India - Tarai Seeds Project

Document of The World Bank F FILE PYr FOR OFFICIAL USE ONLY Report No. 2551 PROJECT PERFORMANCE AUDIT REPORT INDIA TARAI SEEDS PROJECT (LOAN 614-IN) June 18, 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. ACRONYMS ARDC - Agricultural Refinance and Development Corporation ETO - Exploratory Tubewell Organization GOI - Government of India HYV - High Yielding Varieties NSC - National Seeds Corporation NSP - National Seeds Program SBI - State Bank of India SEB - National Electricity Board SPP - Seed Processing Plant TDC - Tarai Development Corporation UCB - United Commercial Bank UP - Uttar Pradesh UPAU - Uttar Pradesh Agricultural University GLOSSARY Kharif - Wet Season (May - October) Rabi - Dry Season (November - April) FOR OFFICIAL USE ONLY Project Performance Audit Report INDIA TARAI SEEDS PROJECT (Loan 614-IN) TABLE OF CONTENTS Page Preface i Basic Data Sheet ii Disbursement Table iii Highlights iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. Project Summary I II. Main Issues 4 New Construction versus Expansion of Existing 4 Facilities Overdesign of Irrigation Component 5 Income Distribution Effects of the Project 5 Coherence with Follow-on Projects 6 PROJECT COMPLETION REPORT I. Background 7 II. Project Processing 8 III. Implementation 13 IV. Project Impact 26 V. Institutional Aspects 31 VI. Special Issues 34 VII. Bank Performance 37 VIII. Conclusions 38 ANNEXES 1 - Itemized Farm Machinery Expenditures 39 2 - Cash Flow and Financial Rate of Return for 40 100 Acre Farm Model 3 - Economic Return Calculation Based On Change 41 In Income To Seed Growers 4 - Economic Return Calculation Based On Value Added 42 from Incremental Foodgrain Production Map 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 TARAI SEEDS PROJECT (Loan 614-IN) PREFACE This report presents the results of a performance audit of the Tarai Seeds Project (the Project), which was supported by a Bank loan (Loan 614-IN) of US$13 million, effective in September 1969, of which US$1.55 million was cancelled at closing, December 1977, as it remained unutilized. The audit was based on interviews with Bank staff involved in appraisal and supervision of the Project (and to a lesser extent of the two follow-on projects), and on a reading of the Project Completion Report (PCR) prepared by the Bank's South Asia Projects Department, the appraisal and supervision reports, correspondence files, legal and other relevant project documents, and appraisal reports of the two follow-on projects. In accordance with normal procedures, the draft audit report was sent to the Borrower for comments on February 16, 1979; however, none were received. The audit finds the PCR a comprehensive and accurate document and concurs with its main findings. The Audit Memorandum however expands the discussion on project design and highlights the income distribution effects of the Project and its impact on follow-on projects.  - ii - PROJECT PERFORMANCE AUDIT BASIC DATA SHEET INDIA TARAI SEEDS PROJECT (LOAN 614-IN) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Estimate Total Project Cost (US$ million) 22.4 21.8 Loan Amount (US$ million) Disbursed ). 13.0 11.5 Cancelled - 1.5 Repaid ) 0.0 Outstanding )- 11.5 Date for Completion of Physical Components 12/74 9/78/1 Proportion of Time Overrun ) - 50 Economic Rate of Return (M) without downstream benefits 17 18 with downstream benefits - 42 OTHER PROJECT DATA Original Actual or Item Plan Revisions Current Estimate First Mention in Files 06/31/64 Government's Application 06/19/67 Negot iat ions 11/12/68 Board Approval 05/28/69 Loan Agreement Date 06/18/69 Loan Effectiveness Date 09/12/69 Closing Date 12/31/74 12/31/76 12/31/77 12/31/77 Final Disbursement Date 02/24/78 Borrower India Executing Agency Government of India Fiscal Year of Borrower April 1 - March 31 Follow-on Projects National Seed Project, Loan 1273-IN, of U1S025.0 M, signed 06/18/76, Second National Seed Project, Credit 816-IN, of US$16.0 M, signed 07/17/78 MISSION DATA Month, No. of No. of Date of Item Year Weeks Persons Manweeks Report Identification 11/66 Preparation 02/67 Appraisal 05/68 3 6 18 05/23/69 Total 108 Supervision 1 02/70 2 2 4 03/31/70 Supervision Il 07/70 (2 days) I - 07/30/70 Supervision 111 03/71 2 3 6 04/27/71 Supervision IV 11/71 1 1 1 12/31/71 Supervision V 07/72 1 1 1 09/13/72 Supervision VI 02/73 1 1 1 02/08/73 Supervision VII 09/73 1 1 1 12/14/74 Supervision VIII 09/74 2 2 4 10/14/74 Supervision IX 07/75 1 2 2 07/23/75 Supervision X 02/76 1 2 2 02/27/76 Supervision X 08/76 1 1 1 09/13/76 Supervision XII 11/77 1 1 1 12/30/77 Total 24 Completion 11/77 2 1 2 Total 2 COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) Indian Rupee (Rs) Year: At appraisal (1968) Exchange Rate: US$ Rs7.50 At closing (1978) US$1 = Rs8.60  - iii - Project Performance Audit Report INDIA TARAI SEEDS PROJECT (Loan 614-IN) DISBURSEMENT TABLE IDA Appraisal Actual Semester Estimated Actual as % of Ending Disbursement- Disbursement Estimate US $ million 1971 June 7.3 2.1 29 December 2.3 1972 June 10.1 2.7 27 December 2.7 1973 June 12.4 3.0 24 December 3.1 1974 June 13.0 5.5 42 December 5.6 1975 June 5.6 43 December 5.9 1976 June 6.0 46 December 9.2 1977 June 9.3 72 December 11.4 1978 June 11.5 88 1/ Derived from "Investment Costs", Annex 9, Appraisal Report.  - iv - Project Performance Audit Report INDIA TARAI SEEDS PROJECT (Loan 614-IN) HIGHLIGHTS The project, over five years, sought to expand foodgrain seed production by about 46,000 tons, through levelling, irrigating, and mechanizing seed farms and through processing and marketing of the seed produced. The project has been generally successful in achieving its objectives, although only after considerable delays. The project is expected to produce, process and market 42,000 tons of certified seed at full development in 1981/82. Tubewells serving the target area, associ- ated power transmission lines, farm mechanization and seed processing capacity were provided largely as planned (though seed processing capacity was provided by expanding existing facilities rather than by the projected new construction). However, less than half of the target land area was levelled and none of the planned underground water distribution systems in- stalled. The quality of the seed produced and processed under the project has been outstanding. The project's reestimated economic return is slightly better than at appraisal. The principal lessons of the project, which have been applied to the two follow-on projects, were that the organizational structure and the compact-area approach proved successful. The following points may be of special interest: - Capability of two persons, in key project positions, was decisive for project success (PCR para. 5.01); - New construction of processing facilities was abandoned in favor of less costly expansion of existing facilities (PPAR paras. 10 - 14, PCR 3.07 - 3.10); - Irrigation component was apparently overdesigned (PPAR para. 15, PCR 3.27); - Lack of comprehensive farm budgets make it difficult to put the project's income distribution effects in perspective (PPAR paras. 16 and 17); - A seed demand-and-supply study in an earlier project phase would have been desirable (PPAR para. 19); - Working capital was used for investments, because of commercial credit shortage (PCR para. 6.01); and - Both Borrower and Bank were flexible in adjusting project design to new findings (PCR para. 7.02).  Project Performance Audit Memorandum INDIA TARAI SEEDS PROJECT (Loan 614-IN) I. PROJECT SUMMARY 1. The possibility of a seeds project in the Tarai area was first brought to the Bank's attention in 1965/66 through the efforts of the Uttar Pradesh Agricultural University (UPAU) and its Vice-chancellor. The project was identified and prepared by two FAO/IBRD missions in November 1966 and February 1967, with a preparation report issued in June 1967. This report was generally well received by GOI except for the organiza- tional proposals. GOI also wished to include some additional invest- ments. The project was appraised after 10 months (April 1968) and ap- praisal proposals alleviated GOI's concerns to a large extent. Sub- sequently, negotiations were held in Washington in December 1968. The project was approved in May 1969 and became effective in September of the same year. 2. The project was prepared during a time when lending to India moved slowly and IDA funds were limited. The project was therefore financed by a Bank loan, rather than an IDA credit. It was the first project lending to India since September 1967 and the first lending for an agricultural project since 1962. 3. The project, at the time of its negotiation, played an important role in the development of Bank policies because at the core of the extended discussions between GOI and the Bank were the preferences to be given to domestic suppliers of project goods and services and whether to disburse local currency. The Executive Directors requested a paper on the subjects, preparation and discussion of which took about six months. Policies for the treatment of both aspects evolved from these discussions. 4. The objective of the project was to increase agricultural production by expanding the availability of seeds of high yielding varieties of foodgrains in India as part of GOI's "new strategy" for agricultural development introduced in 1965, which sought to promote rapid increases in the annual growth rate of foodgrain production. At appraisal it was assumed that the project, at full development would provide seeds for some seven million acres annually. 5. Under the project high yielding dwarf wheat varieties, new maize hybrids, sorghum and pearl millet and high yielding rice varieties adapted to local conditions were to be grown for seed. Seed production requires specialized technology both in the field and in processing which had to be developed and perfected under Indian conditions if high quality seeds were to be marketed. The project aimed at the develop- ment of some 46,000 acres from which, with double cropping and adequate - 2 - fertilization, about 46,000 tons of seeds would be produced, processed and sold each year. Full development was expected to be reached after five years. As assumed at appraisal on-farm development for seed production entailed provision of tubewell irrigation, land levelling, installation of underground water distribution systems in some of the fields and more intensive mechanization of farming operations. This development was to be made possible by provisions of credit to selected farmers through the State Bank of India (SBI). Bank financing also included provision of foreign exchange to meet the fertilizer needs of seed crops. Seeds produced were to be processed by two plants to be built and operated by the Tarai Development Corporation (TDC)in the project area. TDC, a private company to be owned by farmers, the National Seed Corporation (NSC), the Uttar Pradesh Agricultural University (UPAU) and commercial organizations, would be responsible for seed marketing and for providing services to seed producers. 6. The project has generally been successful in achieving its objectives, although only after considerable delays. The project's annual certified seed production, which has averaged 28,000 MT since 1973, is expected to increase to 42,000 MT (80% of total TDC processing capacity) at full development in 1981/82, as against 46,000 projected for full development, at appraisal. To produce this quantity of seed, 17,000 acres of land were leveled by the end of 1977 (versus 36,000 acres project- ed at appraisal), tubewells were constructed to serve 26,000 acres (as envisaged at appraisal), but the underground water distribution systems were not installed (a command area of 41,000 acres was assumed at apprai- sal). In addition, 196 km of power transmission lines were built (in line with appraisal estimates). The farm mechanization program as implemented was successful (though different from appraisal projections) and fertili- zer could be provided only at much higher prices. Processing capacity was provided for 53,000 tons of seed by expanding three existing plants at lower cost instead of constructing two new plants with a total capacity of 45,000 tons (as envisaged at appraisal). The number of TDC distributors and dealers for seed marketing exceeded 2,500 in 1977 (no specific target was given during appraisal). 7. Major problems and changes of original plans occurred and resulted in the postponement of the closing date by three years from December 1974 to December 1977. These problems and changes include: - The idea of constructing new seed processing plants was gradually abandoned in favor of expanding capacities of existing seed processing plants. - The absence of suitable contractors for land-levelling forced the project to arrange for alternatives (UP University and State Agro-Industries Corporation to carry out the work). - Legislation introducing ceilings on farm sizes contributed to lower than projected investments in land levelling and no investment at all in underground irrigation water distribution systems. -3- - Crop acreages deviated from plans as follows: (i) for maize, because the Bhabar area, where maize was to be grown, was much less developed than anticipated due to the decision to drill fewer deep wells there, while in the main project area maize suffered from disease; and (ii) for soya beans, due to a change in seed demand. Wheat and rice acreages were much larger than projected. - Actual mechanization differed from appraisal assumptions partly due to a larger number of smaller farming units participating and partly due to other, cheaper funds available for financing farm machinery. - High standards of TDC management declined temporarily during the second half of the project period. - Towards the end of the project farmers shifted from seeds production to other crops (including sugar cane) due to a weakening of seed prices making them less competitive with alternative crops. - Considering the very slow pace of disbursements under TDC, GOI suggested that the Tarai loan balance (US$1.55 million) be cancelled and that future requirements be addressed in the context of the follow-up projects. 8. The project has been successful in achieving its objective, despite the above mentioned problems and changes from the appraisal con- cept. The project's recomputed economic rate of return is 18%, compared to 17% estimated at appraisal. If downstream benefits are taken into account, the economic rate of return increases to 42%1/. Incomes of the 1,150 participating seed growers have increased as a result of the Pro- ject, as well as those of the 2,500 TDC dealers and the estimated 400,000 smallholders using the improved seeds. The project also served as a model for two follow-on projects2/ covering all of India. 9. Several factors contributed to the project's success. First of all, most of the time, demand exceeded available supply of seeds. Second, the Tarai area generally has37 suitable climate and good soils conducive to providing quality seeds- . Third, highly motivated and qualified project management and the project's organizational structure were also factors in assuring good seeds. Fourth, some seed production was already in progress in the Tarai area prior to the project, providing UPAU's Seed Production Division, progressive farmers and UP State marketing agencies with some experience. Fifth, the project received support from virtually every group or agency affected by it. Sixth, the concept and timing of the project were basically appropriate. 1/ PCR, paras. 4.15, 4.16 and Annex 4 2/ See Basic Data Sheet 3/ Except maize: See para. 7. II. MAIN ISSUES New Construction versus Expansion of Existing Facilities 10. The potential of expanding existing seed processing plants in the Tarai area versus new construction was not realized during project preparation and appraisal. At the time of appraisal, the Bank opted for, and GOI (TDC) agreed to, a turnkey contract for building two new pro- cessing plants to secure high processing standards. It was then thought that TDC's limited experience could be more easily overcome, by providing new plants with less risk of possible quality problems of processed seeds. However, in September 1972 TDC rejected the offer from the firm selected for the turnkey contract. TDC thought the bid too costly, and felt that it failed to meet specifications and that the firm was inexperienced (PCR para. 3.08). While revised tenders for a new processing plant were prepared, excluding the turnkey clause, TDC proposed, and the Bank approv- ed (January 1973) an expansion of the existing Nagla plant as a stop gap measure to avoid further delays in the expansion of seed processing capacity (appraisal projected that construction of new processing facili- ties would start in 1969). In 1974 and again in 1975 TDC proposed, and the Bank agreed, to further expand existing processing capacity. Con- sequently, there was no more need for constructing new processing plants. 11. The firm initially selected but later rejected for the turnkey contract charged that TDC, from the very beginning, had never intended to sign the contract. In fact, demand for wheat seed had been slackening in 1969/70 and again in 1972/73 and this made TDC very cautious in in- vesting in new processing plants. Also, as a private, profit-oriented corporation, TDC was quite conscious of costs. The high cost of new plants appears therefore to have been the main reason for TDC rejecting the turnkey contract and giving preference to expanding existing seed processing plants. 12. In 1973/74 wheat suffered a rust attack in many parts of India, leading to a sudden increase in demand for wheat seed. Since TDC did not have the anticipated processing capacities (because of the changes in project design mentioned above), it could not satisfy this demand. Consequently TDC had to forego a share of potential benefits, and some downstream benefits to the entire economy were lost. The change in project design also led to some waste of manpower and funds: consultants had been engaged to design the new plant, TDC and Bank staff had been involved in discussing differences in plant design and tender documents for extended periods of time. 13. In this situation TDC proved to be extremely cost-conscious. The Bank, in advocating the turnkey contract, appeared to be particularly concerned with seed quality in this first seed project. In retrospect the alternative of expanding the existing plants might have been examined more carefully at appraisal. - 5 - 14. In any event, one result of expanding existing capacity was that capital investments in seed processing have been less than two-thirds of what it was estimated to have been for new construction and seed quality is still outstanding. Overdesign of Irrigation Component 15. The project, as appraised, provided for underground water distribution systems for 41,000 acres of land and land-levelling for 36,000 acres. The water distribution systems were not installed, and only 17,000 acres of land were leveled. Despite these lower achievements TDC's physical seed production is estimated to reach over 90% of the appraisal estimate. This raises the question of whether this component was overdesigned. The appraisal report states that "land-levelling is essential to make best use of irrigation water, thereby ensuring even growth of seed producing crops", and the... "underground pipe system has been demonstrated by UPAU to result in reduced maintenance expendi- ture, reduced land wastage, ease of cultivation operations, improved weed control and comparable installation costs with open, lined canals". The appraisal, by implication, assumed that there were constraints on the optimum use of water and land. Apparently the anticipated constraints were much less severe than expected. This experience, combined with the experience with increasing the capacity of the seed processing plants, suggests that the project was originally overdesigned. Income Distribution Effects of the Project 16. The project was appraised in 1968, well before Bank activities increasingly focused on the poorest 40% of the population. The objective of the project was then to increase agricultural production by expanding the availability of HYV seeds and this will be accomplished once full development is reached in 1981/82. What was not included in the economic returns estimated at appraisal but what has been an additional project benefit is that the output and incomes of the estimated 400,000 seed users must have increased considerably. The recomputed economic rate of return for the project is 42% if benefits estimated to accrue to seed users (that is the value added from incremental foodgrain production) are taken .into account. However, no comprehensive farm budgets are presented either in the appraisal report of the Tarai project or in those of the follow-on projects. An analysis of farm budgets, particularly for typical seed users, would have put the project's income distribution effects in a better perspective. 17. Net income from a typical 100 acre seed producing farm, was projected at appraisal to increase from US$11,000 to US$19,000 due to the project. However, due to slightly lower (2%) cultivation costs and slightly higher (3%) producer prices for seed actual net income is higher than anticipated. It should be noted that although the project benefited about 400,000 small farmers using improved seed, it also contributed to supporting a group of farmers who were well off, and that many of the participants tried to use the argument of the need for larger farms to produce quality seeds in support of the project to maintain the size of their farms. - 6 - In the follow-on projects this problem no longer exists: seed growers are mainly small farmers. And although a much larger number of seed growers is now involved, the "compact area" approachl/ will to a great extent help reduce the inherent disadvantage of dealing with large numbers of small producers. Coherence with Follow-on Projects 18. During project implementation two more seed projects (First and Second National Seed Projects) were started in India, covering between them most of the country's needs for improved seeds. The project's organization and success, particularly TDC's, served as a model for the two follow-on projects. The State Seed Corporations (SSCs) were created with a structure similar to that of TDC. The "compact project area" approach has also been taken over from the project. Another project contribution was that preparation of the follow-on projects proceeded faster than that of the Tarai Project. For the first National Seeds Project it took nine months from identification to appraisal, while the same activities for the Tarai project took 17 months. First indica- tions from the follow-on projects are, however, that they may not be as successful as the Tarai Seeds Project: they are multi-state projects with implementation problems that hamper many of those projects in India. 19. Delimitation of seed markets is an issue in this context. Tarai seeds have been marketed in areas as distant as Assam and Andhra Pradesh and even in Bangladesh, Vietnam and the Middle East. Considering the high transport costs for these distant areas it is safe to assume that in the near future these markets will be served by SSCs which are located closer to those markets. TDC in turn will have to develop new markets nearby. While local markets may well develop the fact that the UP State Government has put a relatively low ceiling on retail seed prices, may, by reducing farmers' incentive to produce seed reduce the project's expected profitability. Lower seed prices would also enlarge seed demand, but with falling supply, such demand could not be satisfied. The present expansion in seed production in India and its implications were probably difficult to foresee at preparation and appraisal but it would have been prudent, in anticipation of possible developments if a seed study had been included to better understand the relations between price and demand/supply and the appropriate location of seed production. Such a study is now being carried out under the follow-on projects (First and Second National Seeds Projects). 1/ Approach to get the potential seed growers in compact areas around each processing plant. - 7 - INDIA TARAI SEEDS PROJECT - LOAN 614-IN PROJECT COMPLETION REPORT I. BACKGROUND 1.01 The Tarai Seeds Project represents the Bank Group's first lending operation in the field of seed production. The importance of adequate supplies of good quality seed was emphasized by the Government of India in its new agri- cultural strategy set forth in August 1965. A prominent element of this strat- egy was the nationwide High Yielding Varieties (HYV) Program, under which certified, high yielding seed varieties are channeled, as a matter of priority, to areas of assured water supply, together with complementary inputs, such as fertilizers, pesticides and extension service. 1.02 The Program is dependent for its successful implementation on the organized multiplication of improved varieties produced by plant breeders, such that genetic purity is maintained and seed planted by farmers is viable and reasonably free from adulteration. These conditions presuppose a high degree of coordination between plant breeders and farmers. To facilitate such coordination and to stimulate growth of the seed industry, GOI estab- lished the National Seeds Corporation (NSC) in 1963. By initiation of the project in 1969, NSC occupied a dominant position in India's growing seed industry. Improvements brought about by NSC did, however, not yet assure seed production of highest quality. Typically, seed production and distribu- tion was undertaken by scattered inefficient farms which were rarely equipped with proper processing or storage facilities. Arrangements for breeder stock and foundation seed stock were also weak. The Tarai Seeds Project was con- ceived to address these weaknesses through the combined efforts of NSC, the Uttar Pradesh Agricultural University (UPAU) at Pantnagar and neighboring pro- gressive farmers. 1.03 . As appraised, the project entailed the development over five years of seed farms on about 46,000 acres of land in the Tarai region of Uttar Pradesh (UP) in north India. 1/ At full development, certified seed produc- tion was projected to reach 46,000 tons per annum - sufficient to sow over two million acres each of rice and wheat, and over three million acres of other foodgrains. Project components included: - on-farm development (i.e., land levelling, irrigation and tubewell electrification) on some 400 farms; 1/ The Tarai is a fertile belt of land sandwiched between the foothills of the Himalayas and the open plains of Uttar Pradesh. - 8 - - provision of farm machinery, fertilizer and foundation seed to project farms; - execution of soil and topographic surveys of seed farms; - preparation of farm plans; - extension of electric power distribution system; - provision of seed processing facilities; and - credit and technical assistance for the above purposes. 1.04 The Bank loan (614-IN) of US$13.0 M was for 30 years, including 10 years grace, at an interest rate of 6-1/2%. It was signed on June 18, 1969, with project completion expected by December 31, 1974. Subsequently, the project Closing Date was extended twice, in the first instance for two years (to December 31, 1976) and in the second instance for one year (to December 31, 1977). Total withdrawals through December 1977 amounted to US$11.4 M, leaving an undisbursed loan balance of approximately US$1.6 M which was cancelled. II. PROJECT PROCESSING Identification and Preparation 2.01 The possibilities of the Tarai region as a center for a seed pro- duction project were first brought to the Bank's attention in the mid-1960s through the efforts of the Uttar Pradesh Agricultural University (UPAU) and its Vice-Chancellor. The basic project design developed as a result of missions by the FAO/IBRD Cooperative Program in November 1966 and February 1967 1/. The latter mission prepared a Draft Report 2/ which was issued on June 19, 1967. The report was well received by GOI, with the exception of certain issues relating to project management and organization. In particular, GOI wanted to reduce expenditures on proposed technical assistance services. This was to be accomplished by (a) decreasing the number of technical per- sonnel; (b) transferring certain personnel from the Project Authority to TDC; and (c) obtaining on a no-cost (advisory) basis essential expertise through USAID and the Ford and Rockefeller Foundations. 2.02 In additi6n to reducing technical assistance services, GOI sought to ensure that a majority of share capital in the Tarai Development Corporation 1/ The project design (and, more generally, the Bank's lending strategy for seed development) appears also to have been influenced by a report entitled The Prospects of the Indian Seed Industry (August 1967), pre- pared by R. Pantanali of the Resident %ission in New Delhi. 2/ Draft Report of the India Seeds Development Project. FAO/IBRD Cooper- ative Program. Report No. 9/67 IND.8. June 19, 1967. - 9 - would be held by public sector organizations, such as the NSC and Pantnagar University. On both issues the mission put forth strong arguments in favor of the original proposal (as presented in the Draft Report). The mission pointed out that the success of the project would depend largely on efficient seed farm management, for which a well-trained and full-time technical staff would be needed. On the question of shareholdings, the mission indicated that the Bank would want to see the private sector take up a substantial proportion of equity in TDC. 2.03 While seeking to reduce technical assistance expenditures, GOI simultaneously recommended that three new cost elements be incorporated in the project. These included: (a) extension of power transmission lines for tubewell electrification; (b) a groundwater survey; and (c) foreign exchange for imported fertilizers required for crops planted with the certified seed. 2.04 The power line component was subsequently included in the project. With respect to the groundwater survey, however, it was pointed out that such costs are usually not charged to farmers. Accordingly, GOI was advised to "find and provide adequate resources for this development outside the pro- ject." In the case of fertilizers for seed users (i.e., in seed reception areas), the Bank maintained t!,at the project was specifically intended for seed production and that financial provisions for purchases of inputs for general farming would be outside the project's scope. 2.05 By 1968, GOI issued a new report setting forth proposals for what it then called the "Tarai Seeds Development Project." The new report followed the outline of the FAO/IBRD draft, without changing technical aspects and physical targets. However, proposals on project organization were modified along the lines discussed in paragraph 2.01. Of particular significance was a provision to reduce the technical group from 34 to 8 and the number of expatriate seed experts from five to one. There was no mention of provisions for consultants' services, even though this had been an important part of the FAO/IBRD draft. The Bank's reaction to the GOI report was one of general concern. In essence, it was felt that the report revealed a failure to understand the complexity of the Tarai project and the need for mobilizing development efforts in the project area. 2.06 To clarify GOI's position on the above-mentioned issues, a Bank staff member was sent to India for two weeks in late January 1968. This mission was successful insofar as it re-established momentum in the project cycle. Of particular importance was the Secretary of Agriculture's commitment to proceed with incorporation of the Tarai Development Corporation by April 1968. The capital structure of TDC was contemplated as follows: farmers 30%; UPAU 30%; UP Agro-Industries Corporation 20%; NSC 20%. This distribution of equity capital was designed to protect farmer-interests (against undue pres- sure from both the State and the Center) through the participation of NSC and UP Agro-Industries Corporation. - 10 - 2.07 The January mission also secured the involvement of the State Bank of India (SBI) as a credit channel for project farmers and succeeded in getting UPAU to volunteer the services of its survey and engineering departments to help in preparatory on-farm planning. Given these develop- ments, together with the growing interest enjoyed by the project, the mission recommended that steps be taken to mount an appraisal mission within two months' time (by April 15, 1968). Appraisal 2.08 As recommended, an appraisal team began field work in the Tarai in mid-April. The five-man mission included three Bank staff, one FAO/IBRD Cooperative Programme representative and one consultant, the latter engaged to examine seed production and processing questions. The mission's work was supplemented by a Bank energy specialist who reported on tubewell electrifica- tion and seed processing plant power requirements. 2.09 The findings of the appraisal mission reflect several noteworthy differences in what was presented in the project preparation report. 1/ Where, for example, the preparation report set forth a rather complex man- agement structure, including a fully staffed Project Authority 2/ operating as part of the NSC, the appraisal report proposed a less cumbersome inter- agency "Advisory Committee" to review progress and provide technical advice. Representation on the Committee would include GOI, State agencies and founda- tions involved in agriculture and seed research, SBI, NSC, project area farmers and TDC itself. In eliminating the Project Authority, the appraisal mission was also able to reduce considerably the number of technical assist- ance personnel associated with the project. As mentioned earlier, this had been a source of concern for GO. 2.10 Another difference in the appraised project is the absence of a training component, though this was much emphasized in the preparation report. Such a component, however, was not essential to the viability of the project, particularly insofar as a number of training opportunities were found to be available through GOI and the University. 3/ 2.11 A significant aspect of the preparation report was the elabora- tion of certain "minimal conditions" to be met by all project participants. One condition in particular - that participants "must have a compact holding of at least 50 acres" - was regarded as biased against small farmers and was accordingly dropped by the appraisal mission. Subsequently, seed growers with as little as three acres became TDC farmer shareholders. 4/ 1/ The Draft Report of June 19, 1967. 2/ In addition to TDC, which also would be fully staffed. 3/ ' Additional training was supplied by the-Ford and Rockefeller Foundations and through bilateral channels such as USAID. 4/ However, some 70% of project farmers had holdings over 45 acres. - 11 - 2.12 In assessing project benefits, the appraisal mission used a some- what more conservative approach than that found in the preparation report. The mission based its economic return calculations on incremental changes in seed growers' net income and purposely did not include major downstream bene- fits such as increased foodgrain production associated with quality seed. Also, seed growers' crop yields were estimated to be 6% to 20% lower than was assumed in the preparation report. The rate of return of the appraised project was estimated to be about 17%. Negotiations 2.13 Project loan negotiations were held in Washington from November 27 to December 13, 1968. During negotiations assurances were obtained, inter alia, that: (a) TDC would consult the Bank about changes in the position of Chairman and appointments to senior management posts; (b) SBI would employ and UPAU would - if required - provide a farm appraisal unit for the technical appraisal of project loan applications; (c) SBI would keep clearly identifiable accounts of project loans and borrowings under the Subsidiary Loan Agree- ment; and 1/ (d) TDC would consult the Bank about proposed second p-yments 2/ to farmers and proposed distribution of net income. 2.14 The loan was to become effective once: (a) the Subsidiary Loan Agreement was signed; (b) Rs 4 M of TDC share capital was paid in; (c) TDC employed adequate staff to enable it to carry out its project obligations; (d) _SBI provided its technical unit with staff satisfactory to the Bank; and, (e) GOI had obtained commitments from UPAU and the State Electricity Board that these entities would undertake their project obligations. 1/ The Agreement provided for the on-lending of certain loan proceeds from GOI to SBI, utilizing the Agricultural Refinance and Development Corporation (ARDC) as a financial channel. 2/ Second payments would be made to seed growers only in cases where TDC received a price for processed seed in excess of its operating costs. - 12 - 2.15 An overriding issue during negotiations involved procedures to be followed in the procurement of items required for the project. GOI was pre- pared to accept international competitive bidding over a wide range of items on the understanding that, subject to approval by the Executive Directors (EDs), the Bank would allow a margin of preference to domestic suppliers of 27.5% or the prevailing customs duty, whichever was lower. Several EDs, however, questioned the wisdom of such a high domestic preference and directed management to prepare a paper on the subject as a basis for further action. 2.16 Preparation and discussion of the paper on "Preferences for Domestic Suppliers" delayed consideration of the project by approximately four months (to June 1969). By that time, GOI agreed to presentation of the project on the basis of a 15% preference level, a rate acceptable to the EDs. 2.17 Another issue which arose during negotiations concerned whether proceeds of the Bank loan would finance goods with components produced in countries not members of the Bank. 1/ At a Special Loan Committee Meeting, it was decided to inform GOI that the goods (tractors) in question would be eligible for full disbursement under the loan, provided the non-member com- ponents did not exceed 50% of total value. 2/ GOI accepted this proposal, clearing the way for presentation to the Board of Directors. 2.18 The Board presentation emphasized that the project was notable insofar as: (a) it represented the first lending operation for project purposes in 18 months, largely because of delays in IDA replenishment; (b) it was the first financing anywhere by the Bank Group of seed production, an operation vital to India's new agri- cultural strategy; and (c) it was the first step in what the staff hoped would be a return to agricultural project lending to India on a large scale (this was the Bank's first lending operation for Indian agriculture since 1962). 2.19 In concluding the Board presentation, it was pointed out that the project would be economically sound even if the product of the farms were not sold as seed; "that is, even if this were viewed as simply a land development project." The EDs approved the proposed US$13 M loan on June 10, 1969., 1/ Specifically, a locally assembled tractor (Escorts) was found to include components of Polish origin amounting to some 25% of total value. 2/ The total value of all non-member compoennts to be procured under the loan was not predicted to exceed $50,000. Based on available records, this prediction appears to have been quite accurate. - 13 - III. IMPLEMENTATION Effectiveness and Start-up 3.01 The project was declared effective on September 12, 1969, some three months after the Loan*Agreement was signed. In the interim, all legal conditions had been satisfied, with the exception of the appointment of TDC's Chief Marketing Officer and Chief Financial Officer. Considerable efforts had been made to recruit suitable candidates for both positions, but with- out success. The Bank felt that insistence on filling the two posts before loan effectiveness would unduly delay project implementation and therefore declared the loan effective. Simultaneously, GOI was informed that the Bank would "continue to be most concerned about these appointments and expects prompt, effective action on them." 3.02 In reviewing project implementation, one finds this incident to be the first of a series of problems associated with recruiting and keeping competent managerial personnel at the helm of TDC. At the outset, few quali- fied managers were anxious to work in the Tarai, which was regarded as an isolated and rather backward area. To reverse this trend, Bank staff urged project authorities to advertise nationally for key positions and to offer competitive salaries and benefits. 1/ While such an approach was helpful in attracting talented managers, it was only partially successful in reducing the high turnover among TDC staff. 3.03 To assess project start-up problems, including staffing, an ini- tial inspection visit to the Tarai was carried out by Resident Mission staff in late November 1969. The mission discovered a number of significant prob- lems. With respect to recruitment, the mission found that the marketing and financial officer posts remained vacant, although recruitment efforts were still under way. There were also procurement difficulties, with farmers unwilling to be specific about desired equipment in the absence of inform- ation about prices (which TDC, as the designated bulk purchasing agent, had failed to provide). 3.04 In assessing the status of engineering, the mission noted that the consultants for preparing seed processing plant (SPP) specifications and tender documents had yet to be specified. Once again, TDC was attempting to "economize" on necessary personnel by utilizing, on a part-time basis, for- eign advisors attached to the NSC. 3.05 An unexpected finding of the mission was that seed production targets had been altered (without the required Bank approval) to include 7,500 acres of soybean cultivation. The Bank reacted to this development by informing GOI and TDC that it was not opposed to variations in seed production targets 1/ One of the arguments for establishing TDC as a private corporation was that it would not be'bound by civil service salary levels for any post. - 14 - per se; however, any future variations would be subject to Bank approval and should be based on an analysis of crop demand, national production esti- mates and processing capacity. TDC responded by providing the Bank with its rationale for shifting to soybean production and the Bank subsequently agreed (in May 1970) to include soybeans in the list of seed crops which could be grown under the project. 3.06 Resolution of the other problems mentioned above was first attempted through periodic communication with GOI and TDC. Then, in mid-February 1970, a supervision mission was sent to the field to deal with project implementa- tion issues directly. The mission's full supervision report states that "some of the problems and delays that have been encountered could probably have been avoided had there been a review mission shortly after the loan was signed (June 1969)." Mid-Course Revisions 3.07 An examination of supervision reports and other pertinent docu- ments reveals the following revisions in project'design: (a) TDC's decision to modify existing seed processing plants rather than to construct new plants; (b) the addition of tubewell drilling rigs to the list of equipment to be financed under the project; and (c) significant increases in farm mechaniza- tion requirements. 3.08 On the issue of processing-plants, it was envisioned at apprai- sal that two new SPPs would be constructed, providing a total annual capa- city (both plants) of 45,000 Lons. 1/ However, only one acceptable bid was received for the first plant (out of 11 submitted), and TDC had major reser- vations about signing a contract with the firm selected, Lang Engineering of Florida. Eventually, negotiations broke down completely, with TDC alleg- ing that the bid was too costly, that it failed to meet specifications, and that Lang was inexperienced in the installation of complete SPPs. As a stop- gap measure (before reinviting bids for a new plant), TDC proposed to double the capacity of its Nagla SPP from 10,000 to 20,000 tons. 3.09 Once the Nagla expansion was successfully completed, TDC decided in June 1975 to expand and modernize other existing SPPs as well, as opposed to the original plan of constructing new plants. TDC argued, that moderni- zation was less costly and could be done incrementally (i.e., it did not entail the lumpy turnkey investments proposed at appraisal). TDC further explained that it was now taking a more conservative view of market prospects, particularly since the proposed National Seeds Project (also financed by the Bank) would significantly increase competition. This line of reasoning was ultimately accepted by the Bank and the processing capacity targets shown in Table I were established. 1/ Clean seed basis. In terms of raw seed, total annual capacity would equal approximately 60,000 tons. - 15 - Table 1: SEED PROCESSING CAPACITY (tons per year, clean seed basis) Capacity Full Development 1970 Target Nagla SPP 10,000 20,000 Haldi SPP 6,000 18,000 Matkota SPP 4,000 15,000 20,000 53,000 3.10 The total proposed capacity of 53,000 tons per annum 1/ agreed to by the Bank is well in excell of the 45,000 tons originally envisioned in the appraisal report. As part of its expansion program, TDC also requested (and the Bank approved) a 1,000-ton cold store. Such a facility would permit TDC to keep on hand a strategic reserve of foundation seed without risking a loss of germination potential. Since a cold store is a normal ancillary of a seed processing plant, it was decided that it would fall under an existing loan disbursement category and that no reallocation of funds would be required. It was estimated that the cold store would cost approximately US$300,000, half of which would be financed by the Bank. 3.11 The second revision in the project concerns tubewell drilling rigs which were requested in February 1970. This iequest was a surprise in that the appraised project was based on irrigation by shallow tub'wells. Soon after appraisal, however, an artesian aquifer was discovered. While project farmers were anxious to exploit the "new" aquifer, its greater depth neces- sitated the use of heavy-duty drilling rigs not available at the time. In this regard, the Exploratory Tubewell Organization (ETO) estimated that three additional drilling rigs capable of sinking wells to depths of about 700 feed would be required .to optimally exploit the aquifer 2/ within the envisaged project period. Total investment (rig) costs were estimated to entail a foreign exchange expenditure of approximately US$300,000. 3.12 Before amending the loan documents to allow for drilling rig pro- curement, the Bank requested ETO to provide copies of its aquifer investiga- tion. ETO delayed on this matter for over a year before supplying the necessary data, at which time (in December 1971) the Bank agreed to include heavy-duty drilling rigs in the List of Goods which could be procured under the project. 3.13 The tubewell drilling issue surfaced again in August 1972 when TDC informed the Bank that the price of drilling rigs had escalated to about US$250,000 each (from the original estimate of US$100,000 per rig). Further, it was not estimated that it might take as long as three months to drill each 1/ Includes capacity which existed prior to project implementation. 2/ ETO proposed that some 150 artesian wells be sunk in the western sector of the project area. - 16 - deep well. The Bank, on receiving this news, once again asked for more inform- ation to determine if the utilization of deep tubewells was still a cost- effective option. After receiving the required data, the Bank agreed to the procurement of the more expensive rigs. By February 1976, however, the ETO (since renamed the Central Ground Water Board) had still not ordered the rigs. Their reasons for delaying were twofold. First, they were worried about high operating costs and cost recovery, and second, there was a lingering "doubt as to the ability of the formations in question to yield sufficient tubewell water for irrigation purposes." 3.14 At this point (after six years of delay), the Chairman of TDC suggested that the proposal be dropped, though TDC was still anxious to purchase the rigs. In view of the Central Ground Water Board's inability to make up its mind on the issue, the mission fully supported the Chairman's suggestion. Thus, what started out as a potentially beneficial change in the project ended up as a frustrating and inconclusive episode. 3.15 The third major revision of the project - an increase in farm mechanization procurement - took place in April 1970.when the number of inde- pendent .farming units had increased to 700, as opposed to 400 units envisaged at appraisal. The larger number of farms was not an oversight of the apprai- sal team; rather, it was indicative of the rapid progress achieved in the early 1970s in conve rting leasehold farms into freeholds. As a result, it was argued (and the Bank concurred) that approximately 375 additional tractors were justifiably required for planned seed pioduction operations. 1/ This increase in numbers was partly offset by lower unit prices btained through bulk procurement, but the total cost of tractors still exceeded the US$1.7 M estimated at appraisal by about US$1.0 M. Physical Progress - Seed Production and Processing 3.16 In reviewing seed production, one notices that both the area and output of maize, sorghum and millet are well below the targets envisioned at appraisal (Table 2). Over the years, these crops were found to be not well suited for the main part of the project area. For maize, disease resistance (to downy mildew and stalk rot) was a major problem. Lodging was also a problem, affecting up to 90% of the crop in some years (due to high winds in the area). 1/ In order to avoid over-investment, TDC drew up, and the Bank approved, guidelines for the amount and type of machinery according to farm size. - 17 - Table 2: SEED CROP AREA AND PRODUCTION AT FULL DEVELOPMENT (since 1973/74) Appraisal Estimates vs. Actual Performance I. Area (acres) Appraisal 1973/74 1974/75 1975/76 1976/77 Estimates Actual Actual Actual Actual Maize, Sorghum & Millet 19,000 3,028 3,828 4,318 5,272 Rice 5,000 4,987 9,289 9,986 9,003 Wheat 16,000 17,815 21,353 27,213 18,945 Soybean - 1,373 3,702 2,927 1,679 Pulses & Others - 1,298 1,433 1,379 325 Total 40,000 28,501 39,605 45,823 35,224 II. Seed Production /a (tons) Maize, Sorghum & Millet 15,200 823 2,111 1,296 1,342 Rice 7,250 5,667 10,638 10,696 8,325 Wheat 23,200 14,902 17,471 20,135 14,513 Soybean - 160 527 534 - Pulses & Others - 614 1,330 971 532 Total 45,650 22,166 32,077 33,632 24,712 /a All production figures on clean seed.basis. 3.17 Rice and wheat production in the Tarai has been generally successful. Loose Smut has attacked certain wheat varieties, but this problem has been actively addressed by UPAU agricultural research personnel. More recently (1976/77), wheat seed production dropped sharply as a result of: (a) untimely rains; and (b) seed growers' resistance to take up sufficient production. The resistance of growers has been due to delayed payments by TDC as well as TDC's recent low intake prices. In 1977, for example, wheat seed growers were paid Rs 150 per quintal, versus payments of 200 per quintal in 1974 and 1975. The responsibility for low intake prices rests in large part with the UP Government, which placed a ceiling on retail seed prices since 1976. In response to the cut in intake prices, some farmers have felt inadequately remunerated for the cost and risk associated with seed production and have subsequently shifted to grain production. 1/ 1/ The formulation of an appropriate seed-pricing policy is currently being considered within the context of the Bank-financed National Seed Project. - 18 - 3.18 As mentioned earlier (para 3.05), soybean seed production was not envisioned at appraisal, though TDC farmers began cultivation in any case in 1969/70 (without consulting the Bank). Subsequently, it was learned that the impervious soils and high rainfall of the Tarai were generally adverse to soybeans, with the result that seed production targets were consistently not achieved. Although some progress was made in shifting production to up- land areas where soil and climatic conditions were more favorable, production was terminated in 1976/77 as a result of a sharp decline in demand. 1/ The cultivation of pulses and other seed crops was undertaken in response to request from farmers for a wider range of TDC seed. This aspect of diversi- fication, though of limited scale, has been generally successful, with TDC now producing seed crops such as potato, peas, oats, berseem and mustard. 3.19 The various issues surrounding development of TDC's processing plants have already been discussed (paras 3.08 - 3.10). By project comple- tion (December 31, 1977), all three expanded SPPs were operational, although final completion is not envisioned until late 1978. Work on the Haldi cold store is also behind schedule, with completion now set for September 1978. The slow pace of construction of all seed-processing facilities has been attributed to heavy monsoon rains, lack of materials (steel and cement) and power shortages. Finding qualified contractors willing to work in the somewhat isolated Tarai region has also been a problem. At present, all remaining work on overhead conveyors and drying-cum-storage bins is being undertaken by a single contractor working on one SPP at a time. Marketing and Distribution 3.20 Since project inception, TDC has worked diligently to establish an efficient marketing network for its seed. By late 1977, this network included 47 distributors and 2,482 dealers. In addition to their distribu- tion function, dealers undertake pre-season surveys of their respective areas to determine effective demand. The estimates provided by dealers are sub- sequently used by TDC's management to establish seasonal seed production targets. 3.21 The recent removal of restrictions on the interstate movement of foodgrains has helped rationalize TDC's marketing operations. In addition, TDC was able to persuade the Indian Railways to give certified seed a higher priority (from "C" to "B" category), thus assuring the provision of freight space on most trains. Physical Progress - Farm Development 3.22 As appraised, the project features a comprehensive program of on- farm development. The specific elements of the farm development program include: (a) land levelling; (b) tubewell construction and electrification; 1/ Specifically, demand in the principal market of Madhya Pradesh shifted toward locally grown (black-grain) varieties of soybean. Since these varieties have not been released by the Central Seed Committee, TDC under its statutes cannot produce them. - 19 - (c) underground irrigation water distribution systems; (d) farm mechanization; and (e) fertilizer procurement. 3.23 The provision for land levelling was included to optimize the use of irrigation water and promote the uniform growth of seed crops. At project inception, farmers were quite enthusiastic about the levelling program, which included the reclamation of virgin land, as well as rough levelling and fine levelling. However, suitable contractors were difficult to find and work was delayed until early 1973, at which time the levelling target was reduced from 36,000 to 26,000 acres. Initial land levelling operations were performed by Rajasthan State Agro-Industries Corporation with execution of work on an hourly basis, rather than on a volume basis as envisioned in the original tenders. Eventually, the University and a second contractor (UP State Agro- Industries) also took up land levelling. As of the extended project Closing Date, some 17,000 acres of land had been levelled for seed cultivation. 1/ 3.24 An important disincentive to land levelling, indeed to all on-farm investment, materialized in 1971 with the enactment of nationwide land ceiling legislation. This legislation restricted the amount of land under single ownership to 45 acres. At appraisal, however, some 70% of project farmers had holdings exceeding 45 acres, with the average farm size amounting to approxi- mately 185 acres. Not surprisingly, the imposition of ceilings triggered a scramble among Tarai farmers to re-register their holdings among various relatives as a means of circumventing the restrictions. Although this in many instances prevented significant reductions in holding sizes, the restric- tions nevertheless had a decidedly negative impact on achievement of the targets envisioned in the farm development program. 3.25 With respect to .tubewell construction, the appraisal target was 325 cusecs (145 wells of I cusec and 90 wells of 2 cusecs). The wells were to be sunk over an area of 26,000 acres. Drilling was largely confined to the sinking of medium and shallow tubewells in the central Tarai region. In the Bhabar area (also farmed by TDC shareholders), progress was extremely slow due to the inability of TDC to obtain percussion drilling rigs (para 3.11). Nevertheless, 20 out of 43 deep tubewells were sunk in the Bhabar between 1971 and 1977. 2/ By project completion, total discharge capacity (Bhabar and Tarai) amounted to 286 cusecs, or 88% of the target envisioned at appraisal. 3.26 The electrification of tubewells was achieved by expanding the UP State Electricity Board's (SEB) planned program of rural electrification. Specifically, SEB technicians installed an additional 100 km of 33 kV trans- 1/ Approximately 50% of land levelling took place on the 10,000 acre UPAU farm. 2/ Each well took seven to eight months to complete. TDC officials maintain that progress would have been faster had the Exploratory Tubewell Organi- zation imported rotary-cum-percussion drilling equipment. Because of their high cost, the wells are operated by the State. - 20 - mission lines and 96 km of 11 kV transmission lines. This component of the project was smoothly and rapidly implemented, with the only difficulty being that connection of distribution lines had to be postponed on occasion due to delays in completion of tubewell construction. 3.27 Along with tubewell development, the project provided for construc- tion of underground irrigation distribution systems. Although the underground system utilized proven technology and was demonstrated by UPAU to have many advantages, its use never caught on among project farmers. In this regard, TDC reported in 1974 that there was simply no demand for such systems and that none was likely in the future. In retrospect, the reluctance of farmers to invest in underground water distribution is quite understandable when viewed in the context of the previously mentioned land ceiling restrictions. 3.28 The farm mechanization program received a tremendous response from farmer shareholders, with all 750 tractors delivered by late 1971. In order to avoid over-mechanization, TDC adopted the following guidelines: Size of Holding Tractor Allocation 30- 50 acres 1 tractor of 35 HP 50- 75 acres I tractor of 45 HP 75-125 acres I tractor of 50-65 HP 125-2,50 acres 1 tractor of 65-80 HP 3.29 For areas above 250 acres, a combination of any one of the above was allowed, provided average horsepower did not exceed 1 HP per every three acres. 1/ 3.30 Shortly after tractor procurement was complete, global tenders were issued for combine harvesters. By 1977, some 130 harvesters (both self- propelled and pull-type) were procured, making possible the rapid harvesting of the seed crop and, it is said, improving threshing quality of the seed. The capability to quickly harvest seed turned out to be a major benefit, particularly with wheat (because of its susceptibility to damage from an early onset of the wet season). The following guidelines were enforced with respect to combine procurement: Model Minimum Acreage Optimum Acreage 5'- 8' pull type 50 200 6'- 9' pull type 75 300 5'- 8' self-propelled 100 250 8'-12' self-propelled 150 450 3.31 Similar guidelines were set forth for other types of farm machinery, such as corn pickers, threshers, shredders and plantqrs. 1/ A limited number of imported 90 HP tractors were authorized for the largest farms, of 500 acres and over. - 21 - 3.32 Another element of the farm development program provided for the imoortation of fertilizers for the first five years of the project (until full development). Thereafter, it was felt that estimated needs could be met by domestic supply. While the appraisal mission recognized the key importance of fertilizers for seed growers, they did not foresee that prices would sky- rocket between 1969 and 1974. The price of urea, for example, reached $400/ ton (CIF India) in 1974, more than double the price at appraisal. The Bank reacted to this by increasing (in late 1974) the loan allocation for fertil- izer imports from $3.6 M to $5.6 M. On several subsequent occasions, TDC requested further increases in the fertilizer import category. These requests were denied on the grounds that the project objective was to finance only those incremental fertilizers which were directly needed by TDC farmer/share- holders for the production of seed crops. The financing of fertilizers for other activities, such as commercial grain production, was judged to be beyond the scope of the project. 3.33 It originally was envisioned that fertilizer would be procured on the basis of a separate tender, following conventional ICB procedures. However, GOI soon found that its comparatively small Tarai Seeds purchase ($3.6 M) would conflict with the much larger ($200 M) all-India fertilizer tender.. This problem was eliminated when the Bank agreed to allow Tarai requirements to be included in the all-India tender. The revised procedure worked well until 1976, when fertilizer supplies became relatively plentiful and better prices could be obtained by purchasing outside the so-called "central fertilizer pool." Accordingly, GOI and the Bank amended the loan agreement to give TDC, as well as farmer sharaholders, greater flexibility in fertilizer procurement. The result was a surge in purchases between July and December 1976. However, of the $1.4 M imported during this period, approximately $1.0 M was in excess of the fertilizer ceiling (loan category A-1) and thus not eligible for Bank financing. TDC and GOI officials were not aware of their excess purchase until December 1977, prompting yet another (eleventh hour) request for reallocation of unexpended loan proceeds to the fertilizer category. This request was denied by the Bank on the basis of the arguments set forth earlier. Procurement 3.34 The project had a wide range of procurement problems. Since these problems are discussed elsewhere in the paper, they will only be summarized in this section. At the very outset of the project, negotiations were slowed considerably by questions regarding a suitable preference margin (para 2.15), and whether the Bank could finance the procurement of machinery having compo- nents produced in non-member countries (para 2.17). The resolution of these issues was relatively straightforward in comparison to the longer term prob- lem of tubewell drilling rig procurement, which persisted until it was finally decided to abandon acquisition of the rigs in 1976 (para 3.14). 3.35 Given the large numbers involved, the procurement of farm machinery was relatively problem free. On occasion, GOI was slow in providing foreign exchange and import licenses for certain equipment, but this was not a serious constraint. In the last year of the project GOI did refuse to approve the - 22 - importation of additional 90 HP tractors for land levelling operations. This refusal was justified, however, since the tractors were to be used to level a new University farm not provided for in the appraisal and not essential for the realization of stated seed production objectives. 3.36 Delays in the procurement of seed processing machinery can be traced back to unforeseen changes in plant design (para 3.08). However, once the number and type of plants were agreed upon (1975), procurement went ahead relatively quickly. Subsequent delays in installation were, for the most part, attributable to poor contractor performance. In this respect, the location of the Tarai, coupled with the piecemeal nature of the installation work, made it difficult to secure fully qualified contractors. Financing 3.37 The Bank loan of US$13.0 M was made available to finance some 50% (US$9.0 M) of the estimated total investment cost of US$18.4 M and 100% (US$4.0 M) of the estimated cost of imported fertilizers (Table 3). Table 3: SUMMARY OF PROJECT FINANCING --Millions-- US$ Rs % Bank Loan Investment Items 9.0 67.5 40 Fertilizer 4.0 30.0 18 Subtotal 13.0 97.5 58 SBI 3.0 22.9 14 TDC Investors 1.7 12.4 7 Farmers (other than as investors in TDC) 3.5 25.9 16 Farm Machinery Importers (spare parts) 1.2 9.2 5 Total 22.4 167.9 100 3.38 The US$9.0 M provided for investment items was transferred from GO to the SBI for on-lending to the TDC, farmer shareholders and other project entities. The ARDC was utilized as a financial channel for the transfer of funds from GOI to SBI. This aspect of project financing was spelled out in a Subsidiary Loan Agreement, the satisfactory negotiation of which was made a condition of loan effectiveness. 3.39 The US$4.0 M provided.for fertilizer was disbursed against import documentation, accompanied by certificates from TDC that the fertilizer concerned had been or would be distributed to farmers in accordance with contracts made with them for seed production. The "certificate" requirement was useful in helping ensure that fertilizers were used for project purposes (seed growing), rather than the cultivation of cash crops grown on an indivi- dual basis. - 23 - Costs and Disbursements 3.40 The planned disbursement of loan proceeds suffered several consider- able delays due to changes in project design. In this regard, the original Closing Date (12/31/74).was extended by two years (to 12/31/76), mainly to allow completion of TDC's revised construction program for seed processing facilities. Subsequently, in May 1976, GOI requested a further (six-month) extension of the Closing Date. The Bank responded by granting a full year's extension (to 12/31/77) to allow a reasonable margin of slippage in construc- tion schedules and to ensure completion of disbursements. It should be noted that granting of the second extension was by no means automatic; rather, it followed a thorough review of the performance of project authorities against benchmarks established by supervision missions in 1976. 3.41 A final accounting of disbursements (Table 4) shows total reimburse- ments (through 12/31/77) equal to some $11.45 M, or 88% of the loan amount. Two noteworthy actions were taken by the Bank to facilitate full utilization of the loan. First, a major reallocation was agreed upon in late 1976 which increased the financing available for imported fertilizers, seed processing machinery and "other capital works." Second, the disbursement rates for "other capital works" (B-4) and rural electrification (C-4) were increased to 85%. The impact of both actions on loan utilization was not felt until shortly before the extended Closing Date. Typically, TDC was slow in pre- paring reimbursement requests and GOI was equally slow in approving the requests and sending them to Washington for processing. 3.42 A comparison of actual expenditures with appraisa! cost estimates reveals several significant differences. With respect to seed processing plants, the appraisal report estimated some US$3.0 M for turnkey construction, while TDC's total expenditure for SPP modernization was less than $2.0 M (Table 5). 1/ Not only did TDC cut costs by opting for modernization, it was also able to increase processing capacity by 18% above appraisal targets. On the other hand, the decision to modernize significantly delayed implementation of this project component. The appraisal report envisioned the construction of two completely new plants, both of which were expected to be fully opera- tional by 1973. In contrast, modernization turned out to be a much longer, incremental process which was not 100% complete even in 1978 (though all plants were operational by 1977).. TDC, however, could afford to "go slow" with respect to SPP development, particularly in view of the market uncer- tainties prevalent in the Indian seeds industry. 1/ SPP expenditures financed by the Bank were reimbursed under loan disbursement categories B-2 and B-3. - 24 - Table 5: CAPACITY AND INVESTMENT COST OF SEED PROCESSING PLANTS Capacity Investment Cost ('000 tons/year) (US$'000) I. Appraisal Estimate Pantnagar SPP 22.5 1,473 Kashipur SPP 22.5 1,473 Total 45.0 2,946 II. Actual Nagla SPP 20.0 544 Haldi SPP 18.0 771 Matkota SPP 15.0 524 Total 53.0 1,839 3.43 Not included in Table 4 is an additional US$550,000 expended for construction of the Haldi Cold Store. The cold store was not provided for in the appraisal project, though the Bank subsequently agreed to its inclusion as a normal ancillary of seed processing facilities. TDC originally estimated some US$300,000 would be required for cold store construction. As of late 1977, however, the price had escalated to US$550,000, with full completion set for September 1978. 3.44 For farm machinery (B-2), disbursements have also been lower than anticipated. In its original form, the loan included US$5.8 M for the pur- chase of farm machinery and spares - the largest amount of money in any of the loan categories. As of the extended Closing Date, however, the total amount reimbursed in Category B-2 stood at only US$3.5 M (itemized in Annex I). While tractor and combine harvester purchases were greater than planned, the procurement of other farm implements was well below levels estimated at appraisal (Table 6). - 25 - Table 6: FARM MACHINERY - ESTIMATED REQUIREMENTS vs. ACTUAL PURCHASES /a Estimated Actual Type of Equipment Requirement Purchase Tractors 35/49 HP 240 587 50/69 HP 120 120 70/90 HP 40 68 Combine Harvester 145 180 Low-Volume Sprayer 200 0 Power Sprayer 50 0 Fertilizer Spreaders 285 60 Seed/Fertilizer Drill 255 30 Corn Planter 210 30 Mounted Corn Picker 100 0 Stalk Shredder 255 0 Reversible Plow 365 10 Cultipack/Cambridge Roll 230 0 Disc Harrows 285 125 Land Planes 212 10 Tipping Trailers 460 0 Rotary Slasher 0 3 /a Does not include purchases of spares. In the case of tractors, the additional purchases are related to an increase in the number of project participants as a result of converting leasehold farms into freeholds (para 3.15). The reasons for the large discrepancies found elsewhere in Table 6 are less clear. At least part of the difference between estimated requirements and actual purchases may be due to the fact that certain implements were purchased with cash, or by utilizing non-project credit channels. A less sanguine, though nonetheless realistic interpretation is that farmers simply failed to mechanize to the extent envisioned at apprai- sal. In view of the many uncertainties which materialized after appraisal (e.g., land ceilings, low intake prices, etc.), it does not seem surprising that farmers would be reluctant to invest in machinery they considered non- essential. Covenants 3.45 A review of the loan covenants does not reveal any serious instances of noncompliance. Early in the project, GOI refused to issue a license for the importation of certain farm machinery spares (i.e., agricultural discs), but this matter was rectified in discussions with a project review mission. To more precisely ascertain water resource availabilities, Section 5.06 (b) of the Loan Agreement directed the Exploratory Tubewell Organization to carry out "investigations of the aquifers and other hydrogeological features of the -_ 26 - Project area." While subsequent research was undertaken, it appears a case of "too little, too late." In this respect, fuller compliance might have been possible with more frequent supervision missions during the initial phase of project implementation. Exchange Rate Adjustments 3.46 Loan disbursements under-the project were made in 11 different currencies, for a total of US$11.45 M equivalent. 1/ Since 1969, the US dollar exchange rate has increased from Rs 7.5 to Rs 8.6, representing an appreciation of approximately 15%. Repayment of the loan is not scheduled to commence until 1979. IV. PROJECT IMPACT Seed Producer Income and Financial Returns 4.01 Within the seed production area, the impact of the project has been felt chiefly in terms of increased farm income. The rise in net farm income is attributable to three factors: (a) a major shift in cropping patterns away from commercial grain production and toward the production of higher valued seed crops; (b) higher than anticipated producer prices for certi- fied seed; and (c) lower than expected seed cultivation costs. 4.02 . With respect to the shift in cropping patterns, the expectations of the appraisal report were fully realized, with project area farmers react- ing swiftly to the financial opportunities associated with seed growing. By way of example, the largest single cane grower in the Tarai shifted 70% of his land into seed production once the project become operational. On an aggregate basis, the shift to seed growing is no less impressive, going from only 420 tons of processed seed in 1966 to an average of over 28,000 tons per annum at full development (since 1973). 4.03 The impact of producer prices on farm income is illustrated in Table 7, which shows that average (inflation adjusted) prices have been some 3-1/2% above the price levels envisioned at appraisal. In this connection, the most significant spread in prices has occurred with respect to wheat, which accounts for approximately two thirds of seed production. Although rice seed prices have been generally lower than expected (Rs 781/MT vs. Rs 950/MT estimated at appraisal), the effect on farm income has been less pronounced because rice accounts for only 28% of average seed production. 1/ The undisbursed balance of US$1.55 M was cancelled as of the extended project Closing Date. - 27 - Table 7: SEED PRODUCER PRICES - ACTUAL vs. ESTIMATED (Rs per MT) Maize Rice Wheat Appraisal Estimate 1,600 950 1,000 Actual Prices /a 1969/70 1,650 850 1,050 1970/71 1,371 743 1,143 1971/72 1,722 741 1,111 1972/73 1,752 778 1,111 1973/74 1,518 809 1,418 1974/75 1,676 883 1,117 1975/76 1,537 718 1,130 1976/77 1,588 729 924 Average Price 1969/70-1976/77 1,602 781 1,126 Average Seed Production Mix 1969/70-1976/77 5.7% 27.8% 66.5% Weighted Average Seed Producer Price /b Estimated Actual (all crops) 1,020.3 1,057.3 /a Adjusted for inflation. /b Based on adjusted seed prices and average seed production mix. 4.04 Net farm income has been further enhanced by slightly lower cultivation costs. As shown in Table 8, the weighted average seed cultiva- tion cost of Rs 531/acre (for maize, rice and wheat) is some 2% below the estimated cost of Rs 542/acre. - 28 - Table 8: SEED CULTIVATION COSTS -ACTUAL vs. ESTIMATED (Rs per acre) Maize Rice Wheat Appraisal Estimate 625 595 505 Actual Cost /a 562 379 586 Weighted Average Seed Cultivation Cost /b Estimated Actual (all crops) 542 531 /a Based on a 1976 survey of seed production costs undertaken by TDC and UPAU. All cost figures have been adjusted for inflation. /b Reflects average seed cropping pattern for maize, rice and wheat of 12%, 25% and 63%, respectively. 4.05 The combined effect of the three factors mentioned above was tested by recomputing the financial rate of return on a 100 acre farm model (Annex 2). This exercise yielded an increase in financial returns of five percentage points, from 16% (estimated at appraisal) to 21% using actual data. 4.06 Despite their apparent success, some seed growers have begun to shift portions of their land back into commercial grain production. In large measure, this shift was triggered by a recently instituted ceiling on retail seed prices put in effect by the UP State Government. The Bank learned of this action only shortly before the extended Closing Date, whereupon it imme- diately urged GOI to take steps to develop an equitable seed pricing policy to protect the interest of producers as well as consumers. Such a policy is now being developed within the context of the Bank-financed National Seed Projects. Technological Change 4.07 . Implementation of the project involved the application of a broad range of agricultural technologies. While farm mechanization and fertilizer usage are among the most visible, both technologies were relatively well-known prior to project implementation. In contrast, high quality seed growing was a completely new venture for almost all project participants. Preparation and maintenance of a typical seed farm required specialized operations such as land levelling, agro-chemical applications, roguing off-types, and detasselling. In the event of disease, not uncommon in seed plots, additional crop isolation procedures were necessary. These needs were anticipated early on, with the project providing for close involvement of the UPAU. - 29 - 4.08 The project provided for a sub-surface irrigation distribution system to service some 41,000 acres. 1/ The system was never constructed for reasons linked more to farmer uncertainties (para 3.27) than to technological constraints. In contrast, significant progress was made on the drilling and electrification of shallow tubewells (paras 3.25-3.26) which, by providing an assured supply of water, increased the number of farmers able to partici- pate in project seed growing activities. Employment 4.09 Job creation was not an important objective of the project. Rather, funds were provided mainly for machinery, farm development, inputs and exten- sion to be supplied to TDC farmer shareholders. The number of TDC shareholders has averaged 1150 per annum since project inception. On occasion, TDC has accepted seeds from non-shareholders (e.g., jute seed in 1973 and wheat seed in 1977), but the number of individuals involved has not been particularly significant. 4.10 The combined workforce at the various TDC plants is less thin 100. If managerial personnel are included, TDC's total staffing approaches 160. Additional seed testing and extension support is obtained by utilizing UPAU personnel. 4.11 A hardly considered, though possibly significant employment effect concerns changes in the demand for seasonal labor. When sugarcane was the principal crop, harvesting was typically done only once per annum. With the transition to seed growing, however, it has become commonplace to cultivate two or three crops a year, for which seasonal labor is utilized during both the sowing and harvesting stages. In the weeks immediately following the kharif (wet) and rabi (dry) season harvests, labor demand reaches a peak, with farmers engaged in seeding and land preparation, and processing plants seeking to employ additional laborers for seed intake, cleaning and storage. The benefits from such added employment, particularly among low-income group-, may be significant, though this issue was never taken up during appraisal. Similarly, the appraisal report did not highlight the additional employment created at harvest time to handle the increase in crop yields realized by more than 400,000 farmers using project seed 2/. Economic Performance 4.12 As appraised, the project economic rate of return (ER), was cal- culated to be 17%. Perhaps because it was the Bank's first seeds venture, the estimation of benefits was restricted to those perceived as arising directly from planned investments, namely: 1/ Approximately one quarter of the Tarai was irrigated prior to imple- mentation. 2/ Assumes a peak seed reception area of 830,000 ha, with the average farm using project seed on about 2 ha. See para 4.15 for details. - 30 - (a) increased value of seed over commercial grain production; (b) increased seed production due to on-farm investments in land levelling and irrigation; and (c) reduced seed farm operating costs due to mechanization. 4.13 Following this approach, a recomputation of economic costs and bene- fits based on actual data (Annex 3) gives an internal rate of return of 18%, slightly higher than the rate estimated at appraisal. The slight upward shift in the ER is in large part attributable to the higher producer prices and lower unit cultivation costs discussed earlier (paras 4.03-4.04). Al- though gross project benefits (i.e. processed seed sales) were lower than expected due to a reduction in planned seed production, this loss was more than offset by investment cost savings stemming from modifications in the design of seed processing plants (para 3.42) and from cancellation of the proposed irrigation distribution system (para 4.08). 4.14 A major long-term benefit--higher all-India foodgrain production-- was purposely not included in computation of the project economic return during appraisal. In subsequent years, however, it was recognized both with- in the Bank and GOI that benefits to the economy from seed projects could be more appropriately measured by quantifying the value added associated with incremental foodgrain production. Consequently, this broadened approached has been used in appraisal of all subsequent Bank-financed seed projects in South Asia, including the two National Seed Projects in India. 4.15 The estimation of dGwnstream production benefits in the Tarai Seeds Project leads to a significant increase in project returns. In this respect, the average TDC seed output of 28,000 MT since 1973 has been sufficient to service about 555,000 hectares of cropland annually 1/. If TDC production continues to expand, it is estimated that the seed reception area will reach some 830,000 hectares by 1981/82 2/. The incremental yields on such land stemming from the use of high-quality seed have been estimated as follows: maize 225 kg/ha, wheat 150 kg/ha and rice 135 kg/ha 3/. Based on these co- efficients, current incremental crop production can be estimated at about 88,000 MT per annum (Table 9). If the impact of soybean, pulse and other TDC produced seeds is taken into account, 4/ this figure increases to approximately 90,000 MT per annum. 1/ Based on the following seeding rates: wheat, 0.100 MT/ha; paddy, 0.030 MT/ha; and maize, 0.015 MT/ha. 2/ Annual seed production throughput is expected to increase by 6 - 7% per annum before stabilizing in 1981/82 at about 42,000 MT (80% of total TDC seed processing capacity). 3/ - Derived in India-Second National Seed Project, Report No. 1725a-IN, May 17, 1978. 4/ Average production of these seeds has been about 1,150 MT per annum since 1973 (see Table 2). - 31 - Table 9: ESTIMATED INCREMENTAL PRODUCTION FROM THE USE OF CERTIFIED "TARAI" SEEDS Average 1973/74 - 1976/77 Incremental Yield Seed Reception Incremental Coefficient /a Area Production (kg/ha) (ha) (MT/annum) Maize 225 92,867 20,895 Wheat 150 167,550 44,160 Rice 135 294,400 22,619 Total 554,817 87,674 /a For wheat and rice, yield increases extend over a four year period on a decreasing scale, as reflected in Annex 4. In the case of hybrid maize,.annual replacement of seed is necessary. 4.16 In monetary terms, the estimated net benefit associated with added foodgrain production ranges from between Rs 13.7 M (US$1.8 M) in 1970/71 to Rs 47.8 M (US$6.3 M) in 1988/89 1/. Inclusion of these benefits in project economic calculations (Annex 4) boosts the rate of return to 42%, compared with 17% estimated at appraisal. Additional benefits from farmer to farmer exchange of subsequent seed generations produced from TDC seed would also be significant but have not been quantified in project benefits. V. INSTITUTIONAL ASPECTS Organizational Performance - TDC and UPAU 5.01 Successful implementation of the project was dependent on actions taken by a number of organizations. These included TDC, UPAU, SBI, SEB, ARDC and the GOI Ministries of Agriculture and Finance. Day to day manage- ment decisions rested with the Chairman and Managing Director (MD) of TDC. Since project inception, all TDC Chairmen have served concurrently as Vice- Chancellors of UPAU. Similarly, TDC MDs have concurrently occupied the position of UPAU Dean of Agriculture. This arrangement has worked well in linking the top management of TDC and the University, though the Bank was somewhat skeptical at the outset about the practicality of an interlocking directorate. Fortunately, the initial choices for Chairman and MD (respectively, Dr. D.P. Singh and Dr. R.L. Paliwal) were extremely capable individuals who managed to handle the responsibilities of their dual roles with considerable success. Both Drs. Singh and Paliwal remained at the helm of TDC/UPAU for over five years, during which time the reputation of both organizations was significantly enhanced. 1/ Grain values based on international prices expressed in 1969 constant dollars. The estimation of value added is net of any incremental pro- duction costs associated with the use of certified seed. - 32 - 5.02 In subsequent years (from 1975), TDC experienced difficulty in finding and keeping high-caliber management talent in the project area. Between August 1976 and December 1977, for example, TDC's turnover included three Chairmen, two Managing Directors and one Chief Financial Officer. To some extent, the high turnover appears related to political factors over which neither TDC nor UPAU had much control. A possible contributing factor, especially for TDC staff, is the Corporation's status as a private under- taking. Most employees consider it advisable to transfer to a government or semi-government body if the opportunity arises, since these are perceived to offer greater employment security. 5.03 The management difficulties experienced by TDC were cushioned to some extent by the integral involvement of UPAU in project implementation. As a state institution, the University offered employment stability to a corps of specialists who provided TDC with a range of essential technical support services. The UPAU Land Development Division, for example, assisted in the preparation of farm development plans for project participants. 1/ Other staff, mostly plant breeders, were given responsibility for the main- tenance of breeder seed and the production of foundation seed, key functions in any seed production endeavor. Even the credit component of the project benefitted from the expertise of UPAU personnel, who were assigned to local banks (SBI and UCB) to ensure the proper evaluation of loan applications. 5.04 One of TDC's greatest assets is the reputation it has acquired over the years for high quality seed. Here again, the University (specifically, the UPAU Seed Testing Laboratory) deserves some of the credit. Each year the laboratory samples some 10,000 seed lots for germination, purity and other indices of quality. TDC's reputation has been further enhanced by the efficiency of its sales and distribution network, which has expanded from some 200 dealers in 1970 to over 2,500 by project completion (para 3.20-3.21). 5.05 An indirect benefit of UPAU participation in the project is that such participation has become a vehicle for introducing promising students to the seeds industry. UPAU cooperated in this respect by setting up special graduate and undergraduate programs in Seed Production Technology. Not surprisingly, a number of program graduates today hold positions in TDC and other seed production enterprises throughout India. Supporting Services - Credit Institutions 5.06 At appraisal, it was envisioned that the SBI would be the primary channel for credit distribution. Potential credit recipients were identified as land levelling contractors, TDC, the State Electricity Board (SEB) and project farmers. From the outset it was made clear that project participants were not in any way obligated to use SBI. This policy was directed primarily toward some of the larger Tarai farmers who had built up credit over the years with private financial institutions, such as the Punjab National Bank and the United Commercial Bank (UCB). The numbers were few, however, and SBI was exp'ected to pick up most of the.business. 1/ The preparation of such plans was required in the loan agreement. - 33 - 5.07 Under the terms of the project agreement, SBI was restricted to financing 75% of farm machinery and 80% of land development expenditures. On seeing this restriction, private banks in the project area attempted to increase their share of the market by offering to provide 100% financing for machinery and/or farm development expenditures. By 1970, SBI lodged a complaint with the Bank that this policy amounted to an unfair practice, not to mention "bad bank- ing", and that steps should be taken to make SBI the exclusive credit channel for project participants. The Bank's position on this issue was one of concern about the 100% financing policy; at the same time, however, it was felt unwise to support action whereby SBI would be given a virtual monopoly over banking activities in the project area. 5.08 Before the Bank could respond formally to SBI's complaint, GOI decided to nationalize all Indian banks. Once the nationalization became effective, no bank in the project area was allowed to offer better financing terms than SBI. The Bank reacted to this new set of circumstances by amend- ing the project documents to provide for the participation of UCB on an equal footing with SBI. 1/ 5.09 During the period 1970-78, SBI increased its branches in the project area from 13 to 17 (as envisioned at appraisal), while UCB branches grew from 2 to 5 (versus no additions envisioned at appraisal). The increase in agri- cultural lending generated by the project led SBI to assemble a special unit at its Kanpur Regional Office to facilitate loan processing. The eight-person team covered a wide area of agricultural expertise and appears to have had a positive effect on SBI's project-related credit activities. At the local level, each of the 17 SBI branches was eventually provided with a technical officer (M. Sc. in Agriculture) for loan appraisal and five loan officers for post-lending supervision. 5.10 Types of financing offered by project credit institutions included working capital advances (seven months), installment credit (five years), medium-term loans (10 years), and debentures. The latter instrument was provided to enable SBI to finance SEB investments under the project. SEB usually borrows through debentures which are guaranteed by the UP State Gov- ernment and repayable in 10 years. 5.11 Despite the efforts of SBI and UCB officials to manage a smoothly functioning credit system, occasional problems arose which required changes in lending policies. By 1974, for example, certain project participants discovered that they could fully repay short-term loans bearing higher interest (e.g., 13%) at the expense of overdues on medium-term loans which bear lower interest (e.g., 9%). Loopholes such as these were typically exploited after poor crop years, at which time the monitoring of credit procedures was found to be particularly important. On balance, however, the repayment experience has been very satisfactory, with no complaints having been received from lending banks in the project area. 1/ The full significance of UCB's active participation in the project was not recognized until sometime later, when additional resources were required to satisfy TDC's working capital and investment financing needs. -'34 - Training and Technical Assistance 5.12 At the preparation stage, a wide-ranging program of training and technical assistance was envisaged. By appraisal,.however, GOI had made clear its lack of enthusiasm for such an approach and the training component was dropped altogether. In subsequent years, a handful of seed technicians were trained abroad, but these were funded on a grant basis by USAID and the Australian Government rather than from loan proceeds. 5.13 The use of expatriate technical assistance personnel was similarly limited at the behest of GO. Thus, even though the project included a loan category for "Consultants", none of the allocated funds ($150,000) was ever utilized. When it became apparent in 1976 that these funds would remain un- disbursed, the allocation scheme was amended to transfer the proceeds to category A-1 for additional fertilizer procurement. 5.14 The Bank utilized the services of two short-term consultants in 1970-71, but these were needed in connection with processing plant design issues, rather than for training purposes. A subsequent three-month visit 1/ to the Tarai by a University of Illinois seed certification specialist did involve training,-for which USAID provided grant support. Reporting Requirements 5.15 Soon after effectiveness, the Bank transmitted to project author- ities a comprehensive "Outline for Quarterly Progress Reports". A reading of the outline shows that the indices of performance established by the Bank were almost exclusively quant:.tative (e.g., bags of fertilizer used, tons of seed processed, etc.), with little or no emphasis placed on monitoring the impact of the project on target income groups. One can argue, however, that this was the first time the Bank had dealt with the complexities of a seed project, and that the notion of monitoring broad, socio-economic benefits was not a high priority issue at the time. In this respect, subsequent Bank- financed seed projects in India have placed greater emphasis on project monitoring. VI. SPECIAL ISSUES TDC Capital Budgeting 6.01 While TDC's capital expenditures have risen dramatically since 1974/75 (Table 10), the Corporation has had difficulty in securing the neces- sary funds to support such an expansion. In the face of impending shortfalls in commercial credit, TDC has, on occasion, utilized working capital to finance planned investments. This practice ultimately contributed to delays in payments to growers for raw seed in 1976 and 1977, prompting TDC's manage- ment to reassess corporate financial policies. 1/ April-June 1972. Table 10: TARAI DEVELOPMENT CORPORATION CAPITAL EXPENDITURES, 1969/70-1977/78 Year Amount 1969/70 0.56 1970/71 1.27 1971/72 1.73 1972/73 0.21 1973/74 1.06 1974/75 3.08 1975/76 4.93 1976/77 7.49 1977/78 4.63 (estimated) 6.02 The new Chairman of TDC has attempted to address the financial prob- lems cited above by: (a) reducing TDC's 1977/78 capital expenditure budget from Rs 11.1 M to Rs 4.6 M 1/; (b) negotiating with SBI and UCB an increase in TDC's cash credit ceiling from Rs 20 M to Rs 25 M; and (c) increasing the Corporation's share capital base by Rs 5 M. While the full impact of these actions remains to be seen, evidence available thus far suggests that TDC is moving toward sounder financial management policies. Research and Technical Services 6.03 An important aspect of the project concerns the seed research and development role of the.University. Thanks to the linkage f.orged between TDC and UPAU plant breeders, the latter were rapidly made aware of farmer needs and problems in seed reception areas throughout India. This feedback in turn formed the basis for determining priorities in the UPAU breeding program, which received wide acclaim for its practical relevance within several years of ini- tiation. The principle achievements of the program are summarized below. 6.04 Wheat. A recurrent problem was that several of the most important varieties were highly susceptible to rust (i.e., yellow, brown and black-stem rust). Accordingly, developing resistant varieties became a major research objective. By late 1973, three resistant varieties had been bred by UPAU and were.being marketed by TDC, namely: (a) UP-301: highly resistant to all three rusts and very popular in several grain growing areas; (b) UP-310: high yield, high resistance to black and yellow rust but only moderate resistance to brown rust; (c) UP-215: good yield, high resistance to all three rust varieties. 1/ The reduction was approved by the Board of Directors on September 18, 1977. - 36 - 6.05 Another rust-resistant and high yielding variety, Sonalika (RR-21), was extremely popular in the early 1970s and was widely distributed by TDC. Unfortunately, Sonalika became susceptible to a seed-borne disease (Loose Smut), and by 1976 urgent steps were under way to replace it with other varieties. Here again, the University filled a crucial role in setting forth procedures (principally Vitavax application) to prevent the further spread of the disease. 6.06 Rice. A principal production constraint has been infestation by bacterial leaf blight, bacterial streak and blast in upland areas. To tackle this problem, UPAU scientists have developed and tested over a dozen new varieties, some of which have been released to TDC for multiplication on a large scale. IR-24, a fine grain variety of paddy which has the same high yielding potential as IR-8 and Jaya but is earlier in maturity, was given to TDC farmers for the first time in kharif 1972. Other fine grain paddy varieties added to the TDC production program include Sona (IET-1991), Ratna, Pusa 2-21 and Kavari. Pre-release field trials are also under way on UPRI 71-12, which is resistant to bacterial leaf blight. 6.07 Maize. The multiplication of maize seed in the Tarai has always been uncertain due to the possibility of excessive rain during the growing season. Nevertheless, the UPAU breeding program has contributed to produc- tion improvements with the recent development of two early maturing composite varieties: Tarun and Vikash. Both varieties are now being distributed to selected farmers throughout India. Table 11: UPAU PLANT BREEDING RESEARCH PROJECTS, 1957-1976 Year Project Initiated Principal Investigator(s) Maize Breeding 1957 Dr. B.D. Agrawal/Dr. I.S. Singh Wheat Breeding 1966 Dr. T.B. Singh/Dr. P.L. Gautam Soybean Breeding 1968 Dr. B.B. Singh Oilseeds Breeding 1968 Dr. B. Rai Sugarbeet Breeding 1968 Dr. P.S. Bhatnagar Rice Breeding 1969 Dr. J.S. Nanda/Dr. R.C. Chaudhary/ Dr. B.N. Singh Forage Sorghum Breeding 1969 Dr. D.L. Singhania Pulse Breeding 1970 Dr. B.P. Pandya/Dr. M.P. Panday/ Dr. D.P. Singh Triticale Breeding 1971 Dr. K.P.S. Chauhan/Dr. S.C. Mani Sunflower Breeding 1971 Dr. Basudeo Singh Brinjal & Peas Breeding 1971 Prof. R.D. Singh Bajra & Millets Breeding 1972 Dr. Amarjit Singh/Dr. D.V.S. Tyagi Forage Legumes & Oats Breeding 1972 Dr. S.N. Mishra Fruit Breeding 1972 Dr. I.D. Singh Cauliflower & French- bean Breeding 1973 Dr. Hari Har Ram Sugarcane Breeding 1974 Dr. A.Q. Khan Sorghum Breeding 1976 Dr. Rameshwar Singh - 37 - 6.08 Other Crops. Aside from the three crops discussed above, UPAU research projects have had an important bearing on TDC's soybean and pulse seed multiplication capabilities. In the case of soybean, UPAU has developed and released the Ankur variety, a strain reputed for its superior germination potential, high yields and resistance to rust. With respect to pulses, two lentil varieties, Pant L406 and Pant L209, have been recommended for release in the northwestern plains. Research on the production of vegetable crops is conducted at the University's Horticultural Research Centre near Pantnagar. The results of this research have enabled farmer shareholders to diversify their seed growing to include tomatoes, eggplant, radish, spinach and chillies. With vegetable growing on the increase, TDC has recently ordered special equip- ment (e.g., gravity separators and cleaners) for commercial scale vegetable seed processing. VII. BANK PERFORMANCE 7.01 An overall assessment of Bank performance would have to be charac- terized as positive, particularly when one considers that projects staff had no previous experience with the design and implementation of seed project.s. 7.02 The "newness" of the project concept significantly influenced the way in which the Bank responded to important implementation questions. For example, when TDC announced its decision to modernize existing SPPs rather than to construet new plants as previously agreed (para 3.09), the Bank adopted an open-minded approach to the issue and ultimately agreed to the TDC plan. While other examples of flexibility may be cited, the point which emerges is that the project was clearly a learning experience for the Bank, the lessons from which are now being applied to other Bank-financed seed projects in India 1/. It can also be agrued that the willingness of Bank staff to consider seriously a number of proposed modifications in the pro- ject was encouraging to both GOI and TDC and contributed to a mutual respect essential for successful implementation of the project. 7.03 The Bank's ability to resolve implementation problems may have suffered somewhat as a result of poorly timed and/or infrequent supervision missions. Although missions every five or six months would have been desir- able, considerably longer intervals elapsed between some visits (Annex 5). Compounding this problem was the high turnover among Bank staff, most of whom never visited the project more than once before responsibility was delegated to someone else. In contrast to the lapses in supervision support, the quality of administrative support provided by loan officer staff appears quite satisfactory. An examination of project correspondence reveals that enquiries from GO, TDC and other interested agencies were answered on a timely basis and, if necessary, followed up in the course of scheduled field visits. 1/ Specifically, the two National Seed Projects (which commenced in 1976 and 1978) are based on experience gained from TDC's success- ful operations. - 38 - VIII. CONCLUSIONS 8.01 In the final analysis, the success of the Tarai Seeds Project appears related to three key factors: Design. The conceptual framework of the project was sound. Farm development, processing plants and all other components were oriented toward the fulfillment of a single, well- defined objective: to grow, process and distribute high- quality seed for Indian farmers. Timing. The timing of the project was extremely appropriate when viewed in the context of the so-called Green Revolution. By the early 1970s, thousands of Indian farmers were anxious to purchase quality HYV seed but were unable to do so because of shortages in supply. Thus, the project helped eliminate a critical foodgrain production bottleneck. Support. The project was well received by virtually every agency and group affected by it. This included GOI, TDC, UPAU, farmer shareholders and seed users. Widespread insti- tutional support kept bureaucratic constraints to within acceptable limits, while extension and farmer support helped ensure the continued production of high quality seeds. Though project execution was delayed on several occasions and seed output was lower than envisioned, these difficulties did not jeopardize the overall viability of the project. As indicated earlier, an accounting of actual costs and benefits shows that the project exceeded both the economic and financial returns estimated at appraisal. - 39 - ANNEX I INDIA TARAI SEEDS PROJECT Itemized Farm Machinery Expenditures Quantity Country Cost Purchased Item of Origin CIF Delivered Tractors 59 IMT 555 (55 HP) Yugoslavia 93 IMT 535 (35 HP) " 177 MF 135 (47 HP) UK 45 MF 165 (62 HP) " 43 MF 178 (75 HP) " 141 IH 276 (40 HP) 176 Ford 3000 (46 HP) " 16 Ford 4000 (60 HP) " 25 Ford 7000 (90 HP) "o 775 Subtotal 18,633 20,461 Combine Harvesters 18 John Deere 630 West Germany 1,951 3,102 22 Laverda M-84-R Italy 1,232 2,002 10 Claas Jr. West Germany 345 550 80 Vicon MS-70 India - 1,956 180 Subtotal 3,528 7,610 Implements 50 Offset disc harrows Australia 170 259 10 Cornplanters (2 row) " 23 35 20 Cornplanters (4 row) " 89 134 60 Fertilizer spreaders " 137 203 10 Reversible disc plows Spain 59 89 30 Fert./seed drills UK/Australia 220 355 3 Rotary slasher UK 14 22 75 Harrow plough " 270 454 4 Bailers West Germany 93 148 7,576 Discs UK - 506 10 Land levellers " 228 388 n.a. Miscellaneous spares various 4,000 4,600 7,848 Subtotal 5,303 7,193 Total 27,464 35,264 NOTE: Rs 27.5 M at 0.13/$l = US$3.5 M (the amount actually reimbursed for farm machinery). In contrast, the appraisal estimate for farm machinery (foreign exchange) was Rs 39.0 M, or US$5.2 M. Including local costs, the original loan allocation for farm machinery and spares (B-2) was US$5.8 M, as shown in Table 3. - 40 - I I§ § s \*.J A.- ' I i .'." §§ L~ T. ~ ~ . , -e - ~r si - s 8 I.: • 6n.e ae... i.i.s.••• .s p.sies se,..on se wsves see -s ..oo.. es g Iq pesetse •••.q a*e (es toje seeeea T ) *••I *0e 0 0!s.-- I!Y ••inse se . ete ers see05- a as . s .*on 1e;.•q isu .p••e e .'. .,w o em • 6in'e denlee o.n> wa ..'ou sa.ro -..d .1.,... -" p.g / *u»,100 9-,g*nense pue :m~ Jo $'q 'tnt IM oo0'e9 ese *O pe010ne IslO."-• 11Oj se ,e'é.We teue -0/561 th. -- ..d wes .n a .. 6q .~...p O e-secre.d *t ..h-Po.es o5no.Io., een 'e..J o•+o..'.o. om -AA/9L: 0V'-'9 -,•. r-.r.. -"neu"', i.""' - '•* I P¥% ~ ~ ~ LJa -a/16 *fllU w~f. ~~4 .1UJ61~qIIZIýr uama gr uni 03 gia 'wmntu of MMg9 0 oPLf 9 069(11 o 0 '"'LT o,1m9t91 ot*N9'LI 06, C1 g'6ti'9 1 oc'o19,11 o9,9LE1C oL,og', (OCIL' g) (09 14) oc 198 00*'. (WIMC1) (o'LJA,) (O6Cw*' (nLZ<4'8) .'tsjc -.~•• I 1 09 coC 09U .a8 0169'#8 0g9egC'9 09-169,48 os tLL'ca 09-169'48 og-r',98 06694,48 oCgao'co 09t19C'9 GL-LEE'a 0'.eol' s oIVtIt ot9u'gs oo'og'Za oV619* o t og ovgCt -3*n' t-& 0 "6 4a 09,16948 09,169'48 09, 169'N 09-169'4 09 169,4a 09*1694 o9169' 06C 0 kua 09v'E e8 CK•*k 'ol 0g'6xi'0a cuL'90ts o(*go t o9LC9'LI OL610*91 00'909'91 OV W'It 00019't -3 T-&''b 10 oWO wo o o oem w owoo oo o coW - Wco5 WM Wm oo o-o03 o-oa c-ona ä oE- cWo-. Wt wm- Eo- wä- W(x go-0U OL -:n"å J ~ le'e .ne E 29°Ct o ZL9°1 0< Lt9'C1 0( E19'C1 O<UE 9'CI o< a°CIt '9°C1 o(CL9'CI of 09' og0a 'C5 og'CC9'9t 000') 'of j01 0 26t' coW' 0o616'11 00L;)6'01 o:'C11 000CN'6 iYCWL 1 09,C4 09Clt ogC4L og C4t ogCNL og'CkL 1 09C4L 09Ct op'C4L 09CNL 09,c4 og.C4L 4. * .96 OC-36L 000C 000o 0'C 0%'69, 0NN o' Qq£ '. "..... C c9 li -C1 og<< og CaL NCi tg't 09*cC Cg ^C4t 09C1 09°Ct1 tCi, eg*to1 0C-916 oC6W oo-C 01-6tCOLC 00,0cc OuNc 01'649 09t " ' - C-C1 06<CC' Ci'< C<CC' C ' CC 6C oWIWC cCCNCC 06L'1N* 2*tCg' 06-oC4V og-1il' ooVfe678 0 -7'1 •1 '0"0'1 09'2"( a-' G"IN OVM .." 09-f11'f C9-191' C9-19f Vg'XQV( 09TLO 09*19gI 0ggL't 0gg'C 09'19C' vg*IJL% oQg9I o19V' ol,N9'L 04110' OL'9g's oT*919' g-g(0' ologo'S OL999'ut r"g, w"ne* 00 919,6) o-wn01 - u G6' c - 09,90'1 - o6(6't 0181 00'691c9 oMIilt o9EK19 09-9"', 09L16%'1 0 oCI'9 00'109'9 08,0'c'9 Og 'L 00'9%8' s«(3 5 e* t l - - - - - - - - - - - w'on oo,0 coM-o co0"a ' o- et"r :- S - - - - - - - - - - C00t91 000? 00-co9 0001C w m090'4 e1 1 s "o i ('OeCC'6) wo-<g9C - o99'9 - 0oeCE' - .9%9'4 - 016911 - - - O-06C 00?1 0so4 06'oza9 og* ' o(M' - WNI • . - - - oL-LoI oLLgo' 1 Lgo'I '1 ' oL'go' - - 5* 'l Loo-, i -C -- - - - - - -oo co-occ'f o-C's c- c •1 o-oCCa coC'a - - - - - - - - - - - 08116'1 09 999'8 cogl' oo'11 0z0' og-CC6 - - - - "*ntG6 * " o1C6'*1 OCC6C o-C6's oCC614 OLCC6'1 OLC61I oLCCU'14 oVCG619 oC6OVg( oo'669'9( 08-a6'i 01'L 0 9 otnwIo' o696vgn 0o9'94 eL6o6a O U c4 WsC 06-INCC o0C649C11 oc".it 113-1 1- 1 fl-& oi'6.o' 09 690 9g -09 O9 0I669 00t49'w oW9W 7U«90h "5og Ig 070669 Mg9919g -oga "-w -g'wwg - 09 o00 o5"06 ow '9'9 o o<MW9119 wöXiL91 0o O-OVC-c o( 6 Wo-'7... OU-CC9,0c, OV-C9,0c olIGC9'oc 0o CC9,0c 01 CC9'oc OU-CC9'oc or*co'oc OUCC9'oc o1-6gV'I4 oq'6'CI# 09*619'oq og-~aCK 06'966'9W og·oEt'14 oa-doq'vf OCCW'ac v9'wlt'Lö OCI"s'18 on*U,91 o,, M9'cl M.."- 1.1. - - - - - - - - - - - - - - ea- orgtC co-I9T og-g-r ••'.9L 1e1-a--• jo•* · 0T'CW*'c OUCC9,0 9'C 01' C, 9'0 C DUCC 9'o01 9, OV *9 C o0 01C*9 G*oC 01 69'L 0460 6 0 *>C4 09'619'0 0g-og'C 06o966*98 090 '14 o606'g of*Cggl'a 0o0C'98 00*c99,06 o C 909V'41 s" 'e 6e/ge~1 ge-90/0:i7¢iit967Wis i7Wis Eg/nom lTÖWi 6-97=1 é=n ii7 Zzn9igu frTf= U7TWii EVUM trelrg/u iTog01z r iM ingii ii -to.,ep.d Jo e1*0* t*.Eoad eneq.s 30 se 'et 1-en I*1*en Iju . *I et ** on o. * -ennene.na. do's te" - ""J •®" l., 1" ~ ~ . ,.i *(eanoin at J. A~ lum lno) F 7j. , --"-- u r ,i P'- (iI,i 1 ouel> .•n,u.T Ls..sj aofew anj easodo-r noin~ was ,..:a., .••td ••s 'eserTop lVeIs'* 6%t VI 9••••sax ••s tet'sI" 'eses 9.'e ••nt.. esss lIT '(01*ne 0o*onead t' e•ru 3* 0001) rec -*1" e 0' ' u ,te ~ 'Ii .,m .1 . e. .. . . .e o . , .. '- Imed -e. we---t-P 696 .• • ---&- I -• - -• • o n - x ."-n e -( -a1d '-. 'o to Oot• Q n , u 0nn teleet el, t :Q 1,1 lignmL gq 'E0.-iu Jo -YO 'IYintILI- t-994' 646L*L VICt'S C NL'Lli 11 L'1gt'g 8'69 9 L'Eaz'U 6't'6a L'610e! L*l69 so'1 EW' UVI 6'' 9'tt (L' ) (9*llVo'(V (W9(a'C) (WoLIC) ('o't) till"'t l.1 (NCC5 CoL'g C-9t'l;l L%'lg 6*9t' rgc-' C'9(t'4 L- 60'9 5'C69*' 4·t6'ý C'066'9 - 'o0'L VoS'6 9-t4L'o1 gg6'ot 6eKI'g o,gCg 'gCg'L 6,160'0 O090 ' 59 ' £,9ct'% 59 '4 ' i' C 1'c%, C'95t'4 f.*9(.1' 14 C91.1'N ý'c6g's , o'I9'E 0,1tWc'i "'o' £·o'e o* If.'EC o' is' 0·,J' L'r.' O,QNC'l c'' ¥ o',11'0 '94t' CØ951' 691* 6Ø'9£ 1°9<14 1'96'% C'95t' C%'94 UCT' O'19 Ct%* Ooo r-j o'LiI ol''4%' 6,,20 01': oU,0 Cy01W o'LIO1' *9o'jt. (O919,61) o' ' • ' g' - 9,16t 0,6Wu o l6C' licl 9'KL19 91'97" g1 L Loit'9 o ? >ID9'9 E015'9 'JV O' ' *L 0'9 .~~~- .0 ., - -, -Qz - -. - ~o1oo0ook -0 . . - - - - - - o'Ogi % a009 o'o 0'i. oo ' l ooG6~ o'og9' - o'999't - o'5SE' - o'%C's • 0(60'1 - • -6 s't' g'o' 6'oL'9 gæ 't ~ ('*gt'Is. ~ LLI.,- - * - -'a' l'I' L-Lgo'I igo' L"O'tUZZ. - ' - - • -• • • • (0'000) o ' o'o61'g o'0oS' o'of6'a o- i'o 0."$8 0-056' • - - - - - - - -T6 '99'E O'gli O'LI' Or' 9'66 z *w 'L6 o-964'9 9•LI%ø9 o-56t'00 o'Cos'C0 g'6tt'E 0·Czs'0 ~6:1'r. I6tt't 1-Co'U .Const 9'o9o'%u r'tee 0-6to'C vi00'z z'o'g a·gtz' g•ty' 0'tar'o (1'ECo't *tueJ.i ti. i 1~1 69- "k4 9 -' 69 g N199 M669'9 V669' T 6699 669' 99 9 459 9 ,-669's It' C''6'_ 0s''6 E '9 'CW, EoC9 o'liC T' d o-go'6 l161'(9 6010oo'9 116'li9 4O'-9 C-611'e9 t-Cr'6 ggt'90 '619*r C Cli 9*-L'It 0'096'Jc o'9 '9 'ab'E 6'49'l t..DC'I UeaS'g 6'99C'" 9g,' t - - - ----- - - • 000 .9o.rOlE'915 -91 1 O9'1'*9 qe T,f 6-uLx& 9 4,46N,9 VN o'9 a*611'eg I*ct5 66 99o', 8 619',n C,ogc 9% lt g0«o96'e - ¥<', -6L9 ¥'^C15 <911 6I1Lo0 G-Ll,11.L'9 - I> &A ILøfl L IX:e 69/ggst gg96 1i7/[, 6 1 99/496 l 7i i g6 Eg/5j6: 29/196 :0/060 og/6MtZ 6I7/gut gl./W4 ITE1i 9/7f5Mt (QALi TIZi (27i1Ki #<tA: U7iTl Eiff7 gåli"" I. .": J.ww0tlU W1gT i0 v Ww iJ'rii 叩