The investments would be mainly in buildings appropriate for the different classes of stock, feed storage, water and drainage facilities, and purchase of breeding stock. The majority of loans are expected to continue to be in the "Bajio" (central plateau) region where commercial production has concentrated because of proximity to grain production and main consumption centers. Mainly through an increase in the annual farrowing rate from 1.6 to 1.8 litters per year, total pigs sold per sow per year would be expected to increase from 12 up to 14 by year 2. These changes would increase pork production from around an initial 890 kg to 1040 kg per sow annually. Agricultural Development Lending Program 3.12 Technical parameters for the agricultural activities are based on actual farm development projects financed in 1976, while costs and prices have been updated to June 1977, based on the most recent information avail- able at FIRA's regional offices. 3.13 Rainfed Annual Crops. This operation represents the prevailing traditional farming practices by which basic food crops, mainly maize and beans, are produced under rainfed conditions. The projected investment is intended to upgrade soils by means of subsoiling as well as to extend the effective cultivable area by about 20%. It would finance the purchase of farming equipment, the clearing of additional lands and the improvement of yields on the presently cultivated area. Through the investment program  - 19 - and application of improved farming practices yields of maize and beans are expected to rise from present levels of 1.6 tons/ha and 0.6 tons/ha to 2.1 tons/ha and 0.8 tons/ha, respectively, in a period of 3 to 4 years. The area served by this type of subloan would tend to fall in the range of 50 ha to 300 ha. The representative model is based on a 180-ha case (Annex 1, Table 12). 3.14 Tubewell Irrigated Annual Crops. This activity refers to the development of irrigation facilities in small units. The financing provided by the subloan would be complementary to investments in infrastructure and works financed by fiscal funds for the development of groundwater resources, a program which has been successfully executed by the Government for some years now. In addition to power installation and well equipment, the subloan would also finance other on-farm investments such as land preparation, buildings and agricultural machinery. This type of operation would take place throughout the country. The subloan would permit up-grading of the cropping pattern, at present based on traditional crops--wheat, beans, maize and sorghum--with an intensity of 80%, by the introduction of higher value crops, such as cotton and soybeans, as well as increasing the cropping intensity to 125%. Yields of wheat, beans and maize would be improved from present levels of 2.0 tons/ha, 0.7 tons/ha and 1.6 tons/ha to 3.2 tons/ha, 1.3 tons/ha and 3.0 tons/ha, respectively, while cotton and soybeans are expected to attain yields of 2.2 tons/ha and 2.0 tons/ha. Full development would be reached in about five years. This type of operation would mostly take place in ejidos, although it could also be on the aggregated area of a group of smallholders. Individual subprojects would most likely be between 200 ha and 600 ha; the analysis is based on a 400 ha model (Annex 1, Table 13). 3.15 Gravity Irrigated Annual Crops. This type of subloan would include financing of on-farm irrigation and drainage infrastructure, roads and land levelling, farm buildings, and agricultural equipment. Changes in cropping pattern would mainly involve a significant increase in area devoted to higher value crops as well as a more intensive use of the land, resulting in a cropping intensity of about 135%. Yields of beans, maize and soybeans would reach similar levels to those projected for the previous model; new crops, safflower and sorghum, are expected to yield 1.5 tons/ha and 3.5 tons/ha, respectively. Full development would be attained in about 4 years. It is envisaged that such subloans would be made in numerous irrigation districts, and would permit fuller use of the potential generated by main infrastructure works. Subloan beneficiaries would be either ejidatarios or groups of small- holders, mostly likely on extensions ranging from 250 ha to 600 ha. This analysis is based on a 400-ha model (Annex 1, Table 14). 3.16 Irrigated Orchard Development. This model describes the replacement of irrigated annual crops with orchards which, in addition to yielding higher returns, have substantially lower irrigation water requirements. The project would finance on-farm civil works, replacement of pumping and farming equip- ment, and the establishment costs of the new plantation. Walnut, a rapidly growing crop in the northern regions of Mexico with a well established inter- national market, has been selected as representative of this activity, which is expected to be undertaken mainly by private producers, although it may also  - 20 - be carried out by low-income producers within ejidos. At full development, yields would be similar to those presently obtained in existing orchards--1.5 tons/ha--and would be achieved in a period of about 12 years. The area for such subloans would most likely fall in the range of 20 ha to 60 ha. The representative model assumes 40 ha (Annex 1, Table 15). 3.17 Rainfed Orchard Rehabilitation. This activity refers to demand for credit to rehabilitate plantations of tropical fruit trees whose present yields are well below their potential. The subloan would finance the cost of replacing about 20% of the existing plantation, with a minimal component of capital goods; it would include planting materials, fertilizers and pesticides, minor farming equipment (sprayers) and labor input for orchard rehabilitation. Cocoa plantations have been chosen as an illustration; these are generally owned by small private farmers whose properties are in the 1 to 3 ha range and for whom they represent the main source of cash income. Yields are expected to reach 1 ton/ha over a development period of about five years. Farmers are expected to form groups with areas ranging between 50 ha and 150 ha. The analysis here is based on a 100-ha model (Annex 1, Table 16). 3.18 Rainfed Sugarcane Rehabilitation. Subloans would also be provided for the rehabilitation of sugarcane farms, mainly in the humid tropical zones. As the equipment needed for the replanting work is provided by the sugar mills under contract to the numerous small scale growers, subloans would cover the fees that mills charge the farmers for carrying out the works and furnishing the materials. With ratoon currently lasting six or more years, present yields are about 42 tons/ha of cane, with a sucrose content of 12%. With the subloans, the cane would be replanted once every four years by rotating 25% of the land from the fifth year onwards; this, together with improved agricultural practices would increase the average yield to about 67 tons/ha and would bring about an increase in sucrose content of one-half of a percentage point. Taking into account the more frequent replanting, full development would be attained in nine years. Beneficiaries of such subloans would be predominantly ejidatarios, but an appreciable proportion of private farmers, both low-income producers and others, are also expected to participate in this activity. Size of these sublending operations would most likely vary between 3000 ha and 5000 ha, depending on the respective mill's capacity and jurisdiction. The analysis here is based on a model with 3600 ha (Annex 1, Table 17). Agroindustries 3.19 Approximately 22% of the proposed investments for agroindustries would be related to the livestock subsector (slaughterhouses, milk plants, feedmills, Annex 1, Tables 18 to 20) and about 7% for crop processing, such as fruit grading and packing (Annex 1, Table 21). Provision has also been made for a few logging and sawmill operations (Annex 1, Table 22) to be established by ejidatarios in the forest areas of Guerrero, Durango and Chihuahua (8%). Items to be financed for agroindustrial enterprises include engineering design, site preparation, provision of utilities (water and power installations), buildings, machinery, and equipment. It is envisaged that about 300 subloans would be made for renovation and expansion of small existing agroindustrial plants. All the processing units are expected to be organized by groups of producers, small farmers and ejidatarios.  - 21 - 3.20 Table 3.1 summarizes the projected number of sub-borrowers, the average investment and subloan values and the volume of investment and lending under the credit program. Training, Monitoring. Demonstration and Technical Assistance 3.21 Training and Monitoring. FIRA would continue to provide training to its own staff as well as the staff of participating banks and the Govern- ment extension service. In addition, FIRA is playing an increasing role in conducting specific training programs for selected farmers. For its own staff, FIRA offers a two-year scholarship program for master-level training at national and foreign universities in the major areas where staff expertise requires strengthening. New staff, anticipated to be about 300 professionals during the execution of the project, attend a six-month course covering credit operations and including practical field experience on topics appropriate to their future responsibilities. Moreover, field staff attend short-term courses to keep them abreast of modern technology and farming practices as well as administration. For the staff of participating public and private banks, FIRA operates training programs at both the professional and technician level. It is anticipated that this latter group will play an increasingly important role in providing technical assistance to LIP beneficiaries. These courses are intended to upgrade and expand these staff and permit assigning them greater responsibility for subloan preparation and supervision and providing improved technical assistance to sub-borrowers. For farmers benefitting from subloans, FIRA operates a number of short-term courses, covering a wide variety of topics, and carries out a number of field day seminars at FIRA demonstration farms on the application of new production methods. Under the proposed project the Bank would finance the foreign exchange component of the graduate training program outside Mexico for approximately 240 technicians in the areas of crops, livestock, agroindustry and producer organization. Moreover, provision is made for financing up to 60 man/months of consulting services during the project to assist with local training and further improve- ment to dairy production grazing systems and 12 man/months of consulting services for the FIRA monitoring program. (The monitoring program is discussed in more detail in paragraphs 4.12 and 4.13.) In addition to these foreign exchange costs, the Bank would finance US$2.7 million equivalent or 35% of the local costs of the training program, so that its total financing of US$5.1 million will represent 50% of the total estimated expenditures of US$10.2 million for this component. Table 23 in Annex 1 summarizes the budget and throughput of the FIRA training program under the proposed project. 3.22 Demonstration. Through its close contact with production problems at the farm level and its activities in production oriented demonstration studies, FIRA has been an important complement to the Government's research and extension services. Initiated in 1973, the demonstration program comprised 39 centers by 1976: 16 for dairy production, 9 for forage investigation, 5 in horticulture and the remaining 9 in crops, irrigation, small animals and beef, strategically located in the zones of major present or potential produc- tion throughout the country. A main priority of FIRA's applied research and demonstration program has been the development of pasture/herd management systems to achieve improved levels of milk production per ha at costs below  MEXICO SIXTH AGRICULTURAL CREDIT PROJECT Number of Subloans, Average Investment and Average Subloan Estimated Number Investment Lending Program of Subloans Average Total Average Total Investment Activity - - - - Y e a r - - - Total US $ Subloan US $ 1 2 3 US$'000 million US$'000 million Livestock Stabled dairy production 500 550 650 1,700 28.3 48.2 24.1 41.0 Grazing "i 100 150 250 500 44.5 22.3 37.8 18.9 Dual purpose livestock 1,000 1,150 1,350 3,500 21.2 74.3 18.0 63.1 Beef production 600 700 800 2,100 22.8 47.9 19.4 40.7 Swine 200 250 350 800 20.4 16.3 17.3 13.9 Other livestock 225 275 300 800 24.0 19.2 20.4 16.3 Subtotal 2,625 3,075 3,700 9,400 228.1 193.9 Crops Rainfed annual crops 800 900 1,050 2,750 26.2 72.1 22.2 61.2 a Tubewell irrigated annual crops 30 40 40 110 231.0 25.4 196.4 21.6 Gravity irrigated " " 75 100 125 300 231.0 69.3 196.4 58.9 Walnut groves 25 25 - 50 85.6 4.3 72.8 3.4 Cacao rehabilitation 30 35 35 100 48.0 4.8 40.8 4.1 Sugarcane rehabilitation 3 4 3 10 1,591.3 15.9 1,352.6 13.5 Other crops 150 200 250 600 25.0 15.0 21.3 12.8 Subtotal 1,113 1,304 1,503 3,920 206.8 175.5 Agroindustries Cattle slaughterhouse 2 2 - 4 1,319.7 5.3 1,055.8 4.2 Milk processing plant 3 4 - 7 310.7 2.2 248.5 1.7 Feed mill 2 2 - 4 922.6 3.7 738.1 3.0 Fruit packing plant 3 3 - 6 552.5 3.3 442.0 2.7 Logging and sawmill operation 2 3 - 5 760.7 3.8 722.7 3.6 Other agroindustries 100 100 100 300 110.0 33.0 88.0 26.4 Subtotal 112 114 100 326 51.3 41.6 TOTAL 3,850 4,493 5,303 13,646 486.2 410.3 Note: Figures have been rounded. March 2, 1978 a  - 23 - those prevailing in stable-fed dairies. Recent results have been impressive and are being implemented by farmers on a pilot scale under the ongoing project, and sublending for such activities would continue under this project (paragraph 3.08). Continuation of the demonstration program under the Sixth Project will confirm and extend these results, provide an invaluable training and demonstration base for both technicians and farmers, and allow further collection of both physical input/output and financial parameters for program evaluation. Under the project, 13 existing demonstration centers for the grazing dairy system will be expanded and two new ones initiated; one existing beef production demonstration center will be expanded; and five existing forage demonstration centers will be expanded and two new ones initiated. Items to be financed under the Bank loan include constructions, pasture establishment, purchase of cattle, fencing, machinery, equipment and training aids. The Bank loan will finance 50% of the total non-recurrent cost of the demonstration centers programs, including the US$1.2 million estimated foreign exchange component and local costs equivalent to US$0.6 million. Table 25 in Annex 1 summarizes the FIRA demonstration center budget under the proposed project. 3.23 Technical Assistance. Under the previous and ongoing projects the technical assistance to project beneficiaries by FIRA personnel and the staff of the participating banks has proven to be a basic ingredient for project success. The role of such technical assistance has become increasingly impor- tant as the volume of sublending to low-income producers has expanded. The FEGA program (para 2.07) to reimburse participating banks for the staff costs of providing this technical assistance to low-income producers (LIPs) has been an appreciable incentive to lend to such groups and will continue to expand under the proposed project. The cost of providing technical assistance to low-income producers under the proposed project has been calculated by FIRA at the rate of approximately 3% p.a. of the outstanding balance of LIP sub- loans during the three-year disbursement period. These are by and large recurrent costs. The total outlay is estimated at about US$8.7 million equivalent during the span of the project. US$400,000 is included in the Bank loan to cover the estimated foreign exchange costs of providing the vehicles and training aids for FIRA to expand its technical assistance to low-income producers under the project. D. Project Cost 3.24 The total cost of the proposed project is estimated at approximately US$627 million equivalent, of which US$175 million, or 28% represent foreign exchange requirements. Baseline project costs were estimated as of June 30, 1977, with a 5% provision for physical contingencies in the various credit models. Expected international price contingencies were added at the rate of 4.5% for the second half of 1977, 7% for 1978, 6.5% for 1979 and 6% for 1980 and 1981. These price contingencies are based on Bank forecasts for interna- tional inflation and it has been assumed that the rate of exchange will fluc- tuate to compensate for any differences between the Mexican rate of inflation and the international rate of inflation. Cost estimates are summarized in Table 3.2.  MEXICO SIXTH AGRICULTURAL CREDIT PROJECT Total Project Costs"/ % Total % Foreign Components Local Foreign Total Local Foreign Total Project Cost Exchange --- Mex$ millions ------ -------US$ millions------ Farm Investments Livestock 3,795.0 1,451.3 5,246.3 165.0 63.1 228.1 45 28 Crops 3,348.8 1,407.6 4,756.4 145.6 61.2 206.8 41 30 Agroindustries 878.6 301.3 1,179.9 38.2 13.1 51.3 10 26 Subtotal 8,022.4 3,160.2 11,182.6 348.8 137.4 486.2 96 28 Training, Demonstration, Technical Assistance and Monitoring Training and monitoring programs 144.9 43.7 188.6 6.3 1.9 8.2 2 23 > Demonstration program 43.7 23.0 66.7 1.9 1.0 2.9 1 34 Technical assistance program 167.9 9.2 177,.1 7.3 0.4 7.7 1 5 Subtotal 356.5 75.9 432.4 15.5 3.3 18.8 4 18 Baseline Project Cost 8,378.9 3,236.1 11,615.0 364.3 140.7 505.0 100 28 Price Contingencies 87.9 34.3 122.2 .26L X2 Total Project Cost 452.2 175.0 627.2 124 28 1/ Figures have been rounded. March 29, 1978  - 25 - E. Financing 3.25 The proposed Bank loan of US$200 million would finance 32% of total project cost, representing the estimated foreign exchange cost of US$175 million and US$25 million equivalent of local costs (6% of local costs). Local cost financing from the Bank loan would cover US$21.7 million equiva- lent, or 9%, of the local costs of on-farm investments carried out by low-income producers, and US$3.3 million equivalent, or 32% of the local costs, of the FIRA training and demonstration centers program under the Project. Sub-borrowers under the lending program of the project would finance US$97 million equivalent, or 15% of the total project cost, the participating banks would finance US$86 million equivalent, or 14% of total project cost, and FIRA would finance US$244 million equivalent or 39% of the total project cost. In addition, participating banks would provide short-term loans to project sub-borrowers for necessary working capital. Table 3.3 summarizes the proposed financing plan. 3.26 The loan would be made to Nacional Financiera, S.A. (NAFINSA), a Government agency designated to borrow from the Bank, at the current Bank lending rate at the time of loan approval, for 17 years, including four years of grace. As under the Fifth Project, NAFINSA would on-lend the loan proceeds to FIRA on the same terms and conditions as those governing the Bank loan to NAFINSA, plus a service charge expected to be one-eighth of 1% of IBRD dis- bursements. The Government would assume the foreign exchange risk and re- sponsibility for the payment of interest and other charges in respect of the Bank loan proceeds used for training, demonstrations, technical assistance and monitoring. Suitable assurances as to the provisions outlined for proj- ect financing and availability of necessary working capital in the preceding paragraph and the on-lending stipulation of this paragraph were obtained dur- ing negotiations. F. Procurement 3.27 Since the agricultural lending activities would be implemented over a three-year period, be widely distributed geographically, cover a variety of farm/ranch investment activities, and involve a large number of sub-borrowers, bulk purchasing under ICB would not be feasible. An adequate selection of machinery, tractors and other agricultural equipment and inputs is available to sub-borrowers through local and international suppliers and permits can be obtained to import types of machinery and tractors in the horsepower ranges required if they are not available ii. Mexico or in the event of a real scarcity. Tractors up to 125 horsepower are manufactured in Mexico under franchises from five major international firms and adequate competition and satisfactory service and maintenance facilities are assured by the range of choice offered and the extensive network of dealers. Equipment for the demonstration and technical assistance programs would be procured through the Guarantor's ordinary procedures, which are acceptable to the Bank. The services of consultants to be financed under the project would be arranged in consultation with the Bank.  MEXICO SIXTH AGRICULTURAL CREDIT PROJECT Project Financing (US$ millions) Total Lending Participating Total Project Sub-borrowers Program Banks FIRA IBRD Cost Amount % Amount Amount % Amount % Amount % Amount % A. Low-income Producers Crops 3.7 5 71.3 7.1 9 32.1 43 32.1 43 75.0 100 Livestock 4.1 5 79.1 7.9 9 35.6 43 35.6 43 83.2 100 Subtotal 7.8 5 150.4 15.0 9 67.7 43 67.7 43 158.2 100 B. Other Sub-borrowers Crop 36.3 20 145.7 29.1 16 65.3 36 51.3 28 182.0 100 Livestock Development 40.2 20 160.9 32.2 16 72.4 36 56.3 28 201.1 100 Subtotal 76.5 20 306.6 61.3 16 137.7 36 107.6 28 383.1 100 C. Agro-industries 12.6 20 50.0 10.0 16 22.5 36 17.5 28 62.6 100 D. Training, Demonstration, and Monitoring Programs - - - 6.7 50 6.7 50 13.4 100 Technical Assistance - - - 9.4 96 0.5 4 9.9 100 Subtotal 16.1 68 7.2 32 23.3 100 Total Cost 96.9 15 507.0 86.3 14 244.0 39 200.0 32 627.2 100 Note: Figures have been rounded.  - 27 - 3.28 As in earlier projects, FIRA would require sub-borrowers to obtain quotations from several sources of supply, whenever practicable, for goods, works and imports of breeding livestock to be financed under subloans. Machinery and equipment for agroindustrial units would be purchased by private producers or groups of producers. Whenever the cost of a single item of equip- ment or an assembly delivered by a single manufacturer exceeded US$250,000 equivalent, quotations would be solicited from not less than three suppliers. Moreover, FIRA would make available to prospective sub-borrowers a list of suppliers of agroindustrial machinery, including in it those from Bank member countries and Switzerland as well as their agents or representatives or partners in Mexico. After the proposed loan is signed, FIRA would advertise locally and notify all the embassies in Mexico of Bank member countries that manufacture agroindustrial equipment, and Switzerland of the details of the agroindustries component of the project, giving broad particulars of the types of machinery likely to be required. Assurances on these points were obtained during negotiations. G. Disbursement 3.29 Bank loan disbursements would be spread over approximately three and one-half years. The Bank would reimburse FIRA for (a) 50% of its rediscounts of already disbursed project subloans for low-income producers, (b) 44% of its rediscounts of already disbursed project subloans for other beneficiaries, (c) 50% ef its eligible expenditures for the training and demonstration centers programs, and (d) 100% of its eligible foreign exchange expenditures for the monitoring and technical assistance programs. Disbursements for the lending program and works carried out under force account for the training and demonstration centers program would be made against corresponding certificates of expenditure, the relevant supporting documentation to be retained with FIRA and be available for inspection by the Bank during the course of Project super- vision missions. All other expenditures under the loan will be fully documented and submitted to the Bank along with the respective withdrawal application. The estimated phasing of subloan commitments is given in Table 3.1 and the expected project cash flows for FIRA/FEFA are presented in Table 7.2. The estimated schedule of disbursements by the Bank is given below, assuming that the date of effectiveness of the proposed loan would be August, 1978:  - 28 - Estimated Schedule of Disbursements (in US$'000) Disbursed during Cumulative Disbursement at IBRD Fiscal Year and Quarter Quarter End of Quarter 1979 December 31, 1978 9 9 March 31, 1979 9 18 June 30, 1979 10 28 1980 September 30, 1979 11 39 December 31, 1979 15 54 March 31, 1980 16 70 June 30, 1980 16 86 1981 September 30, 1980 17 103 December 31, 1980 21 124 March 31, 1981 20 144 June 30, 1981 20 164 1982 September 30, 1981 16 180 December 31, 1981 12 192 March 31, 1982 8 200 IV. PROJECT IMPLEMENTATION A. Project Administration 4.01 As in earlier Bank-financed credit projects, the proposed project would be carried out by FIRA, which administers FONDO, FEFA and FEGA. The policies and operations of each trust fund are determined by a technical committee, which, for FIRA, includes the representatives of th'e Finance and Agriculture Ministries, the Bank of Mexico, BNCR, private bankers and farmers. FIRA has established a reputation for efficient organization and technical competence in the area of term credit based on well designed investment plans. Personnel, who generally serve all the three funds, number more than 1,176, of whom 716 are technicians. In addition, 807 technicians in the public banks and 425 in the private banks have been authorized to approve subloans. 4.02 FIRA's technical assistance has played a major role in the formulation, appraisal, and supervision of investment plans under the Fifth Project and its continuance will be crucial for the proposed pioject. FIRA's promotional functions include the training of staff of banks, demonstrations and monitoring of project results. FIRA would continue to carry out these responsibilities, and assurances were obtainLed that, since the functions entail increasing costs, Government would provide sufficient funJs.  - 29 - 4.03 Apart from loans from the Bank and IDB, FIRA's principal sources of funds are capital provided by the Government and the Bank of Mexico's short- term rediscounts. Government equity has increased over the years through the arrangement by which as external loans are repaid by the Government, the local currency equivalent is accumulated in these funds as additional equity. Assurances were obtained during negotiations that the Government would con- tinue to augment the resources of FIRA as required and to compensate FIRA, as well as eligible participating banks, adequately through FEGA assistance for their expenses for technical assistance for low-income producers. B. Sublending Policies and Procedures Beneficiary Categories 4.04 As in the previous projects there would be two groups of bene- ficiaries, low-income producers (LIPs) and other sub-borrowers. A beneficiary is considered to be a low-income producer if his main income is derived from agriculture and if his family's net annual earnings are less than 1,000 times the relevant regional minimum daily rural wage. At the beginning of 1978, this wage ranged from Mex$50 (US$2.17) in the poorest states up to Mex$117 (US$5.09) in the northern border states; the unweighted national average minimum daily rural wage was Mex$79 (US$3.43). This definition of low-income producer was utilized under the Fourth and Fifth Credit Projects, in the PIDER I and II rural development programs and in the recently approved Tropical Agricultural Development Project; it embraces about 80% of the rural population. Out of the proposed lending program for farm and ranch credit operations, about one-third would be to low-income producers, as defined in this paragraph. In view of FIRA's success in mobilizing increased resources for the low-income producer lending program in recent years, the participation of these beneficiaries in the lending program of the Sixth Project has been estimated at 33% (US$150 million), down from the 56% (US$157 million) foreseen in the Fifth Project. The increased resources to support the FIRA program to rediscount subloans to LIPs include US$63.4 million from the Bank (para 2.13) and a recently approved IDB loan for US$60 million; the IDB also recently lent an additional US$60 million to the Banrural for a LIP lending program. Suitable assurances regarding the application of this definition of low-income producers were obtained during negotiations. Subloan Appraisal and Supervision 4.05 As under earlier Bank-financed projects, subloans would be made on the basis of farm, ranch and agroindustry development plans appraised by FIRA technicians or FIRA-approved participating bank technicians. In step with the progress made by participating banks in improving their technical staff, FIRA has been simplifying procedures for processing rediscount requests and progressively delegating subloan approval powers to its own technicians in the field and to selected bank technicians. Assurances were obtained that, as in the past, approval of subloans for financing under the project would be FIRA's responsibility and that FIRA would continue to review its subloan processing procedures with a view to further simplification and delegation of powers of approval.  - 30 - 4.06 As in previous projects, banks would meet a set of eligibility criteria, established by FIRA and acceptable to the Bank, in order to partici- pate in the project. Participating banks would grant subloans for on-farm investment and agroindustries and FIRA would rediscount these subloans. The sublending terms and conditions would be as summarized below: Annual Minimum Sub-borrower Maximum Rediscount Minimim Contribution Percentage Rates to Annual Subloan (percentage of cost of Subloan Participating Interest Sub-borrowers Amount of project financed) Rediscounted Banks Rates 1. Low-income sub-borrowers (a) on-farm investment projects (i) low-income sub-borrowers not more than 3 90 7.50 11.00 who receive institutioirial 350,000 pesos credit for the first time (ii) other low-income not more than 5 90 10.50 13.50 sub-borrowers 1,000,000 pesos (b) agroindustrial projects not more than 5 90 11.00 14.00 2,500,000 pesos 2. Other sub-borrowers' on-farm (i) not more than 15 90 13.00 16.00 investment projects and 2,500,000 pesos agroindustrial projects (ii) more than 2,500,000 15 80 13.25 16.00 pesos and not more than 5,000,000 pesos (iii) more than 5,000,000 15 70 14.50 17.00 pesos The rediscount and interest rates specified above represent a substantial simplification from the more complex schedule applied in the previous proj- ect in which on-lending rates ranged from 7.6% to 14.5% p.a. The preferen- tial rates to LIPs, especially those which for the first time are receiving institutional credit, reflect the Government's policy of favoring this tar- get group. In view of the Bank's projection of inflation rates in Mexico declining from the 1977 level of about 20% to around 10% by 1981, many LIP sub-borrowers and almost all other sub-borrowers would be paying positive real interest rates over the life of the project. 4.07 In the case of LIPs, sub-borrower contributions to investment cost could be in the form of labor and material or financial resources. For sub- loans to groups of LIPs for on-farm development, the upper limits on the subloan sizes specified in paragraph 4.06 would be determined by dividing the subloan amount by the number of beneficiaries in the group receiving the subloan. For other subloans, the maximum subloan amount is independent of the number of beneficiaries of a given subloan. No subloans for purchase of breeding livestock will be made to sub-borrowers to finance the replacement of breeding livestock financed under the Project or prior projects. In the  - 31 - case of subloans to non-LIP beneficiaries for livestock development, the purchase of breeding livestock will not, in the aggregate, exceed 53% of the total investment for livestock development by such beneficiaries financed under the Project. 4.08 The repayment period for subloans would be based on expected cash flows and would range from three to 15 years (including grace periods) except for special cases when, with prior Bank approval, it could go up to 20 years. Grace periods would be at least one year for development subloans for annual crop enterprises and at least three years for subloans for perennial crop, livestock and agroindustrial enterprises. Repayment schedules for the partic- ipating banks to FIRA would be in line with those of the sub-borrowers. 4.09 Under the Fourth Project, FIRA was required to obtain prior Bank approval for project subloans in excess of US$400,000 (Mex$5 million). However, in view of the rising investment costs and the increased size of individual LIP subloans resulting from group activity, this limit was raised to US$800,000 million (Mex$10 million) for the Fifth Project. This US$800,000 equivalent free limit would continue to apply in the Sixth Project. FIRA would continue to limit the borrowing of any beneficiary under the project, together with the outstanding balance on any earlier subloan to him, not to exceed Mex$5 million. 4.10 Assurances were obtained during negotiations that these lending terms and conditions and other procedures outlined in paragraphs 4.06 to 4.09 would be observed and that detailed data would be furnished to the Bank for all subloans to be approved by it. C. Accounts and Auditing 4.11 The accounting system of FIRA trust funds, FONDO, FEFA and FEGA, is satisfactory. FIRA would maintain separate project accounts as established for the preceding project (Loan 1217-ME). FIRA accounts are audited by a unit directly responsible to the Audit Department of Bank of Mexico and also by an independent public auditing firm. An audit, satisfactory to the Bank, would be made of the accounts and financial statements of FONDO, FEFA and FEGA and the project accounts for each fiscal year in accordance with sound auditing principles consistently applied by independent and qualified auditors appointed by Bank of Mexico. Certified copies of the statements so audited, together with the report of the auditors, would be sent to the Bank not later than three months after the close of the fiscal year. Adequate accounts and records would also be maintained within all the participating banks so as to identify all transactions pertaining to the project and periodically Bank supervision missions would have access to such accounts and any other docu- ments pertaining to lending operations within the project. Assurances on these points were obtained during negotiations.  - 32 - D. Monitoring and Reporting Monitoring 4.12 The monitoring system initiated in FIRA under the preceding Loans 910-ME and 1217-ME would be strengthened and improved under the proposed project. Known as the "Sistema de Registros Continuos," this monitoring program has begun operations with a sample of low-income producers. As an initial exercise in monitoring project performance, the present system has provided insights on both the methodological and substantive difficulties in evaluating a large complex credit program comprising national coverage and a wide range of investment activities. Data collection on the financial aspects of project implementation have been monitored closely by FIRA's own reporting system; bowever, the evaluation of the impact of the credit program has been difficult to follow at the on-farm level. Enterprises financed include a variety of mixed activities under widely differing ecological as well as land tenancy and organizational conditions. The volume of subloans as well as the rapid disbursement of funds has constrained the collection of adequate time-series data on subloan impact. On the basis of this experience, the project monitoring process would be simplified and increasingly focused on the major issues relating to measuring the impact of FIRA's program as a whole. 4.13 The basis for subloan monitoring would be the subloan investment program in which the beneficiary's use of credit and its impact were projected. From a representative sample, data for the monitoring process under this project would be gathered by FIRA and particpating bank technical staff in the course of their normal technical assistance duties. The data to be gathered would refer to key technical operating parameters highlighted in the subloan evaluation. These data would be of sufficient quantity and quality to permit an assessment of the effects of the program, including comparisons with key project indicators set out in Chapter V. In addition, those organized farmer groups which are being provided with technical assistance and are maintaining adequate records would, as a matter of course, also feed infor- mation into the monitoring system. The monitoring system would be designed to provide a rapid feedback of information to project beneficiaries and technical assistance staff. Provision has been made in the loan for funds to be used to contract up to 12 man/months of short-term technical assistance to work with the FIRA monitoring unit. In order to ensure progress of the development of the monitoring system, an increased input of Bank super- vision time during project implementation would be required. Annually a report would be submitted to the Bank based upon these monitoring activities, not more than four months after the end of the Government's fiscal year. 4.14 Reporting. Quarterly reports summarizing progress in project execution, subloan disbursements by categories and FIRA rediscounts would be submitted to the Bank not later than 60 days after the end of each quarter. Detailed reports, which would include subloan commitments and disbursements  - 33 - by size, lending categories and regions, number of beneficiaries, bank parti- cipation in subloan financing and FIRA rediscounts and a summary of reports submitted to FIRA by participating banks showing overdue payments on out- standing subloans would be prepared semi-annually and submitted to the Bank not later than three months after the end of the semester. Together with the annual audit report to be submitted to the Bank by FIRA (para 4.11), FIRA would also send the Bank a summary statement of changes in its own staffing on an overall and regional basis, the throughput of its training programs, a summary of the activities carried out on its demonstration farms, a review of the scope and effectiveness of the technical assistance activities to low- income producers financed through FEGA, and the status of technical staffing in participating banks. FIRA would also submit to the Bank such other infor- mation concerning the progress of the project, its accounts and related statements as the Bank may reasonably request, and, within six months of the closing date of the Bank Loan, a Project Completion Report prepared in accordance to guidelines which the Bank would reasonably recommend. Monitoring and reporting systems to be implemented by FIRA for this project, along the lines set out in this paragraph and paragraphs 4.12 and 4.13 were discussed during negotiations. 4.15 The monitoring and reporting functions of the ongoing project are presently carried out by several divisions within FIRA and the Bank considers the present distribution of tasks appropriate. However, it has proposed that the final responsibility for overseeing the preparation and submission of these various reports to the Bank, as well as coordinating the various activities relating to the Project, be vested in a senior professional in the Office of Planning of FIRA. V. SPECIFICATIONS OF ENTERPRISE MODELS A. Livestock Development Models 5.01 Production coefficients for these activities are based on the lending experience of the on-going project (loan 1217-ME) and are consid- ered reasonable and attainable under the technical assistance and sub-loan supervision arrangements envisaged for the proposed project. These coeffi- cients will constitute the basic data for project monitoring and are sum- marized in Table 5.1. Weights and prices assumed for the livestock models appear in Annex 1, Table 26. B. Crop Development Models 5.02 Technical coefficients used in the crop models are based on the operational experience obtained under the on-going FIRA projects. Their attainment assumes the availability of adequate technical assistance, application of seeds, fertilizers and pesticides in appropriate quantities  MEXICO SIXTH AGRICULTURAL CREDIT PROJECT Production Data for Livestock Models Type of Livestock Enterprise Dairy Item Unit Stabled Grazing Dual Purpose Beef Swine Calving Without project % 70 70 With project % 75 75 Year 5 5 Weaning Without project % - - 60 60 With project - - 65 65 Year - - 5 5 Lactation jield Without project liter 3150 3150 450 - With project liter 3600 3500 750 - Year 5 5 5 - Extraction Rate Pig Sales Without project x - - 17.7 18.0 Per sow/year 12.2 With project - - 23.0 23.9 Per sow/year 14.3 Year - - 8 7 Year 2 Pasture Productivity-Milk Without project liter/ha 6552 6586 125 - With project Liter/ha 9000 8355 300 - Year Pasture Productivity-Beef Pork Production Without project kg/ha 106 109 57 10 kg/sow/year 886 z With project kg/ha 186 224 93 21 kg/sow/year 1039 X Year 5 5 8 7 Year 2 X March 2, 1978  MEXICO SIXTH AGRICULTURAL CREDIT PROJECT Projected Development of Yields With Project Without ------------------------------Year------------------------------------ Project 1 2 3 4 5 6 7 8 9 10 11 12 Irrigated --------------------------(m ton / ha)-------------------------------- Annual crops Wheat 2.0 2.5 2.8 3.0 3.2 ---------------------------------------------- Beans 0.7 1.1 1.2 1.3 ---------------------------------------------------- Safflower - 1.1 1.2 1.4 1.5 ---------------------------------------------- Cotton - 1.6 1.8 2.0 2.1 2.2 ---------------------------------------- Maize 1.6 2.0 2.5 2.8 3.0 ----------------------------------------- Sorghum 1.8 2.8 3.2 3.4 3.5 --------------------------------------------- Soybeans 1.2 1.2 1.5 1.7 1.9 2.0 ---------------------------------------- Perennial crops Walnut 1.5 - - - 0.3 0.4 0.5 0.7 1.0 1.2 1.3 1.4 1.5 Rainfed Annual crops Maize 1.6 1.8 1.9 2.0 2.1 -------------------------- - Beans 0.6 0.7 0.8 --------------------------------------- -