India - Agricultural Refinance Corporation Credit Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 2702 PROJECT PERFORMANCE AUDIT REPORT INDIA AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION FIRST CREDIT PROJECT (ARDC I) (Credit 540-IN) October 25, 1979 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY CONVERSIONS At Appraisal : US$1 = Rs 8.00 At Completion : US$1 Rs 8.60 ABBREVIATIONS APC - Agricultural Project Course ARDC - Agricultural Refinance and Development Corporation CB - Commercial Banks CGWD - Central Groundwater Department DCB - District Cooperative Banks GOI - Government of India LDB - The Land Development Bank System (PLDB and SLDB) PCCS - Primary Cooperative Credit Societies PCR - Project Completion Report PLDB - Primary Land Development Bank RBI - Reserve Bank of India SCB - State Cooperative Banks SFDA - Small Farmer Development Agencies SGD - State Groundwater Department SLDB - State (Cooperative) Land Development Bank FOR OFFICIAL USE ONLY Project Performance Audit Report INDIA AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION FIRST CREDIT PROJECT (ARDC I) (Credit 540-IN) TABLE OF CONTENTS Page Preface i Basic Data Sheet iii Disbursement Table iv Highlights v PROJECT PERFORMANCE AUDIT MEMORANDUM I. Summary 1 II. Main Issues 3 A. The "Line of Credit" Approach Versus the "State Credit" Approach 3 B. Institutional Aspects 5 1. ARDC: Organization and Staffing 5 2. ARDC: Financial Situation 6 3. Integration of Short and Long Term Credit Systems 7 C. LDB's Lending Operations 8 1. Volume of Lending and Trends 8 2. Sources of Financing 10 3. Overdues 10 4. Cost of Credit, Spread and LDB Profitability 13 5. Other Issues 14 D. Other Issues on Credit Operations 17 1. Financing the Replacement of Assets 17 2. The Small Farmer Definition 17 3. On-Lending Terms 18 E. Credit Disbursement Mechanism 19 Tables 20 Annex I Comments received from the Government of India 22 PROJECT COMPLETION REPORT I. Introduction 33 II. Project Implementation 36 III. The Financing Institutions 40 IV. Economic Impact 47 V. Special Issues 51 VI. IDA's Performance 55 VII. Conclusions and Recommendation 56 Annexes 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.  Project Performance Audit Report INDIA: AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION FIRST CREDIT PROJECT (ARDC I) (Credit 540-IN) PREFACE This is a performance audit of the first Indian agricultural credit project to cover the entire country. Credit 540-IN, for US$75 million, was signed in April 1975, and closed, fully disbursed, in December 1977. The project followed a group of ten agricultural credit projects for individual Indian states financed by IDA between mid 1970 and late 1973. A second ARDC credit (Credit 715-IN, for US$200 million) followed in June 1977 and a third ARDC credit (Credit 947-IN, for US$250 million), in August 1979. Five agricultural credit projects for individual states have been also closed lately. They are the subject of two other project performance audit reports: one combines two projects (Punjab and Haryana) and focusses mainly on tractorization issues; the second covers the remaining three projects (Andhra Pradesh, Tamil Nadu and Maharashtra) and focusses on several issues, especially groundwater development. The first state project (Gujarat), which was closed in March 1975, was the subject of an audit report distributed to the Board on October 4, 1976 (Report No. 1303). The audit report consists of an audit memorandum prepared by the Operations Evaluation Department, and a project completion report (PCR) dated October 1978. The PCR was prepared originally by ARDC and put in final form by the Bank's South Asia Regional Office following a country visit in December 1977. A three-member mission from the Operations Evaluation Department visited India in February/March 1979. The mission held discussions with officials of the Government of India, RBI, ARDC, the governments of six statesl/, their state land development banks (SLDBs), the main commer- cial banks involved in theiprojects (CBs) and other relevant state govern- mental agencies - particularly the state groundwater development agencies and the agencies in charge of tractor procurement. The mission also visited some rural branches of the SLDBs, some primary land development banks (PLDBs) and several participating farmers. Comments of the of- ficials interviewed are fully reflected in the report and are gratefully acknowledged. The information obtained during that mission was utilized to test the validity of the analysis and conclusions of the PCRs. Discus- sions with SLDB and PLDB officials on the peculiarities of the project implementation procedures, and on the relative advantages and disadvant- ages of the project funds as compared with alternative sources of funds previously available to them, were particularly useful. 1/ Punjab, Haryana, Maharashtra, Andhra Pradesh, Karnataka and Tamil Nadu. - ii - The audit memorandum is based on these discussions, on inter- views with Bank staff, and on a review of the PCR, the President's Report No. P-1597, dated April 3, 1975 as well as the Apraisal Report (No. 562A; March 15, 1975) the Development Credit Agreement and Project Agreement, both dated April 28, 1975, and materials available in the Bank's files. A copy of the draft report was sent to the Borrower. Comments received from the Government of India are in Annex 1 of the Memorandum. The suggested corrections and changes on the PPAM have been incorporated or, when the audit mission could not fully agree with Government comments, the difference in views has been footnoted. The PCR, which was found to be detailed and accurate, thoroughly covers most aspects of project implementation, analyzes the project's main achievements and shortcomings, and raises some relevant issues. The memorandum analyzes some issues which are not fully developed in the PCR: particularly, some institutional and financial aspects of ARDC and LDB operations, and some relevant aspects of this kind of national level re-finance operation. The analysis of institutional and financial aspects covers also performance and developments under the previous state-wide projects. The valuable assistance provided by the Government of India, the state governments, the different institutions concerned with the projects, particularly ARDC, and the farmers visited, is gratefully acknowledged. - iii - PROJECT PERFORMANCE AUDIT BASIC DATA SHEET INDIA AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION FIRST CREDIT PROJECT (ARDC I) (Credit 540-IN) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Estimate Total Project Cost (US$ million) 168.5 179.0 Additional Investments (%) 6% Credit Amount 75.0 75.0 Disbursed 75.0 Cancelled Up to 4/30/79 - Repaid - Outstanding 75.0 Date of Project Completion 12/31/77 6/30/77 Proportion Completed by Above Date (%) 100 100 Economic Rate of Return (%) 14 - 49% 18 - over 50% OTHER PROJECT DATA Original Actual or Item Plan Current Estimate First Mention in Files 11/72 Government's Application 11/73 Negotiations 03/75 Board Approval 04/15/75 Credit Agreement Date 04/28/75 Credit Effectiveness Date 08/05/75 Closing Date 12/31/77 12/31/77 Borrower GOI Executing Agency ARDC Fiscal Year of Borrower 04/01 - 03/31 Cooperative Year (of Executing Agency) 07/01 - 06/30 Follow-on Project Name ARDC II Credit Number 715 - IN Amount (US$ million) 200.0 Credit Agreement Date 06/01/77 MISSION DATA Month, No. of No. of Man- Date of Item Year Weeks Persons weeks Report Appraisal FEB/MAR 74 4 7 28 03/15/75 TOTAL 28 Supervision I 12/75 1 4 4 02/04/76 Supervision II 05/77 3 1 3 06/10/77 Completion 12/77 2 3 6 10/78 TOTAL 13  - iv - INDIA: AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION FIRST CREDIT PROJECT (ARDC I) Disbursement Table (US$ million) Actual as a IDA Fiscal Year Appraisal Percentage and Quarter End Estimate Actual of Estimate 1976 December 31, 1975 4.0 3.1 78 March 31, 1976 9.0 8.4 93 June 30, 1976 15.0 18.0 120 1977 September 30, 1976 22.5 26.8 119 December 31, 1976 31.5 31.2 99 March 31, 1977 41.5 37.4 90 June 30, 1977 52.5 54.8 104 1978 September 30, 1977 63.5 75.0 118 December 31, 197711 75.0 1/ Original and actual closing date.  Project Performance Audit Report INDIA: AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION FIRST CREDIT PROJECT (ARDC I) (Credit 540-IN) HIGHLIGHTS The first all-India line of credit project financed on-lending activities by state and land development banks (LDBs) and commercial banks (CBs). Funds were channeled through the Agricultural Refinance and Deve- lopment Corporation (ARDC). It also made provisions for training and studies. Sub-loans were primarily for supporting on-farm investments in minor irrigation, with a smaller share assigned to a wide range of diver- sified investments. The refinancing and on-lending procedures worked as anticipated. Credit funds were fully disbursed three months ahead of schedule. Major project components were in line with appraisal expectations. About 210,000 farmers (more than double the 100,000 anticipated at appraisal) benefitted from the project. About 54% of the lending was to small farmers (appraisal estimate: 50%), defined as cultivators whose pre- development income was Rs2,000 or less at 1972 prices. The project's objectives were surpassed. Production increased about 28% above appraisal estimates. About 40% of the increase was due to the incremental irrigated area being 11% larger than envisaged; the rest of it was due to higher yields, higher cropping intensities and the expansion of higher-value crops. Farm employment expanded significantly. Investments implemented under the projects were financially and eco- nomically viable. The re-estimated rates of return are similar to or above appraisal estimates, although the two sets of estimates are not strictly comparable. Under the project, ARDC continued to expand and improve its refinancing operations, thus leading to the second and third follow-on projects. Further ARDC strengthening is still required, however. The LDBs were strengthened through an intensive training program, but overdues continue at a high level. CB participation showed an increasing share of ARDC refinancing, exceeding appraisal expectations. The state groundwater agencies were. generally strengthened. All covenants were complied with. The following points may be of special interest: - The shift from the previous state-wide credit projects to the all-India line of credit approach was smooth, timely and appropri- ate; individual states have not suffered therefrom (PPAM, paras. 12 to 16; PCR, paras. 5.01 to 5.03 and 7.02). - A gradual integration of the long and short-term cooperative credit systems is desirable, bmt rather than looking for a full- scale integration, measures should be envisaged to coordinate their activities (PPAM, paras. 6 and 23 to 27; PCR, para. 2.14).  - vi - - LDB lending operations expanded considerably during implementation of ARDC I, but even with such recuperation LDB lending in real terms is still below the level it had attained seven to ten years ago, before IDA lending foi agricultural credit started; recovery performance has declined, thus restricting LDBs' ability to borrow (PPAM, paras. 28 to 32 and 35 to 40; PCR, paras. 3.11 to 3.13). - ARDC-provided funds have substituted for other sources of funds previously utilized by LDBs (PPAM, paras. 33 and 34). - LDB profitability on current operations is declining, partly because the 3% spread seems to be inadequate (PPAM, paras. 42 to 46; PCR para. 3.14). - The 50% target of lending to small farmers - though somewhat arbitrary - was exceeded; problems in defining what a small farmer is and in extending credit to them (PCR, paras 2.09 to 2.11, and 5.04 to 5.09; PPAM, para 55). - Satisfactory project monitoring and evaluation by ARDC (PCR, paras. 3.05 and 3.06). - The effectiveness of the debenture eligibility formula has still to be proven and its development implications need a careful study (PCR, paras 1.11; 3.12; 5.10 to 5.14; PPAM, paras. 47 to 49).  Project Performance Audit Memorandum INDIA: AGRICULTURAL REFINANCE AND DEVELOPMENT CORPORATION FIRST CREDIT PROJECT (ARDC I) (Credit 540-IN) I. PROJECT SUMMARYI' 1. The project was the first agricultural credit project to cover the entire country; as such it supplemented some on-going state-oriented credit projects financed by the Bank Group by providing credit in areas not covered by them, and, as funds provided under those projects were exhausted, took over the financing of unfinished schemes. Besides invest- ments in minor irrigation and land development (like the state-oriented credit projects), the project also provided finance for a wide range of diversified investments in dairying, horticulture and fisheries, and for an intensive training program for staff of participating institutions. 2. Project cost and sources of finance were as follows (appraisal estimates are in brackets): (a) Total cost Rs 1,539 M (Rs 1,350 M) of which Rs 1,388 M (Rs 1,234 M) was for minor irrigation, Rs 146 M (Rs 100 M) for diversified investments and Rs 5 M (Rs 16 M) for training and studies. (b) The IDA credit amounted to US$75 M (as planned) and financed about 44% (44%) of the project cost. The remaining part was financed by ARDC 36% (37%), participating banks and state governments 10% (9%), and farmers 10% (10%). The credit became effective in August 1975 and was fully disbursed by September 1977, three months ahead of the appraisal estimate. 3. The project's primary objectives were similar to those of the state oriented credit projects, namely: (a) increase production and employment through on-farm investments, with particular emphasis on small farmers and less developed states; (b) improve institutional credit chan- nels; and (c) strengthen supporting services, especially groundwater development authorities. The objectives were broadly achieved. 4. About 185,000 minor irrigation units were financed. Of these about 47% were pumpsets only, 35% were new and improved dugwells, 17% were shallow tubewells (including filter points) and 1% were deep and medium tubewells and lift irrigation units. Other investments financed under the project included 3,400 hal of land development, 9,200 ha of plantation and horticulture, 16,700 dairy animals, 24,500 sheep, 167,400 poultry, 315 mechanized fishing vessels and 70 agro-service centers. 5. ARDC performance under the project was good and it coped efficiently with the additional responsibility allotted to it. The standard of scheme appraisal and supervision was satisfactory. Studies relating to the in- tegration of cooperative credit systems and training requirements were carried out efficiently. Training targets were largely achieved and, under the project, the land development banks (LDB) were strengthened through an extensive training program which covered about 600 medium and senior staff 1/ Adapted from the PCR. - 2 - and about 6,000 junior staff members. All covenants in the credit agr&ement were complied with. A systematic effort to evaluate the impact of project investments was initiated under the project. Progress, however, has been modest, mainly because ARDC could provide the staff required for this purpose only gradually. In the meantime, ARDC has also suggested to participating banks to set up evaluation cells within their own organizations (see Annex I, page 1). A high level of overdues continues to be the major problem of the LDB. During the first project year (1975/76), there was a general improve- ment in loan recoveries and the percentage of primary banks eligible for unrestricted lending increased. But there was a deterioration in recovery performance during the second project year-!. Under ARDC III, rehabili- tation programs have been prepared for the Maharashtra, Tamil Nadu, Karna- taka, Gujarat and Bihar LDBs. 6. A study conducted under the project by RBI recommended a gradual integration of the long and short-term cooperative credit systems. This would be a first step towards the creation of a cooperative credit system enabling farmers to obtain all their credit requirements (long and short- term) from one source. Besides the obvious advantages to farmers, inte- gration should help to reduce the proportion of non-viable primary banks (presently nearly 1/3 of the total) and improve recovery performance. Under ARDC III, GOI undertook to review the progress made by the coopera- tive credit system towards meeting the agricultural credit needs of the farmers through one source or closely coordinated sources by December 31, 1980.2/ 7. Commercial bank (CB) participation exceeded appraisal projections. They accounted for 59% of the lending for diversified investment, 21% of the lending for minor irrigation and 25% of the total program. Moreover, during 1976/77 (project year 2) loans for agriculture made by CB accounted for 42% of total refinancing by ARDC, and for 51% in 1977/78. 8. The state groundwater departments (SGD) conducted investigations on the availability of groundwater resources, which are required before minor irrigation schemes can be approved. The SGD in each state were generally strengthened during the project. 9. Investments financed under the project were financially and econo- mically viable. Rate of return estimates were largely based on results from field surveys of similar investments which were conducted under the state- oriented credit projects. Financial rate of return (FRR) estimates range from 11% to over 50% for minor irrigation investments, 29% to over 50% for land development and from 33% to over 50% for diversified investments. Economic rates of return (ERR) ranged from 18% to over 50% for minor irriga- tion investments, were over 50% for land development and 39% for diversified 1/ GOI comments that the causes for good or bad recovery performance (as reported in the PCR, para. 3.13) appear to be conjunctural, and that the effectiveness of the linking formula to control overdues is not estab- lished (see Annex I, page 2, and PCR paras. 5.12 and 7.04). 2/ Also see GOI commments in Annex I and further discussion of the issue in paras. 23-27 of this report. - 3 - investments. These estimates are not strictly comparable with appraisal estimates (minor irrigation: FRR 18-43%, ERR 26-49%; diversified invest- ments; FRR 15-33%, ERR 14-25%) since the models are based on different geographical areas and, in the case of diversified investments, the type of investment is different. 10. About 210,000 farmers (appraisal estimate 100,000) benefitted from the project. Incremental irrigated area is estimated at 200,000 ha (appraisal estimate 180,000 ha). The value of incremental production at full development is estimated at about Rs 1,080 M at 1976/77 farmgate prices (appraisal estimate at same prices amounts to about Rs 840 M). About 54% of the lending was to small farmers as compared with appraisal target of 50%. Incremental annual employment generated by project invest- ments is estimated at 42 M man-days (at full development stage). 11. Experience gained under the project has been used to improve the design of the follow-up general line of credit projects (ARDC II and III). II. MAIN ISSUES A. The "Line of Credit" Approach Versus the "State Credit" Approach 12. Up to 1975, all IDA agricultural credit lending to India was through individual state projects. The main issue discussed during the preparation and appraisal of the ARDC credit project was whether the time was opportune to embark instead on a line of credit operation, using ARDC as financial channel, in view of the critically poor and declining per- formance of the credit cooperatives. IDA and GOI eventually agreed that an all-India line of credit was appropriate, that the identified needs for major reforms of the agricultural credit sector could be better met by such a credit and that a consolidation of sector policies, control and financial criteria would result in overall improvement. 13. The new system is based on the close relationship which develop- ed between IDA and ARDC during the execution of the state-oriented credit projects and other specific projects, and on IDA's growing confidence in ARDC's capacity and ability. IDA has handed over to ARDC its former appraisal function and is satisfied with the way ARDC handles the approval of overall lending programs and their subsequent supervision. This approach is similar to the way the Bank and IDA handle DFC operations, and they may find it advantageous to compare their systems of appraisal and supervision for agricultural credit projects with those for DFCs.1. 14. The PCR discusses the main advantages and disadvantages of the line of credit approach and concludes that the former appear to far outweigh the latter (PCR, paras. 5.01 to 5.03), a conclusion with which the audit concurs. Three additional aspects of this approach are here discussed: (i) whether the individual Indian states are now in a better or worse position, (ii) whether specific local issues can be still taken into consideration, and (iii) whether IDA's control of factors affecting project implementation has been enhanced or reduced. 1/ In this respect, the management of a commercial bank suggested that they be allowed a greater measure of independence in the appraisal nf qchPmPA_ within certain limits. as in the case of industrial 15. When analyzing state specific problems it must be distinguished between (a) states which previously had received a credit, and (b) those which had been without credit. (a) States with Previous Credit. In all the states visited, govern- ment and LDB officials unanimously stated that changing to the line of credit projects had no negative consequences for them, and that their states have not been discriminated against while distributing project funds. Under the state oriented projects such discrimination could not exist on the obvious grounds that funds could not be shifted from one state to another. Such a risk does exist on the line of credit projects, however, particularly in regard to cooperative lending. Actually, those states with better organized and more agressive LDBs, and with better credit discipline, have been able to obtain an increasingly larger share of ARDC refinancing resources. A case in point is Andhra Pradesh: with the second best recovery rate in India, it has advanced from eighth place among the states in 1975/76, to fourth in 1976/77, and to second in 1977/ 78 (ranking according to refinance drawn from ARDC). Similarly, Tamil Nadu, where loan recoveries and credit discipline have seriously deteriorated in the last years, has slipped from ninth place in 1975/76 and 1976/77, to twelfth in 1977/781/. Such developments are in line with IDA's and ARDC's institution building objectives: to offer incen- tives for good results and to penalize bad performance. But agricultural development in the states where LDBs do not perform well will be penalized too (unless CBs were able to fully cover the resulting gap). Therefore, there is a clear trade-off between institution building and physical development objectives here. Some states may discover that they did lose financial resources in the change towards the all-India projects.!/ (b) States Without Previous Credit. These states have benefitted from the change; some of them are small and most certainly would have not been considered for an individual state project. 1/ Tamil Nadu also reflects some of the convenient flexibility provided by the line of credit. As its groundwater potential nears full exploi- tation, less additional funds are required each year. The same hap- pens with Haryana, which, although showing the best recovery rate in the country, dropped from fifth place to sixth and then to tenth in the same years mentioned above. For further details on these states see the PPARs on Andhra Pradesh, Tamil Nadu and Maharashtra Agri- cultural Credit Projects, and on Punjab and Haryana Agricultural Credit Projects, Reports No. 2688 and 2684, respectively, of October 5, 1979. 2/ The change per se cannot assume the full "blame" for this development; most certainly some of the states which are bound to receive a smaller amount of funds now would have not had a follow-on state credit pro- ject at all. Further, if some states were not receiving a proper share of the total resources made available to India under the lines of credit, IDA could negotiate with G01 the establishment of a thresh- old or a ceiling for lending to some or all states under follow-on projects. -5- 16. The possibility of and capabilities for taking specific local issues into consideration were not changed by the switch-over from the statewide projects to the ARDC lines of credit. They had already de- clined under the former; an effort to standardize terms and conditions of project lending had been already initiated under the state projects and such effort would have most certainly continued had follow-on state pro- jects been carried out. Similarly, IDA's control of factors affecting project implementation has not been affected by the supersession. The only change is a procedural one: previously IDA had to negotiate with the State Governments and obtain ARDC's (and sometimes RBI's or GOI's) agree- ment; now it negotiates with ARDC (and sometimes with RBI or GOI) which is responsible for obtaining the State Governments' agreement. B. Institutional Aspects 17. The PCR briefly reviews some aspects of institution building since these have been dealt with in detail as part of the appraisal of ARDC II (Credit 715-IN) and ARDC III (Credit 947-IN). The audit, however, would like to draw attention to some particularly interesting points. 1. ARDC: Organization and Staffing 18. The expansion of operations has required ARDC to delegate some powers to regional offices, a satisfactory development which has been taking place progressively and cautiously since early-1979. It implies that regional managers and assistants are suitably experienced and quali- fied to communicate, discuss and defend ARDC's views. On the whole, ARDC has been successful in upgrading its staff, but it still lacks high cali- ber staff in sufficient numbers to represent it in all the states, to guide LDBs and to maintain quality in appraisals and supervision. ARDC III includes remedial measures which will allow ARDC to continue develop- ing its activities and to enlarge the powers of its regional offices. 19. ARDC has some management information and data processing pro- blems. It is difficult for ARDC to obtain accurate and consistent data from the participating banks: there is a large number of banks and branches; information is not collected in a consistent way from all of themiL; sub-loan disbursements are effected in multiple installments, etc. Thus, although ARDC does undertake assessments of physical achievements under individual loans and projects, backed by verification during follow up and evaluation (GO; Annex I, page 40), it is not always in a position to readily provide relevant information requested from it. The audit mission was given some information which was at times incomplete, could not be reconciled with information previously furnished to IDA, or was clearly incorrect. For example, data are not easily available on numbers of loans advanced and loans outstanding for most LDBs, which are the main determinants of lending costs and are essential to assess the credit projects' coverage. Therefore, IDA included an informationmanagement expert in the mission appraising ARDC III, and got assurances that ARDC will continue its efforts to improve the management information and data collecting and processing systems. 1/ It must be also said that ARDC is not the only agency which asks them for information. The participating banks must submit information to several central and state government agencies, which request different data presented in different forms. - 6 - 20. ARDC and LDBs should move towards a career staff structure to improve the current situation where staff is seconded from RBI for ARDC, and from the state cooperative departments for LDBs. Both are controversial issues on which long discussions have been going on in India and between GO, ARDC, RBI and the Bank. ARDC's proposals on building up senior management and on recruiting on its own establishment were agreed during negotiations of ARDC III. 2. ARDC: Financial Situationl/ 21. Resource mobilization by ARDC has been increasing by substantial amounts. Total resources increased sharply after disbursements under IDA supported projects started in 1971/72. The figure for total resource mobilization for 1972/73 is 2.4 times larger than that for 1970/71, while that for local resources is 1.4 times larger over the same period. From then on, both total resource and local resource mobilization have been increasing at about the same rate: 2.6 and 2.7 times from 1972/73 to 1977/78, respectively. IDA's financial support to ARDC increased from nil in 1970/71 to 31.4% of ARDC total resources in 1975/76; 39.3% in 1976/77, and nearly 41% in 1977/78. (see Table 1)1/. 22. ARDC's debt/equity ratio, which deteriorated from 11.2 to about 18 during the implementation of the state projects (1971/72 to 1975/76) (though remaining within the statuatory limit of 20 times the paid-up capital and reserve fund), improved during the implementation of ARDC I to 16.9 in 1976/77 and to 14.9 in 1977/78 (see PCR, Annex 2, Table 2; the figure for 1977/78 is taken from the ARDC III appraisal report.).V 1/ The PCR makes only a passing reference to ARDC's financial situation. It mentions IDA's share in ARDC's finance, the increase in ARDC's share capital and reserves, its satisfactory operating results and the reasonable amount of its administration costs relative to disburse- ments (para. 3.09). However, the subject is dealt with in the ARDC II and ARDC III appraisal reports. 2/ A number of figures furnished by ARDC conflict with those of ARDC cash flow as in the PCR (Annex 2, Table 3). The year 1970/71 was selected as base year because it is the last without any IDA funds. 3/ The words "deteriorated" and "improved" have only an analytical con- notation here. No real deterioration or improvement exists, because ARDC's perspective plan provides for issue of share capital for RBI subscription in appropriate amounts whenever this would be required to ensure that the debt/equity ratio remains within statutory limits. The exemption from income tax for a period of 5 years commencing from 1977/78 granted by GOI to ARDC will enable ARDC to allocate larger funds to reserves. - 7- 3. Integration of Short and Long Term Credit Systems 23. The problem of meeting the farmers' overall credit requirement has been a matter of concern for IDA ever since the state oriented agri- cultural credit projects were identified. Part of the problem arose from the existence of a dual system of cooperative agricultural credit outlets, with the cooperative banking system (SCB) providing short- and medium- term credit (together with various banking facilities), and the land development banks (LDB) providing long-term credit. 24. Disadvantages of this dual system have been recognized for a long time. There are indeed inconveniences for borrowers in applying to different institutions, and having to produce twice the same information to satisfy administrative requirements. But as the long-term coopera- tive credit system objectives became development-oriented, instead of being aimed primarily at refinancing old debts and financing land purchase through mortgage loans, those disadvantages have become more apparent. There is no certainty that borrowers who have obtained long-term credit will be able to secure also the short-term loans they need for seasonal inputs. This may actually jeopardize a significant part of the benefits that would otherwise accrue from their investments.1 25. The problems of the two systems have been complicated by their loose relationship. For instance, no, or sporadic, cooperation exists on such subjects as exchange of information on borrowers and borrower dues, and no attempt has been made to establish common services to save manpower and costs. 26. A study of the advisability and feasibility of integrating the two parallel systems was agreed upon at negotiations of ARDC I (PCR, para. 3.20). A committee (the Hazari Committee) was set up to carry it out. The committee's report, published in 1978, recommended a gradual merger of the two systems at all levels but neither GOI nor RBI have endorsed the committee's approach and conclusions.! 27. Rather than looking for a full-scale integration, which seems to be strongly opposed and not feasible for the time being,!/ measures should be immediately envisaged to coordinate the activities of the two 1/ The PCR on the Punjab and Haryana Agricultural Credit Projects, which was prepared by ARDC (see PPAR on those projects, Report No. 2684, of October 5, 1979, para. 6.06,c), states that, although there is no evidence that farmers obtaining long-term credits from SLDBs were denied the required short-term credit, project implementation would be more efficient if.farmers obtained all their credit requirements from a single source. Also see 001 comment in Annex I, pages 3 and 6. 2/ Under ARDC III, GOI and RBI would review progress towards integration of the cooperative structure in all States and, by December 31, 1980, would make proposals as to the timing of future steps. 3/ For current status and prospects of integration in three states, see PPAR on Andhra Pradesh, Tamil Nadu and Maharashtra Agricultural Credit Projects, op. cit., para. 48. -8- systems and to arrange for cooperation between them. There are no serious obstacles between the two systems preventing arrangements either for ex- change of credit information or for sharing common services at various levels. Also LDBs could provide finance for working capital as an inte- gral part of any investment loan. It was agreed at negotiations for ARDC III that GO would review the progress made by the cooperative credit system towards meeting farmers' credit needs through one source or closely coordinated sources. C. LDB's Lending Operations 1. Volume of Lending and Trends 28. LDB lending operations can be viewed in three more or less distinct time periods. In the three years preceding IDA involvement - 1968/71, the lending volumes (commitments) in both real and nominal terms were higher than in the next period of four years - 1971/75 - when state- wide projects were financed by IDA. The first ARDC operation was approved in April, 1975. This was followed by an improvement (1975/78) in lending volume over the period immediately preceding. Whether this improvement represents a reversal of the longer term downward trend is too early to say. LDB lending operations have expanded by 8.5% between 1974/75 and 1975/76 and by 21.4% in the following year in nominal terms, and by 1.4% and 18.6% in real terms, respectively. The current levels are however still substantially below the pre-1971 level (see Table 2)_1/. And, in terms of the overall expansion of agricultural credit in India, LDB's share in overall direct finance for agriculture between 1974 and 1978 - the period of fastest growth in LDB lending - declined from 38% to 28%2/. 29. For the six year implementation period of the statewide and ARDC I projects, 1971/72 to 1977/78, the increase in absolute value of loans outstanding was Rs 5,900 million (see Table 2). This increase is quite high, taking into account the total amount of loans advanced during the 1/ This analysis is based on figures provided in the ARDC/GOI Preparation Report for a proposed Third ARDC Project, June 1978, and other informa- tion provided by ARDC. The comments sent by GOI on a draft of this report (Annex I, page 6) contain figures which are different from those in the ARDC III preparation report. Since the former were officially provided by GOI to IDA to support the request for IDA funds, and were reproduced in the SAR the Bank staff submitted to IDA's Board of Execu- tive Directors to support the President's recommendation for Credit 947 (SAR, Supplementary Data Volume, Annex 5, Table 1; Report No. 2404a-IN), they have been retained here as the formal basis for analyzing LDB lending operations. However, even the figures provided by GOI confirm the decline of LDB lending in real terms (see Table 2, footnote 3/). 2/ The figures for total institutional credit for term credit, a compara- tor suggested by GOI (Annex I, page 6), are not readily available. Even if they were available, the comparison would become almost tautological, for LDB provides most agricultural term credit in India. -9- period (Rs 13,583 million) and their average duration (about 8 years). This reflects the declining overall performance of LDBs in terms of loan recoveries, lending and overdues - even taking into consideration that the longer maturity terms of loans extended from 1973 on somewhat contributes to such increase. But even before the IDA credits were started, there was a sharp decline each year in the ratio of loans advanced to loans outstand- ing (Table 2, ratio A/B, which is indicative of the turnover of the loan portfolio) from 0.37 to 0.21; the decline continued until 1973/74 when the ratio was as low as 0.16. Thereafter it fluctuated around 0.20. 30. Based on information provided by ARDC, published by RBI, or available in the Bank files, the audit mission had concluded that the number of loans granted annually by LDB, and, therefore, the number of beneficiaries, was declining as well. However, figures now provided by GOI (see Annex I, page 7) do not show any such decline over the period analyzed. 31. Although the trend has been reversed in recent years, LDB lending in real terms is still far below the pre-1971 levels (see para. 28 above). The main contributing causes to LDB's reduced volume of lending would appear to be: (a) Limitations in the LDBs' physical capacity to handle a sub- stantial volume of loans. (b) Due to the linking formula, increase in overdues (see para. 35 to 40) restricts LDBs' sources of funds, including LDBs ability to borrow through the issue of debentures. (c) Minor irri- gation has remained the main field of lending, but with decelerating growth in some states, especially in Tamil Nadu. There is still a great potential for groundwater irrigation development in many other states. The current rigid spacing and density restrictions imposed to prevent water over- exploitation are likely to be made more flexible and better adapted to local conditions as SGDs accumulate more knowledge about the available groundwater resources. (d) Farmers also find alternative sources of funds more easily than in the past, including CBs and traditional sources, and have more funds available for investments themselves. Some studies have shown that a significant amount of investments, particularly minor irrigation invest- ments, have been made without recourse to institutional sources of credit. (e) Little lending has taken place in the field of land development, apparently because of unresolved problems in land consolidation, delays in completing irrigation infrastructure and underestimation of how much work farmers can carry out themselves without borrowing. Further, many farmers remain unconvinced about benefits of land development. (f) The LDB's diversified lending, an alternative to compensate for the slowdown in demand for minor irrigation loans,, has not developed significantly. This may be traced to statutory limitations, lack of experience in this area, and lack of aggressiveness on the part of LDBsl/. 32. The audit was not able to gather any evidence that technical conditions for lending imposed on LDB under IDA projects, a common source of complaint by most of the SLDBs and PLDBs visited, have slowed down demand and approvals. A different factor, poor harvests, is found to be associated with temporary decreases in loan demand, but this occurs only for specific years and LDBs and not for overall lending. I/ CBs have shown more interest in this field, which looks more "commer- cial" than traditional agriculture and may involve larger farmers. - 10 - 2. Sources of Financing/ 33. During the period 6/30/71 to 6/30/78 ARDC disbursements to LDBs have grown far more rapidly than loans advanced, showing increasing financial assistance by ARDC and a shift from financing through ordinary debentures (RBI) to special debentures (ARDC) (see Table 2). The percent- age of ARDC and IDA disbursements on loans advanced rose for ARDC finance from less than 20% to an average of almost 45% for the whole period, and for IDA finance from nil to about 33%. More detailed data on Maharashtra, Tamil Nadu, Karnataka and Andhra Pradesh LDBs show deviations compared with all-India data, but on the whole confirm the above. For Maharashtra LDB, and to a lesser extent for Tamil Nadu LDB, the decline in lending operations and the increase of financial assistance by ARDC and IDA are more significant than for Andhra Pradesh and Karnataka LDBs, whose overall performance is better. 34. Since LDB lending in real terms today is below the levels before 1971 - the year of first IDA involvement - (see para. 28 and Table 2), ARDC and IDA funds have substituted for local resources previously available to LDBs in the form of ordinary debentures at maturities of 10 to 15 years subscribed by the Life Insurance Corporation of India, RBI, CBs, state government, SLDB, and central and other cooperative banks in the state. ARDC is replacing to a large extent a system of finance which had worked properly as a channel of funds in the past but which, because of new developmental objectives, cannot be justified any longer. The new system entails a more technical approach to lending and a better leverage and more responsibilities for ARDC. 3. Overdues 35. As is underlined in the appraisal reports and the PCR, the most critical problem LDB faces is the high level of overdues. Since bad debts have never been written-off, some of the overdues refer to old operations. However, except for a small improvement between 1975 and 1976, the all-India position has worsened in spite of pressure from IDA and ARDC for increasing efforts in recoveries. Only in three states (Haryana, Punjab and Andhra Pradesh) have overdues been kept at a very low level. Other three states (Kerala, Uttar Pradesh and West Bengal) all have recoveries in excess of 75% of demand, which the Bank has deemed to be acceptable for a country like India. 36. The causes of the overdues have been analyzed at length in the PPARs for several state credit projects2J. They are multiple but the 1/ The available data on all-India LDB lending operations and financing are inadequate to support an indepth, updated financial analysis of LDB performance. Consolidated data published by the Department of Agriculture (GOI) are some years behind. 2/ See PPARs on the Gujarat Agricultural Credit Project, and on the Andhra Pradesh, Tamil Nadu and Maharashtra Agricultural Credit Pro- jects, op.cit. - 11 - most commonly quoted are: a) deficiencies in lending policies and pro- cedures: poor appraisal and supervision (a deficiency solved to a large extent through the IDA-financed projects); poor mechanism for recoveries; poor organization and shortage of experienced staff; inappropriate use of some techniques (such as rescheduling) to deal on a general basis with defaults and defaulters as a result of genuine hardships; b) weather vagaries, natural calamities and adverse economic conditions; c) in- adequate short term financing made available to borrowers; d) inadequate returns both to service loans and to meet farm families' basic needs, mostly in the case of small farmers; d) farmers' efforts to keep the money at hand as there is still a considerable credit shortage; e) public cam- paigns promoting non payment by farmers (these affect mostly but not ex- clusively debts to Government or Government sponsored bodies), and f) political interference in varied forms: (i) pressure to advance loans to certain categories of borrowers, for certain purposes and in given cases without consideration of the economics of the operation; (ii) orders given to LDBs or other cooperatives and government sources not to take coercive actions or press for repayment at certain crucial moments; (iii) mislead- ing declarations regarding terms and conditions of loans, etc. 37. One particular aspect regarding project analysis techniques is worth highlighting. Emphasis is usually put on rates of return and incre- mental returns, while somewhat overlooking the farmers' repayment capacity and the risks involved. Satisfactory financial rates of return or profit- ability of investments are a necessary condition for the farmers to repay promptly. They are not, however, a sufficient condition for good repay- ment rates, especially where small farmers are concerned. Absolute returns on minor investments are bound to be small. Furthermore, and more par- ticularly for small farmers living under difficult conditions, incre- mental returns are first used to cover additional living expenses and to repay current debts. Only after such requirements are met is loan service to institutional credit agencies forthcomingl/. 38. There are considerable differences between SLDBs in their recovery performance. Some studies have been made of the factors which may explain such differences.. The PCR (para. 3.11) properly states that "the most important characteristics of LDBs with good recovery performance (e.g. Haryana, Andhra Pradesh) are a strong LDB management, state govern- ment support of loan recovery efforts, and minimal political interference in LDB banking operations". 39. A further deterioration of SLDB and PLDB collection rates, or a failure to improve them, may prevent a number of SLDBs from meeting at least a portion of their dues. ARDC's capital and reserves represent less than one year of principal repayments by borrowing institutions. However, ARDC faces no legal risk, for the state governments guarantee the redemp- tion of the LDB debentures. Further: a) SLDBs have never defaulted on 1/ LDB officers frequently made this statement to the audit mission. Some studies, however, have shown that small farmers are no worse than others at repaying medium-term debts. - 12 - either ordinary or special debentures; b) dues on special debentures are not yet significant; c) the cushion of SLDB and PLDB resources is not yet exhausted. It should be borne in mind, however, that a portion of PLDBs' own resources is invested in SLDB share capital. In turn SLDBs (and SCBs as well) are holding a significant portion of ARDC share capital (for both SLDBs and SCBs it is about 24% as of June 30, 1978). Thus, the system's "own resources" are interwoven; if individual figures were consolidated they would show that actual, realizable resources are much less than what they appear to be; d) although it is against their statutory rules, SLDBs can delay without sanction the building of the sinking funds for the repayment of ordinary debentures;1/ and e) an increasingly large portion of ARDC refinance is channeled to CBs, whose financial situation is stronger than the LDBs'. 40. Is there any correlation between the implementation of the IDA credits and the increase in overdues? The process leading to increasing overdues had started before IDA financing became available, and in many states it followed the same lines as described for Maharashtra, Tamil Nadu and Gujarat2/. Most certainly, lending under IDA-financed schemes has been neutral in this respect, although additional funds provided by the IDA credits may have contributed to worsening the situation in those LDBs where management was weak and unable to cope with growing operations (this seems to be the case both in Tamil Nadu and in Maharashtra, where this was exacerbated by inhibiting government interference).!/. 41. Commercial banks are also facing high overdues. The magnitude of the problem, however, is quite different from that of LDBs since agri- cultural activities are still a small portion of their portfolio. Commer- cial bank officials explained to the audit mission that high overdues originated because they were not familiar with agricultural credit oper- ations, and that before acquiring some experience they had committed some errors and, notably, had underestimated the risks of farm lending. ARDC has set up a Committee on Agricultural Loans to CBs which is expected to result in improvements in the CBs' recovery performance. 4. Cost of Credit, Spread and LDB Profitability 42. Over recent years, more and more PLDBs have incurred losses and LDB profitability has, in general, decreased. The decline in profitability seems to be the consequence of: (i) the banks' inability to continue expanding their lending operations (see para. 31) and (ii) the current margin between LDBs' borrowing and re-lending rates, which seems to be insufficient to cover their operating expenditures and charges, plus maintaining adequate provisions for bad debts and building up adequate reserves to improve their financial strength and independence. 1/ The Committee on Integration (see para. 26) reported on some such delays. For a more detailed discussion of this particular point, see para. 51.c.ii. 2/ See the PPARs on these projects, op.cit. 3/ The GOI comments that there is a constant dialogue among GOI, ARDC, State Governments and LDBs to improve LDB's recovery performance. Rehabilitation programs for some of the worst affected LDBs are already in hand. - 13 - 41. Commercial banks are also facing high overdues. The magnitude of the problem, however, is quite different from that of LDBs since agricultural activities are still a small portion of their portfolio. Commercial bank officials explained to the audit mission that high over- dues originated because they were not familiar with agricultural credit operations, and that before acquiring some experience they had committed some errors and, notably, had underestimated the risks of farm lending. ARDC has set up a Committee on Agricultural Loans to CBs which is expected to result in improvements in the CBs' recovery performance. 4. Cost of Credit, Spread and LDB Profitability 42. Over recent years, more and more PLDBs have incurred losses and LDB profitability has, in general, decreased. The decline in profitability seems to be the consequence of: (i) the banks' inability to continue expanding their lending operations (see para. 31) and (ii) the current margin between LDBs' borrowing and re-lending rates, which seems to be insufficient to cover their operating expenditures and charges, plus maintaining adequate provisions for bad debts and building up adequate reserves to improve their financial strength and independence. 43. A wide range of margins were quoted by LDBs as necessary to cover their operating costs - usually 3.5 to 5%; generally around 4%1/. No estimates were given for possible losses because of defaults. There are some divergences of opinion regarding the adequacy of such margins because: (i) different portions of LDB administrative charges are borne by state governments; (ii) the cost of credit distribution and recovery and of defaults are not known with sufficient accuracy, and (iii) the volume of lending of an individual PLDB affects its cost structure as well as its profitability. A World Bank document?-/ suggests that lending administrative expenses incurred by LDBs themselves would reach about 5.7% of loans outstanding on average, or about 8% taking into account the services currently provided by Government. Risks are esti- mated to be 0.5% of loans outstanding. These figures are better supported by collected data than others, and seem more realistic. On these assump- tions, the current margin of 3% would fall considerably short of require- ments and would result in operating losses. 44. At present, however, LDB income statements continue to show profits because SLDBs have some funds of their own which have no cost, in many cases no provisions for bad debts are made, and LDB investments earn returns which offset losses on lending operations. Such investments, mostly of ordinary debenture redemption funds, are as high as 50% of loans outstandingI/. They will tend to decline as special debentures redeem- able annually substitute for ordinary debentures redeemable at maturity. There are few recent cases of SLDB having incurred losses (Maharashtra in 1/ CBs provided roughly similar figures. 2/ C. D. Datey (consultant). - The Financial Cost of Agricultural Credit: A Case Study of Indian Experience, World Bank Staff Working Paper No. 296, October 1978. 3/ For further details on returns on investments, see PPAR on Andhra Pradesh...,op.cit. - 14 - involvement in small farmer lending will have to be devised to prevent this socially desirable shift in lending patterns from squeezing the LDB's profit ratio below the viability threshold (see para. 55). 46. It was agreed during negotiation of ARDC II that ARDC, in con- junction with RBI, would carry out a study of interest spreads, with particular reference to the needs of LDBs. The completion of the study, initially envisaged by March 31, 1978, has been rescheduled to June 30, 1979. Under ARDC III, GOI would undertake to review it with particular reference to the LDBs. IDA further requested that the review concentrate on the cost of lending to small farmer's, but no agreement was reached during negotiations because by then the review was almost complete. 5. Other Issues a) Eligibility Criteria for Refinance 47. In order to control the level of overdues of the LDB system, a formula linking the issue of new debentures to the performance in the recovery of loans was agreed upon at the negotiations of ARDC I. This followed some attempts made under the most recent state projects to involve the state gavernments in underwriting overdues and, consequently, to induce them to apply pressure to LDBs to improve collections. The eligibility criteria do not apply to commercial banks. (PCR, paras. 1.11, 3.12 and 5.10 to 5.13). 48. Various amendments have modified the initial formula; the latest amendment was made in January 19791/. They have mostly aimed at reduc- ing the rigidities which had been introduced with the formula, enabling completion of incomplete investments and allowing for new lending to small farmers. These amendments have increased the eligibility of PLDB and SLDB branches which would otherwise have been deprived entirely or partly of new refinance. For example, in Gujarat the number of SLDB branches with unrestricted eligibility had been decreasing from 78 (as of June 30, 1976) to 11 (June 30, 1977) and 5 (June 30, 1978); it was raised to 28 as a result of the newest regulations. Similarly, the number of branches non-eligible for lending had climbed from 25 to 79 and 95 as of the same dates mentioned above; it came down to 68. PLDBs or SLDB branches which had become ineligible for lending because of their poor collection efforts have thus again became eligible. Caution will however need to be ex- ercised that the new flexibility is not misused to frustrate the basic purpose of the linking formula, namely, to help improve PLDB and SLDB loan recovery performance. 49. As the PCR explicitly states, the formula's effectiveness in controlling overdues has yet to be demonstrated. The continuation of its use may be questioned on the grounds of effectiveness. It may also be questioned on the grounds of equity and development requirements. How- ever, as the problem of overdues becomes more acute and no better alter- native is found, it would be a step backward to abrogate the present rules as this might result in a still more unmanageable situation. Investiga- tions should be continued concerning not only the formulation of rules which would not slow down development in nonprivileged areas, but also on practical measures or arrangements to improve recoveries. I/ Under the third project, ARDC undertook to maintain at all times criteria satisfactory to the Bank with respect to LDB eligibility. - 15 - b) Issue of Debentures 50. LDBs obtain their funds through the issue of ordinary debentures controlled by RBI, and of special debentures controlled by ARDC. The former carry no particular restrictions and LDBs can use their proceeds in any way they like within their statutory purposes. IDA-provided funds are transferred to LDBs through special debentures, which are issued by the LDB and bought by ARDC. Funds obtained through special debentures can be on-lent only for the purposes stipulated in the relevant IDA credit; lending must follow the technical procedure required and controlled by ARDC. Previously the two kinds of debentures were issued under different rules and procedures. In accordance with the credit agreement for ARDC I, efforts were continued towards the application of consistent standards by RBI and ARDC in the purchase of debentures from LDBs. Similar criteria for the issuance of all debentures were eventually adoptedi/. Following that, the sharing of responsibilities between ARDC and RBI in controlling their issuance does not seem to be any longer an obstacle for the adoption by LDBs of sound credit procedures and for the improvement of their performance. Since special debentures have become the main source of funds for most LDBs (see Table 2, last three columns), ARDC's technical control on LDB lending operations has become more effective. 51. The credit agreement stipulated also that a study be made on requirements for the issuance of LDB debentures as well as on the most efficient ways of improving the debenture system. The main purpose of the study was to provide the basis for formulating common requirements for the issuance of all LDB debentures. The obligation of making the study was waived by IDA (letter of July 19, 1976) in the light of the developments described in the preceding paragraph. However, the problems of the issue of debentures has not yet been fully resolved: (a) On the whole, the rationale of the debenture system remains questionable; as a case in point, ARDC is utilizing loans to transfer IDA funds to the CBs. (b) There are lengthy procedures for establishing the land mortgage basis to cover the ordinary as well as the special debentures, which in turn may delay the issue of debentures and call for costly interim finance to advance to mortgagees. (c) Although with the growing share of special debentures in LDB lending operations (see previous paragraph) ordinary debentures have become of less significance, their handling carries two other problems: (i) the sinking fund established to redeem them blocks funds which could otherwise be devoted to lending2/, and (ii) the dif- ferent timing of repayment for ordinary debentures, which are redeem- able at maturity, compared to those of sub-loans, which are recoverable 1/ A further step was taken in March 1979. As a result of the lowering of refinancing rates by GOI, rates for special debentures have become almost alike those for ordinary debentures. 2/ Funds other than those invested in ordinary debentures of sister banks or deposits with cooperative benks are lost for agriculture. Cross- subscription of debentures between LDBs, which have been sometimes questioned, results in the spreading of risks and better opportunities to use funds within the system - 16 - annually, may weaken LDB efforts to recover each year loan installments in their entirety, since the funds so collected are not going to be utilized immediatelyl/. 52. The matter of debentures has been dealt with by the Hazari Committee as an aspect of the problems of integration (see para. 26). The report of the committee is strongly critical of the system and suggests a change in the procedures for raising resources on the following lines: a) for schemes approved by ARDC, the committee recommends that ARDC provide refinance in the form of loans on the same basis as it provides for commercial banks and SCB at present; the contributions by the state governments now being made available to the special debentures would be granted in the form of term loans on the same conditions; and b) for other lending programs, debentures may be replaced by the issue of bonds with state government guarantee. The audit largely concurs with these recom- mendations and urges that steps be taken to implement them. The whole issue of LDB's raising resources through the flotation of debentures is currently being considered by RBI; necessary instructions will be issued after consultation with ARDC. D. Other Issues on Credit Operations 1. Financing the Replacement of Assets 53. The issue of financing the replacement of pumpsets was not raised in the PCR. It had been considered, however, by the ARDC II appraisal mission; during negotiations agreement was reached that ARDC would carry out a survey of requirements for probable pumpset replacements over the next 5 years and the most appropriate means of financing them. 54. The matter deserves attention from a general policy point of view. In general, IDA has refused to finance replacement of assets through credit proceeds. Moreover, financing through LDBs replacement of depreciated assets would augment the substitution of IDA resources for locally mobilized resources. But these statements may have to be modified in light of the decrease or levelling off of lending for minor irrigation in some states: lending for replacement of assets may become necessary as one way to continue supporting the expansion and improvement of LDB and ARDC activities. Further, it would contribute to protect long-term well investment. ARDC has recently constituted a committee on projected pumpset requirements and methods of financing. 1/ Further, with the exception of the latest years, funds left unrecov- ered earned higher interests for the LDBs than those recovered and invested in securities to establish the sinking fund. Strictly financially, therefore, non-recovering was almost more attractive for LDBs than recovering. (For further details on the relative profit- ability of investments and loans, see PPAR on Andhra Pradesh,... op.cit., para. 70).) Although the co-operative laws and by-laws of the LDBs' require that sinking fund requirements are to be met, if necessary, by using their own funds, RBI has reported several cases of LDBs not fulfilling their commitments to sinking funds on account of poor recoveries. (Report of the Committee on Integration; see paras. 26 and 39 d.) In some rare cases, LDBs were required to invoke the State government's guarantee to eventually redeem ordinary debentures for which sinking fund provisions proved inadequate (Annex I, page 9). - 17 - 2. The Small Farmer Definition 55. The PCR recalls the evolution of the "small farmer" definition through the history of the agricultural credit projects (paras. 5.04 to 5.09). It is important that a satisfactory definition has been arrived at, but whatever the cut-off point the real problem is extending credit to the greatest number of farmers below it. This problem has been tackled in various ways but it does seem, although data is scarce, that larger farmers continue to obtain most of the benefits from institutional credit, mostly from CB lending. There seems to be little incentive at present for lending agencies to assist small farmers, and few ways to mitigate the cost of lending small amounts to a great number of borrowers. Currently, SFDA grants a 2% "rebatge" to LDBs on the amounts disbursed by these to small farmers (as defined by SFDA; see PCR, para. 5.05). Some CBs have found a solution in indirect lending through cooperative credit societies and sharing the spread (1% for the CB, 2% for the cooperative). Since lending to small farmers is only a minor part of the CB portfolio, a kind of "cross-subsidization" within that portfolio occurs. Some LDB officers suggested that they should be allowed to have a few large loan operations each year to generate additional income to cover the high cost of lending to small farmers. Another solution may be allocating the spread rather on the basis of the number than on the value of loans made. The Bank and GOI should give priority to develop new approaches which would allow LDBs and CBs to cover the actual cost of lending to small farmers and introduce incentives which would encourage them to lend to the small farmer. The study on interest rate spreads mentioned above (see para. 46) may provide some additional information on this issuel/. 3. On-Lending Terms 56. On-lending terms to ultimate borrowers as stipulated under the first state-oriented agricultural credit projects have been softened and made more flexible, particularly to facilitate lending to small farmers. Representations were still made to the audit mission by officials of GOI's Department of Agriculture, state governments, some CBs and LDBs regarding "harsh terms" of IDA-financed loans: a) high rates of interest on loans to ultimate borrowers (in this, GOI's high rates of on-lending to ARDC were at stake); b) requirement of a down payment; c) inadequate repayment periods; and d) inadequate period of grace, especially in the case of minor irriga- tion. It was suggested to the mission that more flexibility should be allowed in order to enable the lending agencies to adapt the lending con- ditions to the borrower and area conditions. 1/ GOI comments that the interest rate structure obtaining for agri- cultural lending in India takes into account all factors including continuing viability of the financing institutions involved (Annex I, page 10). - 18 - 57. Leaving aside the problem of interest rates, the audit does not share the above reported view except with regard to the period of grace. At present, the first annuity is usually due 18 months after the disbursement by the lending agency of the first loan installment (the project conditions require it to be within 2 years). It may happen that the first annuity becomes due only a short period after the completion of the farmer's investment and before he has harvested a single crop. Thus, the question of an appropriate period of grace is being examined. Some ways have been suggested. For example, the repayment period might start after completion of project investments or the first payment could be postponed to allow the borrower to harvest his first crop. It would be necessary to fix limits on the completion of the project (for instance, one year) with possible waiver in case of delays beyond the borrower's control. Another possibility might be a scale of repayments increasing as the farmer gets larger benefits. The present on-lending terms do not forbid such arrange- ments._1/ 58. Although other lending terms appear to be appropriate, there is the problem of highly variable weather and crop yields which may jeopar- dize the farmers repayment capacity for one or several years. Apart from crop insurance, the practicability of which under Indian conditions has yet to be demonstrated, flexibility might be built into the repayment schedule, envisaging the probability of the random occurrence of a calam- ity and providing for the consequent adjustments in loan repayments. Rescheduling after calamities is a common banking practice, but it calls for some caution. The techniques used in Gujarat, Maharashtra and Tamil Nadu during the implementation of the respective state projects were poor and had adverse effects on recoveries. E. Credit Disbursement Mechanism 59. Credit disbursement procedures were not challenged by the lend- ing agencies visited by the audit mission, nor are they mentioned in the PCR. Disbursements against loans made by LDBs are based on a percent- age of total expenditures of each category. In theory, the procedures appear to be simple. In practice, however, they do not work so simply, as they require the consolidation of multiple statements of loan disburse- ments at each institutional level (PLDB, SLDB, ARDC regional offices, and ARDC head office), repeated controls and intervention of the registrar of cooperative societies and state governments with regard to the flotation 1/ Under the ARDC III project, grace periods may be granted at the dis- cretion of ARDC, provided that the repayment period is not exceeded. Also, ARDC has been pressing LDBs to introduce flexibility in repay- ment terms by introducing a system of graded installments in place of equated installments. - 19 - of special debentures. Thus, the procedure is simple for IDA but compli- cated up to the ARDC levell1/. It is somewhat less complicated regarding commercial banks, which are refinanced by ARDC through loans instead through debentures. 60. Whether these procedures and controls guarantee that the opera- tions are carried out in accordance with the credit agreements, and that disbursements have actually taken place, is another issue. In principle, the desk controls, usually supplemented by spot checks in the field, should produce the assurances required. But the disbursements made in Tamil Nadu against "switch-over" loans.2/, which are now difficult to identify, raise some doubt whether existing controls are effective enough3/. 61. Before 1976, there were some delays between disbursements to farmers and IDA disbursements to GOI. There are still some delays because of the procedures of establishment of mortgages and the time required to tabulate disbursements, but they have been reduced and are not a matter of concern any longer. 1/ GOI states (Annex I, page 11) that the forms for disbursements are pre- pared by ARDC but it is correct that the drawal of refinance by the participating banks is delayed as they have to meet the requirements of funds position in individual banks. This trend is however being (re- viewed) with a view to draw refinance against their disbursements without delay. 2/ This was an exceptional operation, whereby GOI and IDA agreed to refinance under several statewide projects (hence the term "switch- over" loans utilized by GOI and ARDC) subloans which did not fully conform with the lending conditions stipulated in the credit agree- ments (these loans were usually referred to as "irregular" loans in IDA memoranda and correspondence), or which had been effected outside the previous projects areas since the credit agreements had become effective ("off-limits" loans). For a detailed discussion see the PPAR on Andhra Pradesh...,op.cit., paras. 73 to 79. 3/ Prompted by this conclusion, the last supervision mission of ARDC II (June 1979) checked five different states. The mission reported having been able to satisfactorily trace payments from IDA through GOI, ARDC and LDB to the participating farmer's accounts in every state. Table 1 ARDC: LOCAL RESOURCE MOBILIZATION 1 (Amounts in Rs million) Sources/Years 1970-71 1971-72 1972-73 1973-74 1974-5 1975-76 1976-77 1977-78 G01-2/ 220 65.5 70.4 RBI ) 203.2 ) 218.7 304.8 311.4 403.8 385 ) ) ) ) ) 940 ) 856 Bonds ) ) 110.0 275 330.0 605 ) ) Subtotal 423.2 284.2 485.2 586.4 733.s 990.1 940 856 Repayments from Borrowers 5.8 5.9 14.2 42.3 92.7 245.9 480 829 Share Capital --- 50.0 50.0 --- 50.0 50.0 100 125 Accretion to Reserves 1.4 2.1 3.7 8.3 12.2 16.7 27 43 Subtotal 430.4 342.2 553,4 637.0 888.7 1302.7 1547 1853 o Repayments of borrowings 42.8 127.5 18.5 45.1 155.9 125.9 158 416 Total Local Resources 387.6 214.7 534.6 591.9 732.8 1176.8 1389 1437 GOI/IDA __38,3 406.8 386.5 331.2 534.7 900 996 Total Funds mobilized (net after 387.6 253.6 941.4 978.4 1064.0 1711.5 2289 2433 Repayments) Local resources as % of total funds 100 84.9 56.8 60.5 68.8 68.7 60.7 59.1 mobilized Index of increase of Local resources 100 55 138 153 189 304 358 371 Total funds mobilized 100 65 243 252 274 442 591 628 1/ This table records data furnished by ARDC to the audit mission. A number of figures conflict with ARDC cash flow 1971/72 to 1976/77 given in Annex II Table 3 of the PCR. L From 1973-74 on funds from GOI were limited to reimbursements of the IDA credits drawn. Information on the origin of GOT funds previous to 1973-74 is not available. - 21 - TABLE 2 LAND DEVELOPMENT BANKS: ALL-INDIA LENDING OPERATIONS (Amounts in Re million) I 1 IIncrease in Loanal I ! I Loans Advanced ILoan Out-1 I Outstanding I ARDC Disbursemental I Years I During Year Istanding I I lin % of I to LDBs I I ! 1/ IIn NominallIn real I end of I I in ! Loans I lUnder IDAI I 1 I Terms ITerms2/31Period I A/B I AmountslAdvanced I Total IProjects I D/A ! E/A ! E/C ! I I I I I I 1 1 4/ 1 I I (A) I I (B) II (C) 1 1 (D) I (E) I I I I II I II 1967/68 1 N/A 1 1 3,560 1 I I 1 I 1 1 I I I II III 1968/69 I 1,900 1 2,145 I 5,060 1 0.355 1 1,500 1 78.9 I N/A I - I I I I I I IiI I I II 1969/70 1 1,900 1 2.034 1 6,220 ! 0.305 1 1.160 1 61.1 1 267.5 1 - ! 0.148 I I I II I ! ! 1970/71 1 1,940 1 1,940 1 7,250 1 0.264 1 1,030 1 53.1 1 266.6 I - I 0.137 I I I 1 1 I I I I ! 1971/72 1 1,457.5 I 1,542 1 7,670 I 0.212 1 420 1 25.8 I 283.9 1 53.7 1 0.174 ! 0.033 ! 0.128 ! II ! I I I I I I 1972/73 1 1,714.3 1 1,499 I 8,492.1 I 0.202 1 822 I 46.1 ! 861.4 I 635.8 I 0.502 ! 0.371 1 0.773 ! I I 1 I I I I 1 ! ! ! 1973/74 ! 1,467.0 1 1,122 1 9,140.5 1 0.160 1 648.41 46.4 1 777.6 1 529.2 ! 0.530 1 0.361 I 0.816 I I 1 1 I 1 1974/75 1 1,887.3 1 1,147 1 9,925.4 1 0.191 ! 784.91 41.5 I 770.6 ! 519.8 I 0.408 1 0.275 ! 0.662 ! I I 1 1 III 1975/76 I 2,049.4 1 1,163 110,687.0 1 0.192 1 761.6! 37.2 I 990.9 1 906.9 I 0.483 I 0.442 ! 1.191 ! I 1 1 I I 1 1 1976/77 I 2,487 I 1,379 112,123.9 1 0.205 I 1,436.9! 57.7 ! 1,267.0 1 1,005.3 ! 0.509 ! 0.404 ! 0.793 ! 1 1 1 I1 1 1977/78 1 2,348 I 1,264 113,150.0 I 0.178 11,026.1! 43.7 1 1,119.4 858.0 1 0.477 ! 0.365 1 0.836 I I 1 ! I 1 ! ! ! ! Totals 113,583 1 I 5,900 1 ! 5,900 1 I 6,070.8 ! 4,508.7 ! 0.4475/ ! 0.332-/ ! 0.7431/ i 6/30/71-! I 1 ! I I I I I 6/30/78 1 1 I I I I I I ! ! Source: 1967/68 to 1971/72. - Figures provided by ARDC 1972/73 to 1976/77.- ARDC/GOI - Preparation Report for a proposed Third ARDC Project, June 1978 (figure for 1976/77, as amended by IDA in SAR of ARDC III, Annex 5, Table 1). 1977/78. - IDA - SAR of ARDC III; Report No. 2404a-IN; May 16, 1979; Annex 5, Table 1. 1/ End of Periods: as of June 30. 2/ Base 1970/71 - 100; Deflator: Implicit Price Deflator of Gross Domestic Capital Formation, as reported in IBRD - Economic Situation and Prospects of India, Report No. 2431-IN, April 9, 1979, Table 6.12. 3/ Figures provided in GOT comments (Annex I, page 6) conflict with some of the figures in this table, but they confirm the decline of LDB lending in real terms: Amount in Amount in Year Nominal terms Real Terms 1969/70 Ra. 1555 H Re. 1665 IH 1970/71 Re. 1706 M Rs. 1706 M 1977/78 Rs. 2781 M Rs. 1497 M 4/ Disbursements under ARDC I in the period 1974/75 to 1976/77 slightly overlap disbursements under the late statewide projects. The overlapping refers to the aggregate, all-India figures. There is no overlapping in either individual State; ARDC I funds were utilized only after the state credit project funds had been exhausted. Disbursements of ARDC I and II overlap in 1977/78. 5/ Average for period 6/30/71 to 6/30/78. Annex I Page 1 Comments Received from the Government of India Z-140 NEW DELHI 2800 151115 ETAT PTY MOST IMMEDIATE INDEMBASSY WASHINGTON REDDY FROM SIBAL ECOFAIRS NO.F.6(31)79.F.R.2 REFERENCE DRAFT AUDIT REPORT SENT BY OED *ON CLOSED 0 CREDIT PERTAINING TO ARDC I PROJECT FOR COMMENTS TO GOI. THE FOLLOWING OBSERVATIONS MAY KINDLY BE PASSED ON TO OED:: QUOTE AAA PARA 5 OF THE AUDIT MEMORANDUM MENTIONS (D THAT THE EVALUATION OF THE IMPACT OF PROJECT INVESTMENTS "1 HAS BEEN MODEST BECAUSE ARDC HAS NOT COMMITTED ADEQUATE 3 MAN-POWER. IT NEEDS TO BE MENTIONED IN THE AUDIT FEPCRT THAT A SYSTEMATIC EFFORT IN CARRYING OUT EVALUATION STUDIES WAS INITIATED WITH THE COMMENCEMENT OF ARDC AND IN THE VERY NATURE OF THINGS ARDC COULD PROVIDE THE STAFF FOR THE PURPOSE ONLY GRADUALLY. ARDC HAS NOT STRENGTHENED - ITS EVALUATION CELL AND DETERMINED FIRM PROGRAMME FOR THE CELL AND HAS SUGGESTED TO PARTICIPATING BANKS TO SET UP C EVALUATION CELLS WITHIN THEIR OWN ORGANISATIONS AND UNDERTAKE SPECIFIC EVALUATION STUDIES. - 23 - Annex I Page 2 BBB IN PARA 5 OBSERVATION IS MADE THAT GOOD RECOVERY IN 1975-76 COULD BE ATTRIBUTED TO THE FORMULA LINKING LDBS ELIGIBILITY WITH RECOVERY PERFORMANCE. BUT IN 1976-77 LESSER RECOVERY COULD BE ATTRIBUTED TO POLITICAL CLIMATE. 0 r THE CAUSES FOR GOOD OR BAD RECOVERY PERFORMANCE APPEAR TO C? BE CONJECTURAL PARTICULARLY WHEN WE LOOK AT THE OBSERVATIONS N N MADE IN PARA 5.12 OF THE PCR WHICH POINTS OUT THAT THE i EFFECTIVENESS OF THE LINKING FORMULA TO CONTROL OVERDUES . IS NOT ESTABLISHED. IT IS, THEREFORE, SUGGESTED THAT 25 SUITABLE MODIFICATION NEEDS TO BE MADE IN PARA 5 IN THIS CONNECTION AS THE LINKING FORMULA HAD IN FACT BECOME REGRESSIVE BY RELATING THE LENDING PROGRAMME TO THE PREVIOUS YEARS ACTUALS. CCC AGAIN IN PARA 5 WHERE IT IS MENTIONED THAT A HIGH 8 LEVEL OF OVERDUES CONTINUES TO BE THE MAJOR PROBLEMS OF THE o co LDB IT MAY BE APPROPRIATE TO ADD THAT IN TERMS OF THE UNDER- STANDING REACHED IN ARDC III PROJECT REHABILITATION PROGRAMMES HAVE BEEN PREPARED/LAUNCHED IN THE CASE OF MAHARASHTRA, TAMIL NADU, -KARNATAA, GUJARAT AND BIHAR LDBS WHERE THE OVERDUES EXCEED 50 PER CENT OF THE DEMAND. Annex I - 24 - Page 3 DDD. IN PARA 6 A CONCLUSION HAS BEEN DRAWN THAT INTE- GRATION OF LONG AND SHORT-TERM CREDIT SYSTEM IS DESIRABLE. THIS IS IN CONTRADICTION TO THE CONCLUSION MADE IN PCR PER- TAINING TO PUNJAB AND HARYANA AGRICULTURE CREDIT PROJECTS WHERE IT HAS BEEN MENTIONED THAT THERE IS NO EVIDENCE THAT FORMERS OBTAINING LONG TERM CREDIT FROM LDBS WHERE DENIED REQUIRED SHORT-TERM CREDIT. FURTHER IT IS NECESSARY TO NOTE IN THE AUDIT REPORT THE FACT THAT UNDER ARDC III AGREEMENT GOI HAS AGREED TO REVIEW THE PROGRESS MADE BY THE COOPERATIVE CREDIT SYSTEM TOWARDS MEETING THE AGRICULTURAL CREDIT NEEDS OF THE FARMERS THROUGH ONE SOURCE OR CLOSELY COORDINATED SOURCES AND GOI IS EXPECTED BY DECEMBER 31 1980 TO INFORM IDA ON ITS FINDINGS AND RECOMMENDATIONS ARISING FROM THE SAID REVIEW. EEE. THE LAST TWO SENTENCES OF PARA 16 WOULD NEED MODIFI- CATIONS BECAUSE OF THEIR CONTROVERSIAL IMPORT. FIRST IT IS NOT CLEAR WHY A COMPARISON SHOULD BE DRAWN BETWEEN ARDC AND IDA VIS-A-VIS THEIR RELATIONSHIP WITH SLDBS AND THE STATE GOVERNMENT SINCE THEIR RESPECTIVE ROLES IN FINANCING AGRI- CULTURAL DEVELOPMENT ARE DIFFERENT. IN ANY CASE ANY MODIFI- CATION TO THE COVENANTS AGREED TO UNDER THE LINE OF CREDIT APPROACH REQUIRE CONCURRENCE OF I-A AND THERE HAS BEEN NO DILUTION IN APPRAISAL STANDARD UNKER THE GENERAL LINE OF CREDIT. THE QUESTION, THEREFORE, OF ARDC COMING UNDER POLI- TICAL OR OTHER PRESSURES IN VIEW OF ITS GREATER ACCESS DOES NOT SEEM RELEVANT AND THE CONCLUDING SENTENCE IN PARA 16 SHOULD EITHER BE DELETED OR SUITABLY MODIFIED. - 25 -Annex 1 Page 4 FFF. IN PARA 13 OBSERVATION IS MADE REGARDING LACK OF HIGH CALIBRE STAFF IN ARDC REGIONAL OFFICES TO MAINTAIN QUALITY IN APPRAISAL AND SUPERVISION. AS ARDC HAS BEEN MAKING EFFORTS ON A CONTINUOUS BASIS TO IMPROVE THE CALIBRE OF ITS STAFF AS WELL AS RECRUIT NEW STAFF PERHAPS IT WOULD BE MORE APPROPRIATE IF IT IS INDICATED IN THE AUDIT REPORT THAT ARDC STILL LACKS HIGH CALIBRE STAFF IN SUFFICIENT NUMBERS'IN ORDER TO INDICATE THAT AT LEAST SOME STAFF WITH HIGH CALIBRE IS AVAILABLE IN ARDC. GGG. COMMENTS INCLUDED IN PARA 19 ON THE QUALITY OF DATA AND INFORMATION PROVIDED BY ARDC WOULD APPEAR TO INDICATE THAT ARDC HAS NOT BEEN PROVIDED CORRECT INFORMATION TO IDA. WHILE ARDC AGREES THAT THERE IS SCOPE FOR IMPROVEMENT IN MANAGEMENT INFORMATION AND DATA PROCESSING SYSTEM, IT SHOULD BE RECOGNISED AT THE SAME TIME THAT THERE ARE CONSIDERABLE DIFFICULTIES RELATING TO COMPILATION OF ACCURATE DATA SINCE PROJECTS ARE IMPLEMENTED THROUGH A NUMBER OF INSTITUTIONS AND THEIR BRANCHES IN THE FIELD, SIMILARLY LOAN DISBURSEMENTS ARE EFFECTED IN MULTIPLE INSTALMENTS. DESPITE THESE DIFFICULTIES ARDC DOES MADE A CAREFUL ASSESSMENT OF THE PHYSICAL ACHIEVEMENTS UNDER INDIVIDUAL LOANS/ PROJECTS AND THIS IS BACKED BY VERIFICATION DURING FOLLOW- UP/MONITORING/EVALUATION STUDIES. ALSO ARDC HAS INTRODUCED SCHEME COMPLETION REPORTS WHICH ARE FILED BY THE BANKS AND SENT TO ARDC. IT MAY, PERHAPS, BE THEREFORE MORE APPROPRIATE IN ARDCS EFFORTS TO IMPROVE THE MANAGEMENT INFORMATION AND DATA PROCESSING SYSTEMS ARE SUITABLY REFLECTED IN THE AUIT REPORTS. Annex I - 26 - Page 5 HHH. THERE ARE TWO POINTS REGARDING PARA 20. WHICH REQUIRE EITHER DELETION OR FULLER EXPLANATION. FIRST IS REGARDING ARDC NOT AGREEING TO CHANGE ITS STAFFING STRUCTURE UNDER ARDC III NEGOTIATIONS. THE POSITION AS IT EMERGED DURING THE NEGOTIATIONS IS INCORPORATED IN PARA 7 OF THE AGREED MINUTES. THIS NEEDS TO BE SUITABLY REFLECTED IN PARA 20. SECOND THERE IS A REFERENCE TO LACK OF LEVERAGE OF THE BANK WITH INDIVIDUAL STATE GOVERNMENTS ON LDBS STAFFING SITUATION UNDER THE ALL INDIA LINE OF CREDIT APPROACH. THE FACTUAL POSITION IS THAT NO SPECIFIC SUGGESTIONS HAVE EMERGED OUT OF VARIOUS IDA SUPERVISION MISSIONS REVIEWING LDBS ORGANISATIONAL SET UP IN THE STATES VISITED BY THEM EXCEPT PERHAPS REGARDING APPOINTMENT OF PROFESSIONAL BANKERS AS MANAGING DIRECTORS FOR REASONABLY A LONG PERIOD IN A FEW STATES. THE CONCLUSION DRAWN IN THIS REGARD, THEREFORE, IS TOO SWEEPING AND WOULD HAVE SOME RELEVANCE IF THE SUGGESTIONS MADE BY IDA WITH REFERENCE TO SPECIFIC LDBS HAVE REMAINED UNATTENDED. III. PARA 21 DRAWS THE CONCLUSION THAT LOCAL RESOURCE MOBILISATION OF ARDC HAS NOT KEPT PACE IN RECENT YEARS WITH THE TOTAL RESOURCE MOBILISATION. IT HAS TO BE KEPT IN VIEW THAT IDA ASSISTANCE BEING IN THE NATURE OF GENERAL SUPPORT TO INDIAN ECONOMY BUT DIRECTED TO SPECIFIC PROJECTS IN PRIORITY SECTORS CANNOT BE CONSIDERED AS SUBSTITUTING FOR LOCAL CAPITAL WHICH HAS SEVEPAL ALTERNATE PRODUCTIVE USES. Annex I Page 6 - 27 - JJJ. IN PARA 22 DISCUSSION OF DER OF ARDC DOES NOT MENTION THE EXEMPTION GRANTED BY GOI TO ARDC FROM INCOME TAX FOR A PERIOD OF 5 YEARS COMMENCING FROM 1977-78 WHICH WOULD ENABLE ARDC TO IMPROVE ITS DER BY ALLOCATING LARGER FUNDS TO RESERVES. SECONDLY, ARDC PERSPECTIVE PLAN ALSO PROVIDES FOR ISSUE OF SHARE CAPITAL IN APPROPRIATE AMOUNTS TO ENSURE THAT DER REMAINS WITHIN STATUTORY LIMITS. LASTLY THE RELEVANCE OF DER IN CASE OF FINANCIAL INSTITUTIONS LIKE ARDC IS SUBSTANTIALLY DIFFERENT FROM THAT APPLICABLE TO INDUSTRIAL CONCERNS. KKK. THE COMMENTS MADE IN PARA 24 OF THE MEMORANDUM NEED TO BE MODIFIED IN THE LIGHT OF THE EARLIER COMMENTS GIVEN ON PARA 6 ABOVE. LLL. FACTS GIVEN IN PARA 28 OF THE MEMORANDUM NEED CORRECTION AS OTHERWISE THEY LEAD TO ERRONEOUS CONCLUSIONS. IN 1969- 70 TOTAL ADVANCES BY LDBS WERE RS.1555 MILLIONS. AGAIN IN 1970- 71 ADVANCES OF LDBS WERE RS.1706 MILLION. THE FACTUAL POSITION IS THAT THERE HAS BEEN A PROGRESSIVE INCREASE IN THE TOTAL LENDING OF LDBS OVER THE YEARS BARRING ONE OF TWO YEARS AND A PEAK WAS REACHED IN 1977-78 WITH A LENDING OF RS.2781 (2781) MILLION. SECONDLY IT WOULD BE MORE APPROPRIATE TO COMPARE LDBS PERFORMANCE IN LENDING WITH THE TOTAL INSTITUTIONAL CREDIT FOR TERM INVESTMENT IN AGRICULTURE ALONE. Annex I - 28 - Page 7 MMM. THE CONCLUSION DRAWN IN ARA 29 REGARDING DECLING PERFORMANCE OF LDBS NEED ELUCIDATION. FIRST IROM 1969-70 TO 1977-78 LOANDING BY LDBS INCREASED FROM RS.1555 MILLION TO RS.2781 MILLION WHILE THE OUTSTANDING INCREASE FROM RS.5131 MILLION TO RS.11934 MILLION. SECONDLY, MOST OF THE LOANS ADVANCED BY LDBS ARE FOR LONGER DURATION EXTENDING FROM 9 TO 15 YEARS WHILE RECOVERIES WOULD RELATE TO SMALLER LEVELS OF ADVANCES OF EARLIER YEARS. NNN. THE FACT AND CONCLUSION IN PARA 30. APPEAR TO BE INC-CORRECT. THE FACTUAL POSITION IS THAT THE NUMBER OF BORROWERS DECLINED FROM 4,80,000 IN 1969-70 TO 3,61,000 IN 1971-72 AND INCREASED TO 4,79,000 IN 1972-73, 5,00,000 IN 1974-75 AND OVER 6,52,000 IN 1976-77. FURTHER COMPARISON BETWEEN VARIOUS STATES MERELY GOING BY THE NUMBER OF LOANS WOULD NOT BE CORRECT UNLESS THE FIGURES ARE COMPARED ONLY WITH REGARD TO SIMILAR PURPOSES FOR WHICH LDBS ADVANCE LOANS, FOR EXAMPLE. FARM MECHANISATION BEING CAPITAL INTENSIVE THE NUMBER OF RORROWERS WOULD BE SMALL AND IN CERTAIN AREAS EVEN INVESTMENTS IN MINOR IRRIGATION WOULD BE RELATIVELY HIGH COMPARED TO OTHER AREAS DEPENDING SOIL CONDITIONS. IN CONCLUSIONS THERE APPEARS TO BE NO EVIDENCE TO SHOW THAT THER HAS BEEN ANY DECLINE IN THE OVERALL LOANING BY LDBS. 000. IN PARA 31 NECESSARY MODIFICATIONS NEED TO BE MADE AS THERE IS NO CONCLUSIVE EVIDENCE TO SHOW THAT THERE IS A DICLINING TREND IN VOLUME OF LENDING OR NUMBER OF LOANS. FURTHER SUITABLE MENTION NEED TO BE MADE OF THE EFFECT OF THE LINKING OF ELIGIBILITY CF PLDBS LENDING PROGRAMME TO THE PREVIOUS YEARS ACTUALS UNDEr THE NORMS AGREED TO UNDER ARDC I. Annex I - 29 - Page 8 PPP. IN PARA 35 IT MAY BE RELEVANT TO NOTE THAT THE LDBS IN U.P. WEST BENGAL AND KERALA HAVE ALSO IMPROVED CONSIDERABLY IN RECOVERY. QQQ. IN DISCUSSION REGARDING CORELATION BETWEEN IMPLEMENTATION OF IDA CREDITS AND INCREASE IN OVERDUES IN PARA 40 IT NEEDS TO BE KEPT IN VIEW THAT THERE IS A CONSTANT DIALOGUE BETWEEN GOI. ARDC, STATE GOVERNMENT AND LDBS TO IMPROVE THE RECOVERY PERFORMANCE OF LDBS. IN THIS REGARD REHABILITATION PROGRAMME FOR SOME OF THE WORST AFFECTED LDBS IS ALREADY IN HAND. RRR. IN PARA 41 NEEDS TO BE NOTED THAT ARDC HAS ALREADY SET UP A COMMITTEE ON AGRICULTURAL LOANS TO COMMERCIAL BANKS TO EFFECT IMPROVEMENTS IN THE RECOVERY PERFORMANCE OF COMMERCIAL BANKS. FURTHER THE REFERENCE IN PARA 41 TO THE USE OF COOPERATIVE SOCIETIES BY COMMERCIAL BANKS APPEARS TO BE EXAGGERATED. THE FACTUAL POSITION IS THAT LAST YEAR DISBURSEMENTS BY COMMERCIAL BANKS FOR AGRICULTURAL CREDIT WERE NEARLY RS.770 CRORES WHILE THE LOANS DISBURSED BY THEM THROUGH COOPERATIVE SOCIETIES WERE ONLY RS.27 CRORER. THE RATIONAL FOR INVOLVING COOPERATIVE SOCIETIES AS CHANNELS FOR COMMERCIAL BANKS CREDIT IS MAINLY WITH A VIEW TO HELF COMMERICAL BANKS IN EXTENDING THEIR INVOLVEMENT IN AGRI- CULTURAL SECTOR AND NOT WITH A VIEW TO REDUCE THEIR RISKS AND COSTS AS MENTIONED IN THE AUDIT REPORT. Annex I -30 - Page 9 SSS. REFERENCE IN PARA 48 REGARDING DILUTION OF THE ELIGIBILITY CRITERIA WITH A VIEW TO CIRCUMVENT THE DISCIPLINARY CONSTRAINTS IS MISCONCEIVED. IT SHOULD BE SUITABLY INCORPORATED IN THE AUDIT REPORT THAT THE AMENDMENTS MADE TO THE ELIGIBILITY CRITERIA ARE INTENDED PRIMARILY TO REDUCE THE UNREALISTIC RIGIDITY WHICH HAVE CREPT IN AND AND DESIGNED TO ENABLE COMPLETION OF INCOMPLETE INVESTMENTS AND TO ADMIT LOANS TO SMALL FARMERS. A SUITABLE NOTE ALSO SHOULD BE MADE OF THE MORE STRINGENT ELIGIBILITY NORMS AGREED TO UNDER ARDC III. TTT. THE REFERENCE IN PARA 50 THAT LDBS CAN NO LONGER EVADING CRITERIA IMPOSED BY ARDC BY ISSUING THEIR OWN DEBENTURES IS ALSO BASED ON INCORRECT REASONING AS THERE IS NO EVIDENCE TO SHOW IN THE PCR THAT LDBS HAVE SOUGHT TO EVADE ARDC CRITERIA BY RESORTING TO ORDINARY DEBENTURES. UUU. IN PARA 51 THE OBSERVATION THAT LDBS WOULD NOT FIND IT FINANCIALLY ATTRACTIVE TO RECOVER THE LOANS AGAINST ARDC DEBENTURES IS NOT BASED ON CORRECT APPRECIATION OF THE FACTUAL POSITIONS. UNDER THE COOPERATIVE LAWS AND BY LAWS OF 7HE LDBS SINKING FUND REQUIREMENTS ARE TO BE MET IF NECESSARY BY USING THEIR OWN FUNDS IN THE FORM OF SHARE CAPITAL AND RESERVES. FURTHER INSTANCE WHERE LDBS WERE REQUIRED TO INVOKE STATE GOVERNMENTS GUARANTEE FOR REDEMPTION OF DEBENTURES IN THE ABSENCE OF ADEQUATE SINXING FUND PROVISIONS ARE RARE. WITH THE GROIING HSARE OF ARDC SUPPORTED DEBENTURES IN LDBS LOAN BUSINESS THE MISSIONS OBSERVATION ABOUT THE ORDINARY DEBENTURES SYSTEM BECOME OF LESS SIGNIFICANCE. Annex I - 31 - Page 10 VVV. IN REGARD TO PARA 52 IT MAY MENTIONED THAT THE ISSUES PERTAINING TO THE SYSTEM OF LDBS RAISING RESOURCES BY WAY OF ISSUE OF DEBENTURES IS UNDER CONSIDERATION OF THE RBI AND NECESSARY INSTRUCTIONS TO THE LDBS WILL BE ISSUED SOOM AFTER THESE ARRANGEMENTS ARE FINALISED IN CONSULATATION WITH ARDC. WWW. IN PARA 54 WHILE DISCUSSING FINANCING OF PUNP-SETS REPLACEMENTS A NOTE MAY BE MADE THAT ARDC HAS-RECENTLY CONSTITUTED A COMMITTEE ON PROJECTED PUMPSETS REQUIREMENTS AND METHODS OF FINANCING. THE ISSUE RAISED IN THE AUDIT REPORT COULD BE EXAMINED AFTER THE COMMITTEES FINDINGS ARE AVAILABLE. XXX. IN PARA 55 THERE ARE RECOMMENDATIONS TO GOI FOR DEVELOPING NEW APPROACHES TO ALLOW LDBS AND CBS TO COVER THE ACTUAL COST OF LENDING TO SMALL FARMERS. IT NEEDS TO BE NOTED, HOWEVER, THAT THE INTEREST RATE STRUCTURE OBTAINING FOR AGRICULTURAL LENDING IN INDIA TAKES INTO ACCOUNT ALL FACTORS INCLUDING CONTINUES VIABILITY OF THE FINANCING INSTITUTIONS INVOLVED. YYY. IN PARA 57 NEEDS TO BE NOTED THAT SUITABLE FLIXIBILITY IN FIXATION OF GRACE PERIOD FOR INDIVIDUAL LOANS HAS BEEN PROVIDED FOR UNDER ARDC Ifl. SIMILARLY ARDC HAS ALSO BEEN PRESSING LDBS TO INTRODUCE FLEXI3ILITY IN REPAYMENT TERMS BY INTRODUCING A SYSTEM OF GRADED INSTALMENTS IN PLACE OF AEQUATED INSTALMENT. Annex I Page 11 - 32 - ZZZ. IN PARA 59. IT HAS BEEN OBSERVED THAT DISBURSEMENT MECHANISM IS COMPLICATED AND CUMBERSOME UPTO ARDC LEVEL. THE BASE FOR THIS OBSERVATION IS NOT CLEAR, THE FORMS FOR DISBURSEMENTS ARE PREPARED BY ARDC. IT IS CORRECT, HOWEVER, THAT THE DRAWAL OF REFINANCE FROM ARDC BY THE PARTICIPATING BANKS IS AT TIMES DELAYED AS THEY HAVE TO MEET THE REQUIREMENTS OF FUND POSITION IN INDIVIDUAL BANKS. THIS TREND, HOWEVER, IS BEING WITH A VIEW TO DRAW -REFINANCE AGAINST THEIR DISBURSEMENTS WITHOUT DELAY. LASTLY. THE EXPRESSION USED IN PARA 60 REGARDING IRREGULAR'AND OFF-LIMITS'LOANING HAVE AN UNHAPPY CONNOTATION WHICH IS INAPPROPRIATE AS THIS TYPE OF LOANING CONFORM ALL SUBSTANTIAL REQUIREMENTS OF IDA CREDITS. THE EXTENSION OF PROJECT AREAS IN DIFFERENT STATES WAS ALSO BASED ON SOUND TECHNICAL CONSIDERATIONS. IT WOULD, THEREFORE, BE MORE APPROPRIATE THAT THESE TERMS ARE SUBSTITUTED BY THE TERM SWITCH-OVER LOANING: UNQUOTE. FOREIGN COLLS Z-140 1967-77 5.12 50 PER CENT 1555 5132 2781 11934 4,80,000 3,61,000 4,79,000 5,00,000 6,52,000 MEA/NEGI 151315 PL ACK - 33 - Project Completion Report INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) I. INTRODUCTION Background 1.01 The Agricultural Refinance Corporation Credit Project (ARDC I) was the first India-wide agricultural credit project. _/ It followed a series of ten agricultural credit projects which were appraised during 1969-1973, each of which financed agricultural credit in a single State. 2/ Its main objectives were similar to those of the State-oriented projects, namely: (a) support of on-farm investments, mainly in minor irriga- tion but also for other agricultural purposes ("diver- sified" lending); (b) strengthening the long-term institutional credit struc- ture, with the Agricultural Refinance and Development Corporation (ARDC) as the refinance agency; and (c) strengthening supporting services, especially ground- water development authorities. 1.02 The typical State-oriented project financed a three to four year lending program to farmers covering investments in minor irrigation (about 80% of investments), land leveling, and farm mechanization. Each of these projects was separately appraised by IDA missions in order to determine the suitabil- ity of the proposed project area, examine the economic and financial viability of the investments to be financed and assess the capability of the participat- ing banks and supporting technical services. Most of the lending was through the Land Development Bank system (LDB) 3/ in each State but Commercial Banks (CB) also participated. ARDC refinanced loans made under the projects and supervised their implementation. 1/ It has been followed-up by a second general line of credit project (ARDC II) and a third one (ARDC III). 2/ In the States of Gujarat, Punjab, Andhra Pradesh, Haryana, Tamil Nadu, Karnataka, Maharashtra, Madhya Pradesh, Uttar Pradesh and Bihar. 3/ In this report the term LDB refers to the Land Development Bank system which consists of the State Land Development Bank (SLDB) at the apex and the Primary Land Development Banks (PLDB) or SLDB branches at the district or block level. - 34 - 1.03 The institutional developments sponsored under the State-oriented projects strengthened both ARDC and LDB. These improvements, and the close working relationship which developed between ARDC and IDA, opened the way to the line of credit type of project, under which ARDC was allotted a greater measure of independence regarding appraisal and supervision of the investment program, than under the State-oriented projects. 1.04 The preparation of this Project Completion Report (PCR) was facilitated by a completion report prepared by the ARDC. Identification and Preparation 1.05 In November 1972, an IDA Credit Review Mission evaluated IDA experi- ence with the on-going State-oriented agricultural credit projects and suggested guidelines for future Bank Group supported agricultural credit operations in India. In accordance with these guidelines, GOI requested ARDC to prepare a plan for a credit project covering the whole country in May 1973. The project preparation report was sent to IDA in November 1973. 1.06 The project was appraised during February-March 1974, and negotia- tions took place in February-March 1975. Main topics of discussion during negotiations were: (a) Amount of credit. The IDA Credit was agreed to be US$75 M, to be disbursed over two years at a rate of 55% of esti- mated ARDC disbursements during the project period. (b) Small Farmer Development Agencies (SFDA) schemes and electricity connections. IDA did not agree to include SFDA schemes under the project because it considered investments financed through them to be excessively sub- sidized. IDA also rejected inclusion of finance for electric connections because such a component had not been appraised and its inclusion could conflict with arrangements now existing and proposed IDA Credits in the power sector. (c) Groundwater overexploitation. The Indian Delegation objected to a list prepared by IDA, of districts in which ARDC was to seek detailed groundwater data before sanction- ing schemes, pointing out that in parts of many of these districts there were ample groundwater resources. It was agreed that GOI would provide a revised list, to be agreed with IDA, of districts in which such detailed data would be a prerequisite of sanctioning. (d) Overdues criteria. A formula gearing the volume of LDB lending under ARDC-supported schemes to recovery perfor- mance was agreed upon (para 1.11). It was also agreed that the same criteria would apply to lending refinanced through the Reserve Bank of India (RBI). - 35 - (e) Allocation of credit to small farmers. The Indian Dele- gation objected to allocation of a fixed portion of the credit to small farmers; it was agreed, however, that 50% should be an indicative target for the proportion of loans to be made to small farmers. 1.07 Questions raised during Board presentation in April 1975, referred mainly to ARDC's financial situation. The Credit Documents were signed on April 28, 1975 and the Credit became effective on August 5, 1975. The Project 1.08 The project was planned to support two years of ARDC lending by meet- ing 55% of ARDC disbursements during this period. Total cost was estimated at US$168.5 M of which US$154.0 (91%) was for minor irrigation, US$12.5 M (8%) for diversified investments and US$2.0 M (1%) for LDB training and for studies. At least 50% of project loans (by value) were to be made to small farmers. 1.09 The IDA Credit of US$75 M was to finance about 44% of project ccsts and cover 55% of ARDC refinance for minor irrigation and diversified lending and 50% of the cost of training and studies. The other sources of finance were to be ARDC/GOI 37%, farmers 10% and participating banks and State governments 9%. 1.10 Groundwater control arrangements which were introducci under the State-oriented credit projects were to continue, but the spacing and density formulas were simplified. Investments for additional groundwater developnant., within designated districts containing intensively developed and/or potential problem areas (para 1.06) was to be contingent upon ARDC's prior approval based on detailed survey reports. 1.11 The overdues in the LDB system were to be brought under control through a formula linking new debenture (see description of types of deben- tures in para 3.10) eligibility with recovery performance. 1/ Moreover, after October 1, 1977 ARDC was not to refinance any PLDB (or SLDB branch) if its collection plus State government contribution (limited to 10% of demand) was less than 75% of the demand. 2/ 1/ The debenture eligibility formula was as follows: Overdues on June 30 Debenture Eligibility (% of demand) (% of previous years issue) 0-25 unrestricted 26-35 80 36-45 70 46-55 60 56-60 50 61 and above nil 2/ In India, overdues at the end of any financial year are expressed as a percentage of "demand" (principal and interest falling due during the year plus overdues from previous years). - 36 - 1.12 The project included provisions for an extensive training program in order to improve the standard of operations of the participating banks. Some 750 senior and middle-level LDB officials were to be trained and a study to identify training needs of LDB junior staff was to be conducted and their training was to commence during the project. 1.13 Rates of return were not calculated specifically for the project. However, estimates of the potential financial and economic viability of the different activities to be financed were provided by investment models devel- oped under the State-oriented credit projects. These estimates ranged from 15-43% for financial rates of return and from 14-49% for economic rates of return. The increase in irrigated area was estimated at about 180,000 ha and the number of beneficiaries at over 100,000. II. PROJECT IMPLEMENTATION Cost, Finance and Disbursements 2.01 Total project cost was about Rs 1,539 M, as compared with appraisal estimate of Rs 1,350 M. A detailed comparison is shown below: Project Cost IDA Credit Appraisal Actual Appraisal Actual ----- (Rs M) ----- -----(US$)------ Minor Irrigation 1,234 1,388 69.0 '67.9 Diversified Investments 100 146 5.0 6.8 Training 16 5 1.0 0.3 Total 1,350 1,539 75.0 75.0 About 44% of the project costs was financed by IDA, 36% by ARDC, 9% by parti- cipating banks and State governments and 11% by farmers. The relative contri- bution of the various sources was about the same as estimated at appraisal (para 1.09). 2.02 Total project cost, in local currency, slightly exceeded appraisal estimates (+14%) mainly due to lower than estimated Dollar/Rupee conversion rates. The relative share of minor irrigation and diversified investments was about the same as estimated at appraisal (90% and 10% respectively). The cost of training and studies was smaller than estimated mainly because of a delay with the start of the training program (see Annex 5). 2.03 Actual disbursements were as scheduled until the end of the first half of FY77. During the second half of that year, however, disbursement accelerated and the project was fully disbursed by September 30, 1977, three months before the original closing date (Annex 3). - 37 - Minor Irrigation 2.04 About 649 minor irrigation schemes 1/ were sanctioned by ARDC under the project. average scheme involved a lending program of about Rs 5.6 M. The number of beneficiaries of the minor irrigation program is estimated at about 185,000 as compared with an appraisal estimate of 100,000. The main types of minor irrigation investments financed under the project were dugwells, pumpsets, well improvement and tubewells. Estimates of units financed, by type of investment, are shown below (details are in Annex 1): Investment Units Dugwells 43,150 Shallow Tubewells 28,500 Improvement of Existing Wells 21,300 Filter Points 3,000 Medium and Deep Tubewells 655 Augmentation Tubewells 175 Lift Irrigation 230 Pumpsets 86,480 Persian Wheels 150 Total 183,640 2.05 As agreed at negotiations, the Central Groundwater Board (CGWB) furnished a list of areas in which there was a.danger of overexploitation (para 1.06). New schemes were sanctioned only in a few of these areas after detailed investigations indicated availability of adequate groundwater resources. Although groundwater control criteria (para 1.10) were generally observed, they were not entirely effective because many minor irrigation investments are apparently financed by non-institutional sources (see PCR for Tamil Nadu and Maharashtra). 2.06 The expansion of groundwater investigations and the strengthening of the State Groundwater Departments (SGD) which were a major objective of the,various State-oriented credit projects, continued under this project. The SGD were under constant pressure to carry out detailed studies regard- ing the availability of groundwater resources, in order to facilitate the preparation of schemes, and their approval by ARDC. Diversified Investments 2.07 About 560 schemes for diversified purposes were sanctioned under the project. The average scheme covered a lending program of about Rs 1.1 M. Estimates of the number of units financed by type of investment are shown below (there are no appraisal estimates for comparison): 1/ A scheme is a small credit project which supports credit to farmers for a fixed number of specified investments (e.g, dugwells, pumpsets, etc.) within a defined geographical area. - 38- Investments Number of Units Land Development (ha) 3,400 Horticulture and Plantation (ha) 9,200 Dairy (animal) 16,700 Sheep (animal) 24,500 Poultry (bird) 167,400 Fisheries (mechanized boats) 315 Agro-service centers 70 The Lending Operation 2.08 ARDC disbursements under the project .amounted to the following: Actual LDB CB SCB Total Appraisal Estimate -------------------------- Rs M ---------------------- Minor Irrigation 883.9 240.9 - 1,124.8 1,000 Diversified Investment 40.7 61.9 2.2 104.8 72 924.6 302.8 2.2 1,229.6 1,072 The LDB accounted for about 75% of ARDC disbursement, and CB for about 25%. State Cooperative Banks (SCB) accounted for only about 0.2%. 1/ Most of the lending for diversified investments was carried out by CB which accounted for about 59% of ARDC disbursements for that purpose. The main reason for this was that many of the LDB were limited in their lending for diversified pur- poses according to their articles of association as their loans must be secured by land mortgages. 2.09 Lending to Small Farmers. Under the project, a "small farmer" was defined as a farmer cultivating land providing an annual pre-development net income of not more than Rs 2,000 (US$250) in 1972 prices. The 50% target of lending to small farmers (para 1.08) was exceeded as ARDC disbursements against loans to small farmers were 54% of total disbursements. Purpose-wise, credit to small farmers amounted to 56% of total lending for minor irrigation and 38% of lending for diversified purposes. The relatively high proportion of lending to small farmers was achieved despite the fact that schemes financed under the SFDA were excluded from the project because they included a capital subsidy component (para 1.06). The percentage of disbursement to small farmers for diversified investment was smaller than for minor irrigation for the following main reasons: (i) relatively high investment costs; (ii) higher risk; (iii) in some States (e.g. Tamil Nadu, Karnataka) loans for plantations were given only to partnerships or corporations; and (iv) the exclusion of SFDA schemes which lay special stress on diversified investments as supplementary sources of income to small farmers. 1/ ARDC refinance of SCB was limited under the project to a maximum of US$500,000 of IDA credit. - 39 - 2.10 There were considerable variations between States in the percentage of loans (by value) made to small farmers as can be seen from the following table: State % Andhra Pradesh 47 Assam 14 Bihar 74 Gujarat 15 Haryana 71 Karnataka 54 Kerala 31 Madhya Pradesh 47 Maharashtra 40 Orissa 57 Punjab 53 Rajasthan 65 Tamil Nadu 67 Uttar Pradesh 58 West Bengal 80 Country 54 The States having the smallest percentage of lending to small farmers were Assam and Gujarat (14% and 15%) while the State with the highest percentage was West Bengal (80%). In about 40% of the States disbursements to small farmers were below 50% while in 60% they exceeded 50% of Lotal disbursements. 2.11 The countrywide target (50%) is somewhat arbitrary and does not require any specific efforts from the lending institutions or State govern- ments. Moreover, conditions vary from State to State with respect to the actual distribution of small farmers, possibilities for diversified lending and other factors which affect lending to small farmers. It would be logical therefore to set up small farmer coverage targets for each State which would take into consideration its special circumstances and be determined at the time the lending program for the State is being prepared. The overall target for the country would then be the weighted average of the targets for the States. Training (Annex 5) 2.12 Some 609 LDB senior- and middle-level staff members were trained by the end of December 1977, as compared with an ARDC target of 736 (i.e. 83% of target). 547 (90%) of the 609 participated in a four week Agricultural Projects Course (APC) which is the main training course for senior and middle-level staff. The APC has been widely recognized as the centerpiece of training in agricultural credit in India. 6,.056 LDB junior staff were trained as compared with ARDC target of 8,788 (69%). A late start to training in one State (Maharashtra) accounted for much of the shortfall. Project - 40 - achievement was therefore satisfatory. The target for junior staff was based on a study of their training requiremntc conri:cted b, ARDC 1,ndpr the project (para 1.12) 1/. 2.13 The training program (especially the APC courses) has been instru- mental in spreading the production oriented approach to agricultural lending (as against the security oriented approach which was predominant in the past) among the lending institutions. The intensive training for all tiers of staff has resulted in a slow but perceptible improvement in the formulation and appraisal of schemes. Study of Cooperative Credit Systems 2.14 As agreed at negotiations, a high-level committee was appointed in September 1975 to study the feasibility of integrating the short- and long- term cooperative credit institutions. The Committee recommended 2/ that the two credit systems be integrated at all levels. Details of the recommenda- tions are in the appraisal report of ARDC II. 3/ Compliance with Covenants 2.15 All the covenants included in 'the Credit documents were complied with. III. THE FINANCING INSTITUTIONS 3.01 The main financing institutions involved in the project were the ARDC, LDB and CB. The LDB accounted for about 75% of ARDC disbursements under the project and the CB for about 25%. State Cooperative Banks (SCB) accounted for a negligible proportion (about 0.2%) of ARDC disbursement under the project. A detailed description of these institutions and their methods of operation is incorporated in recent IDA reports 4/ and is, therefore, not 1/ The study recommended that about 50% of the 17,600 junior staff of LDB, be trained in 1976 and 1977. 2/ See the "Report of the Committee on Integration of Cooperative Credit Institutions", September 1976, published by ARDC. 3/ Report No. 1520-IN, Annex 3. 4/ The most recent comprehensive description is in the Appraisal of Second Agricultural Refinance and Development Corporation Project Report No. 1520-IN dated May 12, 1977. - 41 - repeated in this report. Case.descriptions of the performance of the institu- tions in three different States are incorporated in the PCR for Maharashtra, Tamil Nadu and Andhra Pradesh agricultural credit projects. 1/ Agricultural Refinance and Development Corporation (ARDC) 3.02 The ARDC has been the refinance agency for all the agricultural credit projects and for many other IDA-supported projects. The refinance operation is basically as follows: (a) schemes (small projects involving a number of individual investments in a limited area) are submitted to ARDC by the lending banks for appraisal and approval; (b) following approval, the bank is notified of the lending terms and conditions to be used under the scheme; (c) ARDC refinances a pre-determined percentage of the bank's lending under the scheme; and (d) during implementation, the scheme is supervised by ARDC. The coverage of the ARDC program is country wide. By the end of June 1977, thpre was at least one scheme in 349 of the 387 districts in the country (90%). 3.03 Appraisal. Under the project, ARDC appraised and approved some 1,209 schemes which amounted to about 47% of the total number of schemes (about 2,562) sanctioned by ARDC during the project period (FY76 and 77). The total number of schemes approved by ARDC increased from 909 in 1975/76 to 1653 in 1976/77. There was however a decline in ARDC average commitment per scheme, from Rs 3.3 M in 1975/76 to Rs 1.9 M in 1976/77. This decline has been mainly a result of the increase in the proportion of CB schemes (which are generally smaller than LDB schemes) and ARDC's efforts to reduce the geographical cover- age of individual schemes, in order to improve implementation standards. 3.04 Of the schemes approved in 1976/77, about 19% required and received technical feasibility studies prior to appraisal and 17% financial feasibil- ity studies. Both categories involved intensive field visits by ARDC staff. The remaining 64% involved only "desk appraisal" as they were mostly either $repeater' schemes or schemes similar to ongoing ones in adjacent areas; special feasibility studies were considered unnecessary in such cases. Tech- nical desk appraisal was carried out by the Technical Division of ARDC at the head office, in respect of each proposed scheme on the basis of data provided by SGDs, banks or other institutions. In fiscal year 1976/77, about 97% of the schemes sanctioned were approved by the Managing Director and 3% by ARDC 1/ Dates of PCR: Maharashtra - January 78; Tamil Nadu - October 78; and Andhra Pradesh - June 78. - 42 - Board of Directors 1/. About 11% of the schemes approved in 1976/77 consti- tuted single loans to individuals or Corporations e.g. tea or coffee planta- tions, processing plants, forestry; ARDC commitments under these schemes amounted to about 9% of its new commitments during that year. 3.05 Monitoring and Evaluation. The level of ARDC supervision was satis- factory. During the project implementation period ARDC conducted about 300 supervisions of schemes. During a supervision, lending bank branches imple- menting the scheme are visited, loan records are inspected, and beneficiaries are interviewed. It is estimated that about 6,000 beneficiaries were visited in the course of ARDC supervisions conducted during the project period, about 3% of total beneficiaries. 3.06 Under the evaluation program, four evaluation studies were com- pleted during the project period 2/, and four studies started during the project which will be completed in the implementation period of ARDC II. 3/ ARDC has assisted participating banks in setting up monitoring and evaluation systems and has conducted three short training courses for CB on project monitoring and evaluation. The evaluation cell of ARDC has helped to design and carry out farm benefits surveys, in conjunction with preparation for PCR in the State-oriented credit projects in Tamil Nadu and Andhra Pradesh, and conducted such a survey in Maharashtra. It also assisted with the preparation of completion reports for those projects. While the time devoted to monitoring and supervision by ARDC during the project was satisfactory, ARDC did not commit adequate resources to evaluation. Head office staff especially was insufficient for developing evaluation and monitoring methodology and most of the regional offices did not have an agricultural economist to initiate and coordinate evaluation studies in the State. However, the evaluation system is now being strengthened through recruitment of senior staff to head office and agricultural economists to each regional office (under provisions made in ARDC II). 3.07 Training. The study of junior staff training requirements and the establishment of training programs for banks' staff (para 2.12) were well handled by the training cell. The training cell was not, however, adequately staffed to monitor the junior staff training program and organized only one of the five planned workshops for trainers of junior staff (Annex 5, para 7). 3.08 Recruitment. Presently, ARDC staff is seconded from RBI and ARDC has only limited control of the quality of this staff. This arrangement may have been adequate in earlier years when ARDC was a relatively small institu- tion. However, with the rapid expansion of ARDC, it is essential that it has freedom to hire its own staff. 1/ The Managing Director may approve schemes requiring refinance of Rs 5 M or less, while schemes exceeding this sum require Board Approval. 2/ Minor irrigation (two studies) and land development (two studies): the studies cover investments prior to ARDC 1. 3/ The studies which are being conducted, cover the following investments: citrus (1), dairy (2), and fisheries (1). - 43 - 3.09 Financial Aspects. Total refinance disbursed during the project period amounted to Rs 3,920 M. IDA (through GOI) was the source of about 37% of ARDCs finance and the remaining 63% was financed through local resources (details are in Annex 2, Table 3). ARDC's share capital and reserves increased (85%) from Rs 227 M at 6/30/75 to Rs 421 M at 6/30/77 when the project ended. Its operating results were satisfactory; administrative costs n 1976/77 amounted to a very reasonable 1.1% of disbursements (0.8% of commitments), and profits before tax amounted to Rs 58.5 M in 1975/76 and Rs 78.5 M in 1976/77. ARDC accounts are regularly audited by independent GOI auditors and, in both years, received unqualified reports. The Land Development Banks 3.10 The long term cooperative credit system is two tiered with the State Land Development Banks (SLDB) at State level and Primary Land Development Banks (PLDB) or branches of the SLDB at district or taluk level. 1/ The main sources of SLDB funds are: (i) ordinary debentures; and (iiT special debentures. Issue of ordinary debentures is regulated by RBI. The special debentures are purchased by ARDC and the State government concerned to re- finance lending under the ARDC program, and are regulated by the ARDC. The bulk of LDB lending is now financed through special debentures. 3.11 The perennial problem of LDB has been high overdues which, at the beginning of the project (1974/75) reached a level of 40% or more in nine out of 17 LDB (Annex 2, Table 7). The main reasons for overdues vary from state to state. 2/ However, the most important characteristics of LDB with good recovery performance (e.g. Haryana, Andhra Pradesh) are a strong LDB management, State government support of loan recovery efforts and minimal political interference in LDB banking operations. Since in most states the State government either nominates LDB top management or influences its selection, government policy has a considerable effect on the success or failure of recovery operations. 3.12 In order to control overdues, a formula gearing the issuance of special debentures to recovery performance was introduced (para 1.11). 3/ Under the formula a PLDB (or SLDB branch) was not eligible for ARDC refinance unless it had a recovery rate of 40% or more, while those having a recovery rate of 75% or more were unrestricted. For recovery rates between 40% and 1/ There are slight differences in the structure of the LDB system in the various States. In most States (e.g., Andhra Pradesh, Tamil Nadu) the system consists of SLDB as the apex bank of independent PLDB (federal system); in seven States/Union territories (e.g., Maharashtra) it con- sists of SLDB and its branches (unitary system) and in two States it is,partially federal and partially unitary. 2/ A detailed discussion of the reasons for overdues is in the PCR for Maharashtra Agricultural Credit Project (dated January 1978). 3/ For the purpose of this formula, recovery percentage was calculated as the proportion of the actual recovery plus State government contribution to share capital (limited to 10% of demand), of total demand. - 44 - 75% the eligible lending program increased proportionally with the increase in recovery. The same formula was adapted by RBI for the issuance of ordinary debentures. 3.13 During the first project year (1975/76) there was an improvement in recovery and the average rate of overdues for LDB throughout the country decreased from 39% to 35%. This improvement may have been a result of the application of the above formula but could have also been affected by the exceptional harvest in that year. In the second project year (1976/77) recovery rates declined and the average rate of overdues increased to about 40%. Moreover, the number of PLDB/SLDB branches not eligible for any lending increased from 82 in 1975/76 to 242 in 1976/77 (+195%) while the number of those entitled to unrestricted lending declined during the same period from 1,135 to 756 (-33%). The increase in the level of overdues is probably explained by the effect of political intervention in collections which accompanied the national and State elections. The effectiveness of the formula as a means for improving recovery rates is still unproven, although the general impression is that it is an important control measure. Various aspects of the formula are discussed in paras 5.10-5.14. 3.14 Financial viability. In 1975/76, all the SLDB (except SLDB in Tripura) showed profits, but about 33% of the PLDB recorded losses as did many SLDB branches in unitary-structured States. The profitability of a PLDB depends mainly on administrative efficien,v and the volume of lending and therefore on the recovery performance (which affects the volume of lending). The volume of lending largely depends on the demand for investments in minor irrigation. Therefore, the proportion of non-viable PLDB/SLDB branches can be expected to further increase in the future as investments in minor irrigation reach saturation and, as a result, the volume of lending for minor irrigation declines. This trend may be offset in two main ways: (i) greater diversifi- cation of LDB lending program; and (ii) amalgamation of long- and short-term credit operations (para 2.14). 1/ Under the project IDA did not require that SLDB and PLDB, participating in the project be viable, in order not to con- strain lending for development purposes. However, given the expansion of CB network of branches lending to agriculture, this approach should be recon- sidered. Commercial Banks 3.15 Some 33 CB participated in ARDC's lending program during the project period. ARDC total disbursements to CB in 1975/76 and 1976/77 amounted to about Rs 1,638 M (US$190 M). About 50% of the disbursements were made to the State Bank of India (SBI) and its subsidiaries (Annex 2, Table 9). The SBI has dedicated considerable technical resources to agricultural lending, has gained a considerable experience, and is ahead of most CB in this field. 1/ But not necessarily through a merger of the short and long-term coopera- tive credit institutions into one; allowing each of them to extend long and short-term credit may also be a satisfactory solution. - 45 - 3.16 CB participation in the project exceeded appraisal estimates 1/ principally due to the following reasons: (i) RBI instructed CB to increase their lending to priority sectors (which include agriculture); (ii) under the State-oriented credit projects CB gained experience and strengthened their agri-cultural staff and that facilitated expansion; (iii) the attractiveness of the relatively cheap funds provided under the ARDC program; and (iv) progressive implementation of Talwar Committee recommendations 2/ in several States that removed many fiscal or procedural obstacles to CB lending. 3.17 There are no satisfactory statistics on CB overdues related to agricultural credit but the limited data available indicate that they are high. 3/ As agricultural lending forms a small share of CB lending, they have been better able to withstand arrears than the LDB, whose total lending is to agriculture. However, in the long run a failure to achieve satisfactory recovery rates is bound to reduce CB interest in further expanding their agricultural lending. 3.18 The volume of ARDC refinance to CB has not been linked with their recovery performance as in the case of LDB and SCB for two main reasons: (i) overdues on term agricultural loans have little effect on their financial situation; and (ii) CB experience in this area began only in 1968, and bad loans made in the first few years made them hesitant to expand their agri- cultural activities. At the time the project was planned, any restrictions imposed would have slowed down exnansion of CB agricultural lending. With CB presently accounting for nearly 50% of ARDC disbursements, and given future prospects of further expansion, the situation has changed and now is the time to initiate steps aimed at improving CB recovery performance. For example, a minimum recovery requirement set up at a level which every branch with an average level of management can achieve (e.g. 40%) is unlikely to restrict lending but will cause CB to maintain and provide records to prove eligibility. Such information is required before further measures can be recommended. Short Term Credit 3.19 The cooperative credit system (CCS) is the major source of short term institutional credit to farmers in India. It consists of three tiers: Primary Agricultural Cooperative Credit Societies (PACS) at the village level, District Cooperative Banks (DCB) at district level and the State Cooperative 1/ The Appraisal Report estimated that CB share in ARDC refinance would increase from 10% at appraisal to about 17% in 1978. The actual share of CB was about 42.% already in 1976/77. 2/ The 'Talwar Committee' (nominated by RBI) recommended to provide CB the same privileges as cooperative banks (LDB, SCB) in matters such as exemp- tion from stamp duty, registration charges, etc. 3/ The method of calculation of overdues is not standard for all CB and is in many cases different from that used by LDB, which makes comparisons difficult. Recently ARDC has conducted seminars on overdues among CB in order to standardize reporting. - 46 - Bank (SCB) at the State level. As with the LDB system, the main problems of the short term system are: (i) high overdues; and (ii) a large number of non-viable primary units (PACS). 3.20 Involvement of the CCS in the project was mainly indirect through the provision of short-term credit to the LDB borrowers (LDB provide only long and medium term loans). At appraisal, it was considered important that farmers be able to obtain long and short-term credit from the same organi- zation and a study of the feasibility of integrating short and long-term cooperative credit operations was conducted under the project (para 2.14). 1/ Future Prospects 3.21 The share of CB in the ARDC program is likely to increase further as can be seen in the following table: ARDC Commitment During the Year /a 1974/75 1975/76 1976/77 Rs M % Rs M % Rs M % SLDB 1,146.9 56.0 1,766.2 59.5 1,408.8 45.9 CB 874.1 42.9 1,194.5 40.2 1,561.1 50.8 SCB 22.9 1.1 8.4 0.3 101.6 3.3 Total 2,043.9 100.0 2,969.1 100.0 3,071.5 100.0 Number of Schemes 623 909 1,653 /a Source - ARDC Annual Reports. The proportion of ARDC commitment through the CB increased from about 40% in 1975/76 to 51% in 1976/77. At the same time commitments through SLDB declined in both relative and absolute terms, probably as a result of the fact that in some areas (e.g. Tamil Nadu) investment in minor irrigation has reached its peak and has started declining. The indications are that a new era, in which CB will account for most of ARDC refinance, is due to begin. It is therefore imperative to give a greater attention to CB and to closely follow-up the expansion of their agricultural credit operations. 1/ This problem is peculiar to the cooperative credit systems; the CB provide their borrowers with both short- and long-term credit. - 47 - IV. ECONOMIC IMPACT 4.01 Investments of the type financed under the project require four years or more in order to reach full production stage. Because the project duration was only two years hardly any of the project investments have reached full pro- duction. Completion estimates of economic input are based on results from completion reviews of three State-oriented credit projects undertaken by IDA, ARDC evaluation surveys and calculations prepared in the appraisal of ARDC II. Rates of Return and Farmers' Benefits 4.02 Rates of return were not calculated specifically for this project at appraisal (para 1.13) because of its wide coverage and the diversity of cost and benefits for even the same type of investment. Models presented in the appraisal report were taken from appraisal reports of other IDA projects which included similar investments (mostly State-oriented credit projects). The main purpose of these models was to illustrate the order of magnitude of incremental benefits and rates of returns expected under the project. 1/ A similar approach was used in this review, but estimates were based, wherever possible, on factual data. The following table summarizes the results obtained and shows the source of data: 1/ The appraisal. models are not suitable, therefore, for comparison of planned and actual results; such a comparison is meaningful only if both planned and actual models cover the same type of investment, in the same area. The data available at completion was mostly for areas different from those covered by the appraisal models. - 48 - Investment Type and State FRR ERR ---------(%)------------ Dugwell and Pumpset: Maharashtra /a 11-36 18-Uver 50 Tamil Nadu /b 16 26 Andhra Pradesh /c 16 36 Well Improvement: Maharashtra /a 32 45 Tamil Nadu /b 24 41 Pumpset only: Maharashtra /a over 50 over 50 Tamil Nadu /b 33 29 Andhra Pradesh /c 49 over 50 Land Development: Maharashtra /a over 50 over 50 Tamil Nadu lb over 50 over 50 Andhra Pradesh /d over 50 over 50 Karnataka /d over 50 over 50 Dairy /e over 50 39 Mechanized Fishing Vessel /e 33 39 /a Maharashtra PCR. /b Tamil Nadu PCR. /c Andhra Pradesh PCR. /d ARDC evaluation studies. /e ARDC II appraisal report. 4.03 As can be seen, the financial rates of return (FRR) are generally satisfactory; in the models presented they range from 11% to over 50% for minor irrigation and from 29% to over 50% for land development. For diversified investments empirical data are not yet available and estimates presented were taken from the appraisal report of ARDC II. 1/ Estimates of economic rates of return (ERR) range from 18'/. to over 50% for minor irrigation, over 50% for land development and 39% for diversified investments. 2/ Production Impact 4.04 Most of the investments financed under the project were in minor irrigation (90%, see para 2.02). This type of investment was also the focus 1/ Data based on survey should be available shortly from surveys which are being conducted by ARDC (para 3.06) and under the IDA financed Gujarat Marine Fisheries Project (Loan 1394T/Cr. 695-IN). 2/ Appraisal estimates were: Minor irrigation: FRR - 18 to 43%; ERR - 26 to 49% Diversified investment: FRR - 15 to 33%; ERR - 14 to 25% - 49 - of most of the State-oriented credit projects and has had a high priority in GOI's development plans. Farmers are generally keen to invest in minor irri- gation because it enables them to intensify land use, increase the proportion of high value crops, intensify the use of inputs and provides them with a full or partial (depending on the area) protection against drought. Increases in crop intensity following investment in dugwells, as measured in a few States 1/, were as follows: Crop Intensity Percentage of Without Project With Project Increase Maharashtra 114 150 32 Tamil Nadu 91 158 74 Andhra Pradesh 106 150 42 A survey of beneficiaries under the Maharashtra Agricultural Credit project showed a 750% increase in post investment value of seed, fertilizer and insecticides used, over the pre-investment level. 4.05 The increase in the irrigated area resulting from minor irrigation investments is estimated at about 200,000 ha as compared with appraisal esti- mate of 180,000 ha. Details of the number of physical units added under the diversified investment program are in para 2.07 and Annex 1. The value of incremental production at 1976/77 financial prices is roughly estimated at 1,080 M as compared with appraisal forecast of incremental proauction valued Rs 840 M at 1976/77 prices. 2/ The higher production estimate is mainly a result of the larger area brought under irrigation. Distributional and Employment Impact 4.06 The number of beneficiaries from the minor irrigation program is estimated at 185,000 and the number of beneficiaries from the diversified lending program at about 33,000. The total number of beneficiaries exceeded 200,000 as compared with appraisal estimate of about 100,000 indicating that 1/ Based on surveys of beneficiaries under the State oriented credit projects. 2/ 'Full development' stage varies by type of investment but, on the aver- age, should be reached at about the year 1980. - 50 - appraisal estimates were in this respect conservative. 1/ As indicated above (para 2.09) about 54% of the value of loans was to small farmers. The percen- tage of small farmer beneficiaries (numbers) was even higher since small farmers' loans are generally less in terms of amount than those made to other farmers 2/. Incremental annual employment at full development stage generated by project investments is estimated at 42 M man-days. Most of the latter was generated on small farmers' farms and thus improved the employment situation there. The landless, who belong to the poorest segments of the population and are the main source of hired labor, also benefitted from the increased demand for labor generated by project investments. Sector Impact 4.07 ARDC-supported lending programs have become a major component of the institutional long- and medium-term credit for agriculture in India. In the project period (1975/76-1976/77) ARDC disbursements amounted to about Rs 3,900 M (US$453 M), compared to total of institutional agricultural credit (long- and medium-term) over the same period of Rs 12,600 M (US$1,465 M). The project (ARDC I) accounted for about 31% of the ARDC lending program (i.e. about 9% of total institutional long-term credit for agriculture). 4.08 Effect of IDA participation. The effect of IDA participation in the agricultural credit program can be divided into two partn (i) quantitative and (ii) qualitative. It would be hard to determine the increase in insti- tutional credit resulting from IDA support of agricultural credit projects (including ARDC I). The decision to support agricultural credit programs was made in anticipation of increased lending to agriculture in India. 1/ The larger estimate of beneficiaries is not necessarily an indication of a greater achievement. Firstly, because of estimation difficulties specific to this type of project which covers the entire country, data have to be collected from a large number of participating banks, there- fore, appraisal and completion estimates such as number of beneficiaries, incremental irrigated area, incremental production and employment should be regarded as indicating orders of magnitude rather than precise figures. Secondly, given the wide range of unit investment costs (a pumpset alone cost about Rs 3,500, while the cost of a dugwell with pumpset is about four times that sum), the number of units which can be financed through a fixed amount of money depends on the relative proportion of 'low cost' and 'expensive' investments which depend on farmers' preference. The overruling objective is that investments financed be financially and economically viable. 2/ For example, in Tamil Nadu Agricultural Credit Project the average loan for dugwell and pumpset given to large farmers (Rs 7,600) was 501 higher than the average loan given to small farmers (Rs 5,050). - 51 - 4.09 The main effect of IDA participation was on the qualitative side, i.e. improvement in the quality of lending, increased targeting of credit towards small farmers, strengthening of credit and groundwater institutions, and introduction of a measure of groundwater control. Especially important was the change in the approach to lending from a security oriented to a pro- duction oriented approach. This change was facilitated by the introduction of the incremental net return criterion which started under the State-oriented projects and continued under this project, and by the intensive training pro- gram (2.12). It can be concluded that without IDA participation the pace of those improvements would have been considerably slower. V. SPECIAL ISSUES The 'Line of Credit' Approach 5.01 The project, with its national scope and line of credit nature, involved a major departure from the State-oriented Lype of project suppDrted by IDA in the past. Advantages and disadvantages of the new style of project, as compared with the State-oriented project, are reviewed below. The main advantages are: (a) Greater coverage. This type of project covers the whole country and any worthwhile scheme conforming with IDA- approved criteria can be included. This enables a greater coverage than in the period prior to this project when many schemes had been too small to justify individual IDA ap- praisal had to be excluded. A related advantage is that because the program is implemented through relatively small schemes, the time lag from identification to imple- mentation is shorter than in the case of large projects. (b) Strengthening of ARDC. The greater measure of independent decision making awarded to ARDC under the project increased its influence in the various States, and, at the same time, induced a further strengthening of ARDC. (c) Manpower savings. The 'line of credit' type of project results in a considerable saving of manpower to IDA in preparation appraisal and supervision. (d) Flexible targets. Under new style project indicative targets were set up for each State. However, these were not binding; thus shortfalls in lending occurring in cer- tain States could be covered by increased lending in other - 52 - States where demand for credit for approved investments existed as well as the capacity of local credit institu- tions to serve this demand. Under the State-oriented projects the States were under some pressure to complete the financial program even if demand and/or capability of lending institutions were less than originally estimated because otherwise India stood to lose the unutilized part of IDA funds. This may have resulted in pressures to reduce the standard of loan appraisals, to include investments which did not conform with appraisal criteria, as well as the retro- active financing of investments. Such pressures are avoided under the national type of project. 5.02 The main disadvantages of the line of credit approach are: (a) The terms and conditions of loans are standardized for the country as a whole and the possibilities of taking into account the special conditions and requirements of each State are limited. In particular, this con- straint adversely affects the less developed States, drought-prone areas, tribal areas and other areas whose special circumstances cannot adequately be addressed within a standardized framework. (b) Smaller leverage-at State level. In the State-oriented credit projects the respective State government and tne SLDB each have had a separate Credit Agreement with IDA; this incorporated State specific covenants such as estab- lishment of a groundwater board, strengthening of LDB staff, etc. Such State-IDA agreements are not feasible under the line of credit approach. 5.03 On balance, the advantages of the new style project appear to far outweigh the disadvantages. The Small Farmer Definition 5.04 The project aimed at directing at least 50% of the credit to small farmers (para 1.08). The objective was to reach the poorest segments of the farming population. However, while the term small farmer is usually taken as synonymous with that of a poor farmer this is not the case where the farmer has off-farm income or where he cultivates high value crops which can generate a relatively high income from a small area. Considering the large number of borrowers involved, and data and administrative constraints, any small farmer definition for credit projects has to compromise elements of accuracy for administrative feasibility. Definitions which have been used by GOI and in IDA financed credit projects are reviewed below. 5.05 SFDA definition. Under GOI's Small Farmers Development Agencies (SFDA) a small farmer is defined as a cultivator having 2.0 ha or less of - 53 - non-irrigated land or 1.0 ha (or less) of irrigated land. The definition does not take into account the differences in cropping patterns and produc- tivity between different parts of the country. 5.06 IDA/ARDC definition under the State-oriented credit projects. Under the State-oriented projects, a small farmer was defined as a cultivator whose potential post development income is Rs 2,400 or less. Since most investments in minor irrigation increased the net income beyond the Rs 2,400, only a small proportion of the farmers qualified (e.g. 10% in Maharashtra, 21% in Tamil Nadu). Subsequently, the definition was changed under ARDC I to a pre-development one (see below). 5.07 IDA/ARDC definition since ARDC I. Under ARDC I, a small farmer was defined as a farmer cultivating land which provides an annual pre-development net return not exceeding Rs 2,000 in 1972 prices. The adjustment to current year prices is through the use of the Consumer Price Index for Agricultural Laborers for the State in which the land is located. The application of the definition is through calculation of net income per acre norms for each dis- trict. The upper area limit of the small farmer category is then determined through dividing Rs 2,000 (adjusted by price index) by the normative net in- come per acre. The upper area limit varies from place to place. It has been as low as 0.8 ha in Kerala where high value crops can be grown, and as high as 6 ha in Madhya Pradesh, where land productivity is low. 5.08 Discussion. A comparison between the upper area limit (on a rainfed basis) of the small farmer category according to the three defini- tions as applied to one of the regions in Maharashtra is presented below: Definition Upper Limit (ha) SFDA 2.0 State Oriented credit projects 1.5 ARDC I 4.7 Although in a few areas the upper area limits of the ARDC I definition may be lower than those used under SFDA and the State-oriented projects, in most cases they are higher. While ARDC I definition enabled a wider category of farmers to benefit from small farmers' concessions, there were no provisions to ensure that the lower brackets of this category obtain an adequate share of the loans. This has been corrected to some extent by the inclusion of SFDA farmers under ARDC II, since SFDA farmers consist largely of the lower brackets of the IDA/ARDC definition. 5.09 The exclusion of off-farm income is a deficiency of all the above small farmer definitions. The reason is a practical one; it is extremely difficult to obtain reliable data on off-farm income, therefore its inclusion may open the ways to subjective judgments by banks' staff and may result in exclusion of some small farmers as a result of an incorrect estimate of off- farm income. This appears a less justifiable risk than that of including a few, strictly speaking, ineligible farmers by ignoring this off-farm income. The Debenture Eligibility Formula 5.10 Under the State-oriented credit projects, it was stipulated that loan recovery plus any State government contribution to share capital was to be at least 75% of the repayment demand in order to qualify a PLDB/SLDB branch for"participation in the project. There were no limits on the amount of the State government contribution. Under ARDC I the following changes were intro- duced: (a) PLDB/SLDB branches were allowed to participate even if the percentage of recovery plus government contribution was below 75% of the demand provided that it was higher than 40%; (b) State government contributions were limited to 10% of the demand; (c) lending eligibility increased proportionally with recovery (see para 3.12); and (d) the above arrangements would apply to both ordinary and special debentures. 5.11 As the viability of a PLDB/SLDB branch depends largely on the volume of lending, the eligibility formula gave banks a strong incentive to improve recovery. Moreover, the limit en State government contributions ensured that improvements were achieved through actual recoveries rather than through financial manipulations 1/. The debenture eligibility formula was to be used for a limited period (initially until October 30, 1977) during which LDB were expected to improve recovery to reach a minimum recovery rate of 75% (65% if State government contributed 10% of the demand). Thereafter, PLDB/SLDB branches not achieving the above recovery rate were not to be eligible for refinance, and as this applied also to regular debentures they would not be able to do any new lending until they achieved the minimum recovery level. 5.12 While, as indicated above (para 3.13), the formula's effectiveness to control overdues has still to be proven, its development implications need a careful study. Firstly, the farmers penalized are not the defaulting ones but rather those who have not yet taken loans. It is even possible that small farmers suffer more from the application of this formula since they generally lag behind the larger and more progressive ones in the adopLion of new technologies. 1/ For example, under Maharashtra Agricultural Credit Project, the State gov- ernment transferred its contribution from SLDB branches having a small lending potential under the project (thus excluding them from a further participation in the project) to branches with a greater lending poten- tial. This way it was able to utilize all project funds without any substantial improvement in actual recovery rate. Such manipulation was not permitted under ARDC I. - 55 - 5.13 Secondly, the formula may restrain development in areas prone to natural calamities (drought) more than in others, because recovery rates in the less affected areas are generally better. It has been assumed that rescheduling of loan repayments will generally take place following a natural calamity 1/. However, some LDB (e.g. Andhra Pradesh, Rajasthan) object to such practice, claiming that rescheduling raises farmers' expectations for further rephasing of loan repayments and has a bad effect on credit discipline. Experience in Maharashtra and Tamil Nadu, where rescheduling failed to improve actual loan recovery supports the above argument. 2/ 5.14 In order to avoid a conflict between recovery and development objec- tives it is essential to accelerate the creation of a competitive institutional credit system in all parts of the country. In such a system a CB or DCB would be able to replace a poorly managed PLDB/SLDB branch (and vice versa) and thus the enforcement of credit discipline, which is essential for successful credit operations, would have a minimal effect on development. Until such a system is applied, the present debenture eligibility formula adjusted from time to time should remain in force. 3/ VI. IDA'S PERFORMANCE Appraisal 6.01 By the time the project was appraised none of the State-oriented credit projects had been completed, but the lessons learned during their imple- mentation were utilized in the planning of ARDC I. The success of efforts to strcngthen the indigenous refinance institution (ARDC) was recognized by allowing it greater independence. On the whole, the appraisal report pres- ented a balanced development program with only a few deficiencies reviewed below. 6.02 Although it was recognized at appraisal that effectiveness of ground- water control through credit institutions is limited, no measures for progress towards generalized controls were included under the project. Besides polit- ical difficulties, the lack of reliable data to quantify the consequences of 1/ This can be regarded as a common banking practice if done on a case by case basis; however, in India rescheduling by LDB has been done en bloc, and covered all the borrowers within areas declared as affected, because a case by case investigation is difficult. 2/ Although in those specific cases (a) rescheduling involved delaying pay- ments without increasing the overall maturity period thus increasing repayment burden in the years following the resheduling; and (b) the poli- tical climate was such that credit discipline would have broken down with or without rescheduling. (See details in the PCR for Maharasthra Agricul- tural Credit Project). 3/ The overdues criteria for regulating LDB lending are reviewed period- ically by a Standing Committee on Debenture Norms in which RBI, GOI, LDB and ARDC are represented. - 56 - the absence of effective groundwater controls, has been often mentioned as a main obstacle to legislation. An inclusion of studies and surveys to generate such data could have facilitated persuasion of State governments of the importance of legislation and might have opened the way to progress in follow- up projects. This has been corrected in ARDC II project under whicii a study of private investment in overdeveloped areas is being conducted. 6.03 A study of ways to give farmers access to short- and long-term credit from one source, of whatever kind, would have been more appropriate than a study of a formal merger of the cooperative credit systems. The merger is just one administrative mechanism for achieving this aim, and not neces- sarily the most suitable one in every State. For example, another possibility would be to enable LDB to provide short-term loans to its long-term borrowers. 6.04 The project was to be the first in a series of general line of credit projects. The project investment period was only two years, a period which is too short to complete improvements in fields such as groundwater control, institutions, etc. Therefore, project design may have benefitted had the project been planned as the first phase of an agreed (at negotiations) long term (say six years) multiphased program. This would have facilitated a gradual but consistent progress and inclusion in each phase (project) of measures which would facilitate achievement of objectives planned for future phases. Su ervision 6.05 IDA's supervision level was adequate. Two supervision missions were sent specifically to review progress of this project. A mid-term review of project progress was conducted in the course of the appraisal of ARDC II (see below). In addition, every supervision of a State-oriented credit project included a review of the activities of the respective ARDC regional office and was therefore also a supervision of ARDC I at the level of that State. Design of Follow-Up Projects 6.06 An appraisal of a follow-up line of credit (ARDC II) took place at about mid-term of this project. The appraisal involved an indepth review of all aspects of ARDC activities and the experience gained was utilized for improving the design of ARDC II. The findings made in the course of prepara- tion of PCR for this project and for completed State-oriented credit projects, were utilized for improving the design of the third line of credit project (ARDC III). Specifically, IDA has begun a dialogue with GOI aimed at reviewing measures of groundwater control during ARDC III. VII. CONCLUSIONS AND RECOMMENDATIONS 7.01 Physical achievements. The project achieved its physical objec- tives and, although the investments financed have not yet reached full pro- duction stage, indications are that the incremental production generated by the project will meet and exceed appraisal expectations. - 57 - 7.02 Transition from the State-oriented to the national line of credit type of project. Transition from the State-oriented type of project was smooth and successful; project funds were utilized to finance approved invest- ments and the standard of appraisal and supervision by the national refinance agency, ARDC, was satisfactory. 7.03 ARDC. The key to the good performance achieved under the project was ARDC. Its performance under the project was good. It strictly enforced project terms and conditions and acquired an influential position in the various States. However, the fast expansion of its operations has not been sufficiently supported by an adequate strengthening of staff. Recommendations: (a) A continuous adjustment of ARDC organization, staffing and approach to changing situations; special attention to strengthening of the evaluation staff (para 3.06) and the Training Cell (para 3.07); (b) ARDC should control recruitment of its staff (para 3.09). 7.04 The LDB system. The high level of overdues continues to be a serious problem as is the high proportion of non-viable PLDB/SLDB branches. A reduc- tion in the level of overdues requires strong action by the State governments and LDB managements concerned. The impression is that the debenture eligi- bility formula helps to control overdues.. However, due to the short period in which it has been applied and the special events (elections) which took place during this period, it is too early to arrive at definite conclusions regarding its effectivenes-s. 1/ Improvement of the viability of PLDB will require an increase in the volume of lending per PLDB through a reduction in the number of branches, better management and greater diversification of leading operations. Recommendations: (a) The debentures eligibility formula should remain in force, amended as necessary (para 5.14) until sufficient data to examine its effectiveness is available (about 2-3 years). (b) The effect of the formula on development, especially in less developed regions should be studied (para 5.13). (c) Exclusion of non-viable PLDB from IDA supported program should be considered (para 3.14). (d) LDB should diversify lending operations (para 3.14). (e) The progress towards merging short and long- term credit operations should be followed up (paras 2.13, 3.14). (f) Steps should be taken to minimize political interference in LDB banking operations (para 3.11). 7.05 Commercial Banks. Expansion of CB lending has been beyond expec- tations and they are likely to soon replace the LDB as the main channel for ARDC funds. Data on CB activities is inadequate particularly with regard to loan recovery rates. Recommendations: (a) CB should be required to improve data on loan recovery (para 3.17). (b) A minimum recovery require- ment should be imposed on CB (para 3.18). 7.06 Small farmers. The present definition is administratively manage- able and although covering a wide category of farmers, the lower brackets of that category should be adequately covered because SFDA farmers (who mostly belong to the lower brackets) are now financed under the line of credit projects. 1/ Under the terms of ARDC II the formula remains effective until October 31, 1978. - 58 - Overall coverage of small farmers depends on efforts made at State level. Conditions differ between States and it would be desirable therefore to determine a specific small farmers' coverage target for each State. Recommendation: In future projects, overall small farmers coverage target should be based on individual State targets (para 2.11). 7.07 Future line of credits should include an agreed long-term (say six years) multiphased program of institutional development (para 6.04). £ana~LcUrAcg, IuCE NbORATION CDr rOECT (ARDC I) Minor Irrigation Investments Financed Under the Croject Due11e Dug-cn. Dug-wII Shallov Bamboo Meamt/ Persian Develop- Angmant- Lift irri- Titav PSa~ Z9ing of 9t~s borwelle with pmup- tuibevels Shallov deep hla sent of a~ton gation points al~.. pipeIM: get@ Tubewelle tubewells exigting tuaffe1l1 nts surfaco wells a ~nder 2 3 4 5 6 7 8 9 10 .13 2 14 --------------------------- (units)--------------------------------------------------------- EW,rya - - 1,100 5,600 - 100 - - 150 - - - - ?radeh - - - - - - - - - - - . Punjab - - - 2,750 - 100 - - -- - - - .. Rajasthan 3,040 - - 100 - - 100 600 - - - 2,300 3~5,00 sa- - - 50 50 - - - - - - - - 1,210 - - 7,050 - - - - 25 - - 5,00 - O~esa 1930 - - 100 -100 - 1,750 - ~ast Beaul 130 - - 1,150 - 5 - - - - - 300 22,000 N~6ya treaid 3,570 - - 300 - - 50 750 - 100 - 5,900 30,000 Uttcr Praes~ - 2,710 - 9,480 - - - - - - - 1,830 - - - - - - - - - - - - - - Oujarat - - 2,000 - - 50 - - - 10 - 1,000 33,000 Nahrashtra 15,970 - - - - - - 3,000 - 20 - 16,350 - Anara Prtdesh 6,810 150 - 50 - 300 - 7,500 - - 200 32,800 - Karnataka 2.310 400 - - - - - 2,800 - - - 11,500 - Kerala 200 - - - - - - 50 - - 25 50 - Tamil Nadu 5,90 7700 1-800 6.600 - _-_ 2.775 7.10- Total 29,100 3,250 10,800 28,430 50 655 150 21,300 175 230 3,000 86,48 500,000 5 AGCULTU.AL REFIRNC CORPORATION CMCDIT PRJECT (ARDC 1) Diversified InvestmenS Fnanced Under the Project Land Agro Dair ee2 Poultry Fisheries Horti- Cf g0gg 0bftieulture a Develop- service (Asi- (Animals) (Birds) (Boats culture Plantation combine State ment Centte mal@) Trawlers) (Acres) (Uits) (Acres) (Acres) (Are) (Acres) (Asses) (Acres) 1 2 3 4 5 6 7 8 910 1 12 13 Rayans - - 105 - - - - - - - - - Kimachal 4/ Pradesh - - - - - 50 - * - - - Punjab 1,700 - - * - - - - - - - - Rajasthan - - 820 - - * - - - - * * Asaam . . - - - - 490 - * * Tripura - - - - - - 95 * * Bihar - - - - * - Orissa1,800 - - - - -- West Bengal -;S 3-4 - -M-- S Madhya Pradeh - - - - * Uttar Pradesh - - 2,560 - - -1/ God - - - - 1,410 SW- - - - - - - Oujarat 15 3,020 - - Maharashtra - 6,355 - 60,000 w * Andhra Pradesh 4,200 - 3,560 24,530 58.335 - - - - .15 Karnataka 700 - 40 - 25,425 11 3/ lo - 375 110 SA,Si - Kerala - - 130 - 70 - - - - - Tamil Nadu - 100 - 2 - - Total 8,400 70 16,720 24,530 167,390 315 220 930 375 110 14,410 6,725 Conversion: 1 be - 2.47 acres 11 Trawlers and Mechanised boats 4/ Apples Trawlers / Pineapples Mechanised boat 6/ grapes - 61 - ANNEX 1 Table 3 INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) ARDC Disbursenent by Type of Investment (Rs M) ARDC ARDC Minor Irrigation Disburserent Diversified Investment Disbursement Dugwells 307.3 Land Development 10.8 Dugwell-cum- borewells 7.4 Agro-service centres A.5 Dugwell with pump- sets 106.3 Dairy 22.8 Pumpsets alone 336.2 Sheep 1.3 Shallow tubewells 156.8 Poultry 9.0 Bamboo shallow tubewells 0.2 Fisheries 18.9 Medium/deep tubewells 39.3 Plantation and Horticulture 37.6 Persian wheels 0.2 i) Horticulture 1.1 Development of wells 86.3 ii) Tea 6.0 Augmentation wells 11.3 iii) Coffee 1.6 Lift irrigation units 50.9 iv) Rubber 0.4 Filter points 21.7 v) Coconut 22.1 Laying of vi) Plantation and pipeline 1.5 horticulture combined 6.4 Sub-total 1,125.4 Sub-total 104.9 Total ARDC disbursement Rs 1,230.4 M - 62 - -A AGRICULT L REFINANCE CORPORATION CREDIT PROJECT (ARDC I) Table 4 Estimated Total Lending and ARDC Disbursement by Category of Investment and by State Region/State Total Irrigation Diversified Purposes Total ]/ ARDC 2/ Total / ARDC 3/ ending Disbursements Lending Disbursements (Ra M) Haryana 85.7 77.1 0.3 0.2 Himachal Pradesh - - 0.1 0.1 Punjab 31.4 28.3 3.0 2.4 Rajasthan 36.2 32.6 2.5 2.0 Assam 0.7 0.6 4.6 3.7 Tripura - - 0.4 0.3 Bihar 91.4 82.3 - - Orissa 49.5 44.6 3.1 2.5 West Bengal 13.0 11.7 3.1 2.5 Madhya.Pradesh 81.2 73.1 4.5 3.6 Uttar Pradesh 106.8 96.1 3.5 2.8 Goa - - 6.0 4.8 Gujarat 39.7 35.7 7.0 5.6 Maharashtra 216.6 194.9 15.3 12.2 Andhra Pradesh 216.6 194.9 22.4 17.9 Karnataka 72.7 65.4 40.7 32.6 Verala 2.1 1.9 5.2 4.2 Tamil Nadu 206.9 186.2 9.4 7.5 Total 1,250.5 1,125.4 'd 131.1 104. 9 1/ Assumed as 90% of project cost. 4/ Actual 7/ Assumed as 81% of project cost. r/ Assumed as 72% of project cost. - 63 - ANNEX 2 INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) The Financing Institutions 1. The main financing institutions involved in the project were the ARDC, the LDB and CB. The LDB accounted for about 75% of ARDC disbursements, the CB for about 25%, and the State Cooperative Banks (SCB) accounted for a negligible proportion (about 0.2%). A detailed description of these insti- tutions and their method of operation is incorporated in various IDA reports of which the most recent one is report No. 1520-IN. 1/ This annex consists mainly of updated financial tables on the operations of the above institutions. Case descriptions of the performance of the institutions in three different States are incorporated in the recently prepared PCRs for the Maharashtra, Tamil Nadu and Andhra Pradesh agricultural credit projects. 2/ 1/ Appraisal of Second Agricultural Refinance and Development Corporation Project Report No. 1520-IN dated May 12, 1977. 2/ Dates of PCRs: Maharashtra - January 1978; Tamil Nadu - October 1978; Andhra Pradesh - June 1978. INDIA AGRICULTURAL REFINANCE CORPORATION CRET)IT PROJECT (ARDC I) ARDC - Condensed Statement of Income & Expenditure - 1971/72 - 1976-77 1971/72 1972/73 1973/74 1974/75 1975/76 1976/77 Income (Rs M) Interest earned on: i) IDA Loans/Debentures 1.0 12.0 50.9 79.4 133.2 224.4 1i) Other Loans/Debentures 57.8 76.6 99.0 132.1 154.0 175.5 Other income 1.8 3.8 5.4 9.9 11.9 9.6 Total Income 60.6 92.4 155.3. 221.4 299.1 409.5 Expenses Interest paid on: i) GOI/IDA Loans 31.2 43.7 66.3 85.1 107.6 146.5 ii) RBI Loans 0.5 6.1 15.7 30.1 48.5 75.4 iii) Bonds & Debentures 12.4 17.7 31.4 47.0 64.4 84.4 Salaries and Staff Benefits 3.9 4.9 7.1 10.1 12.6 14.7 General expenses 1.7 3.0 3.9 5.0 7.5 10.0 Total Expenses 49.7 75.4 124.4 177.3 240.6 331.0 Profit before tax 10.9 17.0 30.9 IA.1 5A.5 78.5 Ttansfer to Special Reserve 1.1 1.7 3.1 4.5 5.9 19.6 Tax 5.8 8.9 16.0 23.0 30.9 34.0 Profit after tax 4.0 6.4 11.8 16.6 21.7 24.9 Dividend 3.0 4.4 6.7 8.9 10.9 17.3 Net Surplus 1.0 2.0 5.1 7.7 10.8 7.5 INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) ARDC - Condensed Balance Sheets 1971/72 - 1976/77 1971/72 1972/73 1973/74 1974/75 1975/76 1976/77 Assets Cash on hand - 1 1 2 4 2 IDA Loans - - 43 139 535 980 IDA Debentures 54 660 1,250 1,676 2,483 3,293 Other Loans 138 205 339 492 701 988 Other Debentures 1,043 1,296 1,465 1,756 1,775 1,961 Interest accrued on loans 3 3 8 15 33 70 Interest accrued on Debentures 31 50 83 114 158 206 Other Assets 3 11 2 12 10 54 Total Assets 1,272 2,226 3,191 4,206 5,699 7,554 Liabilities & Capital Liabilities Bonds & Debentures 277 387 662 992 1,377 1,817 Deposits .10 12 14 18 23 44 * Loans from GOI: 1) IDA Loans 45 452 839 1,170 1,705 2,604 2) Other Loans 727 796 796 796 796 796 Loans from RBI: 1) Long Term 50 345 540 882 1,384 1,726 2) Short Term 34 37 116 45 17 Other Liabilities 25 39 59 76 103 146 Total Liabilities 1,168 2,068 3,026 3,979 5,405 7,133 Capital Paid up Shares 100 150 150 200 250 350 Reserves & Undistributed Income 4 8 15 27 44 71 Tbtal Capital 104 158 165 227 294 421 Total Liabilities & Capital 1,272 2,226 3,191 ,206 5,699 7,554 [p INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) ARDC - Cash Flow, 1971/72 - 1976/77 1971/72 1972/73 1973/74 1974/75 1975/76 1976/77 Resources: GOI/IDA 203.8 477.2 386.5 331.2 534.7 915.2 RBI/Bonds 166.4 447.0 621.0 775.0 1,002.0 940.0 Total Borrowings 370.2 924.2 1,007.5 1,106.2 1,536.7 1,855.2 Repayments from Borrowers 5.0 14.2 42.3 92.7 245.9 480.5 Share Capital 50.0 50.0 --- 50.- 50.0 100.0 a Sub Total 425.2 988.4 1,049.8 1,248.0 1,832.6 2,435.7 Accretion to Reserves 3.3 5.5 9.2 16.1 21.8 '33.4 Total Cash Inflow 428.5 993.9 1,059.0 1,265.0 1,854.4 2,469.1 Disbursements: IDA Schemes 54.5 636.2 563.5 618.7 1,323.3 1,466.2 Non-IDA Schemes 295.3 305.2 414.9 445.3 388.2 743.2 Total Schemes 349.8 941.4 978.4 1,064.0 1,711.5 2.209.4 Pcpa,ments To: GOI --- --- 1.4 --- 0.9 Bonds --- --- RM7 75.2 38.9 72.0 174.0 143.0 175.0 Total Cash Outflow 425.0 980.3 1,051.8 1,238.0 1,854.5 2,385.3 > Excess (+) or Deficit (-) of cash inflow over outflow +3.5 +13.6 +7.2 +27.0 (-0.1) +83.8 - 67 - Annex 2 Table 4 INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) Follow-up (supervision) Studies Conducted in the years 1975/76 & 1976/77 July 1975- July 1976- State June 1976 June 1977 Total Andhra Pradesh 22 29 51 Assam - 2 2 Bihar 11 26 37 Goa Gujarat 3 3 6 Haryana 18 18 36 Himachal Pradesh N.A. 1.A. N.A. Karnataka 13 17 30 Kerala - Madhya Pradesh N.A. N.A. N.A. Maharashtra - 4 4 Orissa 6 14 20 Punjab 9 15 24 Rajasthan 10 10 20 Tripura N.A. N.A. N.A. Tamil Nadu 14 22 36 Uttar Pradesh 17 13 30 West Bengal N.A. N.A. N.A. 123 173 296 N.A. - not available -68- AM .EX 2 Table 5 INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC 1) Estimated State-wise Percentage of Small Farmer Holdings1/ Income Ran-" nf Typical Airr'ximate State equivalent acreage acreage percentage of at 1974 norms norms small farmers2/ prices (Upper limit) within the limit shown in Rs coluin 4 (1) (2) (3) (4) (5) 1. Andra Pradesh 2,900 8.25-14.50 11.50 87 2. Madhya Pradesh 3,200 13.50-15.00 14.00 81 3. Assam 3,000 6.00-10.00 8.00 91 4. Punjab 2,800 7.00-10.00 7.00 70 5. Haryana 2,800 7.50-11.00 9.00 78 6. Maharashtra 2,900 6.50-19.00 12.00 72 7. Rajasthan 3,300 11.00-41.50 15.00 76 8. Bihar 3,600 7.20-8.10 8.10 89 9. Tamil Nadu 3,500 3.10-10.00 7.70 89 10. Orissa 2,800 2.50-13.00 5.50 78 11. West Bengal 3,000 3.50-5.00 4.50 78 12. Karnataka 3,100 6.00-15.00 13.00 82 13. Kerala 3,000 2.00 only 2.00 82 14. Uttar Pradesh 3,300 7.00-9.00 8.00 92 15. Gujarat 2,800 5.00-12.00 7.50 56 1/o estimate roughly the percentage of small farmers in each State, the definition of small farmer under ARC Credit Project has been applied. Based on this definition, the income equivalents at 1974 prices (column 2) have been converted into acreage limits for various agro-climatic regions/districts in the States. The intra-State range of upper limit on land holdings of small farmers is indicated in column 3. Within this range, a typical acreage norm has been chosen which represents the -upper acreage limit for small farmers in most of the districts/agro-climatic regions (column 4). These typical acreage norms have been made use of to estimate the percentage of small farmers to the total holdings in the State as given in the All-India Report on Agricultural Census, 1970-71. However, it may be mentioned that the size classes in the Census do not agree precisely with the estimated upper acreage limits. 2/The holding of many of the small farmers included under this column is too small for making them potential borrowers under ARDC program. If the lower area limit of the small farmer category (determined by the requirement that investments financed be financially viable) is considered, then the size of the small farmer category, which is relevant as a target group for ARDC program, is reduced considerably. For example, in the case of Tamil Nadu about 40% of the small farmers holdings (Col. 5) were of 0.5 ha or less (generally too small for minor irrigation investments) and the holdings of 64% were 1.0 ha or less (based-on the Agricultural Census, 1970-71). Annex 2 INDIA Table 6 AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) ORGANIZATION CHART AS ON 30, NOVEMBER 1977 HEAO OFFICE CAIRMAN DEmry MANAGING DIRECTOR C0 SEcto~ SENIOR SENIOR SENIOR OfRECTO-t DiRECTOR 81RECIOR OIRECToR CCCT-TR ArNG . MAL-O 1 3 AND PUONNG RAJASTMAN A S FNING D R CT OR -I-ECTO ------ - OIRECTO ERS GO C10TA DRECTOR :CCOUNT3 DRECTR1RECTC EVAtuTron C0 PRADESM OMATION ORISSA A.. SES ----- CTOR GUJARAT DRCTORDRECTOR PROGRAMMG RCTR CORVAT BONAR ERRy E NT " PURJAR RARRATARA E CTOR .... RCO DIRECTO SURVES RCTOR FIE ARTANA KERALA "UALWACUT £»RECTOR ACA ----A CA TSL 9 GArl,HYCRERAB&D,MJPJR,LLICXMOW.~RARASK~W SCLXI.-TM^tl-J%U], CORSUII'@RECTORT - 70 - Table 7 1A AGRICULTURAL REFINASCE CORPOAnl_ CRET PRJECT (ARDC I) State Land Development Bagts & Primary gnd Development Banks - SumaJ, of overdue-z- 1971172 - 9711177 State SLDB 1971/2 1972/3 1973/4 1974/5 1975/6 1976/72/ Primary Andhra SLDB 7.5 36.7 14.2 9.2 8.7 9.6 Pradesh Primaries 11.2 19.6 23.3 22.0 25.2 15.7 Assam BIDS 60.0 71.4 62.3 65.0 40.3 43.7 Primaries 66.7 76.3 72.0 72.0 63.0 44.6 Bihar SLDB 39.1 42.2 41.6 33.7 34.8 41.9 Gujarat SLDB 25.4 57.1 47.5 72.4 56.3 61.9 Haryana SLDB nil nil nil nil nil nil Primaries 0.4 1.1 0.6 1.9' 2.1 5.0 Rimachal Pradesh SLDB 68.0 71.1 56.1 58.5 50.4 63.3 Jammu & Kashmir SLDB 27.3 27.6 35.8 33.3 38.1 50.5 Karnataka SLDB 27.7 25.6 22.7 20.2 25.5 40.2 Primaries 43.0 42.0 39.3 40.1 41.9 54.6 Kerala SLDB 26.6 27.5 39.2 20.6 12.3 5 7 Primaries 33.3 34.4 45.8 29.1 20.7-I 20.8 Madhya SLDB 37.2 20.6 39.5 24.4 22.5 45.0 Pradesh Primaries 48.7 38.3 50.1 47.0 41.0 50.0 Maharashtra SLDB 32.1 83.3 44.5 58.6 60.5 62.7 Primaries 39.2 1/ Orissa SLDB 24.1 42.2 50.8 57.4 NA 60.0 Primaries 54.7 48.9 54.2 56.9 32.0 49.7 Punjab SLDB nil NA 7.2 1.4 2.7 9.5 Primaries 2.7 10.8 11.5 17.0 20.3 22.1 Rajasthan SLDB 32.4 47.9 50.6 45.4 20.0 27.7 Primaries 47.4 47.9 48.7 36.6 26.7 35.4 Tamil SLDB 2.2 5.7 5.1 9.8 17.7 36.0 Nadu Primaries 18.8 19.1 20.0 40.0 36.3 51.1 Utter Pradesh SLDB 12.7 25.7 21.0 25.4 17.02' 23.9 West Bengal SLDB 32.8 28.8 24.9 20.0 0.6 13.9 Primaries 64.0 NA 43.8 29.5 21.3 17.5 1YExpressed as percentage of demand. (Principal and interest falling due during the year plus overdues from previous years). 2Provisional S.Converted into unitary structure since 1972/73. Source: Statistical statements relating to cooperative movement in India (ACD, Rhll) - 71 - INDIA ANNEX 2 Table 9 AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) State Cooperative Banks Summary of Overdues 1971/72 - 1975/76 State 1971/72 1972/73 1973/74 1974/75 1975/76 --- --(--%)------M-------------------------------- Andhra Pradesh 5.0 6.0 1.5 2.0 N/A Assam 92.0 93.0 97.1 98.9 94.1 Bihar N/A 29.0 29.0 37.0 N/A Gujarat Nil Nil Nil Nil 4.0, Haryana Nil Nil Nil Nil Nil Himachal Pradesh 52.0 72.0 67.5 68.0 N/A Jammu & Kashmir 10.0 19.0 23.4 23.0 N/A Karnataka 2.0 1.0 0.6 1.3 N/A Kerala Nil 1.0 0.1 0.05 Nil Madhya Pradesh 4.0 5.0 4.5 1.0 5.4 Maharashtra 1.0 5.0 9.6 3.0 2.0 Orissa Nil 7.0 3.3 2.6 0.5 Punjab Nil Nil 0.1 N/A Nil Rajasthan 32.0 12.0 2.4 1.0 0.3 Tamil Nadu Nil Nil 0.2 013 0.6 Uttar Pradesh 5.0 4.0 4.3 3.1 Nil West Bengal 59.0 35.0 20.7 6.1 5.2 Others 17.4 38.2 45.5 30.0 70.0 *Expressed as percentage of demand **Provisional - 72 - ANNEX 2 Takle 9 INDIA ARICULTURAL REFINANCE CORFORATION CREDIT PROJECT (ARDC I) Bankwise Position of ARDC Disbursements to Commercial Banks in in 1975/76 and 1976/77 Name of the bank Amount 1975/76 1976/77 A. State Bank of India & Subsidiaries (Rs M) 1. State Bank of India 281.3 359.4 2. State Bank of Bikaner & Jaipur 14.8 21.4 3. State Bank of Hyderabad 15.7 19.0 4. State Bank of Indore 8.5 19.8 5. State Bank of Mysore 6.7 4.6 6. State Bank of Patiala 27.2 27.4 7. State Bank of Saurashtra 3.3 3.0 8. State Bank of Travancore 0.7 0.3 Total 358.2 454.9 (A) B. Nationalised Commercial Banks 1. Allahabad Bank 8.3 12.1 2. Bank of Baroda 18.5 54.6 3. Bank of India 63.4 36.3 4. Bank of Maharashtra 6.1 8.9 5. Canara Bank 14.0 26.3 6. Central Bank of India 69.2 73.6 7. Dena Bank 3.5 7.1 8. Indian Bank 0.5 3.6 9. Indian Overseas Bank 0.2 3.3 10. Punjab National Bank 36.9 42.4 11. Syndicate Bank 29.0 24.6 12. Union Bank of India 34.5 57.9 13. United Bank of India 4.2 20.2 14. United Commercial Bank 24.4 42.8 Total 312.7 413.7 (B)___ C. Others 1. Andhra Bank Ltd. 1.9 11.8 2. Karnataka Bank Ltd. 0.2 0.9 3. Punjab & Sind Bank Ltd. 23.6 32.6 4. New Bank of India Ltd. 7.4 10.7 5. Sangli Bank Ltd. 1.0 1.0 6. Vijaya Bank Ltd. 0.7 1.9 7. South Indian Bank Ltd. - 0.1 8. Corporation Bank Ltd. 0.5 1.0 9. Vyasa Bank Ltd. 0.8 0.4 10. Bank of Madura Ltd. 0.5 0.7 11. Bank of Rajasthan Ltd. - 0.7 Total 36.6 61.8 (C) Grand Total (A)+(B)+(C) 707.5 930.4 MRWfITIAMt urvINAMT NIrmATI,ln CVnI,T PRfIFi ARN' T1 C.enduepof ScheJuled Ct~teal häsnko 1973-1976 8tate/Upnion åt9 e Sector Sanks Omor Indian schedled A11 Indian Scheduled C.eretal 1urtory C~eret Ranks4nke ~ormenta: of överduac ffi m o urftag* of overdues to J n" JunJ Junam "-. June Jun Jun June. .12 2i 1,75 1,76 2l fÅ7 175 1976 19u3 "12 1475 1476 S3 4 5 6 7 § 9 10 11 12 13 1. NORME REGION Rärymna 37.8 30.9 31.3 32.4 45.7 43.6 43.9 43.4 40.0 33.2 33.8 33.6 i=mal Pradech 42.4 39.7 43.3 48.3 • 12.9 33.3 42.4 39.6 42.9 44.2 Jamma and Kauir 40.1 30.0 33.2 59.3 33.3 13.6 49.7 36.9 37.9 26.3 33.0 37.7 pumjah 44.4 42.3 36.8 31.0 26.8 27.6 26.3 24.6 41.9 40.0 33.2 30.1 Rja«han 36.2 33.0 33.5 49.9 32.0 30.8 53.4 45.3 36.0 32.9 33.3 40.7 Cenath 33.2 39.9 66.1 67.8 48.1 13.8 38.8 3.6 32.6 37.3 64.5 37.5 Delhi 56.4 61.9 64.7 56.8 36.1 63.3 73.3 89.8 31.3 62.4 66.0 62.2 För the Rögton 48.6 46.2 43.7 41.2 38.3 39.0 40.2 33.9 47.1 43.4 44.3 40.7 II. 302f1 LASTERN REIO Asa 71.2 81.9 74.6 69.8 - - 30.0 93.0 71.2 81.9 74.6 69.8 ~ipur 67.4 59.2 60.4 64.9 - - 96.7 59.2 60.4 64.9 maghalaya 81.3 52.3 67.4 70.9 • • • 81.5 32.3 67.4 70.9 Naaland • - 87.0 18.9 - • • - • 82.0 18.9 Tripura 87.0 36.6 48.1 45.3 - - - 87.0 36.6 48.1 45.3 Asunach&l Prådesh • - - - Hizoram • • • • - - 7o the Region 72.3 74.1 .67.7 64.0 - - 30.0 93.0 72.3 74.1 67.7 64.1 11I. STERN REGION 3Ihar 63.8 63.3 59.8 34.3 - • 60.0 27.3 63.8 63.5 59.8 54.3 Orlsa 36.0 49.3 50.7 57.1 77.5 63.2 91.3 85.9 58.9 30.5 32,2 58.2 Vet sengal 78.3 75.1 68.4 66.3 - 11.1 26.1 78.3 73.1 68.4 66.3 A~i and Nicobar Ibland - - 14.3 - - • • • 14.3 - For the Region 70.3 67.4 62.6 60.1 77.3 65.2 90.7 84.2 70.4 67.4 62.8 60.2 ITV. CENTRAL REGIDN ~adya Pradeab 38.4 56.6 55.0 54.2 67.4 11.0 12.9 36.6 38.3 36.3 54.9 34.2 Uttar Pradesh 43.0 44.4 47.7 42.0 43.2 54.8 51.2 50.7 43.0 44.6 47.8 42.2 Por the Region 48.1 48.4 50.2 46.2 48.4 53.6 50.0 50.3 48.1 48.4 50.2 46.3 V. WESTERN R:CION ajarat 51.9 48.0 58.0 58.6 - - - - 61.9 48.0 52.0 58.6 Maharaahtra 64.2 67.5 61.7 60.7 59.7 42.1 56.2 54.8 63.9 66.5 61.6 60.5 doa, Dnan *and Diu 24.1 37.4 27.7 35.1 •- - 37.2 37.4' 27.7 35.1 Dera and Nagar Revalt - - 26.5 52.3 - - - - 26.3 52.3 ror the Region 60.1 60.3 60.3 59.7 59.7 42.0 56.2 54.8 60.1 59.8 60.2 59.6 V%. SOUDIERN RECION Adra Pradeah 45.8 39.8 33.0 34.1 32.6 29.0 35.4 38.0 42.8 38.2 33.3 34.7 Karnataka 46.8 49.3 46.1 46.1 34.4 41.1 44.4 43.9 43.1 48.3 46.0 46.1 Kerala 40.0 37.3 34.5 37.2 19.2 12.3 15.5 16.5 29.8 26.3 25.2 28.7 7Gmal Nadu 41.8 43.7 49.0 49.7 10.3 19.2 26.5 45.1 34.8 37.8 45.3 48.9 Pondichery 15.9 28.2 36.0 27.7 - 13.0 20.0 84.1 15.9 28.1 36.0 27.8 Lkähadeep• 3.3 • • • 3.3 loö the Region 44.2 43.1 39.9 41.4 26.2 23.0 28.3 34.8 39.7 39.1 38.1 49.4 All Indla 52.0 51.8 49.8 49.1 30.7 27.1 32.4 36.9 49.1 49.3 48.4 48.1 On all1iult &.l d 1.e. abort and [on-term. loa... - 74 - ANNEX 4 Page 1 INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC 1) Summary Findings of Evaluation Studies Conducted During the Project 1/ 1. The ARDC had recognized the need for establishing a system of eval- uating, on a continuous basis, the financial and economic benefits from projects assisted by it and for this purpose set up an Evaluation Cell in January 1974. The evaluation studies undertaken by the Corporation seek to assess the economic benefits accruing from investments refinanced so as to compare earlier expectations with actual achievements, particularly at the farmers' level. 2. In the first phase, the Corporation conducted the following four evaluation studies, the reports on which have now been published: (a) Minor irrigation scheme - Construction of New Wells and installation of Pumpsets thereon in Sholapur District - MaharashLra (1969/70-1972/73); (b) Minor irrigation scheme - Installation of shallow tubewells in Karnal District, Haryana (1967/68-1971/72); (c) Bhadra Land Development Project - Scheme for Reclamation and Development of Land - Karnataka (1966-1972); (d) Land Development under Nagarjunasagar Project, Miryalguda Taluka - Andhra Pradesh (1964-1969). 3. These studies undertaken for the first time are in the nature of pilot studies and will help in evolving appropriate techniques in sampling and evaluation methodology. The techniques will have to be suitably modified on the basis of experience gained during the initial studies to suit different projects financed by ARDC. It is hoped that these studies will be found useful and will provide a broad framework in evolving a methodology for evaluating benefits from agricultural development schemes. Brief summaries of the above four study reports are given below: A. Minor irrigation scheme - Construction of new wells and installition of pumpsets in Sholapur District - Maharashtra (1969/70 1972/73). 1/ Printed verbatim from the completion report prepared by the ARDC. - 75 - ANNEX 4 Page 2 4. The scheme was implemented by the Maharashtra State Cooperative Land Development Bank (SLDB) during the years 1969/70-1972/73. The scheme ( originally envisaged long-term finance to agriculturists in four talukas of Sholapur district, viz. Sangola, Akkalkot, North Sholapur and South Sholapur, for construction of 330 new wells, renovation of 780 old wells and purchase of 1,110 pumpsets for installation on these wells. The financial outlay worked out at Rs 6.93 M. By the close of the scheme, loans amounting to Rs 4.15 M had been advanced to cultivators for construction of 308 new wells, renovation of 472 old wells and purchase of 480 pumpsets for instal- lation on 129 new wells and 351 renovated wells. 5. The scope of the evaluation study was restricted to 129 cultivators in the 69 villages of the four talukas, who availed themselves of composite loans under the scheme for construction of new wells and installation of pumpsets thereon. 6. A two-stage random sampling design was adopted with village as the first stage unit and cultivator as the second stage unit.. A sample of 22 villages was selected and all the 59 borrower beneficiaries were selected for the field study. For the purpose of reflecting the position of borrower beneficiaries in the pre-investment period a control sample of 45 rainfed area cultivators was selected. A sample of 16 non-borrower beneficiary cultivators with similar investment was also selected for comparative analysis. A general schedule in four parts was canvassed among the selected cultivators and farm data and other particulars were collected ior the reference year July 1, 1973 to June 30, 1974. 7. The data emerging from the study and the results therefrom are sub- ject to limitations such as the pilot nature of the study and after effects of the drought spell in the earlier three consecutive years in the reference year. 8. In the course of the field investigations, it was found that of the 59 beneficiaries selected for the study, 20 loans (34%) did nou appear to have been utilized for the purpose for which they were availed of. Hence, the results of the study given below relate to the 39 borrower beneficiaries who had undertaken the investment: (a) The total cultivated holding of the average borrower beneficiary was 7.13 ha. The area of the plot in which the well was situated was 2.53 ha of which 2.0 ha was irrigated by the well during the reference year as against an area of 3.24 ha assumed to be irrigated by the well in the Economics of the Scheme. 1/ 1/ The economics of a scheme are calculated by ARDC at the time the scheme is appraised. - 76 - ANNEX 4 Page 3 (b) The cropping intensity on the benefited area worked out to 114% for borrower beneficiaries and 177% for non- borrower beneficiaries, as against 188% assumed in the Economics of the Scheme. (c) Incremental income per ha of net cropped area worked out to Rs 1,284 for the borrower beneficiaries and as much as Rs 5,457 for non-borrower beneficiaries. (d) The net present worth, benefit-cost ratio and internal rate of return worked out on the basis of the data collected to assess the financial feasibility of the investment showed that the investment was worthwhile. (e) The repayment capacity worked out after making allowance for increased consumption was Rs 2,047 for the average borrower beneficiary. Though the annual installment towards LDB loan was Rs 1,284 the amount actually repaid by the average borrower beneficiary during the reference year was only Rs 558. Earlier drought period appeared to have affected the actual repayment performance of the borrower beneficiaries. (f) From the data collected on employment it was found that an estimated 1,700 man days of employment was created in constructing a new well and a further 151 man-days per ha of benefited area or about 300 man-days per investment was created per year from increased farm activities. (g) The total cost of the composite investment, in the case of the average borrower beneficiary, worked out to Rs 9,066 of which about 93% was financed through borrowings from the LDB under the scheme. (h) The data on time taken at various stages on disbursement of the loan showed that the LDB took, on an average, more than 5 months from the date of application to sanction of the loan, about 3 months more for disbursing the first install- ment from the date of sanctioning the loan and about 15 months more for disbursing the last installment thereafter. B. Minor irrigation scheme - Installation of shallow tubewells in Karnal District - Haryana (1967/68-1971/72). 9. The minor irrigation scheme sanctioned by ARDC in September 1967, and rephased in November 1970, provided for the installation of 3,600 shallow tubewells and relevant water channels in the eight blocks of Karnal, Gurgaon - 77 - ANNEX 4 Page 4 and Ambala districts. The total financial outlay involved was Rs 2.70 M with ARDC's refinance at Rs 2.43 M. 10. The physical targets with regard to the installation of tubewells were fulfilled but there was a shortfall in the construction of water channels. In financial terms, an amount of Rs 2.57 M was disbursed by the Haryana State Land Mortgage Bank with ARDC's refinance at Rs 2.32 M. 11. The evaluation study was confined to the investment program taken up in the Karnal district, which covered a major portion of the scheme area. A sample of 118 borrowers was selected from 28 villages for the field inves- tigation. 12. The main findings of the study are given below: (a) The work of installation of tubewells and construction of water channels was left entirely to the initiative, skill and convenience of the beneficiaries who obtained guidance from knowledgeable persons and local ministries in carrying out the work. (b) All the selected respondents reported inspection of the investment at various stages by the PLDB staff. (c) Of the 118 respondents selected, 73 had installed electric motors and the remaining 45 had installed diesel engines. About three-fourths of electrically operated tubewells were reported to have been energized after a time lag of 3 months. (d) Of the 118 selected beneficiaries studies, 20 did not avail themselves of the loans for the construction of water channels; while another 22 appeared to have not constructed the water channels even though they had drawn the first of the two installments of the loan amount. The average cost of water channels actually incurred by the 76 beneficiaries fell con- siderably short of the loan amount of Rs 2,000 advanced for the purpose. Some of the beneficiaries were reported to have diverted a part of the loan amount given for construc- tion of water channels to other items of work such as land development and switch room, which were not included in the anticipated cost of investment under the scheme. (e) The study revealed that the area benefited by the invest- ment was 3.60 ha for an average borrower beneficary, slightly less than 4.05 ha anticipated in the scheme. (f) The cropping intensity in the benefited area at 179% after investment was about the same as anticipated (180%) under the scheme. - 78 - ANNEX 4 Page 5 (g) Wheat, paddy, maize, sugarcane and fodder were the prin- cipal crops grown by the beneficiaries before as well as after investment. A large portion of the total benefited area (69%) was used for the cultivation of wheat and paddy after investment, as compared with the position before investment (45%). (h) The financial rate of return worked out to more than 50%, indicating that the investment was highly profitable to the beneficiaries. (i) The employment of labor involved in the construction of 2,553 tubewells in the scheme area covered by the evalua- tion study is estimated to be about 1.2 M man-days. The additional employment opportunities for agricultural labor arising from intensive cultivation practices were esti- mated at 96 man-days per ha of benefited area and about to 8.79 M man-days per annum in the scheme area under study as a whole. C. Land development project - Scheme for reclamation and devel- opment of land - Karnataka (1966-1972). 13. The scheme for reclamation and development of land under Bhadra project sanctioned in November 1965 and rephased in November 1966 and August 1970, envisaged development of 20,730 ha of land in 6 talukas of Shimoga, Chitradurga and Bellary districts in Karnataka State. The scheme area was very largely drought-prone before the construction of the Bhadra dam and advent of irrigation facilities. The total financial outlay involved under the scheme was Rs 12.36 M with ARDC commitment at Rs 92.7 M. 14. As against an area of 20,730 ha proposed to be developed, an area uf 20,235 acres was actually developed. Total disbursements under the scheme amounted to Rs 12.11 M. 15. The evaluation study was confined to the program of land development in Harihar and Davangere talukas of Chitradurga district, which accounted for 70% of the total area developed under the scheme. 16. For the purpose of the study a sample of 78 borrower beneficiaries was drawn from 10 selected villages. With a view to obtaining the pre- development position of borrower beneficiaries, a control sample of 50 cultivators with only rainfed area was selected. A sample of 40 non-borrower beneficiaries who had undertaken the investment on their own was also selected for comparative analysis. 17. Some of the important findings of the study are as under: (a) As many as 35 of the 78 selected borrower beneficiaries had taken 5 months or more for development of land. - 79 - ANNEX 4 Page 6 (b) In the absence of a systematic survey of the slope of land in the command area, the slope of half the lands was assumed to be below 1% and the other half between 1% to 2%. However, most of the lands fell between 1% to 2% slope and a majority of the cultivators who dev- eloped their lands did not depend on bulldozers or tractors. (c) The scale of finance for land development originally fixed at the inception of the scheme was found to be inadequate and appeared to have affected the pace of development. The scale of finance was, no doubt, revised later but by that time, the bulk of the land development was already over. (d) The 'contour borderstrip' method of land development recommended for the area was not adhered to and most of the cultivators at the head reaches of the canals developed their lands in the traditional 'compartment' method, fit for paddy cultivation. As a result, tail-end cultivators were not getting adequate water for cultivation. (e) The field channels in the individual fields were not cleared of weeds and silt in many cases, while most of the respondents had not provided for adequate drainage channels. (f) The operational efficiency of PLDB working in the scheme areas was unsatisfactory during the scheme period due to organizational inadequacies. They were also burdened with mounting overdues. (g) Under the scheme, an average beneficiary developed an area of 3.07 ha of his cultivated holding as against 2.73 ha assumed in the Economics of the Scheme. (h) The cropping pattern, as proposed by the State Government on the developed area, was that on lands at the head reaches of the canal only light irrigated crops like ragi, jowar and groundnut should be raised and on the low lying areas, heavy irrigated crops like paddy and sugarcane should be raised. However, it was actually found that-paddy was grown as kharif as well as summer crop, wherever possible. (i) The cropping intensity on the benefited area worked out to 176%. The average yield rate per ha of paddy grown by the borrower beneficiaries was around 3.0 tons as compared with 2.5 tons assumed in the Scheme. As assumed at 1.5 tons per ha of groundnut grown, the actual vield worked out to 1.9 tons per ha. - 80 - ANNEX 4 Page 7 (j) The incremental income per ha of net cropped developed area worked out to Rs 2,584 and the repayment capacity of the beneficiary cultivator was found to be nearly 20 times the annual equated instalment of the loan amount taken from the PLDB for the investment. (k) The financial rate of return worked out to more than 50%, thereby indicating high profitability of the investment. (1) The process of land development in the scheme area under study generated employment opportunities for an estimated 3.8 M man-days. The additional employment opportunities for agricultural labor in the selected scheme area arising from intensive cultivation prac- tices were estimated at 744 man-days per annum per average borrower beneficiary and at about 3.4 M man-days per annum on the whole. D. Land development under Nagarjuna Sagar Project, Miryalguda Taluka - Andhra Pradesh (1964-1969). 18. ARDC sanctioned, in December 1964, the scheme submitted by the Andhra Pradesh State Coopeeative Land Mortgage Bank (CLMB) envisaging long term finance to cultivators in four talukas of Guntur district viz. Guntur, Narasaraopet, Palnad and Sattenapalle on the right bank and one taluka viz. Miryalguda of Nalgonda district on the left bank. The area to be developed under the scheme (repnased in 1967) was about 75,700 ha at an estimated cost of Rs 62.1 M. The evaluation study was confined to Miryalguda taluka of Nalgonda district. 19. Out of 5,824 borrower beneficiaries under the scheme spread over the villages in Miryalguda talika, a sample of 76 was drawn from 8 selected villages for the study. In order to obtain the pre-development position of borrower beneficiaries, a control sample of 30 cultivators with only rainfed area was drawn. In addition, a sample of 30 non-borrower beneficiaries who had undertaken similar investment during 1964-9 without resorting to PLMB loan, was drawn for comparative analysis. 20. Some of the important findings of the study are as follows: (a) The actual cost of development of land of the average borrower beneficiary turned out to be higher than that envisaged in the scheme. The cost escalation was mainly on account of increase in wages paid and higher cost of initial manuring. The dif- ference in cost at Rs 270 per ha was met by the beneficiary himself. - 81 - ANNEX 4 Page 8 (b) The initial insistence on the use of bulldozers for levelling and the later emphasis on the use of manual and animal labor (as it was found that enough machines were not available to cope with the work) delayed development of land. Enthusiasm for land development was also dampened to some extent by o delay of 14 months in the release of canal water. (c) Not much technical assistance was available from the CLMB to cultivators for land shaping. (d) A handicap in the approach was that there was no agency like the Command Area Development Department which could attend to planning, technical guidance and if necessary, undertake execution of work on behalf of cultivators. (e) The study revealed that the area developed by the average borrower beneficiary was 3.06 ha as against 2.43 ha assumed in the Economics of the Scheme. (f) The cropping pattern as revealed by the survey data bore out the assumption that the cultivators would switch over to paddy cultivation during kharif. But during rabi the actual position was not as envisaged, the general attitude of cultivators being overwhelmingly to grow paddy instead of the recommended crops such as groundnut, wheat, ba'ra and blackgram. (g) The cropping intensity achieved by the average borrower beneficiary in the developed area was 185% as against 200% anticipated. The actual yield rate of paddy per acre was slightly higher than 3.0 tons assumed in the Economics of the Scheme. (h) The incremental income per ha of net cropped area worked out to Rs 1,899 and the repaying capacity of the average borrower beneficiary was found to be more than 13 times the annual equated instalment due to PLMB. (i) The internal rate of return worked out to more than 50% thereby indicating that the investment was worthwhile. (j) An estimated 3.0 M man-days of employment was created dur- ing land development work and 4.4 M man-days per year from intensive on-farm activities under irrigated conditions. During the busy season, an acute shortage of labor was reported by cultivators-in the scheme area and migration of laborers from non-project areas of the taluka took place to take advantage of additional jobs created because of intensive irrigated cultivation. - 82 - ANNEX 4 Page 9 (k) Because of the field-to-field system of irrigation prevalent in the area, there is need for field drains and since not much attention has been paid to this aspect, about 15% of command area had been rendered unfit for cultivation due to waterlogging and salinity conditions. (1) Excessive use of water for cultivation of two crops of paddy in the upper reaches has resulted in nearly one- tenth of the command area located at the tail-end of the main distributories of the canal not getting suffi- cient water for cultivation of even one crop of irrigated paddy. - 83 - ANNEX 5 Page 1 INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) Training 1. Under the project, ARDC had the following responsibilities with regard to training: (i) organization of training for senior and middle level LDB staff; (ii) to conduct a study to identify the training needs of LDB junior staff, (iii) to organize a training program for the LDB junior staff based on the findings of the above study and commence training. Organization 2. The following organizational arrangements were made to facilitate achievement of training objectives: (a) Committee for Training. The committee consists of 15 members representing GOI, RBI, ARDC, LDB and CB. The committee advises ARDC on various aspects of training. (b) Internal Group on Training. The group consists of representa- tives of the College of Agricultural Bank (CAB), ARDC, RBI, Maharashtra SLDB and IDBI. The group reviews the training program, training materials, etc. (c) Steering Group. This group consists of five members represent- ing ARDC, ACD, CAB and the LDB Federation. Its main functions are to plan, guide and supervise the training program of LDB junior staff. (d) Training Cell. ARDC set up a training cell at the head office. The cell is responsible for organization and monitoring of the training program. Training of Senior and Middle,Level LDB Staff 3. Actual achievements are compared below with targets set in October 1976 for the years 1976 and 1977. These targets are higher than the estimate made at appraisal. - 84 - ANNEX 5 Page 2 Actual as Target /a Actual % of Target Four-week APC /b 530 547 103 Short APC 100 9 9 Other courses 106 53 50 Total 736 609 83 /a To be achieved by December 31, 1977. 7- Agricultural Projects Course. Overall achievement was 83% of the target. However, participation in the most important course, the four-week APC, exceeded the target. Training of LDB Junior Staff 4. Of the 19 LDB in the country, 14 operate one or more training centers (Andhra Pradesh has four, Bihar two, and Gujarat two). Of the other five LDB, all but two are so small that they send their trainees to training centers of other LDB; Jammu and Kashmir and Assam are indeterminate. 5. The comprehensive study of the training needs of the junior staff of LDB (para 1) divided the juniors into the following three groups, accord- ing to their duties and responsibilities: Group I - Technical Officers, Land Valuation Officers, Agricul- tural Officers, Agricultural Engineers, Recovery Officers, Accountants, Enquiry Officers, and Internal Auditors. Group II - Sub-branch Managers, Loan Superintendents, Investi- gators, Supervisors, Legal Assistants, and Head Clerks. Group III - Assistants, Clerks, etc. (excluding Typists). The study determined that about 8,800 of the 17,600 (50%) junior staff were in "Critical Categories" and should be trained during 1976 and 1977, the remainder to be trained later. The priority trainees would be all of Group I, half of Group II, and a third of Group III. 6. By December 31, 1977, about 69% of the above targets had been achieved as can be seen from the following table: - 85 - ANNEX 5 Page 3 Actual as Targets /a Actuals /a % of target Nationwide 8,788 6,056 69 Largest LDBs: Andhra Pradesh 1,606 1,452 90 Maharashtra 1,541 209 /b 14 Bihar 768 613 80 Tamil Nadu 768 706 92 Gujarat 730 359 49 /a Targets and actuals for the years 1976 and 1977. /b Started very late (September 1977) but is running two channels for catching up. It is estimated that the training program has covered about 30% of junior staff from Group I (see classification above). 7. Workshops for trainers were to be a vital element in building high quality into the junior staff training program. The expectation was to mount a series of five workshops before the training got fully underway. Each workshop would concentrate on one feature of training - course design, teach- ing methods, or another feature with participation of 30 trainers for one week. Since 30 was a greater number than had yet been appointed as trainers, it would give everyone a taste of training problems, to be followed by other concentrations in the series. All trainers would have had from two to four of these exposures before the courses began (since then, one workshop has been held in August 1977 and five others are planned for early 1978). Because the study report was not published until April 1976 and the trainers were not appointed until later, the 1976 period when workshops could be held was already past. This period had to be in the first half of the year because the "lean period" (July-November) is the period when training is at its peak and the trainers busiest. The workshop held in August 1976 brought together 25 trainers and none was held in the first half of 1977, for reasons that are not entirely clear, but which seem to stem from lack of staff at ARDC head- quarters to organize these in addition to their more immediate duties. The effectiveness of the workshops could have been considerably increased through a greater utilization of available funds (see below). With greater use of funds for personnel, the trainers could have had more than the mere ten hours of hardcore training provided by this one-week workshop. The notable defi- ciency of good training material and the rarity of excellent instruction in the current program might have been largely avoided in many cases. - 86 - ANNEX 5 Page 4 Cost.,of Training 8. Actual cost is compared below with appraisal estimates: Estimates Actuals Actual as % of Estimate Study of Junior-level training needs 400,000 15,000 4 Training: 1,200,000 560,000 47 Senior and Middle- level /a (400,000) (320,000) /b (80) Junior-level /c (800,000) (240,000) (30) Total 1,600,000 575,000 36 /a Includes common expenses /b Subject to possible reduction, as bills were paid to CAB by ARDC at a higher level than ARDC claims is justified. /c Excludes common expenses of ARDC Training Cell, which is included in /a. 9. Cost of the study was greatly over-estimated. Good desicn and effi- cient organization by ARDC placed the most time-consuming fact-gathering work in the hands of the LDB. ARDC then collated and analyzed the comparable data. Cost of the senior and middle-level training was low because several regional APCs that ARDC expected to organize and finance were not held because ARDC could not find sites. 10. Cost of the junior-level training was considerably below appraisal estimates for three reasons: (a) Delays in completing the study and normal slippage in organizing the training centers and the courses caused the training under this program to begin in the second half of calendar 1976, which was more than halfway through the funding biennium; one major LDB (Maharashtra) did not commence training until after the close of ARDC I. (b) A few courses conducted in 1976/77 were charged to ARDC II because bills were not cleared before June 30, 1977. (c) Only one of expected five workshops for trainers was actually held (para 7). This was due to restriction of manpower at ARDC Head Office. - .87 - ANNEX 5 Page 5 Evaluation of Training 11. Senior and Middle-Level Staff. The APC has received wide recogni- tion as the centerpiece of training in agricultural credit in India. Its curriculum closely fits the needs of the LDB, who constituted the majority in each course throughout ARDC I. The course's length and detailed content could be debated, but the trainees have been generally satisfied with it. The level of trainees has been satisfactory to the trainers and virtually all graduates have returned to jobs where they use the material covered directly or indirectly and have given some type of formal instruction to others. The training is prestigious enough to develop an "esprit de corps" that increases its effectiveness in a social and administrative structure like that of India. 12. Although there is no practical way to evaluate the effectiveness of APC objectively, there is no doubt about two judgmental conclusions: (i) it has had a profound and spreading influence toward betterment of agri- cultural lending throughout the country; and (ii) it would be still more effective if teaching methods were improved without a radical change in the Indian training tradition. If morale has some relationship to institution- building, it is significant that APC graduates seem to believe they are on the right track and, therefore, presumably have more concern about doing a good job in the field. Yet only 15% (547) of total LDB senior and middle- level staff have been trained in APC. If it stops too soon, it will keep the value of exclusiveness and loqe the value of universality; on balance, near-universality would be better. 13. Progress with training at CAB is dissappointing. Lack of some combination of inclination, ingenuity and time has prevented the faculty from (i) converting many of the existing "raw" materials into more useful teaching instruments and (ii) developing new materials. Equipment for taking advantage of visual training materials has not changed noticeably since a year ago, when it could be described as primitive. Physical facilities are very good and the core faculty shows generally a high capability. 14. Junior Staff. Over-all, the junior-level training in LDB is well worth the effort and inve'tment although effectiveness is not as great as it should be. The training is being conducted for the number of trainees planned, in the local languages, and according to the basic curriculum approved by ARDC. This is being done under conditions that -in most cases are acceptable. The curriculum adapted to local needs, is suitable even though judgments could differ as to emphasis. The course of four weeks is optimum length. 15. Trainers are.well selected for their background of education and work experience; though not all have had teaching experience this is a less important qualification. The classroom ability of those observed was good and sometimes very good, when considered within the traditional Indian methods, but trainers are hampered by lack of enough first-class training materials and (to a lesser degree) equipment. They are not so pressed for time that more imaginative materials could not be developed locally with more encouragement from ARDC and more training in regional workshops. The trainers are on training duty all year, according to those sampled, and are - 88 - ANNEX 5 Page 6 not :isked to perform any other duties but preparation, instructing, and eval: ition of training. 16. Materials used for instruction are generally scarce, raw in form (untreated for classroom use), non-visual, and often of too high level to be useful enough to be translated. Some trainers regard most of the materials received from ARDC as of too high a level and/or irrelevant and others say most of it is useful. Each LDB translates a different amount of the English material received or produced. 17. Equipment for making audio-visual use of materials is almost entirely absent. This is explained by the absence of enough suitable materials to warrant ownership of some types of equipment. One type, however--overhead projector--can itself promote the generation of materials by the trainers as well as by a central agency such as ARDC or CAB. Ready availability (if not ownership) of other types of projectors can stimulate the search for local materials (slides and movies) and their central production. 18. Trainees are apparently serious and anxious to learn. They seem also to take seriously their own operating problems and those of the farmer. Many cannot concentrate maximum effort on their training, however, because of poor living arrangements. 19. Facilities as a whole--classrooms, libraries, faculty offices, and their locations--are adequate, though better in some LDB than in others. Lodging and board for trainees observed was good in one LDB, inadequate (because of location or lack of capacity) in two, and almost unacceptble because of the absence of a hostel in the other two; one of these was acquir- ing a hostel in a few days. Good living conditions make a real difference in the effectiveness of training, mainly because of time and effort saved but also because of the high value of exchange among trainees living together. Good facilities also heighten morale by giving evidence that LDB management recognizes the importance of training and trainess. 20. Present evaluation of training results, suffers somewhat from the illusion that accurate trainee opinion can be obtained through standard ques- tionnaires. Course-end evaluation by a questionnaire to trainees is useless in the form sent to ARDC. At the LDB the paper response is both supplemented and validated or invalidated by oral feedback and acquaintance with the trainee, but the summaries sent to ARDC are bound to be misleading without elucidation. 21. In summation, the training program, while highly beneficial, will not be as effective as it could be until living conditions of trainees are greatly improved in some LDB and until training materials and equipment in all LDB are 'used to stimulate learning and retention of what is basically a good curriculum, taught with traditional skill. - 89 - ANNEX 5 Page 7 Staffing 22. The Training Cell of ARDC has not had sufficient staff for per- forming all its functions. The study of Junior staff training requirement, the Senior and Middle-level training program, and the establishment of the Junior-level training program were very well handled by ARDC. Once the junior training began, however, the Training Cell was able to keep up with administrative necessities but not fully with pedagogical needs. The Training Cell should be strengthened and staffed with a few more experienced trainers. Conclusion 23. The training element in this project shows that a massive training program can be mounted and carried on with reasonable effectiveness through a decentralized, but centrally financed and monitored, method in a very large country that: (a) believes in training; and (b) has a wide base of educated people to train. Granted some slippage and overlooked opportunities, a good portion of the trainees were undoubtedly learning and retaining enough to improve substantially the performance of their agricultural credit work. It is hard to disagree with a judgment stated by ARDC in its own completion report: "With the commencement of intensive training for all tiers, [there was] slow but perceptible improvement in the formulation and appraisal of sub-projects under ARDC I." 24. The training program has helped to develop a production oriented cadre in the financing institutions. Graduates of the APC have spread the doctrine of production oriented lending (as against the security oriented lending which was prevalent in the past) and facilitated its fast spread within the participating banks. This has been a major contribution towards the institution building objective of the project. - 90 - Annex ) Table 1 INDIA AGRICULTURAL REFINANCE CORPORATION CREDIT PROJECT (ARDC I) LDB Junior Staff Training Program Up to 31 December 1977 Name of SLDB Appraisal Estimates Program achieved by SLDB ----------------(No. of Trainees)-------------------- Andhra Pradesh 1,606 1,452 Bihar 768 613 Gujarat 730 359 Haryana 285 184 Karnataka 494 408 Kerala 107 198 Madhya Pradesh 638 533 Orissa 156 203 Punjab 304 294 Rajasthan 413 194 Tamil Nadu 768 706 Uttar Pradesh 519 497 West Bengal 308 206 Maharashtra 1,541 209 Assam 40 - Jammu & Kashmir 30 - Himachal Pradesh 56 - Pondicherry 10 - Tripura 15 8,788 6,056 IBRD 0483R4 - I ~ f w. Rc i 929 DEMOCRATiC REPU6UC OF AFGHANISAN 1 N D 1 A JAmmu dn KASHMIR r -' Sr nagor- ® St.te and Union Teritoy Capitols sf. ,'~n~* National Capitai H A H C Other Cines PRADESH Stote and Unon Terv.y Bondones m.- International Boundaries PA KISTAN PUINJAB CHINA Dl D-t1, DEIl V p f W OELinnngr p ' rPR NEPAL BH w PRA DESH ý U R AUA ST HSAN --i p.r amppurr .oprrl - Kanpur 7'f Osu~~, N 1 F C50Hý41A4 > " 8 / R IBANGLADESH? s ';l GandhInag., Bh 1 oo GU A-d.eT A BEGA/ MADHY RSA pRA D ES H it< > BURMA MANARASNTRA9 Bhbe Bombay F ®Nydervbed ANDH/RA A ra bian RAIAsEsh s e aGOA4 KARNyA 74A aangolore jj rAMli NADU) - -ý 00 200 'I 6 8 ML S "0 60 390 11! 9 6 NO6 0 KILOOVERS 4 SR LN ÑKA raye