Mexico - Rainfed Agricultural Development Project

Document of I The World Bank FOR OFFICIAL USE ONLY Report No. 3091-ME STAFF APPRAISAL REPORT RAINFED AGRICULTURAL DEVELOPMENT PROJECT MEXICO December 19, 1980 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$23.0 Mex$l = US$0.04 Mex$l 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 1 square kilometer (km ) = 0.39 sq. miles = 100 ha 1 kilogram (kg) - 2.2 pounds 1 liter (1) = 0.26 gallons 1,000 kg = 1 metric ton = 0.98 long ton ABBREVIATIONS ANAGSA - National Crop and Livestock Insurance Agency BANRURAL - National Bank for Rural Credit CONASUPO - National Marketing Corporation for Basic Foods DGCSA - General Directorate for Soil and Water Conservation DGDUT - General Directorate of Rainfed Districts and Units DGPEA - General Directorate for Production and Agricultural Extension DT - Distrito de Temporal (Rainfed District) FERTIMEX - Fertilizantes Mexicanos, S.A. FICAR - Trust Fund for Credit in Irrigated Areas FIRA - Agriculture Trust Funds of the Bank of Mexico INIA - National Institute for Crop Research INIF - National Institute for Forestry Research INIP - National Institute for Livestock Research NAFINSA - Nacional Financiera, S.A. OHDR - Directorate of Hydraulic Works for Rural Development V' PEMEX - Petroleos Mexicanos PIDER - Investment Program for Rural Development PLANAT - National Plan for Rainfed Agriculture Development PRODERITH - Program for Development of the Humid Tropics SAHOP - Secretariat for Human Settlements and Public Works SARH - Secretariat of Agriculture and Water Resources SIH - Sub-secretariat for Hydraulic Infrastructure, SARH SPP - Secretariat of Programming and Budgeting TCC - Technical Coordination Committee  FOR OFFICIAL USE ONLY MEXICO RAINFED AGRICULTURAL DEVELOPMENT PROJECT Table of Contents Page No. I. THE AGRICULTURAL SECTOR . .......................... .. 1 General ............... I Rainfed Agriculture ........... ... ...... . I Government Policies and Strategies ........................ 1 Present Bank Participation in the Sector .................. 2 II. INSTITUTIONAL ASPECTS OF RAINFED AGRICULTURE .............. 3 Introduction .......................................... ... . 3 General Directorate for Rainfed Districts and Units (DGDUT) 4 Technical Services ..................... ... .5 Research ........................................... 6 Soil and Water Conservation .............,..,.., 7 Credit, Insurance and Supply of Inputs . 7 Marketing and Storage ............. . 9 Infrastructure............. 9 III. THE PROJECT AREA.. ......... .... 10 IV. THE PROJECT ................ .. 12 A. Introduction and Objectives ........................... 12 B. Project Description........ ... 12 C. Detailed Features .... . ............. ... ......... 13 Promotion of Technology Packages . . 1 13 Integration of Research, Extension and Credit 14 Soil and Water Conservation Programs . .15 Rural Works ............................ 16 Forestry ........... . .. .... ................... 17 National Support Activities........ 17 D. Project Costs...... ... 18 E. Financing ..... ......... ... .20 F. Procurement. . . .............. 20 G. Disbursements ....21 H. Accounts and Auditing ....22 This report is based on the findings of an appraisal mission consisting of Messrs. R. Milford, D. Myren, J. Fernandez, H. Wagner, J. Fransen, J.D. Von Pischke (Bank) and L. Bartelli, C. Powers, and Z. Edelman (Consultants) who visited Mexico in January/February 1980. A follow-up mission consisting of Messrs. Myren, Fransen and Von Pischke visited Mexico in May 1980. 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. ORGAtIZAI'ION AND MANAGEMENT @......$.o............... 22 General Aspects ...................e ...............e 22 Program Execution .................................. . 24 Monitoring and Evaluation ................................. 28 VI. PRODUCTIONN9 MARKETING AND FARMER BENEFITS ................ 30' Agriculttural Production ................................... 30 Marketing and Storage ................................ . 32 Farmer Benefits ....... ......... . * . . . * . . 32 VII. ECONOMIC JUSTIFICATION .................................... 34 Gen.eral -, ....... .......... ......... .......... ......... 34 Economic Rate of Return ................................. 34 Sensitivity Analysis ..35 Project Rtisks ................................... 35 Environmental Impact .......................... 35 VIII. STJTMARY OF AGREEMENTS REACHED AND RECOMMENDATION .35 Annex 1 Credit: BANRURAL Key Financial Indicators and Present Operating Procedures of FICAR . .37 Annex 2 Table 1 Land Use and Population PLANAT Districts Table 2 Summary PLANAT District Programs Table 3 PLANAT Cardenas District Table 4 PLANAT Coatzacoalcos District Table 5 PLANAT Chilpancingo District Table 6 PLANAT Cholula District Table 7 PLANAT Jalpa District Table 8 PLANAT Martinez de la Torre District Table 9 PLANAT Morelia District Table 10 PLANAT Villahermosa District Table 11 PLANAT Zapopan District Table 12 PLANAT Monitoring and Evaluation: Districts and General Support  Table of Contents (Continued) Annex 2 (Continued) Table 13 Summary Extension District Programs Table 14 Extension Cardenas District Table 15 Extension Coatzacoalcos District Table 16 Extension Chilpancingo District Table 17 Extension Cholula District Table 18 Extension Jalpa District Table 19 Extension Martinez de la Torre District Table 20 Extension Morelia District Table 21 Extension Villahermosa District Table 22 Extension Zapopan District Table 23 Summary of District Research Programs Table 24 Summary of INIA District Programs Table 25 Summary of INIA Regional Center Support for Nine DTs Table 26 Summary Total of INIA District and Regional Center Support Programs Table 27 INIA Field Research Cardenas District Table 28 " " Coatzacoalcos District Table 29 " Chilpancingo District Table 30 " Cholula District Table 31 Jalpa District Table 32 Martinez de la Torre District Table 33 Morelia Distric Table 34 " Villahermosa District Table 35 " Zapopan District Table 36 Summary of INIP District Programs Table 37 Summary of INIP Regional Center Support for Nine DTs Table 38 Summary of Total District and Regional Center Support Programs Table 39 INIP Field Research Cardenas District Table 40 Coatzacoalcos District Table 41 Chilpancingo District Table 42 Cholula District Table 43 Jalpa District Table 44 Martinez de la Torre District Table 45 Morelia District Table 46 Villahermosa District Table 47 Zapopan District Table 48 Summary of INIF District Programs Table 49 Summary of INIF Regional Center Support for Four DTs Table 50 Summary Total of INIF District and Regional Center Support Programs Table 51 INIF Field Research Chilpancingo District Table 52 " " " Cholula District Table 53 " i " Jalpa District Table 54 " " " Morelia District  Table of Contents (Continued) Annex 2 (Continued) Table 55 Summary- Soil and Water District Programs Table 56 Soil and Water Cardenas and Villahermosa Districts Table 57 Soil and Water Coatzacoalcos District Table 58 Soil and Water Chilpancingo District Table 59 Soil and Water Cholula District Table 60 Soil and 'Water Jalpa District Table 61 Soil and Water M4artinez de la Torre District Table 62 Soil and Water MoreLia District Table 63 Soil and Water Zapopan District Table 64 Summary: Livestock Development Programps Table 65 Livestock Cardenas District Table 66 Livestock Coatzacoalcos District Table 67 Livestock Chilpancingo District Table 68 Livestock Cholula District Table 69 L-vestock Jalpa District Table 70 Livestock MBrrinez de la Torre District Table 71 Livestock Morelia District Table 72 Livestock Vil lahermosa Table 73 Livestock Zapopan Table 74 Summary: Forestry District Programs Table 75 Summary: District Investments in Drainage Works Table 76 Summary: District Investments in Irrigation Works Table 77 Summary: District Investments in Roads Table 78 Summary: District Road Maintenance Table 79 PLANAT Support Services Table 80 Extension Support Services Table 81 Research Support Services Table 82 Summary of INIA Requirements for Phase II of PLANAT Table 83 Summary of INIA Total District ancd National Programs Table 84 Summary of INIP Requirements for Phase II of PLANAT Table 85 Summary of INIA Total District ancl National Programs Table 86 Summary of INIF Requirements for Phase II of PLANAT Table 87 INIF Research Support at Headquarters Table 88 INIF Research Support Regional Ceniter (CIFONOR) Table 89 INIF Research Support N.E. Regional Center (CIFNE) Table 90 " It 2 NEW. IF ft (CIFNO) Table 91 " W, W! It (CIFO) Table 92 "i VI Central i t (CIFREC/CENIPROF) Table 93 It 7 Southern Pac:ific " (CIFPAS) Table 94 Summary: Total WIF District, District Support and Regional Center Programs Table 95 Soil and Water Conservation Support Services Table 96 Training and Consultant Services Summary Cost Estimates Table 97 Plan l (180 ha) Cardenas, Martirnez de la Torre - Budget and Cash Flow Table 98 Plan 1 (180 ha) Cardenas, Martinez de La Torre - On-farm Investment Cost, Perennial Crops Table 99 Plan 1 (1a0 ha) - Cardenas, Martinez de la Torre - On-farm Investment Costs, Dual IPurpose Livestock  Table of Contents (Continued) Annex 2 (Continued) Table 100 Plan 1 (180 ha) - Cardenas, Martinez de la Torre - Cropping Patterns, Yields, Total Production and Production Values - Perennial Crops Table 101 Plan 1 (180 ha) - Cardenas, Martinez de la Torre - Operating Costs - Perennial Crops Table 102 Plan 2 - Cholula - Budget and Cash Flow Table 103 Plan 2 - Cholula - On-farm Investment Costs Table 104 Plan 2 - Cholula - Cropping Patterns, Yields, Total Production and Production Values Table 105 Plan 2 - Cholula - Operating Costs Table 106 Plan 3 - Coatzacoalcos, Villahermosa - Budget and Cash Flows Table 107 Plan 3 - Coatzacoalcos, Villahermosa - On-farm Investment Costs - Perennial Crops Table 108 Plan 3 - Coatzacoalcos, Villahermosa - On-farm Investment Costs - Dual Purpose Livestock Table 109 Plan 3 - Coatzacoalcos, Villahermosa - Cropping Patterns, Yields, Total Production and Production Values - Perennial Crops Table 110 Plan 3 - Coatzacoalcos, Villahermosa - Operating Costs - Perennial Crops Table 111 Plan 3 - Coatzacoalcos, Villahermosa - Herd Development - Dual Purpose Livestock Table 112 Plan 3 - Coatzacoalcos, Villahermosa - Operating Costs - Dual Purpose Livestock Table 113 Plan 3 - Coatzacoalcos, Villahermosa - Cropping Patterns, Yields, Total Production and Production Values - Annual Crops Table 114 Plan 3 - Coatzacoalcos, Villahermosa - Operating Costs - Annual Crops Table 115 Plan 4 - Morelia, Zapopan - Budget and Cash' Flow Table 116 Plan 4 - Morelia, Zapopan - On-farm Investment Costs Table 117 Plan 4 - Morelia, Zapopan - Cropping Patterns, Yields, Total Production and Production Values Table 118 Plan 4 - Morelia, Zapopan - Operating Costs Table 119 Plan 5 - Jalpa - Budget and Cash Flow Table 120 Plan 5 - Jalpa - On-farm Investment Costs - Crops Table 121 Plan 5 - Jalpa - On-farm Investment Costs - Dual-Purpose Livestock Table 122 Plan 5 - Jalpa - Cropping Patterns, Yields, Total Production and Production Values - Annual and Perennial Crops Table 123 Plan 5 - Jalpa - Operating Costs - Annual and Perennial Crops Table 124 Plan 6 - Chilpancingo - Budget and Cash Flow Table 125 Plan 7 - Chilpancingo - On-farm Investment Costs Table 126 Plan 6 - Chilpancingo - Cropping Patterns, Yields, Total Production and Production Values - Annual and Perennial Crops  Table of Contents (Continued) Annex 2 (Continued) Table 127 Plan 6 - Chilpancingo - Operating Costs - Annual and Perennial Crops Table 128 Chilpancingo and Jalpa - Dual Purpose Livestock Development Plan (108 ha) - Investment Costs Table 129 Chilpancingo and Jalpa - Dual Purpose Livestock Development Plan (108 ha) - Herd Development Table 130 Chilpancingo and Jalpa - Dual Purpose Livestock Development Plan (108 ha) - Sales and Operating Expenses Table 131 Cholula, Morelia, Zapopan - Dairy Production Unit (23 ha) - On-farm Investments Table 132 Cholula, Morelia, Zapopan - Dairy Production Unit (23 ha) - Herd Projection Table 133 Cholula, Morelia, Zapopan - Dairy Production Unit (23 ha) - Sales and Operating Costs Table 134 Cholula, (Morelia, Zapopan - Benefit Cost Streams Table 135 Cholula, Morelia, Zapopan - Hog Production Unit, Budget Table 136 Cholula, Morelia, Zapopan - Hog Production Unit, Herd Projection Table 137 Cholula, Morelia, Zapopan - Hog Production Unit, Sales, Operating Expenses and Investments Table 138 Annual Phasing of Project Costs Table 139 Disbursement Schedule Table 140 Incremental Economic Analysis Table 141 Economic Prices Table 142 Market Prices Table 143 Economic Rate of Return and Sensitivity Analysis Annex 3 Documents on Project File Chart 1 District Organization Chart 2 Programming and Budgeting Process: SARH and Rainfed Districts Chart 3 Implementation Schedule MAP IBRD 14946  MEXICO RAINFED AGRICULTURAL DEVELOPMENT PROJECT I. THE AGRICULTURAL SECTOR General 1.01 Mexico has a land area of 197 million ha of which about 30 million are considered cultivatable, 78 million are classified as range land, 44 million are suitable for forestry and the remainder is either too dry, too poorly drained, or too steep for any form of food and fiber production. Of the estimated 30 million ha of potentially arable land, an average of 15 million ha is presently cropped, including approximately 5 million ha under irrigations The cropped area has remained constant since 1965 at about 15 million ha, and production increased from 1965 to 1978 at a yearly average of only 1.4%, down from 5% a year from 1945-65. Livestock production expanded during the same period at about 4% yearly. As a share of national GDP, agriculture declined from 14% in 1965 to just over 9% in 1978. 1.02 Despite increasing urbanization, agriculture employs about 40% of the labor force in Mexico, providing a livelihood for some 4 million families, 1.3 million of which are landless. About two-thirds of this labor force have earnings below the minimum wage. 1.03 External trade in agricultural products remains favorable. Agricul- tural exports in 1979 amounted to about 20% of total merchandise exports (nearly 35% of non-oil exports), while the share of agricultural imports in total merchandise imports was about 7%. However, the combined effect of rapid population growth (3% a year on average during the 1970s) and the declining growth in agricultural production has resulted in a significant increase in food imports, particularly of maize, wheat, sorghum and oilseeds. Rainfed Agriculture 1.04 Rainfed agriculture, including crop lands, grazing and forest, covers a heterogeneous assortment of soils and climates, ranging from desert to humid tropics. In terms of public investment, the rainfed areas have been largely neglected; over the past three decades, nearly 90% of public expendi- tures in the agricultural sector went for capital and current costs in the irrigated areas. However, rainfed agriculture still accounts for over 70% of the cropped area, and up to 50% of the value of agricultural output. Government Policies and Strategies 1.05 The rate of expansion of irrigated areas, once the most important underlying factor in the rapid increase of agricultural production in Mexico, has slowed in recent years. Between 1952 and 1965, the area harvested under irrigation went up from 1.6 million ha to 3.5 million ha (an increase of about 150,000 ha/year) but the expansion slowed between 1965 and 1979 to about 100,000 ha/year. While it is hoped that the current rate of expansion can be maintained for at least 10 more years, it is recognized that the scope for opening up new irrigated lands under large-scale schemes is becoming smaller and such development is becoming more difficult and less cost  effective. Concurrently the rate of increase in crop production per unit area has also slowed down due mainly to a lagging in technical innovation and application and inadequate maintenance. As a result more emphasis is being given to rehabilitation of existing irrigation areas, the development of more small-scaLe (local) irrigation programs and the upgrading of tech- nical services in both ongoing and new projects. 1.06 In addition to a shift in its irrigation development strategy in recent years there has been a gradual change in Government policy towards the promotion of rainEed agriculture which up to 1975 received less than 10% of all public funds Spent in the agricultural sector. The reasons for increased attention to rainEed areas are threefold. First, in areas of adequate rainfall (above 700 mm annually) there is still important potential to increase produc- tivity of rainfed crops and livestock through application of available tech- nology and to increase cropped areas through improved drainage. The average costs of bringing about these changes are considered to be lower than those of large scale irrigation. Second, investments in rainfedl agriculture are also considered an important means by which to redress the social imbalance now evident in Mexican agriculture: while rainfed areas produce only 50% of the value of agriculture output, they comprise 87% of all farmers. Third, the Government has been particularly concerned about growing imports of basic staples, especialLy maize, beans and wheat, due to stagnating domestic produc- tion of these commnodities; of the total cropped area under rainfed conditions of eleven million ha, 8.6 million ha is under maize and beans, with average yields that have changed little in the last decade. 1.07 The present administration took the first major step toward increased public investment in rainfed areas in 1976 with the merging of the Secretariat of Water Resources and the Secretariat of Agriculture into the new Secretariat of Agriculture and Water Resources (SARH). This was followed in March 1977 by a presidential decree establishing the General Directorate of Rainfed Districts and Units (DGDUT) within SARH with the specific objective to coordinate and expand line agencies' activities in promoting rainfed production at the district level (paras 2.03 and 2.04). In addition, early in 1980 the Government announced a Food ])evelopment Plan which has, among others, the objectives of promoting basic food production through higher guaranteed prices, better crop insurance and improved input supplies. The Rainfed Districts were designated as the main execulive arm to implement the productive part of the plan. Present Bank Participation in the Sector 1.08 During the past six years (FY 1975-80), Bank participation in the agricultural sector amounted to US$1.5 billion distributed over 13 projects as follows: (a) Irrigat:'on (Panuco, Sinaloa, Fuerte/Sinaloa, Bajo Bravo! Bajo San Juan, and Apatzingan) - US$450 million; (b) Rural Development (PIDER I and II and Papaloapan Basin) - US$280 niillion; (c) Credit 'FIRA Projects V, VI, VII) - US$650 million; (d) Area Development (Tropical Agriculture) - US$56 million; and (e) Small-Scale Agricultural Infrastructure Development - US$60 mi.llion.  1.09 Implementation of the above projects is generally satisfactory, varying somewhat with the implementation capabilities of the agencies involved. Especially relevant to this project is the experience gained under the FIRA credit projects, and more recently that being obtained under PIDER and the Tropical Agriculture Project. Under the PIDER projects and the Tropical Agricultural Development Project, separate funds were provided to finance the agricultural credit requirements of the project rather than obtaining credit through the country-wide FIRA program. The established procedures whereby the public and private banks rediscounted on-farm loans through FIRA were used for these projects. This was done to enable the management of each project to focus more sharply on its credit activities than would have been possible if the funds had been obtained from general FIRA sources. In particular, this approach enabled closer attention to be given to pro- viding additional technical support to those recipients of credit under those projects. Also by involving directly the agencies in SARH responsible for providing technical support, particularly the extension service, better opportunities were provided for improving the coordination between them and the banking institutions than were possible under the FIRA country-wide proj- ects. Increasingly the Bank has stressed lending to low-income producers and greater emphasis on crop rather than livestock production. The proposed project would be the first to provide funding exclusively for the rainfed areas and thus it is congruous with the increased emphasis given by Government to improving the productivity of farmers in such areas. II. INSTITUTIONAL ASPECTS OF RAINFED AGRICULTURE Introduction 2.01 In addition to DGDUT (para 1.07), more than 20 agencies are directly or indirectly involved in promoting agricultural production in rainfed areas. The most important are (a) those integrated in the various secretariats: SARH, with the extension service (DGPEA), the national research institutes for crops (INIA), livestock (INIP) and forestry (INIF); the small-scale irrigation agency (OHDR) and the soil conservation service (DGCSA); the Secretariat of Industry and Commerce with its buying and marketing agency for agricultural produce (CONASUPO) and the Secretariat of Human Settlements and Public Works (SAHOP) for the construction of rural access roads; (b) the public and private banking system for credit to agricultural producers, foremost among them the Government-owned BANRURAL; and (c) Government-owned autonomous organizations such as FERTIMEX for fertilizer supplies and ANAGSA for crop insurance. 2.02 With regard to the agencies mentioned in paragraph 2.01, two per- formance judgments are of general validity: performance on infrastructure (i.e., roads, small irrigation systems, water supply) has generally been better than on other development activities, at least in rainfed areas, and, second, lack of coordination among these agencies at the field level has reduced the productive impact of investments both in infrastructure and new technology generation. In view of this, the Government's decision to create DGDUT with the main objectives of (a) reinforcing extension, credit, and research services in rainfed areas; and (b) obtaining maximum coordination and participation of all groups involved at the local level, was an important step in the right direction.  General. Directorate for Rainfed Districts and Units (DGDUT) 2.03 Following the pattern established earlier in the organization of irrigation districts, the SARH General Directorate of Rainfed Districts and Units (DGDUTJ) estaIished 124 rainfed districts (Distritos de Temporal (DT)), covering 70% of the total area of the country. The main objectives are to rein. force extensio-n, research and credit services and to encourage maximum participation of aLl local groups involved (producers, banks, technical services, inDutt and marketing organizations, and infrastructure agencies) in order to increase olltput of crops, livestock and forestry products. PlanniThg and budge-ing are at the DT level, with budgets developed through discassiorn and agreement among the groups, both informally and formally at the DT executive co=mittee level. Each DT is responsible for managing its owil budget once it has been approved by federal and state authorities. 2.04 Although the DGDUT has been operating for only two years, con- siderable progress 'nas been made in getting the important line agencies to work cooperatively., The state SARH representatives have allocated staff to the DTs in their states and have designated (in collaboration with DGDUT) chiefs for each of the 124 D'Is in the country. DGDUT has also drawn up a National Plan for Eainfed Agriculture Development (PLANAT) which sets up the DT, consisting of several municipios within a single state, as the basic development unit upon which to build a strong planning and execution capa- bility, The operating program in each district is further sujbdivided into Units and Zones (Annex 2, Chart 1) in order eventually to attain direct contact: with and participation of the majority of farm families. This also provides the structure for obtaining maximum participation of all the local groups involved (p^ra 2.03). 2.05 A key feature of the DT organization is that, for the first time -'n rainfed areas ir M.exico, extension coverage by SARH field staff has been decentralized so that it should now be possible to provide technical services 4rn most rainfed areas in the country. Each district is subdivided into units of about 6,000 to 10,000 ha based on municipal boundaries. The units are subdivided into zones of about 500 farm families, with two to four zones per unit. The DT organization provides for one field extension worker to be located in each zone with a unit extension officer (Unit Chief) respon- sible for the zone extension officers in his unit. The Unit Chiefs report to the Director of Operations who is responsible for all day-to-day opera- tions in the district including extension. The Director of Operations reports directly to the District Chief. Technical support for the field staff is provided by resident specialists from each subsecretariat of SARH, Thus, dependinig on the agricultural pattern, program offices for agrictulture, hydrau- lic resources, livestock, and Eorestry could be located in a DT. T2he subject matter specialists from these offices would work closely with unit chiefs and maintain close contact with the field level research teams from the three research institutes, INIA, INIP and INIF, working in the DT. 2.06 The DT organization structure also provides for the followiu.g administrative features: (a) a district executive committee, with member- ship comprising representatives of all relevent line agencies concerned with district development and producer groups. This committee functions as the body responsible for formuilation of the district development plans and budget.  -5- Once these plans have been reviewed and approved by SARH, COPRODE and the Secretariat for Programming and Budgeting (SPP) at the state and national levels, the committee is also responsible for execution; (b) a technical committee to advise on all technical matters related to DT development and provide the backstopping necessary to formulate the district development plan and supervise its execution; and (c) the appointment of a district chief as executive officer to carry out the decisions of the executive committee and a chief of operations to assist him (see Annex 2, Chart 1). Technical Services 2.07 Extension. The SARH General Directorate for Production and Agricultural Extension (DGPEA) is the public body responsible for extension in Mexico. Government support for extension has expanded rapidly since 1970, reaching US$67.8 million in 1979 (US$52.5 million normal budget, US$14.2 million PIDER, 1/ and US$1.1 million PRODERITH 2/). DGPEA now has a total national staff of 8,787 of which 2,530 have professional level training, 4,780 have been trained at an intermediate level and the remaining 1,467 have lower levels of training; on average, each field staff professional is responsible for two or three intermediate level staff and about 2,000 farmers. With the exception of most irrigation districts, the overall reputation of the exten- sion service is not very high in Mexico. First, many of the staff are not sufficiently trained and there is a dearth of subject matter specialists able to back up extension agents. Second, staff is too thinly spread which, together with a lack of transportation, results in too few contacts with the farmer. Third, linkage between research and extension is inadequate and as a result not enough tested technology is available to the extension worker. Fourth, the fact that new budget appropriations do not become available to the service until well into the fiscal year severely cuts down on operating budgets in the agriculturally important early months of the year and further limits effectiveness. 2.08 Livestock and Forestry Services. Although the responsibility for extension work with farmers is vested in the SARH Sub-secretariat for Agriculture and Operations, other development services are provided to farmers by the Sub-Secretariats of Livestock and Forestry. In the livestock sector, the services of direct relevance to this project that are in need of strengthening at the district level include diagnostic services, disease prevention and parasite control. More investment is also needed in the small animal field to improve home production and augment family meat supplies. In the forestry sector, existing services for farmers are very limited. A small amount of nursery material is provided and this activity requires expansion. Assistance in introducing more rational approaches to wood production and exploitation is also needed, including the establishment of firewood lots and distribution of seedlings and technical advice. 1/ Rural development project initiated in 1973 which has had the support of two IBRD loans (1110-ME and 1462-ME). 2/ Tropical Agriculture Development Program initiated in 1978 with funding from IBRD Loan 1553-ME.  -6 Research 2.09 Crop, livestock and forestry research is mainly the responsibility of the SARH agencies: INIA, INIP and INIF, Other specialized single-crop agencies (e.g., coffee and sugar) and the various universities also do some research, but only INLA, INIP and INIF would be closely associated with the rainfed districts project. 2.10 INIA. INIA is the oldest and best developed of the three institutes. Since its founding in 1960, it has had national respornsibility for crop research. It has been allocated increasing funds (1979 budget was Mex$1,096 million (US$47.7 million) , up from Mex$192 million (US$15.4 million at 12.50/ dollar) in 1974), with which it has expanded its research staff and has improved the scope and orientation of its work. In August 1977, significant organiza- tional changes were made within INIA permitting it to tie its work more closely to the needs of the agricultural producers in the rainfed areas. Of the almost 10,000 experiments conducted in 1978/79, slightly more than 50% were carried out under rainfed conditions. Under its decentralization plan, program coordinators have been moved from Mexico City to regional locations. It has therefore delegated more responsibility to its 11 regional research centers each with a network of experimental stations (54 in the 11 regions) and has adopted a multidisciplinary team approach whereby teams work on specific commodities or production systems on farmers' fields, Of the 917 professional staff, 10% have or are studying for the PhD degree, 30% have or are studying for al M.S. degree and the balance have B.S. degrees. Even so, given the urgency of increasing basic food crop production, there is still a need to further increase the number and upgrade the academic qualifications of staff, as well as to improve physical facilities and equipment, continue to strengthen applied research on farmers' fields in rainfed areas, integrate research information into technological packages for the many agroecological regions, and improve interaction with other research institutes and with the extension service. 2.11 INIP. Predecessor agencies in animal research, dating back some 40 years, were mosl:ly concerned with the production of vaccines and other animal health products. However, since 1968 INIP has had national responsi- bility for studying all problems limiting livestock production and today its major emphasis is on animal production rather than animal health problems. National program coordinators are located at its principal laboratories in Palo Alto and coordinate national programs in the seven regional centers which have a network of 25 experiment stations. Scientific personnel number 236 professionals of whom 1.1 have PhD degrees, 45 hold degrees and the balance have undergraduate degrees. INIP has an urgent need to increase and upgrade staff and research facilities and capabilities to permit greater success in removing factors limiting production, particularly of smallholders who follow a mixed livestock-crop production system. However, its 1979 budget was only Mex$290 million (US$12.6 million). 2.12 INIFo Although forestry research was initiated some 50 years ago, INIF is the least developed of the three national research institutes. Begin- ning in 1977 it initiated a process of transformation and growth which resulted in a modification cf its structure and management. Currently it has three principal activities: research, training and technical support services.  -7- Research program coordination continues to be concentrated at the headquarters in Coyoacan. However, it is decentralizing as fast as facilities are made available at the six existing regional centers (two more are planned) and related experiment stations. Field research is thus carried out mainly at six locations, two of which are in arid, two in temperate and two in tropical areas, but most of its efforts are concentrated in the temperate pine forest region. Of the existing professional staff of 184, few have advanced degrees and many were trained in fields other than forestry. The inadequate budget (US$8.2 million in 1978) and paucity of trained staff are main limitations to strengthening the Institute's research capability and to rapid expansion of research and extension programs in forest production for rainfed districts. Soil and Water Conservation. 2.13 The DGCSA is a fairly recently created and still small General Directorate of SARH; it carries out its works through contractors and has a total professional staff of only 125. Its program has been largely limited to a single practice approach: erosion control, and mainly in areas of severe erosion with limited importance from an agricultural point of view. Although the quality of works has been adequate overall, questions have arisen as to benefits. There is a recognized need to (a) shift emphasis in erosion control work to areas w-ith lesser slopes which are still being cropped and where a productive impact could be obtained; and (b) increase training and demonstra- tion activities in the fields of soil and water conservation. Credit, Insurance and Supply of Inputs 2.14 Credit. The National Bank for Rural Credit (BANRURAL) was formed in 1975 through the amalgamation of three state-owned agricultural banks. With a staff of more than 24,000 in 12 regional subsidiary banks and more than 400 branches, it is the biggest bank in the country. At the close of 1979, its total assets amounted to Mex$70,991 million (US$3,141 million) and its total loan portfolio to Mex$55,490 million (US$2,455 million), of which some 27% was in arrears. BANRURAL's 1979 lending program amounted to Mex$37,000 million (US$1,600 million) of which 90% was short term and 10% medium and long term. 2.15 BANRURAL lends to approximately 17% of Mexican farmers nationwide. However, in rainfed areas - as opposed to irrigated areas - coverage is much lower, at around 8% in any given year. The low level of penetration is mostly due to infrastructural deficiencies which limit farmer access to credit and by BANRURAL's restrictions on lending to farmers who are in arrears. Lending is based on farm models specified on a per hectare or per livestock unit basis, containing a calender of activities, a list of goods required for operation, and a budget which implies no risk by incorporating normal year assumptions. These models are used to develop a lending budget for each branch by estimating the number of farmers that will apply and their categorization as to model. Short-term loan appraisal does not evaluate loans in terms of general improve- ments they might bring about in the farming system and to agricultural produc- tion. Also, coordination with research and extension services is generally absent. 2.16 BANRTRAL's overall financial position is weak and each year it has to be supported by large Government subsidies. Interest rates it charges vary fror 12" 5 tcr s_as aai credit and from 14.5% to 21.5% for medium- and lon-termE se:il d;uer its normal operations. Interest income in 1979  - 8 - was 9.:L% of net interest-earning assets. Against this, interest expense in 1979 was 7.2%, leaving it with a positive spread of 1.9%. This, however, was not enough to cove: non-interest expenses, and net incomes from banking operations before subsidies showed a deficit of Mex$3,672 million (US$162 million) in 1979. Nevertheless, this has not restricted the bank's operations because of its access t:o Government resources and guarantees. Annex 1 shows financial ratios based upon BANRURAL's balance sheets, operation and income statements over the past five years. 2.17 The Government subsidy to BANRURAL covers BANRURAJ's reserves for unrecoverable loans and a portion of BAIRURAL's administrative expenses. These costs arise clue to several factors. First, BANRURAL's operations are not well integrated with extension services. Borrowers generally receive weak follow-tp technical assistance, frequently only from the BANRURAL credit technician wFho must be both banker and extensionist because SARH extension services have been nearly non-existent in most rainfed areas. Second, the poor contact between BANRURAL technicians and SARH extension and research technicians means that BANRURAL staff frequently do not have the latest technical recommendations to give to their clientele. BANRURAL lending is based on standard models, with little variation to compensate for local conditiorLs and latest research findings; bank technicians therefore have little flexibility to incorporate new technologies in farm lending. The overall result is that most of BANRURAL borrowers in the rainfed areas get little or no technical service and hence run a higher risk of crop failure and loan default. With increasing demands for credit, BANRURAL's staff is being stretched even thinner in a generally unsuccessful attempt to be both a provider of credit and of technical assistance. Third, the Government uses BANRURAL to channel as loans, payments which are really grants made for social reasons with no expectation of recovery. Fourth, BANRURAL was created by merging three public sector banks. The administrative economies that were expected have not been achieved; staffing levels are excessive as BANRURAL follows a policy of staff reduction through attrition. The correspondingly high staff costs are an important reason for BANRURAL's high administrative costs. 2.18 Crop Instrance. The Government-sponsored crop insurance prog:ram, implementated by the National Crop and Livestock Insurance Agency (ANAGSA), covers all of the main cereals, oil crops and cotton. All crop production loans made by BANRURAL are contingent on the farmer carrying appropriate insurance. The insurance plan covers a maximum of 70% of the estimated production costs for irrigated crops and 50% for non-irrigated crops. The premium paid by the insured varies from 4% to 8.25% of total estimated produc- tion costs and is collected out of the loan at the start of the crop season. However, premiums collected from farmers cover only part of the indemnities paid; on average, some 40% of indemnities are funded by Government. 2.19 Fertilizer Supp