India - Third Agricultural Refinance and Development Corporation Credit Project

Document of The World Bank FOR OFFICIAL USE ONLY FILE COPY Report No. 2404a-IN INDIA STAFF APPRIAISAL REPORT OF THE THIRD AGRICULTIJRAL REFINANCE AND DEVELOPMENT CORPO]RATION CREDIT PROJECT June 21, 1979 South Asia Projects Department Agriculture Division D This document has a restricted distribution and mav be used by recipients only in the performance of their official duties. Its contents may not othierwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS US$1.00 = Rs 8.60 1/ WEIGHTS AND MEASURES Metric System ABBREVIATIONS ACD - Agricultural Credit Department (RBI) APC - Agricultural Projects Course ARDC - Agricultural Refinance and Development Corporation ARDC I - First Agricultural Refinance (and Development) Corporation Credit Project ARDC II - Second Agricultural Refinance and Development Corporation Credit Project CAB - College of Agricultural Banking, Pune CB - Commercial Banks CGWB - Central Groundwater Board DCCB - District Central Cooperative Banks FSS - Farmers' Service Societies GOI - Government of India ICAR - Indian Council for Agricultural Research LDB - Land Development Banks (Generic term for all SLDB and PLDB whether operating on a federated or unitary system) NIBM - National Institute of Bank Management PCR - Project Completion Report PACS - Primary Agricultural Credit Societies PLDB - Primary Land Development Banks PPA - Project Preparation and Appraisal Course RBI - Reserve Bank of India REC - Rural Electrification Corporation RRB - Regional Rural Banks SCB - State Cooperative Banks SFDA - Small Farmers Development Agencies SGO - State Groundwater Organizations SLDB - State Land Development Banks CROPPING SEASONS Kharif - June to September Rabi - October to February Summer - March to May FISCAL YEAR ARDC - July 1 to June 30 GOI - April 1 to March 31 1/ Until September 25, 1975, the Rupee was officially valued at a fixed Pound Sterling rate. Since then, it has been fixed against a "basket" of currencies. As these currencies are floating, the US Dollar/Rupee exchange rate is subject to change. Conversions in this report have been made at US$1.00 to Rs 8.60, the projected exchange rate over the disbursement period.  FOR OFFICIAL USE ONLY INDIA THIRD AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION CREDIT PROJECT Table of Contents Page No. I. INTRODUCTION .......................................... 1 II. THE AGRICULTURAL SECTOR ................................ I A. Agriculture in India ............................. 1 B. Land and Water Resources ......................... 2 C. Output Trends .................................... 4 D. Extension and Research ........................... 7 III. AGRICULTURAL CREDIT AND AGRICULTURAL CREDIT INSTITUTIONS ................................. 8 A. Background ....................................... 8 B. Cooperative Credit System ........................ 9 C. Land Development Banks ........................... 10 D. Commercial Banks .................................. 12 E. Regional Rural Banks .............................. 14 F. Farmers Service Societies and Village Adoption Schemes ............................... 14 G. Coordination ..................................... 15 IV. PERFORMANCE OF BANK GROUP FINANCED AGRICULTURAL CREDIT PROJECTS IN INDIA ................................... 15 A. Introduction ..................................... 15 B. State-Based Agricultural Credit Projects .... ..... 15 C. The First General Line of Credit Project (ARDC I) ............................... 17 D. The Second General Line of Credit Project (ARDC II) .............................. 18 V. THE PROJECT ............................................ 19 A. Introduction .................. ................... 19 B. Project Description ............. .. ............... 19 C. Detailed Features ................ ................ 20 (a) Minor Irrigation ........................... 20 (b) Diversified Lending .......................... 25 (c) Training .................................... 28 (d) Equipment ................................... 29 D. Cost Estimates ................................... 29 E. Financing .................... .................... 31 F. Procurement ................... ................... 31 G. Disbursements ....... ........... .................. 32 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.  - ii. - Table of Contents (cont'd) Page No. VI. PROJECT IMPLEMENTATION ................................ 32 A. Agricultural Refinance and Development Corporation .................................... 32 B. Lending Procedures ............................... 35 C. Terms and Conditions ............................. 35 D. Lending to Small Farmers ......................... 37 E. Lending in Less Developed States .... ............. 37 F. Accounts and Audit ................ 38 G. Monitoring, Evaluation and Reporting .... ......... 38 VII. PRODUCTION, MARKETING, PRICES AND FINANCIAL RETURNS TO PROJECT BENEFICIARIES ............................ 40 VIII. BENEFITS, RISK AND JUSTIFICATION ...................... 43 IX. AGREEMENTS REACHED AND RECOMMENDATIONS .... ............ 46 SCHEDULE A Lending Terms and Conditions SCHEDULE B Guidelines for Selection of Pumps and Prime Movers for Inclusion on State Approved Lists SCHEDULE C Guidelines for Disbursements Against Public Direct Irrigation and Augmentation Tubewells SCHEDULE D Guidelines for Levels of Evaluation of Groundwater Resources Appropriate to Levels of Resource Development SCHEDULE E Small Farmer Definition ANNEX I - Related Documents and Data Available in the Project File LIST OF TABLES AND CHARTS Table 1 - State and Primary Land Development Banks - Summary of Overdues Table 2 - Summary Data on IDA/IBRD Assisted Projects Through ARDC Table 3 - Cost of Equipment and Services for Strengthening SGO, SLDB and ARDC Table 4 - Estimated ARDC and IDA Disbursements. Table 5 - ARDC Condensed Balance Sheets 1973/74 - 1982/83 Table 6 - ARDC Condensed Statement of Income and Expenditure 1973/74-1982/83 Table 7 - ARDC Summarized Cash Flows 1973/74 - 1982/83 Table 8 - ARDC Total Projected Lending Program by State and Subsector 1979/81  Table of Contents (cont'd) Chart 19300 ARDC Organization Chart Chart 19830 Principal Agricultural Credit Institutions Chart 19809 Maajor Institutional Ftnding for Agricultural Term Loans MAP IBRD 12630R - India: GeneraLized Occurrence of Groundwater, Rainfall and Evaporation, and ARDC Regional Offices. SUPPLEMENTARY DATA VOLUME The following annexes are contained in a Supplementary Data Volume to the Staff Appraisal Report, and are available on request from Agriculture Division D, South Asia Projects Department. ANNEX 2 Water Resources and Minor Irrigation ANNEX 3 Principal Categories of ARDC Diversified Lending Program ANNEX 4 Agricultural Refinance and Development Corporation ANNEX 5 Agricultural Credit Institutions ANNEX 6 Training ANNEX 7 Monitoring and Evaluation ANNEX 8 Financial Analysis ANNEX 9 Economic Analysis ANNEX 10 Small Farmer Definition, ANNEX 11 Note on Disbursement Procedures and Statements of Expenditure ANNEX 12 Note on Interest Rates in India   * 1 - I. INTRODUCTION 1.01 In August 1975, an IDA general line of credit (ARDC I, 540-IN) was made to the Agricultural Refinance and Development Corporation (ARDC). 1/ That credit (US$75 M), which supported a project costing about US$180 M, was fully disbursed by August 1977, three months ahead of schedule. In August 1977, a second general line of credit (ARDC II, 715-IN) for US$200 M, was approved. That credit supports a project estimated to cost US$583 M and is expected to be fully disbursed by December 1979. In September 1978, the Government of India (GOI) asked IDA for further financial assistance to ARDC. 1.02 The project proposed for IDA financing would be a continuation of ARDC I and ARDC II, and a follow-up to a number of past and on-going projects in which IDA and ARDC have participated and which are described in Chapter IV. The project would give special emphasis to meeting the medium and long-term agricultural credit needs of small farmers, and to those States where agri- cultural development is lagging. 1.03 Several other country donors including United States Agency for International Development (USAID), Kreditanstalt Fur Wiederaufbau (KFW), Ministry for Overseas Development, Great Britain (ODM), and Canadian Inter- national Development Agency (CIDA), have expressed interest in supporting ARDC's lending program. 1.04 This report is based on an application prepared by ARDC, and on the findings of an appraisal mission which visited India in October/November 1978, consisting of Messrs. R. L. Headworth, C. Diewald, C. Helman, A. Rogerson, G. Slade (IDA), W. Barber, G. Kaddar, A. Stoneham and R. Van Wagenen (Consul- tants). Messrs. J. Beilby and D. Redden (USAID consultants) and F. Anema (CIDA consultant) joined the mission, and contributed to the report. A follow-up mission (R.L. Headworth) visited India in January/February 1979. II. THE AGRICULTURAL SECTOR A. Agriculture in India 2.01 India's population of about 640 M is growing at an annual rate of about 2.1% per year and per capita income (US$150 in 1976) at about 1.0%. Although average per capita income has increased, much remains to be done to improve the living standards of the vast masses of urban and rural poor which, 1/ Known then as Agricultural Refinance Corporation. Its origins, func- tions, structure and growth are described in 6.01 to 6.08, and its assocation with IDA in 4.01 to 4.07 below.  conservatively measured, consist of some 250 M people with annual incomes below the current estimated rural poverty line of US$70 per capita. Thus, Government of India's (GOI) recent development plan (Five-Year Plan (1978-83)) gives emphasis to maintaining self-sufficiency in food, alleviating poverty, and creating employment, especially in rural areas. 2.02 Agriculture (including crop production, animal husbandry, forestry and fishing) is the dominant sector of the economy, contributing about 42% of GNP, but it is also the one with the slowest growth rate. It engages about 70% of the labor force and provides the base for about 60% of exports. During the last decade, GOI has sought to raise foodgrain production by accelerating irrigation development, increasing the use of fertilizers, plant protection chemicals, and improved high yielding seed varieties. In support of this, GOI has promoted modernization and expansion of extension services and agri- cultural credit institutions. 2.03 Because of the limited scope for bringing new land into cultivation, the greatest part of future increments in food production, and foodgrains in particular, must come from increases in land productivity. A prerequisite for this is an assured irrigation water supply which: (a) provides supplemental irrigation in the rainy season (rainfall is often inadequate for optimal production of even a single rainfed crop); and (b) permit multiple cropping (rainfall over most of India permits only one assured crop a year). Conse- quently, GOI's Five-Year Plan (1978-83) provides for major investments in irrigation, both by the public sector, mostly in major surface irrigation schemes, and by the private sector, in groundwater development through minor irrigation involving dugwells, pumpsets and shallow tubewells. The project appraised in this report would help provide institutional credit support for this effort. 2.04 As the domestic foodgrain situation improves and irrigation devel- opment continues farmers have greater incentives to diversify their crops. Such diversification is important to: (i) better utilize India's diverse climatic and soil resources; (ii) produce a wider range of foods and raw materials in the interests of better nutritional standards, for example, milk and fish, and of increased agricultural exports, such as coffee or cashew products; and (iii) improve income and employment prospects of the rural poor who have little or no land, but who can participate in activities such as small-scale poultry and dairying. Diversified agricultural activities sup- ported by the proposed project would include a wide range of investments in horticulture and tree crops, livestock, marine and inland fisheries, market yards, storage facilities, farm machinery, and forestry. B. Land and Water Resources 2.05 India has a total land surface of about 328 M ha of which 175 M (53%) are classified as arable. The possibility of reclaiming additional land is limited. In 1974, about 142 M ha (81% of the arable surface) were under cultivation; allowing for multiple cropping, the total area cropped was about 169 M ha.  - 3 - 2.06 The southwest monsoon begins at the end of May or early June and continues until November/December; its duration is at a minimum in the north- west and at a maximum in the south. The wettest months are June-September when most of the country receives 80% or more of the annual rainfall. Mean annual rainfall ranges from 200-300 mm in the Rajasthan desert to more than 5,000 mm on the Western Ghats, and in the hills of Assam. The crucial months for rain- fed agriculture are July and August: the amount and distribution of rainfall in those months largely decides the fate of kharif crops. 2.07 The volume of annual3average recharge to groundwater is estimated to be 35 M ha m (350 billion m ). Allowing for irregular spatial distribu- tion of groundwater recharge, variation in demand, and the incidence of dry and wet sequences, it is not possible to regulate more than 70% of annual recharge. Therefore, the total utilizable recharge may be taken for planning purposes as about 25 M ha m. Net extraction in 1976/77 is estimated to have been about 6.5 M ha m (26% of utilizable recharge), and by 1981/82, net extraction may reach 8.8 M ha m (about 35% of utilizable recharge). The principal use and consumption of water has been and will continue to be for irrigation. 2.08 Total area with irrigation potential is about 110 M ha of which 76 M ha would be served from surface sources and 34 M ha from groundwater. By 1974, about 44 M ha had been developed of which about 17 M ha (36%) was from groundwater. GOI's Fifth Five.-Year Plan, which was closed one year in advance in March, 1978, is believed to have added an additional 3 M ha of groundwater irrigation thus bringing the total to nearly 20 M ha. The major part of groundwater development is attributable to the private sector. While the all-India groundwater balance indicates considerable scope for groundwater development, the present distribution of development relative to resource potential is variable. States such as Gujarat, Haryana, Punjab and Tamil Nadu, which have undertaken intensive groundwater development programs over the past two decades, have reached high levels of development in many areas, and over-development in some localities. Conversely, in several other States such as Bihar, Orissa and West Bengal, groundwater potential is large and extent of exploitation low. In a few States with heavy rainfall, for example, Assam, the proportion of groundwater development to potential is extremely low. 2.09 GOI's new Five-Year Plan (1978-83) aims at creating an additional 9 M ha to be served by minor irrigation developments of which 7 M ha would be served by groundwater and the balance by surface water. Emphasis would be on groundwater development in the least exploited areas (particularly in the soft rock areas of the Ganges basin and the Brahmaputra valley), and on private sector groundwater development with the support of institutional finance. Public tubewell schemes would be developed in areas where marginal farmers without the means to justify individual investments predominate, and conjunctive use would be made of groundwater and surface water in medium and major irrigation command areas. 2.10 ARDC's proposed five year program (1978-83) of groundwater develop- ment reflects GOI's policy to encourage development in less developed States and in States with high untapped potential. About 85% of ARDC's proposed lending for groundwater development would be in such States. The States  of Gujarat, Haryana, Punjab, and Tamil Nadu, which have seen intensive ground- water development over the past 15 years would account for only about 15%. Each State has extensive areas of high development, and overdraft conditions have occurred in parts of Gujarat and Tamil Nadu. While some potential remains in all four States, special care will be necessary in evaluating groundwater resources for future development (para 5.12). C. Output Trends Crop Production 2.11 Crop production contributes about 82% of sectoral production, as against 15% by livestock, 2% forestry and 1% fishing. Foodgrains constitute about 55% of gross value of production, and 75% of cropped area. Rice (38% by volume) and wheat (26%) are by far the most important foodgrains, followed by pulses (10%) and by other cereals like grain sorghum (9%), maize (6%), and various millets. The major commercial crops are oilseeds (8.1% of gross value), sugarcane (6.7%), fruits and vegetables (6.3%), cotton (3.0%), condiments and spices (2.8%) and tobacco, tea, and related commodities (2.4%). Overall crop production (still characterized by erratic fluctuations from year to year) was pushed steeply upwards in the late sixties - the "Green Revolution" - through an impressive adoption of high yielding wheat varieties, a new effort to expand irrigation facilities, and with simultaneously growing rates of fertilizer application. This rapid growth could not be sustained through the poor grow- ing seasons of the early seventies, but the four years through 1977/78 have witnessed a remarkable improvement, averaging 5% growth per annum, in which the impact of rapidly increasing input use has been enhanced by excellent growing conditions. 2.12 Other characteristic trends in crop production relate to the seasonal distribution of crops. The share of rabi (winter-season) crops, which depend mostly on residual soil moisture and irrigation, has been steadily growing, while the relative importance of kharif crops, which depend mainly on summer monsoon rains and supplementary irrigation, and are more susceptible to climatic vagaries, has been reduced. 2.13 Performance of particular crops. Wheat production and yield rose 7.6% and 4.6% respectively from 1960 to 1976, and its share in cultivated area from about 9% in the early 1960s to about 13% at present. Just over half of the crop is irrigated, and most production is concentrated in the northwestern States, where private tubewell development is most advanced. Rice had shown a disappointing performance until 1975-76, with production growing at less than 2% per year through modest yield increases (1.1%), and lower area increases. The low historical rate of growth is probably due to a high incidence of rainfed cultivation (over 60%), which was subject to alternating moisture stress and/or flooding. The area under HYV rice has, however, grown from 15% in 1970/71 to about 33% in 1977/78, and production has accelerated considerably during the last three years particularly in the northwestern region. Pulses, the traditional source of protein supply for most of the population have lost importance since the late 1950s as production economics  have made their production less attractive to farmers. Output of oilseeds and cotton has lagged behind the general trend of crop production. Oilseeds are still mostly grown on rainfed areas, resulting in only modest yield increases; thus, India is still dependent on imports of oilseeds and edible oils. Despite the introduction of improved varieties after 1970/71, cotton production, yields and quality are still low, mainly due to slow introduction of irrigation, wide price fluctuations, long growing seasons, and high risk of plant disease. Sugarcane, which is almost 75% irrigated and, therefore, less susceptible to rainfall fluctuations has shown significant increases in area and production, with yields growing more rapidly in recent years. 2.14 Prospects. Most recent estimates confirm that the stagnation of crop production in the early 1970s has been overcome. GOI's Five-Year Plan 1978-83 envisages a target growth rate of 4% per annum for agricultural output and most of this growth has to come from crop production. An increase of this order appears attainable in the light of the 4.3% per annum growth record over the past four years. Even so, India will have to face the possibility of a deficit in its foodgrain supply for some years to come if there is to be any sustained improvement in the nutrition of the poor. (Currently, foodgrain imports are negligible, due to high stocks and good harvests.) Extension and improvement of irrigation systems (including groundwater development), increase in seed and fertilizer supply, reorganization and expansion of sup- porting facilities and services (including agricultural credit), are essential to achieve the target growth rate. Plantation and Horticulture 2.15 India has a wide variety of climatic conditions, which favor almost all types of plantation/horticulture crops. Although these crops cover a small part of overall cropped area (about 2-3%), their contribution to the total gross value of agricultural production is considerable (about 12-13%). Coconut, tea, tobacco and banana are the four most important crops, followed by rubber, cashewnut, coffee and pepper. However, a wide variety of other fruits and spices are grown. Tea, coffee, tobacco, cashewnuts and spices, make up the major portion of all agricultural exports. The growth trend in plantation/horticulture production over the last decade has not been uniform. Increases in crops such as coconut and bananas have been disappointing, but crops like rubber, coffee, and tea, have improved faster. Better results * could be achieved by greater use of higher yielding planting material, more organized marketing, and more efficient specialized extension support. Growth prospectives for almost all crops in this category are good to excellent, as there is a reliable growing demand, both domestically and/or for export. Recent IDA projects in this subsector are listed in Table 2. Dairy 2.16 India has one of the highest densities of bovine population in the world, both in relation to human population and disposable land area. Of a total number of about 253 M head in 1972, about 178 M (70%) were cattle held mostly for draught purpose. The buffalo population, much smaller in size (about 57 M), is mainly for milk production, having higher milk yields and lactation efficiency than unimproved cattle breeds. Typically, about 70% of  - 6 - animal protein intake comes from milk consumption, although, at an annual production of 23 Mt, per capita consumption is still low (107 grams/day). The highest density of animals per ha of cropped area is found where human density is also highest, and amongst very small operational holdings, underlining the importance of cattle/buffalo husbandry for the poorest. Most of the milk produced is consumed locally although there is good demand from the cities. The Indian dairy development program, widely known as "Operation Flood," 1/ is attempting to overcome such problems as low productivity, scarce fodder base, inadequate veterinary services, and to improve facilities for the collection, processing and distribution of milk to cities. This has already benefited about I M farm families, organized in over 6,000 dairy cooperative societies at village level, and 28 unions at district level. Those families, which consist mostly of small and marginal farmers and landless laborers, are estimated to have raised their incomes by as much as 50%. The integrated approach is supported by a scientific program of cross-breeding local cows with exotic buils and upgrading buffaloes by selective breeding with high yielding local breeds. Currently, the organized sector provides about 18% of urban milk demand (19 M liters/day), and its share is growing fast. There is a vast demand for dairy production, and milk plant capacities are ahead of available supply. Poultry 2.17 The poultry industry has made considerable progress during the past two decades; it has emerged as a commercial enterprise from a backyard activity. The value of poultry products increased from Rs 650 M in 1961 to Rs 4,320 M in 1976. Annual egg production grew at about 9.6% per annum, from 2,340 M in 1961 to 9,230 M in 1976, raising per capita availability from 5.3 eggs/year to 15.2 eggs/year, which is, however, still far from sufficient. The number of eggs per layer is low, due to the fact that the bird popula- tion is dominated by local varieties. Bird population is increasingly being upgraded by imported exotic pure bred males and improved local parent stock. Major problems to be overcome include availability of reasonably priced feed and distribution to major producing pockets, control of poultry diseases, and a closer link between producers and consumers, the latter being concentrated in the major cities. Fishery 2.18 India Vas about 5,550 km of coastline with a continental shelf of about t15,000 km and an Exclusive Economic Zone in the Indian Ocean of about 2 M km . Total marine fish catch rose from 0.6 Mt in 1950 to 1.4 Mt in 1976. The potential sustainable yield is estimated at around 3.7 Mt. Present catches are almost completely taken from inshore waters, mostly less than 30 m deep. Marine fishing is dominated by a high percentage of shrimp catches (10-18% in quantity, over 50% in value). India has become the world's largest shrimp producer and exporter, with a world market share of 20%, and annual export of crustaceans of 48,000 tons, valued at about US$214 M per year. The rapidly 1/ For further details see the IDA-supported National Dairy Project, May 1978 (Report No. 1964-IN).  -- 7- increasing fleet of 14,000-16,000 mechanized vessels is mostly used for shrimp fishing. 95% of the fishermen operate about 200,000 traditional vessels, and catch about 70% of overall fish production. For shrimp, a sufficient and grow- ing processing and marketing infrastructure has been created but there are signs that shrimp fishing in some coastal areas might already be approaching its safe limits. Non-shrimp fishery still suffers from uncertainties as to the extent and exact location of commercially and profitably exploitable resources. More resource studies and explorations are necessary, together with an effec- tive organization of processing and marketing, to satisfy a growing demand. 1/ Inland fishery is, as yet, not well developed but has substantial potential. Production rose from 0.28 Mt in 1960 to about 0.8 Mt in 1976. Taking into account available and reclaimable water areas and advanced composite fish culture techniques (mostly for carp varieties), potential production is estimated at 3.5 Mt. Brackish water shrimp culture might be another area where development is possible. Practical problems over marketing, lack of extension, breeding of suitable vari,eties for fish seed supply and institu- tional arrangements are being addressed so as to permit faster growth of this activity. D. Extension and Research 2.19 The agricultural extension system, for which individual States are responsible, has traditionally been integrated with a wide range of other community development activities, with the result that multipurpose village level workers were neither able to direct most of their efforts to extension work nor were they efficiently organized and oriented towards a planned and supervised agricultural extension program. Contact with and feedback to agricultural research have been insufficient. In addition to these organi- zational problems, the extension service has suffered from inadequate and outdated staff training. GOI has become increasingly concerned with improv- ing this situation as yields on irrigated and rainfed areas are well below potential. With IDA assistance, a new system of extension, based on frequent and regular farm visits by full-time, adequately trained extension workers, was developed in 1974. The introduction of this system is showing encouraging results, and it has since been adopted in nine States, i.e., Assam, Bihar, Gujarat, Haryana, Karnataka, Madhya Pradesh, Orissa, Rajasthan, and West Bengal. 2/ 2.20 Notable progress has been made in recent years toward the develop- ment of a highly competent, professional, and effective agricultural research system. When fully developed, this system should be capable of generating and developing the scientific and technological improvements required for 1/ The development of the marine fishery subsector is described in the staff project report on the Andhra Pradesh Fisheries Project (Report No. 1896a-IN, May 1978). 2/ See Composite Agricultural Extension Project (Report No. 2171a-IN November 30, 1978).  - 8 - a sustained and rapid growth in agricultural productivity. The system has already demonstrated its effectiveness in several important areas, notably with wheat, rice and maize. Agricultural research is coordinated by the Indian Council of Agricultural Research (ICAR) and carried out at specialized ICAR centers and at 21 State Agricultural Universities in 16 States. There is, however, a growing need to extend scientific advances from the relatively few major research centers and adapt them to other ecological situations. Emphasis in the future should be on strong regional stations in major eco- logical zones, and on development of research programs serving specific zones. This is especially necessary for serving small farmers in rainfed areas. The IDA supported National Agricultural Research Project (Report No. 2083a-IN) would strengthen the capability of Agricultural Universities to do such area- specific applied research, with particular attention to foodgrains under rainfed conditions, and to improved mixed farming systems. III. AGRICULTURAL CREDIT AND AGRICULTURAL CREDIT INSTITUTIONS A. Background 3.01 Private moneylenders still provide the bulk of credit in rural areas although their share of outstanding rural debt is estimated to have declined to about 65%, compared with as much as 90% in the early 1960s. 1/ Their gradually diminishing role can be attributed not only to social pres- sures and borrower protection legislation, but also to growing accessibility of institutional credit for agricultural investments on more favorable terms. Direct agricultural loans by institutions amounted to about Rs 27,550 M in 1978 2/ as against about Rs 11,900 M in 1974. Medium and long-term lending for agriculture nearly tripled over the same period to Rs 9,230 M in 1978 from Rs 3,270 M in 1974, a real increase of over 100%. GOI's current Five Year Plan calls for a doubling (in nominal terms) of total institutional credit for agriculture by 1980/81. 3.02 The main institutions involved in term lending for agricultural development and their principal sources of funds, are shown in Charts 19809 and 19830. The rural banking system is comprised of: cooperative banks (about two-thirds of outstanding agricultural loans in 1978); nationalized and private commercial banks (nearly one-third); and a number of smaller institutions, such as Regional Rural Banks (RRB), which combine features of both cooperative and commercial banking. All these banking windows offer farmers similar terms and conditions under Reserve Bank of India (RBI) guid- ance. Long-term funding of the agricultural credit system is provided mainly by: open market bond operations supervised by RBI; refinance from ARDC; time deposits; share capital contributions and loans from State governments, RBI and public corporations. 1/ All-India Debt and Investment Surveys. 2/ Provisional.  - 9 - B. CooperaLtive Credit System 3.03 The Indian cooperative credit movement, dating back to the late 1800s, was until recently the only institutional source of finance in most rural areas. Since the mid-1920s, when the distinction between short and long term credit needs first crystallized, cooperative banks have been organized by individual States into two independent groups of institutions: (a) land development banks (long-term credit system); and (b) cooperative banks (short- and medium-term credit system). There are land development banks (LDB) in 17 States, and each LDB consists of a State Land Development Bank (SLDB) which lends through its own branches, or refinances affiliated primary land develop- ment banks (PLDB). LDB provide loans for agricultural investments for periods of between 3 and 15 years. They do not make seasonal loans, accept individual deposits, or lend for non-agricultural purposes. The cooperative bank struc- ture is much larger and more diversified. In 1977 (the latest data available), 26 State Cooperative Banks (SCB) provided funds for 344 district-level Central Cooperative Banks (DCCB) which, in turn, supported about 123,000 village-level primary agricultural credit societies (PACS). These PACS, which specialize in seasonal and medium-term (1 to 5 years) credit, are also designed to provide a variety of services to their members, including savings facilities, and provision of consumption loans. 1/ 3.04 Integration. The split cooperative credit structure has obvious disadvantages. PACS tend to have aL closer relationship than LDB with poten- tial borrowers, yet lack experience and trained staff for long-term lending. LDB, on the other hand, do not have such extensive grassroots organizations as PACS, or close contact with their members' other banking activities. More- over, farmers remain unable to obtain all their credit requirements, both seasonal and term, from a single cooperative organization. 3.05 A study of the advisabiliLty of integration of the two systems, con- ducted by GOI under ARDC I, recommended a gradual merger at all levels, and a few States 2/ are experimenting with various methods of achieving the main objectives of integration. It is too early yet to assess to what extent the experiments have been successful but during negotiations, an undertaking was obtained from GOI to review progress made by the cooperative credit system towards meeting the agricultural credit needs of farmers through one source or closely coordinated sources and to inform IDA by December 31, 1980 of its findings and proposals arising from the review. A major obstacle to early changes remains the extreme weakness of a large number of PACS. SCB currently receive less than 1% of annual ARDC disbursements and that is concentrated largely in areas which either have no LDB, or in specialized lending sectors, such as fisheries and dairy, where local cooperative societies have developed 1/ A full description of the State Cooperative Bank structure is given in Staff Appraisal Report on the National Cooperative Development Corpora- tion Project, December 1978, (Report No. 2198-IN). 2/ Such as Punjab, Andhra Pradesh, and West Bengal. These experiments usually involve sharing of staff and facilities on an agency basis, but not joint financing or merger of the two structures.  - 10 - the necessary expertise. However, the SCB share of ARDC refinanced activities is expected to increase gradually as more PACS are reorganized under profes- sional management. C. Land Development Banks (LDB) 3.06 LDB's annual share of ARDC refinance declined from nearly 90% in 1974 to less than 50% in 1978, whilst that of commercial banks grew propor- tionately. In absolute terms, however, LDB disbursements more than doubled in the same period; they still account for two-thirds of ARDC's total out- standing portfolio, and over half of total outstanding medium- and long-term credit. The main sources of SLDB funds are: (i) special development deben- tures, sold to ARDC and State governments, and also guaranteed by the latter, bearing 6.5% to 7.5% interest, and, retired in annual installments over actual maturities (5 to 15 years) of loans to which they relate; and (ii) ordinary debentures, floated on the wider institutional market at a similar rate (currently 6.25% at 10 and 15 years) and repayable in full at maturity. LDB share capital is owned by the respective State government, and by farmers who are normally required to deposit 5% to 10% of the loan amount in their share accounts for the duration of the loan. 3.07 The most critical problem facing LDB is the growing level of their overdues. 1/ State-by-State details are given in Table 1. Regardless of regional variations, a pattern emerges of a noticeable overall improvement between 1974 and 1976 (a period when performance worsened in four States but improved significantly in eight others), followed by a general relapse between 1976 and 1978. In six States (Bihar, Gujarat, Himachal Pradesh, Karnataka, Maharashtra, and Tamil Nadu) overdues from ultimate borrowers are currently close to or more than 50% of demand, and in two of them (Gujarat and Maharashtra), they have been over 50% for at least three consecutive years. 2/ Overdues of these proportions progressively weaken the credit system through- out the State and ultimately threaten the institutions concerned. 3.08 Sliding scale eligibility criteria were introduced under ARDC I and continued for ARDC II, as a condition of IDA financial support. The main objective of the criteria is to provide an incentive to SLDB branches or PLDB with poor recoveries to improve the situation by linking their future lending volume to their recent collection performance. The criteria are amended as appropriate (for example, when abnormal climatic conditions occur) on the recommendations of an ARDC/RBI Debenture Norms Committee which was established for that purpose, and such amendments are agreed with IDA. 3.09 Natural calamities and reluctance by some LDB to reschedule loans in cases of genuine hardship have contributed to the deterioration. In most 1/ In India, overdues at the end of the fiscal year are expressed as a per- centage of "demand" (principal and interest falling due during the year, plus overdues from previous years). Given the greatly varying criteria used in different countries, it is not possible to compare India overdue statistics with those of other countries. 2/ Provisional figures as at June 30, 1978.  - 11 - cases, however, the deterioration from 1976-78 can be traced to direct inter- vention by State and local authorities with LDB collection efforts during elections. The most effective and essential factor for improving collections is political support of SLDB recovery efforts. Following ARDC intervention, some State governments are taking corrective action by supporting LDB recovery activities. However, in a few States (Gujarat and Maharashtra in particular), the situation is deteriorating and corrective action is urgently required. GOI, RBI and ARDC are holding discussions with the State Governments concerned to try and improve the situation. 3.10 ARDC has agreed to establish more stringent criteria, satisfactory to IDA, for LDB to be eligible for refinance. The existing Debenture Norms Committee (para 3.08) would advise ARDC on criteria for all cooperative banks, and a new Committee an Agricultural Loans through Commercial Banks would advise ARDC on commercial bank matters (para 3.18). ARDC eligibility criteria would include the following changes to the current eligibility criteria for LDB: (a) Lending by individual PLDB or SLDB branches would continue to be regulated on a sliding scale basis, but the minimum PLDB/ SLDB branch recovery level would be raised from the present 45% to 50% of demand. (b) If 50% or more of any LDB's lending units (PLDB or SLDB/ branches) became ineligible under the sliding scale criteria, or if overdues at SLDB level were 50% or more, ARDC's refinanc- ing under the Project would be subject to the State Government agreeing to a specific timebound rehabilitation program acceptable to ARDC; (c) The rehabilitation program under (b) above would be subject to an annual review, and the performance of the LDB in carrying out the program would be the basis for continuing ARDC refinancing support. ARDC would keep IDA informed of the agreed rehabilitation program and of the time schedule for implementation. 3.11 An undertaking was obta-ined from ARDC at negotiations that ARDC would establish and maintain eligibility criteria for LDB, such criteria to be acceptable to IDA. During negotiations, IDA and ARDC discussed and reached agreement on new minimum criteria for LDB effective from October 1980. 3.12 Based on current provisional data, Gujarat and Maharashtra LDB would not be eligible to participate in the project under the provisions contained in para 3.10 above unless they took firm corrective action; in four other States -- Bihar, Himachal Pradesh, Karnataka, and Tamil Nadu, overall performance is only slightly better but as specific action plans are under discussion by the respective State governments and ARDC, they would probably be eligible under the sliding scale criteria by the time the proposed credit becomes effective. State government support is the key to LDB performance in every State, and if that support is forthcoming LDB should be able to participate to the fullest extent in the project. Without that support, institutional credit channels for agricultural development would be impaired  - 12 - to the point where further increases could not be justified. Meanwhile, ARDC's projected lending program has been estimated conservatively to allow for likely institutional constraints. 3.13 While considerable progress has been made, a difficulty faced by some LDB remains their lack of sufficiently qualified management and techni- cal staff. Although chartered as independent cooperatives, LDB are sometimes treated for administrative purposes as if they were part of State governments, with the result that State government officers with little or no relevant experience may be posted to senior SLDB management positions, and be trans- ferred again before they can operate efficiently in their new environment. Several State governments are beginning to appreciate that this practice has had a seriously adverse affect on SLDB operations, and are taking steps to recruit management and senior staff for direct appointments to SLDB estab- lishments. However, it will take some time before this becomes standard practice throughout India. ARDC will continue to move LDB in this direction by insisting on adequate supervision arrangements as a condition of scheme approvals. D. Commercial Banks (CB) 3.14 With over 28,000 branches, nearly 12,000 of which are in rural areas, the CB network is expanding at the rate of 3,000 to 4,000 new branch openings per year. However, CB involvement with agricultural lending (and agricultural term loans in particular), is still comparatively recent and such loans occupy only a modest proportion of CB total business. CB lending to the agricultural sector increased from 5% of total operations in 1969 to nearly 11% in 1978. About four-fifths of the number of agricultural loans and nearly half of CB agricultural lending volume are seasonal loans. 3.15 CB involvement with ARDC increased sharply in the last five years, from Rs 45 M (less than 5%) of ARDC refinance in 1973, to Rs 1,200 M (over 50%) in 1978. This change is due to a number of factors, the most important of which are: the relative decline of the LDB system (para 3.07), the gradual absorption of significant numbers of trained agricultural staff in CB branches, and deliberate GOI and RBI policy measures to encourage CB agricultural lend- ing. In early 1978, RBI instructed CB to reduce their agricultural interest rates, then about 13%-14%, and to adopt the ARDC-prescribed onlending interest rates as a maximum for all term loans for specified agricultural purposes, regardless of whether those loans were refinanced by ARDC or from the CB's own long-term resources. Since the latter are relatively scarce, and more costly than ARDC refinance, 1/ CB are expected to turn increasingly to ARDC refinance as the principal means of reaching the priority sector targets set by GOI. As a consequence, a much greater proportion of CB term lending for agriculture would come under ARDC supervision. 1/ See Chart 19809. The effective cost of term deposits is raised by RBI minimum cash reserve and liquidity ratio requirements.  - 13 - 3.16 Commercial banking facilities and expertise in agricultural lend- ing vary greatly from State to State and between institutions. The most prominent CB is the State Bank of India group with over 7,000 branches operat- ing throughout the country, and absorbing over half of total ARDC refinance to CB. Fourteen other nationalized CB each have branch networks covering a number of States. A small number of private banks operate on a much smaller scale, typically in a single State or a few adjacent States. The larger and more progressive organizations, such as State Bank of India, have opened hundreds of branches specializing in agricultural lending, each staffed with several qualified supervisors and equipped with suitable transport facilities. At the other end of the scale, smaller CB continue to offer agricultural loans as a minority activity of semi-urban branches, supervised by only one or two qualified junior officers whose heavy workload and lack of transport precludes adequate contact with borrowers. Supervision and assistance from ARDC and RBI is gradually helping to improve the situation in these cases. 3.17 As CB began to engage in agricultural lending, pursuant to GOI policy, it became clear that a number of obsolete legal and procedural restrictions, originally aimed at protecting cooperative banks from compe- tition by CB, were inhibiting their activities. The major constraints, which were identified by the Talwar Committee 1/, have now been largely or fully removed in most States. 2/ Some of the restrictions which have been overcome are: exemption of stamp duty, simplification of mortgage registration procedures, and removal of cooperative bank priority over charges created on security. 3.18 Under ARDC I and ARDC II, refinance of CB agricultural loans by ARDC has not been linked to specific performance criteria, such as overdues, mainly on the grounds that: (a) lending for agriculture was still a new field of operations into which CB we!re reluctant to expand, and they needed as much encouragement as possible; and (b) term agricultural lending involved such a small part of CB portfolio and resources that even heavy agricultural overdues could be borne without jeopardizing the financial stability of the institutions. Whilst the second argument still holds true, CB agricultural operations have now reached sufficient scale and maturity to justify closer monitoring of performance, including periodic review of overdues indicators. Introducing monitoring and controls based on overdues, however, presents statistical difficulties since, as yet, not all CB calculate overdues on the same basis. ARDC is attempting to overcome this by conducting seminars throughout the country, and by reaching agreement on a standard definition of "demand" as now used by LDB. Nevertheless, statistics available to date indicate that in many States, CB experience in long term agricultural lending is only marginally better than that of LDB and in a few others, it is worse. CB short- and long-term agricultural overdues combined reached an estimated 51% on an all-India average basis in 1978, with a high of about 68% in the 1/ Report of the Expert Group on 'State Enactments having a bearing on Commercial Banks' lending to Agriculture - Bombay, 1971. 2/ Legislation is still pending in Kerala, Tamil Nadu, Gujarat, Andhra Pradesh, and Jammu and Kashmir.  - 14 - northeastern region. A major recent cause of overdues in many States appears to be political pressure on LDB not to take coercive action on recoveries at the time of elections (para. 3.09), and this has had repercussions on the recovery efforts of other financing institutions. For this project, a new Committee on Agricultural Loans through Commercial Banks has been established to advise ARDC on appropriate appraisal criteria, policies and procedures for improving recovery disciplines in commercial banks (para 3.10). The Commit- tee's first recommendations are expected to be made by the end of 1979; in the meantime ARDC would continue to apply its present procedures keeping in view inter alia the adequacy of technical and financial supervision arrange- ments and paying particular attention to recovery procedures. E. Regional Rural Banks (RRB) 3.19 In an effort to combine some of the perceived advantages of CB and of cooperative banks, some 50 RRB have been set up since 1975. These banks are owned jointly by GOI, a sponsoring CB which also provides management expertise, and the State government concerned. RRB accept deposits and make loans of varying maturities under similar RBI regulations as CB, with the major difference that they deal mainly with small and marginal farmers, artisans, and other specified target groups. These RRB have opened about 800 branches which are still in early stages of development: they account for less than 1% of total institutional finance for agriculture. RRB growth is likely to continue to be gradual over the next few years, but as they develop and take increasing advantage of ARDC refinance facilities RRB are expected to begin to make a significant impact on ARDC small farmer coverage. F. Farmers Service Societies and Village Adoption Schemes 3.20 CB are linked to cooperatives through Farmers Service Societies and village adoption schemes. Farmers Service Societies (FSS), of which there were about 350 in 1977, are expected to provide packages of agricultural inputs and management services alongside credit, for which they are sponsored either by a cooperative bank, or a CB. FSS, launched under the aegis of the National Commission of Agriculture in 1974, are intended to cover a wider geographical area than PACS, and employ more specialized staff. Nonetheless, the two institutions frequently overlap and the evidence is that FSS perform- ance with respect to small farmer coverage, overdues, minimum volume of business, and other yardsticks is usually no better than local PACS. Under a typical village adoption scheme, a CB branch provides seasonal loans to PACS for onlending to its members. Some 4,000 societies were financed in this way in 1977. Although some State governments are encouraging CB to route an increasing proportion of their lending through adopted PACS, the administrative difficulties involved have proven to be substantial, and few PACS show significant improvement after CB sponsorship.  -- 15 - G. Coordination 3.21 LDB, which are closely linked to State governments, normally get priority when responsibility for providing area development schemes is distributed amongst institutions. More recently, however, where LDB are weak or they are ineligible for ARDC support, such schemes are increasingly offered by some State governments to other institutions, or schemes are generated by CB themselves. In nearly every block or district, a particular banking insti- tution has established a commanding position in agricultural term lending. This situation has been formalized by RBI into the "lead bank system" whereby one institution is given the responsibility of proposing and coordinating a comprehensive lending program for the district in conjunction with district authorities and other banks. Steps to improve planning and coordination at State and district levels are emphasized as part of the GOI's new Five-Year Plan, and are essential to this project. IV. PERFORMANCE OF BANK GROUP FINANCED AGRICULTURAL CREDIT PROJECTS IN INDIA A. :Introduction 4.01 ARDC's association with the Bank Group dates to the formulation of a series of State-based agricultural credit projects between 1969 and 1974. Eleven of those projects have since been implemented through ARDC: nine are fully disbursed and the remaining two are scheduled to close during FY1980. More recently, ARDC has been entrusted with two general lines of credit, ARDC I (August 1975 to July 1977) and ARDC II (August 1977 to December 1979), which provide the immediate background for the proposed project. In addition, ARDC has taken an active role in the implementation and formulation of another 20 IDA/IBRD projects approved to date (Table 2), mostly by providing long-term finance and loan supervision in support of investments in major irrigation, horticulture, fisheries and seed production. Other projects with possible ARDC involvement are currently under consideration. The total amount commit- ted by IDA for disbursement through ARDC in the 33 projects approved to date exceeds US$850 M (Table 2). With the present project, this amount would increase to about US$1,100 M. As of March 31, 1979, cumulative IDA disburse- ments to GOI for onlending through ARDC were about US$504 M (59% of commitments). B. State-Based Agricultural Credit Projects 4.02 The 11 State-based agricultural credit projects 1/ were designed mainly to support three and four year lending programs by commercial and cooperative banks to farmers for investments in minor irrigation (about 80% 1/ Including a Rural Development project in West Bengal whose main component is an agricultural credit program through ARDC.  - 16 - of the total program), land levelling, and, in six States, farm mechanization. Nearly all the projects required the State government concerned to establish or strengthen an agency responsible for monitoring the scale and quality of groundwater investments financed through institutional credit. Each project required participating banks to adopt specific loan appraisal procedures based on incremental income calculations, to observe uniform lending terms and con- ditions, and usually to implement various improvements in their staffing and management methods. Nine of the projects closed between 1975-1978. Project completion reviews have been carried out on six of them, and are in progress on the remaining three. 1/ The main conclusions to date of this ex-post evaluation program can be summarized as follows: (a) the main objective of the projects, that of achieving increased agricultural production through expansion of irrigation and the use of modern technology, was clearly achieved. Returns to individual farmers and to the economy are in a similar range to those estimated at appraisal (financial rates of return ranging from 11% to over 50%, and economic rates of return from 18% to over 50%); (b) some LDB (Punjab, Haryana, Andhra Pradesh in particular) performed well in terms of substantially increasing lending whilst maintaining good control over loan appraisal and recovery. However, the long-term agricultural credit system in a few States (notably Gujarat, Maharashtra and Tamil Nadu) has been severely damaged by persistent overdues which continued to grow between 1976-1978 (para 3.07), and for which the most convincing explanations remain wilful default, poor management, and political intervention at State and local levels; (c) a gradual change from purely security-oriented lending to appraisals based on incremental income has been achieved. Although this process is not yet complete, it has led to a substantial improvement in the quality and coverage of agricultural lending; (d) the groundwater resource balance in a few limited areas is precarious enough to warrant careful control over new in- vestments. However, the spacing and density criteria in- troduced under these projects are not applicable to privately-financed investments, and are, therefore, only partly effective. The remaining two projects in Bihar and West Bengal are scheduled to close in March 1980. In all projects, the pace of disbursements in minor irrigation 1/ Out of these reviews, two (Haryana and Punjab) were based on ARDC's own completion reports, and ARDC is preparing completion reports for the remaining three (Karnataka, Madhya Pradesh and Uttar Pradesh).  - 17 - categories has matched or exceeded iexpectations, whereas land development loans and loans to groups of farmers have been slowed by organizational or legal difficulties, which are only lnow being overcome. 4.03 Out of the nine projects which have been closed, one Project Performance Audit has been carried out by the Operations Evaluation Depart- ment with respect to the Gujarat Ag.ricultural Credit Project. The main recommendations of the Audit Report (Report No. 1303 dated October 4, 1976) were: (a) the necessity of enacting legislation to control. groundwater resources and development; (b) the need for SLDB to reschedule overdue loans in areas affected by severe weather, including project loans as appropriate; (c) the necessity for SLDB to initiate schemes to diversify its lending in sectors other than minor irrigation; (d) the necessity for SLDB to continue to implement, at the fastest rate possible, its program of improvement of its managerial, financial and organizational operations; and (e) the need for improvement in the extension service. These findings parallel those of the project completion reviews and were taken into account while designing this project. C. The First General Line of Credit Project (ARDC I) 4.04 In April 1975, IDA approved a general line of credit of US$75 M for disbursement over two years on the general understanding that it would be the first in a series of such credits to ARDC. The objective of the project was for ARDC to commit specific amounts for individual schemes (appraised by ARDC and not by IDA) within an IDA-approved overall lending program. The schemes are normally too small for IDA to approve individually. The project comprised: (i) loans for minor irrigation; (ii) loans for diversified agricultural invest- ments; (iii) a study on the feasibility of merging the short/medium-term and long-term cooperative credit institutions; and (iv) a study of the training needs of LDB junior-level staff; training of such staff; and an intensive two- year training program primarily for senior and middle-level LDB staff. The credit became effective in August 1975, and closed September 1977, three months ahead of the original closing date. 4.05 A project completion review has been completed and its findings have been fully considered in the diesign of this project. The main conclu- sions and recommendations are:  - 18 - (a) the project achieved its physical objectives, and although the investments financed have not yet reached full produc- tion stage, indications are that incremental production generated by the project will exceed appraisal expectations. Some 185,000 minor irrigation units (including 85,000 pump- sets and 43,000 dugwells), 3,400 ha of land development, 9,200 ha of plantation/horticulture and 17,000 dairy animals were financed under the project; (b) although the project target for lending to small farmers (some 110,000 farmers, over 50% of the IDA credit) was achieved, in future projects, overall small farmers cover- age should be based on individual State targets to match achievement with regional variations in landholding patterns. (c) with regard to the LDB system, although the high level of overdues continues to be a serious problem, the impression is that the debenture eligibility criteria (para 3.08) is nonetheless an improvement on undisciplined lending. The criteria should be continued, amended as necessary, and its effectiveness studied over a longer period; (d) expansion of commercial bank refinance through ARDC has been beyond all expectations and minimum institutional standards for CB participation should now be introduced (para 3.18); and (e) ARDC's standard of appraisal and supervision was satisfactory. However, there must be a continuous adjustment of ARDC staffing and organization to keep up with its substantial financial growth. D. The Second General Line of Credit Project (ARDC II) _/ 4.06 The second line of credit (ARDC II), which supported ARDC opera- tions throughout India for two years, was approved in May 1977 and has been effective since August 1977, shortly before ARDC I closed. The credit amount was increased from US$75 M in ARDC I to US$200 M in ARDC II (but down from 44% to 34% of project cost) to reflect the substantial increase expected in ARDC commitments and disbursements, an expectation which has been fulfilled. The diversified lending component was increased from US$4 M to US$25 M to reflect the changing composition of ARDC's portfolio. The credit also includes a major training program for staff of agricultural financing institutions. Terms and conditions, including debenture eligibility norms, in most respects are similar to those of ARDC I, although the scope for eligible borrowers has been widened to include recipients of SFDA 2/ capital subsidies. 1/ Detailed features may be found in Report No. 1520-IN, Second Agricultural Refinance and Development Corporation Credit Project, May 1977. 2/ Small (and marginal) Farmers Development Authority.  -- 19 - 4.07 Disbursements under this project as at March 31, 1979 amounted to US$91 M, over 75% of appraisal estimates and ARDC had accumulated sufficient eligible refinance disbursements to cover the balance. Mid-term estimates of small farmer coverage show that 54% of disbursements in minor irrigation, and 52% overall, are accounted for by small farmers. Preliminary indications of physical achievements are also in line with appraisal estimates. As of September 30, 1978, some 160,000 minLor irrigation units had been financed under the project, including 45,000 dugwells and 72,000 pumpsets, nearly half the total units estimated through December 1979 at appraisal. The project includes a considerable strengthening of ARDC's monitoring and evaluation cell; the Cell is beginning to operate effectively, although delays in recruitment and transfer of staff have so far made it difficult for ARDC to keep pace with the rapidly increased workload. A number of studies have been commissioned as required under project covenants. The main ones are: (i) a sample survey of areas likely to be subject to the risk of groundwater exploi- tation; (ii) a study of the adequacy of interest rate spreads, with particular reference to LDB; and (iii) a survey to estimate the magnitude of pumpset replacement demand. The surveys are well advanced and should be completed before the credit closes. Other project covenants are being fulfilled. V. THE PROJECT A. Introduction 5.01 The Government of India (GOI) has requested an IDA credit to help finance: (a) a third two year time slice (approximately December 1979 through December 1981) of the Agricultural Refinance and Development Corporation's (ARDC) lending program; (b) further training of staff, mainly of participating financing institutions; and (c) the purchase of equipment for State Groundwater Organizations (SGO), State Land Development Banks (SLDB), and ARDC. B. Project Description 5.02 The objectives of the project would be to: (a) increase agricultural production and to raise the produc- tivity of farmers, particularly small farmers and those living in less developed areas by supporting ARDC in its financing and developmental activities (paras 5.04 to 5.21); (b) further encourage institution building: (i) in CB, through the introduction, by ARDC, of a Committee on Agricultural Loans through Commercial Banks (para 3.18);  - 20 - (ii) in SGO through provision of additional staff and equipment (para 5.12); (iii) in participating banks, mainly LDB, through inten- sified staff training programs (para 5.23); and (iv) in ARDC itself by initiating changes in staffing and organization patterns (para 6.07). (c) improve the quality of investments, particularly in minor irrigation, through the establishment of a technical support system in all States (para 5.08); and (d) increase small farmer participation by disbursing up to 60% of the amount of the IDA credit to small farmers (50% in ARDC I and ARDC II) (para 6.15), and help remove regional imbalances by promoting development in lesser developed States (para 6.17). 5.03 The project would support the whole of ARDC's lending operations in virtually every district in India, over two years. However, for admini- strative simplicity, IDA would restrict its own disbursements to a few major categories, such as minor irrigation (including land development), plantation, horticulture, livestock and fisheries. Project funds would be channeled through ARDC which would refinance loans made by participating banks in accordance with specific lending terms and conditions (Schedule A). Final beneficiaries would be individuals or groups of farmers and fishermen, coop- eratives, public and private sector companies, and corporations. As under ARDC I, ARDC II, and the State oriented projects, participating banks would develop schemes, normally for a single activity such as shallow tubewells or fisheries, and in a compact physical area in which necessary support services would be provided. Minor irrigation development would predominate, represent- ing about 63% of project costs which are estimated at US$1,005 M. The project would also provide US$2.0 M to continue training agricultural banking staff started under ARDC I and ARDC II, and US$0.6 M for the purchase of equipment for State Groundwater Organizations (SGO), SLDB and ARDC. ARDC would endeavor to ensure that small farmers receive at least 60% of the IDA credit, and an estimated 45-50% of project lending would be disbursed to States where agricultural development is lagging. C. Detailed Features (a) Minor Irrigation 5.04 Development of groundwater for irrigation in India over the past two decades has been rapid and extensive. This has been particularly true of private sector development when assisted by institutional finance. The importance given to groundwater development in this project reflects GOI policy to expand exploitation of this resource.  - 21 - 5.05 While considerable groundwater resources remain for further devel- opment in most States, high levels of resource utilization have been achieved in several States and pockets of excessive development have occurred in others. Recognizing the necessity of achieving full resource development within safe yields of aquifer systems, groundwater programs for institutional finance would, under this project, be linked to controlled resource evaluation. The controls specify increasingly detaiLed levels of investigation required with escalating levels of resource development so as to provide a measure of security to institutional loans by ensuring that scheme appraisals would be made from a fuller data base in the case of highly developed areas. 5.06 There is considerable scope for improvement in the technical stand- ards of private sector groundwater investments and, in particular, of pumping equipment. Such improvement would result in operating cost savings to farmers, and energy savings at State and all-India level. Consequently, the project would define technical standards for water points in each State, and identify technical capability requirements within the banking systems to apply such standards to individual loans (para 5.08). 5.07 Loans would be made to individual farmers, groups of farmers, coop- eratives and State Corporations. Minor irrigation investments would include development of both groundwater and surface water resources. A relatively small part of the project would be for provision of lined conveyance in irri- gated farm areas, and for development of tanks. Loans to individual farmers for investment in groundwater development would be for construction of new wells, improvement or rehabilitation of existing wells ancd provision of pumpsets. Credit for well construction would include construction of open dugwells, dug-cum-bore wells, shallow bores and filter point or cavity tube- wells. Costs of various types of wells for individual farmer investment range from Rs 600 for cavity wells in soft rock, to Rs 12,000 for dugwells in hard rock. Power units would be typically in the range of 3 to 5 hp. Typical costs of electric powered pumpsets range from Rs 3,000 to Rs 5,000, and diesel motor powered sets Rs 4,000 to Rs 6,000. Well construction for individual farmer investments presents few problems. There are many private contractors with skills, local experience and equipment; in addition, some State government departments or public sector corporations offer custom service to farmers for well construction. 5.08 While supplies of both el,ectrical and diesel pumpsets are plentiful and availability of spare parts is generally adequate, recent ARDC studies carried out at the request of IDA sihow that sets purchased by farmers are sometimes substandard in technical specification and, in some cases, in qual- ity. Pumps and power units are often mismatched, pumps are poorly selected for the required duty, and pipework and ancillary fittings are badly designed. Improvement of technical standards of both wells and pumping units would be achieved in each State through the establishment, by existing committees or committees to be set up in each State, of approved lists of pumps and prime movers in accordance with guidelines set out in Schedule B, and through the establishment of technical support organizations operating through the bank- ing systems. ARDC would set up pilot projects to be completed by December 31, 1980 in selected States (Andhra Pradesh, Rajasthan, Tamil Nadu and Uttar  - 22 - Pradesh) for improvement of investment quality control, and from the results would draw up guidelines for future adoption on a country wide basis. By June, 1981 ARDC would provide IDA with a program showing the phasing by which quality control would be introduced on an All-India basis by December 1981. ARDC would also proceed with recruitment of suitably qualified technical officers for posting to ARDC regional offices to support the program and, where necessary, would require banks to recruit or redeploy technical officers for supervision of quality control. Assurances on these points were obtained from ARDC at negotiations. ARDC's proposed action plan for quality control is dealt with in greater detail in Annex 2 of the Supplementary Data Volume. Quality control should result in important savings in energy at State and Union level, but as the program would entail considerable inputs of manpower and organization improvements would be gradual. 5.09 To protect agricultural investments, ARDC would, from its own resources and mainly in coordination with the Rural Electrification Corpo- ration (REC), refinance loans to State Electricity Corporations for spur lines and associated equipment for private sector minor irrigation schemes which are to be provided with electrical power. The estimated average cost of the step down transformer, 11 kv line and low voltage cable is Rs 4,500 per pumpset connection. An assurance was obtained from ARDC at negotiations that ARDC would, prior to refinancing a loan made by a participating bank to a State Electricity Board for pumpset electification, require the participating bank to obtain from the State Electricity Board confirmation that the connection of the pumpsets would not lead to overloading the local power system or to any other effects detrimental to the State electrical power system. 5.10 Loans for medium duty tubewells serving areas of 10 to 30 ha would be made to groups of farmers. Wells would be equipped with either vertical shaft turbine pumps powered by electric or diesel motors, or with electric submersible pumps. Typical costs of such installations would be from Rs 20,000 to Rs 60,000. There are numerous drilling contractors equipped to construct such tubewells and, in some States, government departments or public sector organizations offer custom drilling services. Drilling contractors and public sector organizations usually provide technical advice on selection of pumping equipment and install pumps as part of their service. Technical standards of installations are normally acceptable. 5.11 Loans for deep, high capacity tubewells serving command areas of up to 100 ha would be made to cooperatives and State Corporations. Wells would be designed for discharges in the range of 150 to 250 m /h, and would be equipped with either electric vertical shaft turbine pumps or electric submer- sible pumps. The distribution system would be either lined channels or pipes. Costs of such installations range from Rs 150,000 to Rs 400,000. Past perfor- mance of irrigation tubewells operated by State Corporations or cooperatives has been generally unsatisfactory. Several causes including electrical power rationing, unscheduled outages due to overloaded electrical conveyance systems, long delays in repair of pumping units and inefficient or inequitable distri- bution systems have contributed to this situation. Water rates generally involve high levels of subsidy, and frequently are inadequate to meet even operation and maintenance costs. Collection of water rates has, in many  -- 23 - cases, been poor. For these reasons, disbursements against public irrigation tubewells would be made according to guidelines as specified in Schedule C. Loans would be made to State Corporations for deep, high capacity, augmenta- tion tubewells which discharge directly into surface water irrigation 3canal conveyances. Wells would be designed for discharges of 200 to 300 m /h and would be equipped with electric vertical shaft turbine pumps or electric submersible pumps. Piped conveyance would be provided from well head to irrigation canal. Costs of such installations would range from Rs 100,000 to Rs 140,000. The performance of augmentation wells has been considerably better than that of direct irrigation public tubewells. Disbursements against augmentation wells would be in accordance with guidelines as specified in Schedule C. An undertaking was obtained from ARDC at negotiations that ARDC would disburse against public deep tubewell and augmentation tubewells in accordance with guidelines shown in Schedule C. 5.12 Legislation to regulate groundwater development has been advocated by GOI for several years. However, such proposals have raised major social and administrative issues and remain largely unacceptable to State governments. Well spacing and density control standards, introduced on a large scale as a result of ARDC I, and maintained as guidelines under ARDC II, have been ineffective because they do not regulate privately financed well construction. The danger of over-exploitation remains and though, as yet, development has reached critical levels only in a few well-defined areas, the present pace of groundwater development implies that a measure of control is necessary, particularly as considerable development is taking place outside institutional finance channels. In the absence of legislation, the only practical way of ensuring that institutional finance does not aggravate excessive groundwater drawdowns is through more detailed investigations by SGO to provide a better basis for ARDC appraisals. The more developed a proposed project area, the greater care must be taken in establishing present groundwater balances, and in forecasting future development over the proposed investment life. GOI and ARDC have agreed to introduce an appraisal system consisting of three cate- gories of areas based on the level of development which relates net draft to recoverable recharge. Each corresponds to a level of resource evaluation and groundwater investigation appropriate to the potential risks. The guidelines defining categories and levels of studies are given in Schedule D. The inten- sified studies required by the guidelines would necessitate strengthening many SGO and providing SGO and SLDB with essential equipment. As a condition of ARDC refinance under the minor irrigation category in each State, ARDC would ensure that SGO had been strengthened to the level required to initiate imple- mentation of work programs for groundwater resource evaluations in support of the proposed development programs of each State. This is discussed in more detail in Annex 2, and a list of essential equipment is in Table 3. An under- taking was obtained from ARDC at negotiations that ARDC would apply Guidelines for Levels of Evaluation of Groundwater Resources Appropriate to Levels of Resource Development, as shown in Schedule D. Although it would no longer be required in all cases, SGO and ARDC would continue at their discretion to apply spacing and density criteria adapted to suit local circumstances.  - 24 - 5.13 Investments in groundwater development involve an element of risk in that a well may fail to yield an adequate quantity of water at completion and, moreover, may fail to maintain an adequate yield through the life of the investment. The former risk is easy to ascertain; the latter is not. Indivi- dual farmers (particularly small farmers) require some protection against the possibility of failure. The probability of failure at completion is low when figures are considered at State level, ranging from less than 1% to 3% and averaging about 1%. Failure of a well through time may be identified but the reason for failure would not be easy to define. Several possible interacting factors could contribute to complete or partial failure of a well, including the natural phenomenon of a long drought period. Given this situation, it is not easy to define a rational basis for insurance of farmers against well failure through time. On the other hand, several States presently run success- ful compensation schemes for failure at completion. In light of this ARDC would, by June 30, 1980, review existing failed well schemes presently operated by some States for well failure, and would on the basis of this review encourage all States to adopt a rational scheme. An undertaking to this effect was obtained from ARDC at negotiations. 5.14 Minor irrigation development of surface water resources would con- sist mainly of river lift schemes ranging in size from single farm commands to several hundred hectares. Loans would be made to individual farmers, cooperatives and State corporations. Loans to individual farmers would be for small pumpsets and pipe work to raise water from the river on to adjoin- ing land. The pumps would be of the centrifugal type and would be powered by either electric or diesel motors. Power unit capacities would cost from Rs 3,000 to Rs 5,000. Loans would be made to cooperatives and State corpo- rations (which would charge a water rate to farmers), for financing river lifts which would involve command areas extending up to several hundred ha. Installations would be energized by electrical power taken directly from 33 kv conveyance. Estimated costs of river lift schemes range from Rs 1,000 to Rs 6,000 per ha. 5.15 Other relatively small minor irrigation investments for which loans would be made include water course lining for surface water irrigation systems, at an estimated cost of about Rs 20-25 per m , and provision of buried pipe conveyance for existing public tubewells at about Rs 2,000 per ha. A small provision is made for tank development at about Rs 10,000 per unit. 5.16 Land Development. Command area development (CAD) works in surface irrigation projects financed through institutional credit include mainly investments below the chak 1/ outlet such as watercourses, drainage and land shaping. There has been an increasing gap between the creation of the capa- bility to convey water to the minor and chak outlet (dams, canals) and the implementation of CAD works required for efficient water utilization. As a result, India has been foregoing a large amount of incremental production. 1/ The term 'chak' refers here to a block of holdings averaging about 40 ha, which receive irrigation water from the same outlet on a minor or subminor irrigation canal.  - 25 - 5.17 In order to accelerate implementation of CAD works, it is essential that water supply to each field in the areas developed be reliable. This would facilitate farmers' participation and encourage them to take loans for land development. Additionally, it iis essential to separate the finance and implementation of the communal items of CAD (watercourses, drainage) from that of the individual on-farm items (land shaping, bunding) because their construc- tion can proceed much faster. Moreover, upon completion of the communal items, water distribution to fields can start and farmers' confidence in water supply and their willingness to undertake onfarm development is therefore increased. The present procedure of financing communal items through credit and charging the cost pro rata to farmers, has caused implementation delays since the con- sent and credit eligibility of each farmer must be established before the finance is made available. Ideally, the most appropriate solution would be to finance the communal items through government sources as a part of the irrigation project concerned, and onLy the on-farm portion through individual loans. This is currently under discussion. 5.18 Under the project, loans for land development would be made to indi- vidual farmers only for on-farm deveLopment, i.e., work done on the farmers' land such as land shaping, bunding and water and drainage channels. The cost of land shaping would depend on various factors such as slope, soil type and condition, etc.; the average cost wouald be about Rs 1,500/ha. The work would be done manually or by machine in accordance with the farmer's preference. It is essential that farmers have an assured and reliable water supply in order to generate the incremental income required for loan repayment. Therefore, lending for on-farm development would be limited to areas where project agencies were making every effort to ensure that: (i) flow rates and fre- quency of water supply at the chak outlet were adequate; and (ii) a rotational water supply system was being pogressively established so that each farmer is allocated water according to a pre-determined time schedule. An assurance to that effect was obtained from ARDC at negotiations. Loans would be also made to State Corporations for construction of watercourses and for lining of watercourses and canals in order to reduce seepage losses. (b) Diversified Lending 5.19 Diversified lending would support agricultural and allied activities other than minor irrigation and land development, such as: dairy, poultry, other livestock related activities, plantation and horticulture, marine and inland fisheries, farm mechanization, storage, market yards, and forestry. The share of diversified lending in ARDC proposed program for 1979-81 would be about 33% as against 26% of cumulative disbursements to June 30, 1978. This reflects a rapidly growing demand for such loans, and the increasing import- ance attached by GOI to diversified agricultural development measures parti- cularly as a means of reaching the landless and near landless rural poor. Many of the above-mentioned activities are currently supported by ongoing IDA financed projects through ARDC (Table 2). Those projects have helped to develop, test and revise suitable technologies and lending arrangements. 5.20 For administrative simplicity, IDA disbursements would be limited to the following four categories:  - 26 - (a) Livestock. Generally, loans would provide for purchase of one or two improved milk animals such as cross-bred cows, or higher yielding buffaloes costing about Rs 2,300 per head. Animals would be in milk for the first or second time at date of purchase. An alternative which will probably be of more importance in the future, is a loan for rearing one or two cross-bred calves through artificial insemination of farmer's own stock. The loan would cover all precalving expenditures and subsequent cost, including medical care and insurance, totalling about Rs 2,700 during 40 months of rearing the female calf until lactation stage. The schemes would mainly be undertaken in conjunction with GOI "Operation Flood II" (para 2.16). That would ensure a regular supply of semen of proven quality, of veterinary cover, and close follow-up of farmers' management of animals, as well as an integration into the collection/distribution network established through Dairy Cooperative Societies. These loans would benefit mostly small farmers and landless laborers. Part of ARDC refinance would, if necessary, provide supporting facilities such as milk processing plants, manufacturing units for milk products, as well as breeding and fodder farms. The project would also support piggery development mainly in the north-eastern States, and sheep breeding in most States. The loans for piggery and sheep breeding development would be about Rs 2,700 to Rs 3,000 for typical units of 4 pigs and 20 sheep respectively. (b) Poultry. Loans, either to groups or individuals, would typically finance the establishment of small to medium commercial layer batteries and broiler units. Medium- sized units would be for about 1,000 birds and cost about Rs 60,000, whereas small units for individual small far- mers (50-100 birds) would cost about Rs 4,000 to Rs 6,000. The latter would normally be sponsored by Small Farmers Development Agencies. Loans would also be granted for new or expanded hatcheries, poultry feed plants, cold storage facilities for eggs/broiler meat, broiler proces- sing plants, marketing facilities, etc. It is intended that wherever possible new poultry units would be integ- rated into an organized cooperative processing and market- ing system, the development of which is planned on similar lines as in the dairy sector. (c) Plantation and Horticulture. Loans would be used mainly for planting, replanting or rejuvenation of existing material of a variety of commercial tree and fruit crops, such as coco- nut, cashewnut, rubber, coffee, cocoa, pepper, cardamon, mango, citrus, and pineapple. A typical loan would include the cost of land preparation, improved planting materials, and working capital for proper maintenance up to maturity of the trees. In some cases, the loan may include cost of processing equipment. Investments would normally be by small  - 27 - and medium farmers. ARDC would ensure that each scheme pro- vided for an adequate and reliable extension service (and supervision) particularly to small farmers and especially for crops like tea, coffee and sensitive fruits (citrus). ARDC would ensure that marketing arrangements were satisfactory. Investment costs vary widely from crop to crop and between regions, but it is estimated that average per ha cost (up to maturity of trees) would amount to about Rs 14,000. (d) Fishery. Loans would go mostly into the marine subsector, where they would be for purchase of mechanized fishing vessels (10 to 12 m) including accessories and mainly on the upper east coast, where there is still scope for an increase in shrimp catching. Unit costs vary with size and type of equipment and are estimated at Rs 110,000 to Rs 200,000. Loans would also be granted for traditional fishery operations, such as canoes (8 to 9 m) with onboard or outboard motors, or for mechanization of existing canoes to extend their reach of operation. Average unit costs would be about Rs 34,000 for a mechanized canoe with equipment, and Rs 6,000 for motor only. If necessary to support a scheme, ARDC would also refinance loans to establish ice plants, fish processing and freezing capacity at harbors, and supply of ice by insulated trailers to landing sites for groups of fishing villages. In inland fishery, a typical loan would be for establishing or rehabilitating fishfarm tanks or ponds, either by individual or cooperatives, and would also include the cost of fingerlings, nets and other implements. Average investment would be about Rs 2,000/ha. 5.21 From ARDC's own resources and as part of the project, ARDC would refinance schemes mainly for the following purposes: (a) Storage: Food Corporation of India (FCI) is expected to continue its arrangements for guaranteed lease of privately owned and constructed godowns. FCI leases such godowns for a minimum period of 3 to 5 years, and usually longer, for the storage of buffer stoc;ks, and for distribution of food- grains. ARDC would provide refinance for loans by CB to private parties for construction of such godowns, ranging in capacity from 2,500 to 20,000 tons and costing from Rs 0.6 M to Rs 5 M. Under similar terms and conditions ARDC would refinance storage capacity for lease by Jute Corporation of India in the eastern States, with average capacity of about 7,000 tons costing about Rs 2.3 M each. Other schemes would cover storage capacity for State Warehouse Corporation and State Marketing Federations. ARDC would ensure that storage schemes would be in line with carefully assessed demand for the area and commodities concerned, and would take into consideration FCI and National Cooperative Development Corporation (NCDC) storage programs.  - 28 - (b) Market Yards: ARDC would continue to refinance loans to State Agricultural Marketing Boards and Agricultural Produce Market Committees for construction, remodeling or extension of regulated market yards in major trading centers in order to exercise some control over the transaction, storage and processing of agricultural produce in a notified area and commodity. Cost of such market yards (which normally would include platforms for drying, cleaning, display and auction, trader's shops, storage facilities, weighing scales, access roads and parking space for vehicles, and a number of public amenities such as rest house, bank and post office), would amount to about Rs 100,000 for small primary markets, and up to Rs 30 M for large regional centers. Average cost would be about Rs 4 M for a 7-10 ha market yard. ARDC intends to strengthen its Technical Cell to include specialized and experienced staff to assess the feasibility of market devel- opment in States which request loans, to advise State agencies in planning and preparation of schemes, and to appraise pro- posed schemes, especially in view of changes in modern market yard functional designs. (c) Forestry: In view of domestic and export demand for commer- cial and industrial wood, and of a low degree of current exploitation, ARDC would support a reorientation program of forest management towards commercial use of existing valuable forest areas, and of those mixed forests which could be economically converted into regular plantations of high value stands. ARDC would refinance loans to State Forest Development Corporations, which would cover items such as building of feeder roads and extraction paths, machinery and equipment for clearfelling, extraction and logging, setting up of nurseries for planting material and provi- sion for maintenance and working capital during an initial period. Total investment cost would be about Rs 2,500- Rs 3,500 per ha. (d) Farm Mechanization: ARDC refinancing of loans for tractors, power tillers and other mechanical farm equipment would be selective. Scheme appraisals would take into considera- tion the need to predict and avoid undersirable effects on employment and income distribution. Average costs would be about Rs 70,000 for a 35 hp tractor and Rs 25,000 for a power tiller. (c) Training 5.22 A training program was started under ARDC I in 1975 when the urgent need to train staff of financing institutions, particularly LDB staff, in more modern methods of financing agricultural development was identified. That program was extended into ARDC II and by mid-1979 it is estimated that about 2,300 senior and middle-level staff, and about 11,400 junior LDB staff (both about 95% of appraisal estimates) will have been trained. That represents about 27% of estimated total LDB senior and middle-level staff and about 54% of their junior staff.  - 29 - 5.23 This project, which would continue and expand that program, would provide: (i) training for about 3,000 senior and middle-level staff, includ- ing ARDC staff, and special technical courses for staff of SGO; (ii) training for about 7,800 junior LDB staff; (iii) introductory training in long-term lending for SCB staff; and (iv) refresher courses for about 150 senior and middle-level LDB staff and about 4,000 junior LDB staff. At the same time, RBI, has begun a study of the training needs of CB as it is recognized that current training facilities are insufficient. With assistance from ARDC's Training Cell, the study is expected to be completed by the end of 1980. 5.24 Senior and middle-level staff are trained mainly at the College of Agricultural Banking, Pune, and junior LDB staff at the 19 LDB Training Centers (including five sub-centers) accessible to almost every State and Territory. Details of the programs are contained in Annex 6 of the Supple- mentary Data Volume. The quality of training is generally satisfactory, and the most serious defect in the junior program, i.e. unsatisfactory living conditions for junior trainees which has damaging effects on training, is being addressed by ARDC more vigorously than in the past. Workshops for trainers that started under ARDC II would be continued. 5.25 ARDC would continue to be responsible for organizing and monitoring the program. Its performance under ARDC I was satisfactory. ARDC II carried higher expectations, but the ARDC Training Cell was badly hampered by under- staffing until recently: it is expected to reach an adequate staffing level by the end of ARDC II. Two weaknesses in the junior program, i.e., inadequate training materials and infrequent visits to Training Centers are being recti- fied. Production and distribution of training materials are gaining momentum, and ARDC intends to visit each of the 14 Training Centers at LDB Head Offices at least once every quarter during the project. 5.26 Training costs have been estimated on the basis of experience gained under ARDC II and are summarized below: Element Cost (US$'000) Senior and Middle Level Staff 432.8 Junior Level LDB Staff 1,154.7 ARDC Training Cell 366.5 1,954.0 (d) Equipment 5.27 The project would provide essential equipment such as well loggers, and water level recording and conductivity meters, to strengthen State Ground- water Organizations, State Land Development Banks, and ARDC, to enable them to more efficiently implement programs for groundwater resources evaluations in support of minor irrigation development programs (para 5.12). A schedule of equipment to be purchased is in Table 3. The basic cost of the equipment is estimated at US$0.6 M. D. Cost Estimates 5.28 Project costs are estimated at Rs 8,643 M (US$1,005 M) of which about Rs 447 (US$52 M) are duties and taxes. The costs are summarized below:  - 30 - Foreign Local Foreign Total Local Foreign Total Exchange ----- (Rs M) - ---- (US$ M) (%) I. Minor Irrigation Minor Irrigation 3,203.0 436.0 3,639.0 372.4 50.7 423.1 12 Land Development 428.0 23.0 451.0 49.8 2.6 52.4 5 Pumpset Electrification 682.0 76.0 758.0 79.3 8.8 88.1 10 Soil Conservation/ Reclamation 81.0 4.0 85.0 9.4 0.5 9.9 5 Subtotal 4,394.0 539.0 4,933.0 510.9 62.6 573.5 11 II. Diversified Lending Plantation/Horticulture 202.0 22.0 224.0 23.5 2.5 26.0 10 Farm Mechanizatioi 659.0 220.0 879.0 76.6 25.6 102.2 25 Livestock 436.0 49.0 485.0 50.7 5.7 56.4 10 Fisheries 254.0 28.0 282.0 29.5 3.3 32.8 10 Forestry 224.0 39.0 263.0 26.1 4.5 30.6 15 Storage 308.0 64.0 372.0 35.8 7.5 43.3 20 Yards 239.0 60.0 299.0 27.8 7.0 34.8 20 Others 55.0 6.0 61.0 6.4 0.7 7.1 10 Subtotal 2,377.0 488.0 2,865.0 276.4 56.8 333.2 17 III. Training 17.0 - 17.0 2.0 - 2.0 - IV. Equipment 3.0 2.0 5.0 0.4 0.2 0.6 25 Total before Contingencies 6,791.0 1,029.0 7,820.0 789.7 119.6 909.3 13 V. Price Contingencies 714.0 109.0 823.0 83.0 12.7 95.7 13 Total Project Cost 7,505.0 1,138.0 8,643.0 872.7 132.3 1,005.0 13 Estimates incorporate a price increase contingency at an annual rate of 5% in accordance with Bank estimates of future price movements in India and ARDC experience with specific items. Cost estimates are based on ARDC's indicative lending program for the period 1979/80 to 1980/81 which is drawn from estimates of agricultural investments requiring ARDC refinance, prepared by the various States, and adjusted by the appraisal mission in light of previous disbursement and commitment performance, and known institutional constraints. Individual investment cost estimates are based on ARDC's recent experience with similar ongoing projects. This is a two-year time slice of ARDC's estimated five year program (excluding ongoing IDA projects channelled through ARDC) and, as with most agricultural credit projects, no allowance has been made for physical contingencies. ARDC's projected lending program by State and subsector for 1979/81 is shown in Table 8.  - 31 - E. Financing Financing would be as follows: Borrowers Banks ARDC/GOI IDA Total US$M % US$M % US$M % US$M % Minor Irrigation 76.3 12 84.4 13 265.8 /a 42 207.2 33 633.7 100 Diversified Lending 62.9 17 66.2 18 198.1 Ia 54 41.5 11 368.7 100 Training and Equipment - - _- - 1.3 /b 50 1.3 50 2.6 100 Total 139.2 14 15(.6 15 465.2 46 250.0 25 1,005.0 100 /a ARDC /b GOI 5.29 The proposed IDA credit of US$250 M would be made to GOI on standard terms and would finance about 25% of total project costs or 26% of costs net of duties and taxes. About 46% of project costs would be provided by ARDC and GOI, 1/ 15% by participating banks, and 14% by borrowers. A Subsidiary Agreement, satisfactory to IDA, wou:Ld be executed between GOI and ARDC, under which GOI would make: (i) US$1.0 M of the IDA proceeds available to ARDC, as a training grant and US$0.3 M to ARDC and any State Government as an equipment grant; and (ii) US$248.7 M available to ARDC, repayable partly up to 9 years and partly up to 15 years at 6.25% and 6.75% respectively (with 0.25% rebate for prompt repayments of capital andl interest depending on the repayment period of ARDC refinance to participating banks. GOI would bear the foreign exchange risk. It would be a condition of credit effectiveness that GOI and ARDC had executed a Subsidiary Agreesment satisfactory to IDA. F. Procurement 5.30 Procurement under schemes involving relatively small investments by individual farmers would be based on farmers' choice through local dealers and contractors. Typical items would be pumpsets, tubewe:Lls, dugwells, dairy cattle, bullock and pneumatic tyre carts, motors for fishing boats and farm mechanization equipment such as tractors, power tillers and sprayers. ARDC would establish guidelines for the selection of pumpsets which would include equipment performance standards established by the Indian Standards Institute, and standards relating a given investment item (e.g., size and horsepower of pumpset) to farmers' typical situations (e.g., size of area to be irri- gated) (Schedule B). Normal commercial channels provide an adequate choice of locally manufactured equipment and good servicing facilities, and prices 1/ GOI is in the process of negotiating additional external support for ARDC's lending program from bilateral donors, including USAID, CIDA and KFW. Since the amount and timing of these contributions have not yet been finalized, GOI has agreed to ensure that ARDC is provided with the local resources required to execute this program.  - 32 - are competitive with those on world markets. Procurement under such schemes as construction of storage warehouses, market yards, and agro-processing facilities would be normally based on local competitive bidding procedures satisfactory to ARDC and IDA. International competitive bidding would not be practical given the relatively small size of investments involved and their scattering throughout the country and over time. Procurement of equipment for SGO, SLDB and ARDC (US$0.6 M), which would be spread over two years, would be based on prudent shopping procedures satisfactory to IDA. G. Disbursements 5.31 IDA disbursements would be against ARDC certified statements of loans made by participating banks and refinanced by ARDC, and of expenditures on training and equipment. For ease of administration, disbursement of the credit would be against: (i) 50% of ARDC refinance for: (a) minor irriga- tion lending (minor irrigation, land development, and soil conservation) and (b) diversified agricultural lending (plantation and horticulture, livestock and fishery); and (ii) 50% of the cost of training and equipment. Documents would not be submitted to IDA for review, but would be retained by ARDC and be available for inspection by IDA during project supervision. This proce- dure has been applied to all IDA supported agricultural credit project dis- bursements because of the extremely large number of individual loans involved. ARDC bases its refinancing of participating banks upon their certification of loans made under ARDC approved schemes: each application is verified by ARDC regional offices to ensure that it conforms with agreed lending terms and conditions. Each loan is required to be physically verified by the lending bank and ARDC carries out sample checks. Both ARDC and participating banks are subject to regular audits. A schedule of estimated quarterly disburse- ments is in Table 4. VI. PROJECT IMPLEMENTATION A. The Agricultural Refinance and Development Corporation (ARDC) 6.01 All project funds would be channelled through and supervised by ARDC whose structure and operational record are described in detail in Annex 4 of the Supplementary Data Volume as well as in the appraisal report of the Second ARDC Credit Project (No. 1520-IN, May 1977). Major features are updated and summarized below. 6.02 Functions. ARDC, a subsidiary of RBI, was established by Act of Parliament on July 1, 1963. Its main objective is to provide credit for agricultural investments by making long and medium-term funds available to LDB, CB, and other approved institutions, to refinance agricultural schemes which are economically and technically sound, and are located within a rea- sonably compact area. In its developmental role, ARDC helps formulate schemes, particularly in less developed States, and consequently, is able to influence lending policies and procedures of agencies it assists. Through  - 33 - its regulated refinancing of participating banks, ARDC wouLd continue to control the flow of project funds, and further contribute to development of agricultural credit institutions throughout India. 6.03 Organization and Management. The Managing Director (who is the chief executive) is assisted by some 500 professional staff of whom about 300 are located at Head Office in Bombay. The rest are distributed between an office in Delhi, to liaise with GOI, and 13 State regional offices (Map IBRD 12630R), to maintain contaLct with State governments and financing institutions. Regional offices mainly assist State governments and lending institutions in scheme preparation and implementation in accordance with ARDC policies and procedures. 6.04 Sources of Funds. Other than loan repayments (Rs 829 M in 1977/78), the four main sources of funds available to ARDC are: (i) Share Capital and Reserves: As of June 30, 1978, issued and paid-up capital was Rs 475 M, and reserves stood at Rs 104 M. ARDC has never had any bad debts or overdues, and reserves are adequate given that all of its lending is either guaranteed by State governments, or, in the case of CB, backed by RBI minimnum reserve requirements. The principal shareholders are RBI (55%), LDB and SCB (24%) and CB (19%). (ii) Borrowings from GOI: As of June 30, 1978, borrowings from GOI, mainly representing the rupee counterpart of disburse- ments made on IDA credits, stood at Rs 4,276 M. Withdrawals during 1977/78 were Rs 876 M. (iii) Issue and Sale of Bonds Guaranteed by GOI: This source of funds helps mobilize savings from a variety of institutions such as insurance companies, and commercial banks, etc. During 1977/78, bonds for an aggregate amount of Rs 206 M were issued, raising total market borrowings to Rs 2,023 M; and (iv) Borrowings from RBI: Borrowings from RBI are of two kinds: (a) term loans (currently 10 years) under the National Agri- cultural Credit (Long Term Operations) Fund. The balance outstanding at June 30, 1978 was Rs 2,168 M and net borrow- ings during 1977/78 were Rs 442 M; and (b) short-term borrow- ings (less than 18 months); ARDC has a limit of Rs 150 M to accommodate temporary requirements but has not drawn against this facility in the last two years. 6.05 Use of Funds. Most of ARDC refinance has been for minor irrigation (69.7%). Other purposes of ARDC refinance include farm mechanization (13.8%), storage and market yards (6.1%), and land development (4.3%). The remaining 6.1% mainly comprises forestry, plantation, livestock and fisheries. Since  - 34 - its inception, ARDC has made significant contributions to increased produc- tion through financing of about 250,000 tubewells, 390,000 dugwells, and 580,000 pumpsets. Land developed on irrigation projects and land improved under soil conservation schemes, aggregate about 890,000 ha. Through its investments, ARDC has assisted in bringing nearly 3 M ha under multiple cropping. ARDC currently supports about 40% of direct institutional term lending for agriculture in India. 6.06 Operating Results and Projections. ARDC operating results continue to be satisfactory. Profits before taxes in 1977/78 were Rs 119.2 M; with a statutory 25% transferred to reserves (Rs 30.0 M) and taxes of Rs 51.7 M, this left a net profit for distribution of Rs 37.5 M. After paying cash dividends of Rs 24.8 M the balance was added to reserves. As of March, 1979 ARDC has been granted exemption from corporation tax, and ARDC's interest rate structure has been slightly modified (para 6.12) so as to pass on most of the increased margin to ultimate borrowers and still allow a more rapid build up of reserves than hitherto. From 1977/78 to 1982/83, ARDC portfolio is pro- jected to increase from Rs 8,700 M to Rs 21,500 M, paid-up share capital from Rs 475 M to Rs 786 M and reserves from Rs 104 M to Rs 603 M. Available re- serves would cover 2.8% as against the present 1.3% of outstanding portfolio: though still low, this level of reserves would be sufficient in the special circumstances of ARDC (para 6.04(i)). Even with administrative costs expected to grow significantly faster than portfolio, ARDC's projected net surplus is adequate over the project horizon. The gross margin on new lending, however, will need to be kept under constant review. Projected balance sheets, income statements, and cash flow projections to 1982/83 are in Tables 5, 6 and 7. ARDC accounts for the year ended June 30, 1978, have been audited and have received an unqualified report. 6.07 Medium-Term Reorganization Plans. ARDC's greatly increased work- load (e.g., 1,800 schemes approved in 1977/78 as against 600 in 1974/75) is beginning to strain its manpower resources to the point where the quality of appraisal and supervision is threatened. Even if specialized staff continue to be recruited almost in proportion to the growth of ARDC's refinance (doub- ling every four to five years), substantial changes in working methods and management systems will have to be introduced to alleviate bottlenecks. ARDC and RBl recognize that the present highly centralized organizational pattern should be gradually replaced by one involving delegation of most preparation, appraisal, and supervision responsibilities to regional offices with a cor- responding increase in senior line managers and technical staff: a process which is beginning this year under ARDC II. Over the same period, the Head Office in Bombay should increasingly concentrate on program planning, design and control of technical standards, and economic evaluation. This gradual regionalization process would also involve strengthening of senior management at Head Office, and considerable upgrading of appraisal and supervision systems and procedures. In a complex institution such as ARDC, major changes can, however, only be introduced after careful review and experimentation. During negotiations, IDA and ARDC discussed and reached agreement on a detailed reorganization plan and on a further review of ARDC working methods and procedures by December 31, 1980.  -- 35 - 6.08 More urgently, ARDC needs to: (a) build up its senior management so as to keep pace with its lending program; top priority must be given to the recruitment of a senior official who would gradually replace the managing director in all routine operational matters; and (b) change its staffing pattern to recruitment on its own establishment instead of on deputation from RBI. ARDC's proposals on these lines were discussed and agreed at negotia- tions. ARDC agreed to appoint a senior official to be next in command to the managing director by December 31, 1979. B. Lending Procedures 6.09 Participation in ARDC-Supported Schemes. Schemes would be prepared on the basis of State and (increasingly) district-level banking programs developed by lead banks, and coordinated by ARDC regional offices and senior State government officials appointed for this purpose. Banks would be invited to participate in schemes already prepared by government agencies, or prepare their own with ARDC assistance. 1/ LDB, CB and the other institutions would not be given specific allocations of funds under the project but would compete for projects, subject to area priorities, and eligibility criteria (para 3.11). 6.10 Appraisal. Schemes are submitted to ARDC for appraisal and prior commitment to refinance. Appraisal covers not only technical and financial feasibility, but also an assessment of the administrative and organizational capacity of supporting institutions, including extension, input supply and seasonal credit. Where necessary, time-scheduled undertakings are obtained that services and institutions will be brought up to required standards. In addition, under the proposed project, every scheme proposal would clearly identify supervision responsibilities within the participating bank for sub- sequent ARDC monitoring purposes. Technical appraisal is conducted by ARDC specialist staff supplemented where necessary by a panel of consultants. The overall quality of scheme appraisal is satisfactory. Under the proposed pro- ject, a substantial increase in ARDC technical staff is projected, particularly at regional office level to ensure satisfactory groundwater investigations, and quality control of investments (paras 5.08 and 5.12). To assist ARDC officers in monitoring consistent application of appraisal methodology, ARDC is compiling an updated operational manual to be completed by not later than June 1980. C. Terms and Conditions 6.11 Lending terms and conditions are set out in Schedule A and an under- taking was obtained from ARDC at negotiations that ARDC would apply them to all its refinancing for similar purposes irrespective of the source of ARDC's funds. 1/ Particularly in lesser developed States (para 6.17).  - 36 - 6.12 Interest. The interest rate structure negotiated under ARDC II has recently been modified as under (figures in brackets refer to rates prior to March 15, 1979): Banks to GOI to ARDC ARDC to Banks Borrowers _______________________%________________________- Minor Irrigation All Farmers 6.0 to 6.5 (6.5 to 7.0) 6.5 (7.5) 9.5 (10.5) Diversified Lending Small Farmers 6.0 to 6.5 (6.5 to 7.0) 6.5 (8.0) 9.5 (11.0) Other Farmers 6.0 to 6.5 (6.5 to 7.0) 7.5 (8.0) 10.5 (11.0) These changes do not affect the gross margin of participating banks and their net effect on ARDC's financial position has been compensated by the granting of full tax exemption (para 6.06). The revised rates remove much of the cost advantage of LDB ordinary debentures, and further reduce the attractiveness of CB lending from own resources; both of these effects will tend to increase ARDC's share in total term financing for agriculture. A comprehensive survey of interest margins, commissioned under ARDC II, is scheduled for completion by June 30, 1979. Preliminary indications are that most organizations find the present spread sufficient to cover the costs of general lending operations, but some additional support may be necessary to compensate banks for higher costs of a high proportion of small farmer loans in certain areas. An assur- ance was obtained at negotiations that GOI would, by June 30, 1980, review the results of the survey and advise IDA of GOI's proposals for implementing the survey recommendations. 6.13 Inflation in India over the past three years has averLged 2.1% per annum. Inflation over the next three years is not expected to exceed 5% annually. Therefore, the proposed rate to the ultimate borrowers (9.5% and 10.5%) would be positive in real terms and compare favorably with real inter- est rates to ultimate borrowers in other developing countries. Furthermore, the interest rates to be applied would be consistent with onlending rates for institutional credit in other sectors of the economy, and are already generally applied to all term lending for agricultural purposes. An analysis of cash flows and financial rates of return for typical investments by farmers shows that the proposed interest rates would be within farmers' repayment capacity and would allow for a gradual build-up of their equity (para 7.04). 6.14 Security. Security would be in accordance with arrangements between ARDC and participating banks. CB secure medium and long-term loans by discre- tionary first mortgages on land, chattel mortgages, guarantees, or other acceptable securities. In most States, LDB are required by law to obtain a first mortgage on land for all loans, and in most cases, they also take a lien on any equipment purchased with loans. These conditions are gradually being relaxed.  -- 37 - D. Lending to Small Farmers 6.15 A minimum of 50% (US$125 M equivalent) of the credit proceeds would be earmarked for reimbursing loans to small farmers, as presently defined by ARDC (Schedule E). However, during negotiations ARDC agreed to try to dis- burse at least 60% (US$150 M) of the credit proceeds to small farmers. The definition includes borrowers cultivrating land which produces a maximum annual net return to the farm family, before the proposed improvements, of Rs 2,000 in 1972 prices. This upper limit is updated regularly by ARDC using different cost-of-living indices for each State: the 1977/78 all-India equivalent is Rs 3,100, which amounts to about $65 per capita, equal to or below current Bank estimates of the absolute poverty income level in rural India. 1/ ARDC would continue to ensure that adequate records are maintained by participating banks to clearly identify small farmer loans. Under ARDC II, these are currently running at about 52% of total disbursements as against a target of 50% (US$100 M of the IDA Credit). Under ARDC III, in addition to reaching for a higher over- all small farmer coverage, ARDC has agreed to establish and monitor indicative State-by-State targets which would be adjusted to reflect existing landholding patterns. ARDC provides special downpayment and loan maturity advantages for small farmers (Schedule A). Under this project, ARDC would permit PLDB/SLDB branches with restricted eligibility to draw additional finance for specific supervised small farmer programs. GOI and most State governments offer small farmers (under various definitions), and other priority groups, capital grants. Most of those are arranged under the aegis of the Small Farmer Development Authority (SFDA), Command Area Development (CAD), or Drought Prone Areas (DPAP) programs, and are combined with ARDC-refinanced loans. 6.16 ARDC's policy with regard to subsidies is that it will admit bor- rowers who receive capital subsidies under an approved GOI or State program provided: (a) ARDC is satisfied that all subsidies granted in connection with subloans are channelled through the lending bank; and (b) suitable field supervision and auditing procedures are followed in identifying recipients and disbursing grants. E. Lending in Less Developed States 6.17 The northeastern region, comprising Assam, Bihar, Orissa, West Bengal, and a few smaller eastern States and Territories, is an area where agricultural development lags and rural banking services are less 1/ While such farmers are very poor, they are not always the poorest rural people. The latter are mostly landless laborers. They benefit in- directly from ARDC-supported minor irrigation operations as a result of improved on-farm employment opportunities resulting from intensification of land use. They also benefit directly from small-scale diversified lending activities which requi.re minimum additional resources.  - 38 - developed. 1/ Yet, most of those areas have relatively high potential for development in terms of moderate to heavy monsoon rains, relatively fertile soils, and vast unexploited groundwater resources. ARDC continues, therefore, to make special efforts to devise and promote new lending schemes in this region. Bihar, Orissa, and West Bengal, which together accounted for only 2.2% of ARDC refinance in 1971/72, absorbed nearly 16% of ARDC business in 1977/78. Between 1978 and 1983, the northeastern region is targeted to reach about 22% of new lending. Including other areas which ARDC also lists as less developed, 2/ ARDC will target nearly 50% of its lending to such disadvantaged areas. Development of agricultural extension and research projects, currently being implemented with IDA assistance or under consideration in those States, are expected to contribute greatly to demand for credit and its efficient delivery. F. Accounts and Audit 6.18 Accounting and auditing procedures would be as in other ongoing Bank Group financed agricultural credit projects in India. Audited accounts of participating banks are submitted through ARDC to IDA normally within four months of the end of the fiscal year. CB accounts are audited by commercial accounting firms and scrutinized by RBI; LDB accounts are audited by an Audit Section of the State Cooperative Department. Those audit procedures are satisfactory, and audit reporting under ongoing projects is up to date. ARDC auditors are rotated once every three years, and are appointed with the prior approval of the Reserve Bank of India. ARDC would continue to have its accounts audited by auditors acceptable to IDA, and those audits would spe- cifically include examination of certificates of expenditures submitted to IDA for reimbursement. G. Monitoring, Evaluation and Reporting 6.19 Strengthening of the monitoring and evaluation systems is a con- tinuous process. ARDC is following guidelines agreed under ARDC II which are also valid for this project. However, because of the projected substan- tial increase in the number of schemes, changes in supervision methodology are necessary to facilitate adequate coverage. 1/ Contrast, for example, the average population per commercial bank office in Bihar as of June 30, 1978 (40,000) with Punjab (10,000) or Kerala (11,000). 2/ Jammu and Kashmir, Himachal Pradesh, Madhya Pradesh, Uttar Pradesh and Rajasthan. The last three have several relatively underdeveloped dis- tricts but the practice of including the entire State under the "less- developed" heading tends to distort the definition.  - :39 - 6.20 Monitoring. Reports from participating banks and field supervisions by ARDC are the main sources of information on the progress of ARDC refinanced schemes. Field supervisions are essential for control of loan utilization, and for direct contacts with beneficiaries. During the project period sev- eral hundred schemes would be inspected and several thousand beneficiaries visited every year. To help ensure adequate supervision, ARDC intends to increase the number of staff at its regional offices, and to improve super- vision efficiency by: (i) introducing a system of district level supervision which would utilize special progress reports prepared by partici- pating banks; and (ii) continuing the process of standardization of data collection and presentation of supervision reports. The concept of a "problem scheme" and a "problem district" would be introduced to focus the attention of ARDC management on needed improvement of performance in critical areas of lending. 6.21 Evaluation. Evaluation provides ARDC with data on costs, benefits, and financial viability of the main investments it refinances. During this project ARDC plans to conduct 20 evaluation studies, and to evaluate two com- pleted IDA financed agricultural credit projects and ARDC II, for which ARDC will prepare Project Completion Reports. In addition, about ten evaluation studies would be subcontracted to independent research institutes, and about twenty evaluation studies would be conducted by participating banks under ARDC supervision. In total about 50 evaluation studies covering about 4,000 bene- ficiaries would be conducted during the project period. Special emphasis would be given to evaluation of investments in less developed States. ARDC would intensify its efforts to develop monitoring and evaluation methodologies, and would assist participating banks with establishment and training of monitoring and evaluation units. 6.22 Reporting. ARDC sends quarterly and annual progress reports to IDA in the format established during ARDC II. These are satisfactory and would continue under this project, as amended from time to time. 6.23 Completion Report. ARDC would prepare a project completion report not later than six months after the Closing Date. 6.24 An assurance was obtained during negotiations that ARDC would follow monitoring, evaluation and reporting procedures agreed with IDA, and complete a program of at least 25 evaluation studies as agreed with IDA.  - 40 - VII. PRODUCTION, MARKETING, PRICES AND FINANCIAL RETURNS TO PROJECT BENEFICIARIES Production 7.01 As a result of investments in minor irrigation, land development and land reclamation, farmers would be able to intensify cultivation, raise crop yields, and shift to higher value crops. In an increasing number of States these efforts would be supported by improvements in extension services and research activities, many of which would take place with IDA assistance. The project would induce an annual incremental crop production at full devel- opment estimated as follows: Estimated Incremental Output at Crop Full production (1985) Existing Output (1977/78) (mtM) (mtM) Rice 1.4 52.7 Wheat 1.4 31.3 Pulses <0.1 11.8 Oilseeds <0.1 8.9 Sugarcane 4.8 187.8 Total value of annual incremental output is estimated to be Rs 6,250 M (US$727 M) at 1978 farmgate prices. Part of the incremental output would substitute for imports, or would enhance export of raw or processed agricul- tural goods. Diversified lending would lead to production increments in milk, eggs, wool, fish and shrimps, and of such plantation/horticulture crops as coffee, tea, coconuts, cashew, pineapple, etc., most of which are destined for exports or for reducing import requirements (butteroil, milk powder, copra). The value of incremental output at full development is estimated to be about Rs 830 M (US$97 M) per annum at 1978 prices 1/. However, all estimates are subject to substantial variation depending on the actual phasing and composition of ARDC's refinance approvals, and participating banks' loan disbursements. Marketing and Market Prospects 7.02 Incremental output of foodgrains (about 3.0 M tons) would be handled through existing local channels. No bottlenecks are expected since production would be widely dispersed throughout the country. Given the low per capita availability of foodgrains and the growing population, no difficulty is expected to arise in marketing of the additional quantity. In case of excep- tionally buoyant short-term supply, government procurement policy would 1/ Including only plantation/horticulture, dairy, poultry, sheep and goat breeding, and fisheries programs.  - 41 - guarantee an outlet for most foodgrains at minimum support prices. Major cash crops (cotton, oilseeds, sugarcane, etc.) would also find reliable domestic and export markets through established channels; for some crops this would increasingly involve a direct link between farmers and processing facilities. Plantation crops (coffee, tea, coconut, cashewnut, etc.) are often connected to established and reliable markets; any excess over domestic demand is usually exportable. Coconut/copra production would substitute for imports. For horticulture crops (pineapple, citrus, mango, etc.), marketing arrangements would be closely reviewed during scheme appraisal. The incre- mental production would serve a growing domestic market, mainly in urban areas, for fruits and vegetables. The annual increase in urban milk demand is estimated at 5%. Given an expected continuing shortage of milk supplies, reduced production within urban areas, and excess processing capacity in the organized sector, prospects for milk marketing from cooperatively organized villages are excellent. 1/ Much the same would be true for poultry production (eggs and broiler meat). Incremental output, (about 150 M eggs), would be small compared to existing supply, and could easily be absorbed through exist- ing marketing channels. GOI envisages the establishment of a cooperative-based processing/marketing system for poultry similar to the one existing in the dairy sector. Market Prices 7.03 Financial prices used in this report are based on estimates of 1978 farmgate prices adjusted in line with the Bank's projections of relative price changes through 1985. Estimates were derived from recent wholesale price trends for certain crops, actual peak marketing season price quotations in 1978, and government procurement prices as lower limits. Allowances were made for processing, transport and other costs between farmgate and wholesale point. Financial Analysis 7.04 A wide range of investments would be financed under the project, and the exact composition of the lending program depends on future demand for loans for individual purposes. Since conditions will vary under different regional circumstances and between similar investments of different size, financial analysis has been limited to selected illustrative investment types, represented by typical farm models in each category. Models Nos. 1 to 10 are derived from results of recent evaluation studies and IDA reviews of previous agricultural credit projects in various States. The other two (poultry and dairy) are based on ARDC's experience with these investments, and on its own appraisal models. The underlying assumptions as to changes in cropping patterns, yields, input and requirements and other physical assumptions pertaining to the models are presented in Annex 8 of the Supplementary Data Volume. Financial rates of return (FRR), and incremental income after full development are given below: 1/ See Report on National Dairy Project No. 1964-IN.  - 42 - Ratio of Incremental Income to Investment Unit Size Cost FRR Debt Service a/ (Rs) (%) 1. Pumpset on Existing Well, Uttar Pradesh 1.6 ha 5,500 33 1.2 - 5.1 2. Pumpset on Existing Well, Tamil Nadu 0.8 ha 4,000 30 1.1 - 4.7 3. New Dugwell and Pumpset, Andhra Pradesh 1.4 ha 12,000 18 0.9 - 5.0 4. New Dugwell and Pumpset, Maharashtra 2.4 ha 14,000 29 1.3 - 7.2 5. Dugwell Deepening and Pumpset, Maharashtra 1.4 ha 10,000 19 1.0 - 5.6 6. Borewell and Pumpset, Uttar Pradesh 2.2 ha 8,000 36 1.4 - 3.7 7. Shallow Tubewell and Pumpset, Uttar Pradesh 1.8 ha 8,000 30 1.4 - 3.0 8. Land Development, Madhya Pradesh 0.9 ha 3,800 31 0.5 - 5.3 9. Citrus Garden with Irrigation, Andhra Pradesh 2.0 ha 16,200 24 1.2 - 2.8 10. Mechanized Fishing 10 m Vessel, Karnataka boat 110,000 29 1.3 - 2.0 11. Poultry, Layer Battery 1,000 (ARDC estimate) layers 50,000 40 1.9 - 2.8 12. Cross-bred Buffalo, 2 Haryana buffalo 4,400 53 2.3 - 2.7 a/ In the first and last year of principal repayment, in constant prices. Debt service was calculated on ARDC on-lending terms prevailing at the time of appraisal. The rate of interest to final borrowers has now been lowered by 0.5% to 1.5% thus further increasing the attractiveness of the investments (para 6.12).  -- 43 - VIII. BENEFITS RISK AND JUSTIFICATION 8.01 The principal benefit of the proposed project would be an increase in production of crops, dairy, poultry, and forest and fishery production. At full development, normally four to five years after the investment takes place, the annual value of the incremental production at 1978 farmgate prices is estimated at about Rs 7,100 M (US$825 M). A considerable part of the increment would replace imports, mainly in foodgrains, edible oils and/or oilseeds, copra and high quality cotton. Increased milk production would reduce import of butteroil and milk powder. Other outputs, mainly from plantation crops (tea, coffee, spices, rubber), but also fruits, sugarcane, and some oilseed would add to supply for exports. The same holds true for wool, some varieties of cotton, and for shrimp. 8.02 Project investments would create a permanent increase in rural employment through increased crop, livestock, forestry and fishery production, due to more intensive cropping, higher yields, increased stocks of animals, and capital investment in forestry and fishery. Total incremental recurrent employment from IDA-supported project investments-is estimated at about 115 14 man/days. At least 60% of the loan amount would go to small farmers, includ- ing landless laborers, and the number of direct beneficiari'L's 1/ is estimated to be about 1.1 M rural families. Loans for dairy, poultry, plantation/ horticulture, and mechanized canoes as well as a high proportion of the loans for minor irrigation would predominantly benefit the poorer agricul- tural population. Lastly, there would be considerable numbers of indirect beneficiaries such as neighboring farmers who buy water from new or improved wells, or hire pumpsets from direct beneficiaries. Their number is estimated at about 10% of individual minor irrigation investors (about 60,000). 8.03 Environmental Effects. N4o adverse effects on environmental condi- tions are expected from the project. The dairy sub-program would help to remove cattle holdings from urban to rural areas. Controlled forest exploit- ation would enhance soil conservation. Problems of soil erosion, salinity and waterlogging would also be specifically addressed in a number of schemes. 8.04 Economic Rates of Return., The illustrative models used for finan- cial analysis (para 7.04) have been adjusted for the calculation of the economic rates of return. Details of adjustments and assumptions are in Annex 9 of the Supplementary Data Vlolume. Changes in cost and benefit streams were made to test rates of return on their sensitivity to deviations from assumed values. Results are summarized in the following table: 1/ Excluding programs financed exclusively out of ARDC resources, such as forestry, farm mechanization, and market yards, and including partici- pants in group use schemes such as public tubewells and CAD scheme projects.  - 44 - Economic Rates of Return Economic Rates of Return Best Investment Operational Benefits Investment Estimate Cost + 10% Cost +10% -10% 1. Pumpset (1.4 ha) 52 48 44 33 2. Pumpset (0.8 ha) 45 42 36 28 3. Dugwell and Pumpset (1.4 ha) 30 28 24 20 4. Dugwell and Pumpset (2.11 ha) over 50 49 46 41 5. Dugwell deepening 29 27 24 19 6. Borewell and Pumpset over 50 over 50 over 50 49 7. Shallow tubewell over 50 over 50 49 38 8. Land development 46 43 42 36 9. Citrus 30 29 29 26 10. Mechanized fishing vessel 45 39 36 30 11. Poultry 49 45 32 26 12. Dairy 48 39 32 22 The overall ERR (weighted average) is about 46%. 8.05 These rates, which are consistently higher than the FRR for the same models (para 7.04), already incorporate overall adjustments for the economic cost of loan administration and family labor. The result reflects the relationship between domestic and border prices for project inputs and output: many farm investments bear taxes, imported inputs are a small frac- tion of project costs, and world commodity prices are projected to remain higher than their domestic equivalents over the project horizon. Other Benefits 8.06 The project aims to consolidate institutional development and con- tinue the long-term process of strengthening the agricultural credit structure started under ARDC I and ARDC II. In ARDC itself, a reorganization phase would begin with a build-up of senior management and technical expertise, the design and implementation of area-based supervision systems, and the intro- duction of additional technical appraisal criteria. In the States and dis- tricts, ARDC and State governments would work with participating banks to develop comprehensive banking plans to provide a framework and set priorities for scheme proposals. Participating banks would continue to be assisted directly through training, and indirectly through the enforcement of specific conditions of scheme approval relating to adequacy of technical support services and supervision systems. Project Risks 8.07 A lending operation of this magnitude and diversity involves both institutional and physical risks:  -- 45 - (a) There is a risk that eithe!r ARDC might not achieve the lending targets set forth in its two-year program (and the physical targets attached to it), or that the program would be achieved only at the expense of the quality of appraisal and supervision work. However, measures would be taken to expand ARDC staff, especially at senior level and in technical posts, and to enhance the quality of participating banks' scheme proposals and their loan supervision and monitoring efforts through better trained staff. Also the composition of ARDC's projected lending program was based on careful estimates of probable demand for different categories of loans and the overall size of the program is well within ARDC and participating banks' capacity. (b) A further risk is to be seen in participating banks' ability to enforce satisfactory repayment discipline at higher volumes of lending. The reduced participation and, if necessary, exclusion from the project: of LDB with chronic loan recovery problems, and the increased attention which ARDC would devote to the overdues problems of commercial banks applying for refinance would reduce this risk. There is, however, always the possibility of political or other circumstances beyond institutional control which could cause a serious deteriora- tion in repayments. Under such conditions, reduction of further lending in areas so affected would be considered until adequate measures to restore financial performance of banks had been taken. (c) Agricultural investments are by their nature susceptible to failures due to adverse cLimatic conditions. This risk can never be eliminated, but the build-up of groundwater-based irrigation should at the same time reduce dependency on the vagaries of weather. The related risk of overexploita- tion of water resources would be addressed through specific requirements for water ba:Lance investigations and the degree of precision would rise with higher degrees of existing or potential exploitation. A further risk lies in the possible lack of advice to and supervision of farmers on technical design of their investments and appropriate cultivation techniques, especially in the more sensitive plantation and horticulture crops. ARDC would pay special attention to arrangements for extension/supervision in each scheme, so as to minimize that risk.  - 46 - IX. AGREEMENTS REACHED AND RECOMMENDATIONS 9.01 Assurances have been obtained from GOI that: (a) GOI would review progress made by the cooperative credit system towards meeting the agricultural credit needs of farmers through one source or closely coordinated sources and would inform IDA, by December 31, 1980 of its proposals arising from the review (para 3.05); and (b) GOI would, by June 30, 1980, review the results of the ongoing study of interest margins and advise IDA of GOI proposals for implementing the survey recommendations (para 6.12). 9.02 Assurances have been obtained from ARDC that: (a) ARDC would establish and maintain criteria satisfactory to IDA for LDB to be eligible for ARDC refinance (para 3.11); (b) ARDC would maintain a Banking Committee which would advise ARDC on appraisal criteria for schemes involving commercial banks. The Banking Committee is expected to make its initial recommendations by December 31, 1979, but in the meantime ARDC would continue to apply its present procedures keeping in view inter-alia the adequacy of technical and financial supervision arrangements and paying particular attention to recovery procedures (para 3.18). (c) ARDC would set up pilot projects to be completed by December 31, 1980 in selected States (Andhra Pradesh, Rajasthan, Tamil Nadu and Uttar Pradesh) for improvement of investment quality control and, on completion of a review of the pilot projects, would start to introduce quality control in other States. In order to imple- ment the introduction of quality control on an All-India basis by December 31, 1981 ARDC would prepare, by June 30, 1981, a phased quality control program to be satisfactory to IDA (para 5.08); (d) ARDC would appoint or cause to be appointed by participating banks such number of suitably qualified technical officers as may be required to carry out the quality control program (para 5.08); (e) ARDC would, prior to refinancing a loan made by a participat- ing bank to a State Electricity Board for pumpset electrification, require the participating bank to obtain from the SEB confirmation that the connection of the pumpsets would not lead to overloading the local power system or to any other effects detrimental to the State electrical power system (para 5.09);  - 47 - (f) ARDC would refinance against public deep tubewell and augmen- tation tubewell loans in -accordance with guidelines shown in Schedule C (para 5.11); (g) ARDC would ensure that SGO have been strengthened to the level required to initiate implementation of work programs for groundwater resources evaluations as a condition of ARDC refinance under the minor irrigation category in each State (para 5.12); (h) ARDC would apply guidelines for Levels of Evaluation of Groundwater Resources Appropriate to Levels of Resource Developments, as shown in Schedule D; (para 5.12); (i) ARDC would by June 30, 1980, review existing compensation schemes for failed wells and would recommend to all States the acceptance of a rational scheme based on the review (para 5.13); (j) ARDC would refinance on-farm development loans in command area schemes where the project agencies assured ARDC that: (i) flow rates and frequency of water supply at the 'chak' outlet would be adequate and (ii) a rotational water supply system would progressively be established (para 5.18); (k) ARDC would by December 31, 1979 appoint a senior official to be next in command to the managing director (para 6.08); (1) ARDC would apply Lending Terms and Conditions as set out in Schedule A (para 6.11);; and (m) ARDC would follow monitoring, evaluation, and reporting procedures satisfactory to IDA (para 6.24). 9.03 Condition of Effectiveness It would be a condition of- effectiveness that: (a) GOI and ARDC had executed a Subsidiary Agreement satisfactory to IDA (para 5.29). 9.04 Understandings Reached at Negotiations At negotiations, IDA and AIRDC discussed and reached understanding on: (a) guidelines for the establishment of minimum eligibility criteria for LDB (para 3.11); (b) a reorganization plan for ARDC (para 6.07);  - 48 - (c) proposals for direct establishment of ARDC staff, and for recruitment of ARDC senior staff (para 6.08); and (d) a program of 25 monitoring/evaluation studies to be conducted during the project (para 6.24). 9.05 Subject to these agreements, the project is suitable for an IDA credit of US$250 M on standard terms. The Borrower would be the Government of India.  - 49 - Schedule A Page I INDIA THIRD AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION CREDIT PROJECT LENDING TERMS AND CONDITIONS 1. GOI to ARDC: (a) For ARDC refinancing up to 9 years: (i) annual interest rate of 6.25%, less 0.25% for prompt payment; (ii) repayment at the end of 9 years. (b) For ARDC refinancing for more than 9 and up to 15 years: (i) annual interest rate of 6.75%, less 0.25% for prompt payment; (ii) repayment at the end of 15 years; and (c) GOI to bear the foreign exchange risk. 2. ARDC to Participating Banks: (a) Interest rate to be 6.5% per annum for minor irrigation, 6.5% per annum for diversified lending to small farmers, and 7.5% per annum for all other diversified lending; (b) Installment repayments to coincide approximately with collections from ultimate borrowers; and (c) Refinancing to be by purchase of debentures or by loans, generally up to 90% of individual loans. 3. Participating Banks to Ultimate Borrowers: A. Minor Irrigation and Land Development: a. Interest rate to be 9.5% per annum; b. A once and for all evaluation fee of 0.5% of the cost of project investment may be charged; this fee may be waived with the concurrence of ARDC;  - 50 - Schedule A Page 2 c. Farmers' contribution (including obligatory purchase of LDB shares, own labor, and other contributions in cash or kind): (i) for lending to small farmers, a minimum 5% of investment cost; (ii) for farmers cultivating land providing a pre- development net return to family resources to such farmers and his family ranging from Rs 2,001 to Rs 3,500 based on 1972 prices, a minimum of 10% of the investment cost. (For the purpose of determining the said net returns, the criteria set forth in Schedule E shall apply). (iii) for other farmers, a minimum of 10% 1/ of the cost of pumpsets and 15% 2/ for other minor irrigation investments. (iv) for corporate borrowers, public trusts, etc. a minimum of 10%. d. Repayment periods to be based on the ultimate borrower's repayment capacity, but generally not to exceed: For lending to small farmers: (i) 7 years on loans for pumpsets, whether financed as separate loans or included in other minor irrigation loans; and (ii) 15 years on loans for all other minor irrigation loans; and For lending to other farmers: (i) 7 years on loans for pumpsets, whether financed as separate loans or included in other minor irrigation loans; and (ii) 9 years on loans for minor irrigation, other than pumpsets; and 1/ 7% for two or more farmers in a group loan. 2/ 10% for two or more farmers in a group loan.  - 51 - Schedule A Page 3 For lending to other beneficiaries (i) 15 years maximum for minor irrigation investments. e. Grace periods may be granted at the discretion of ARDC, provided that the repayment period of such loans is not exceeded; f. Technical standards, as laid down by ARDC taking into consideration IDA guidelines where applicable, to be observed. g. If his own cultivated area is significantly smaller than that which can be adequately irrigated by the irrigation unit so financed, the borrower must under- take to sell water stLrplus to his needs, or to hire out his pumpset. B. Diversified Lending: (a) Annual interest rate to be 9.5% per annum for lending to small farmers, ancl 10.5% for all other diversified lending; (b) A once and for all evaluation fee of 0.5% of the cost of investment may be charged; this fee may be waived with the concurrence of ARDC; (c) Farmers' contribution (including obligatory purchase of LDB shares, own labor and other contributions in cash or kind): (i) for lending to small farmers, a minimum of 5% of the investment cost; (ii) for farmers cultivating land providing a pre- development net return to family resources to such farmer and his family ranging from Rs 2,001 to Rs 3,500 based on 1972 prices, a minimum of 10% 1/ of the investment cost. For the purpose of determining said net returns, the criteria set forth in Schedule E shall apply; and (iii) for other farmers a minimum of 15% 2/ of the invest- ment cost. 1/ 7% for two or more farmers in a group loan. 2/ 10% for two or more farmers in a group loan.  - 52 - Schedule A Page 4 For the purpose of this paragraph, the terms "Small Farmer" and "Farmer" shall include all beneficiaries of schemes other than minor irrigation; for the pur- pose of determining the equivalent of small farmers and farmers among such beneficiaries, the definition of small farmer and the description of farmers set forth in sub-paragraph (ii) hereof shall be adapted to mean "any person primarily engaged in an activity which provides a pre-development net return to family resources to such person and his family not exceeding Rs 2,000 and ranging from Rs 2,001 to Rs 3,500" respectively (based on 1972 prices); and "net return to family resources" shall mean gross family income from the said activity less costs actually incurred. (d) Repayment periods to be based on the ultimate borrower repayment capacity, but not to exceed 15 years (including grace periods where necessary). C. General: (a) Participating banks to maintain separate accounts for lending to small farmers; (b) Security to be in accordance with arrangements between participating banks and ARDC; (c) ARDC shall refinance only sound schemes which, on the basis of careful study are considered to be financially viable, have a minimum financial rate of return of 15% to total investment costs, and are backed with satis- factory technical and administrative management to be identified at the time of scheme approval; (d) ARDC shall forward all applications for refinance of diversified lending schemes having total investment costs of US$500,000 equivalent or more together with appraisal reports and all other relevant data, to IDA for its approval; (e) Where necessary, ARDC may approve rescheduling of loans particularly when requests for rescheduling are from districts suffering from adverse climatic conditions. (f) Lending Terms and Conditions from ARDC to participating banks, and to ultimate borrowers, for similar purposes to be identical irrespective of source of funds. 4. These lending terms and conditions may be amended from time to time as agreed between IDA and ARDC.  - 53 - Schedule B INDIA THIRD AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION CREDIT PROJECT Guidelines for Selection of Pumps and Prime Movers for Inclusion in State Approved Lists 1. A committee has been or will be set up in each State, with repre- sentation from GOI, major participating banks, State governments, Indian Standards Institute, SGO, and ARDC to review pumpsets availability in the State. The committees will attempt to establish State-wide standards for inclusion of pumping equipment on an approved list, that would take account of local hydrological conditions as well as pumpset mechanical aspects. The committee should subject equipment to such technical tests as necessary to ascertain their eligibility to be included on the approved list. 2. All pumps and prime movers would be: (a) constructed of materials and designed to ISI standards; (b) provided with appropriate literature defining the equipment's type, design and capacity; (c) in the case of pumps, provided with manufacturer's rating curves for duties under an appropriate range of discharges, working heads and revolutions; (d) permanently labelled showing manufacturer's or trade name, equipment type and mark, construction serial number and basic information on capacity; and (e) provided with a suitable performance guarantee for at least one year. 3. The Committee would reserve the right to independent checking of the manufacturer's performance claims by arranging for tests on a randomly selected sample of any unit. 4. The Committees would satisfy themselves that the sales agencies for any equipment are appropriately distributed and have sufficient spares and personnel to provide satisfactory after sales service.  - 54 - Schedule C Page 1 INDIA THIRD AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION CREDIT PROJECT Guidelines for Disbursement Against Public Direct Irrigation and Augmentation Tubewells A. Public Direct Irrigation Tubewells 1. ARDC would ensure that the Tubewell Corporation obtained an assur- ance from the State Electricity Board that electric power would be available for the requisite number of hours and at the appropriate times and seasons for the economic operation of schemes. 2. Assurances would also be obtained from the State Electricity Board through State governments that the power distribution system serving the wells would not be overloaded by extended rural electrification connections, and that the system would be protected against power surges, low potential and low power factors. 3. Schemes would be designed to provide adequate stand-by water delivery capacity (e.g. by providing for stocks of spare motors). 4. The conveyance system for irrigation water would be in buried pipes or lined channels down to turnouts serving 10 to 15 farmers. 5. Irrigation water would be provided at turnouts according to a fixed and scheduled rotation. On completion of this facility, farmers within turn- out commands would be organized into 'water users groups' to agree sharing of water within the command, and each group would nominate a member to a 'tube- well command committee' which would decide questions of water rotations and allocations. 6. In principle, the water rate charged to farmers would cover at least operation and maintenance costs and capital recovery on components which are renewable at fixed intervals (pumps, motors, switch gear, etc.). Where water rates are insufficient to cover these costs, ARDC would obtain specific assurances from State Governments that for the duration of the loan, adequate funds would be provided for these purposes as well as to meet loan repayment commitments. B. Augmentation Tubewells 7. ARDC would ensure at appraisal that electrical power would be avail- able for the requisite number of hours annually for economic operation of the schemes.  - 55 - Schedule C Page 2 8. The 11 kv power supply would be exclusive to the scheme and would be properly protected against power surges, low potential and low power factors. 9. The State Irrigation Department would reimburse the Tubewell Corporation for water supplies, at a, price which would cover the costs of operation, maintenance, replacement and capital recovery, or the State would subsidize the Tubewell Corporation against any shortfall.  - 56 - Schedule D INDIA THIRD AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION CREDIT PROJECT Guidelines for Levels of Evaluation of Groundwater Resources Appropriate to Levels of Resource Development 1. If the projected net extraction in a scheme area in Year 5 is less than 60% of recoverable recharge, technical approval by the State Groundwater Organization, and appraisal by ARDC would be made on the basis of a taluk or block level water balance. 2. If the projected net extraction in a scheme area in Year 5 is between 60 and 80% of the recoverable recharge, the scheme would be subject to special scrutiny by ARDC at appraisal, and the State Groundwater Organi- zation would be required to provide a block level balance as a minimum, and an extraction projection through a planning period of 10 years. 3. If the projected net extraction at Year 5 is in excess of 80% of the recoverable recharge, the State Groundwater Organization would be required to support its evaluation of resource adequacy with detailed hydrogeological maps showing the distribution of all wells in the scheme area, extraction distribution and recharge distribution together with projected extraction for a planning period of 15 years, and an evaluation of the probable effects of drought periods on water availability and farmers repayment capacity. 4. For the purpose of applying these guidelines recoverable recharge may be taken as 70% of the total mean annual recharge and net extraction may be calculated as 70% of gross extraction that is total pumpage or draft.  57 Schedule E INDIA THIRD AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION CREDIT PROJECT Small Farmer Definition 1. "Small Farmer" shall mean any farmer cultivating land providing a pre-development net return to family resources to such farmer and his family not exceeding Rs 2,000 based on 1972 prices (1977/78 All-India equivalent Rs 3,100). 2. For the purpose of determtining said net return, the following criteria shall apply: (a) "land" shall include all land actually cultivated by the farmer notwithstanding the fact that ownership of such land may vest in one or more persons; (b) "net return to family resources" shall mean gross family income from the land, less costs actually incurred (including cash value of the farmer's own input, including seed, fertilizer, hired human labor, hired bullock labor, feed consumed by family bullocks, irrigation cha-rges, land revenue, interest on crop loan, and rent onl leased land); and (c) the amounts for the current year shall be arrived at by applying the current Agricultural Laborers Consumer Price Index for the State in which the land is located to the 1972 base set forth in paragraph 1 hereto.  58 Annex I INDIA THIRD AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION CREDIT PROJECT Related Documents and Data Available in the Project File It em 1. Preparation Report for a proposed Third Agricultural Refinance and Development Corporation Project, ARDC/GOI, June 1978. 2. ARDC Annual Report, 1977/78. 3. RBI, Report of the Review Committee on Regional Rural Banks, February 1978. 4. RBI, Report of the Expert Group on the Multi-Agency Approach in Institutional Finance, April 1978. 5. RBI, Report of the Expert Group on the Agricultural Credit Schemes of Commercial Banks, April 1978. 6. RBI, Report of the Committee on the Functioning of Public Sector Banks, April 1978. 7. Consultant Report to U.S.A.I.D. (Donald R. Redden, Sr. Consultant) on a proposed Third Agricultural Refinance and Development Corporation Project, January 1979. 8. IDA Appraisal Mission working papers, including a Review of the Operations of Land Development Banks in 5 States, February 1979.  - 59- Table 1 INDIA THIRD AGRICUTITURAL REFINANCE ANI) DEVELOPMENT CORPORATION CREDIT PROJECT State and Primary Land Development Banks Summary of Overdues A/ As at June 30 1974 1975 1976 1977 19789Y I STATE LDB (with unitary structure)