India - Third Rural Electrification Project

Documnt of The World Bank FOR OFFICIAL USE ONLY Report No. 13289 PERFORMANCE AUDIT REPORT INDIA THIRD RURAL ELECTRIFICATION PROJECT (LOAN 2165-IN) JUNE 29, 1994 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS At time of appraisal (1981): 1 US$ = 8.0 rupees At time of completion (1988): 1 US$ = 14.0 rupees ACRONYMS AND ABBREVIATIONS ERR = Economic Rate of Return FRR = Financial Rate of Return GOI = Government of India LRMC = Long-Run Marginal Cost of Electricity OA Ordinary Advanced Group of Villages OB = Ordinary Backward Group of Villages PB = Project Brief PCR = Project Completion Report PAR = Performance Audit Report RE = Rural Electrification REC = Rural Electrification Corporation Rs = Rupees SAR = Staff Appraisal Report SEB = State Electricity Board SI = System Improvement Schemes SPA = Special Project - Agricultural Pumps SRI Special Project - Rural Industry FISCAL YEAR OF BORROWER April 1 - March 31 FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. )ffice of Director-General Operations Evaluation June 29, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on India Third Rural Electrification Proiect (Loan 2165-IN) Attached is the report entitled "India - Performance Audit Report on the Third Rural Electrification Project (Loan 2165-IN)," prepared by the Operations Evaluation Department. This project was the third in a series of closely linked Bank operations in support of time slices of India's Rural Electrification (RE) program. The primary objective of this RE program has been to provide reliable and economical power for irrigation pumps, as the foundation of an agricultural policy to eliminate the risk of large-scale crop failure associated with periodic failure of monsoon rains, and to establish the conditions needed for the long- term growth in agricultural productivity. Although disbursement met with significant delays, project physical implementation was successful. However, financial covenants were not complied with, and the project failed to meet its sector reform objectives of improving the financial condition of participating State Electricity Boards (SEBs). In fact, RE has been a significant factor in the declining financial health of all of India's SEBs; agricultural tariffs, when collected, averaged less than 10% of the cost of supplying electricity to the medium voltage grid, before accounting for the distribution costs. Accordingly, the outcome of the project is rated as unsatisfactory, its institutional development impact on the SEBs, the ultimate borrowers, as negligible, and its sustainability in most states as unlikely. Robert Picciotto by H. Eberhard K6pp Attachment 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.  FOR OFFICIAL USE ONLY PERFORMANCE AUDIT REPORT INDIA THIRD RURAL ELECTRIFICATION PROJECT (LOAN 2165-IN) TABLE OF CONTENTS Page No. PREFACE .................................................... BASIC DATA SHEET ......................................... EVALUATION SUMMARY ...................................... iv 1. BACKGROUND ........................................... 1 2. PROJECT DESCRIPTION .................................. 3. IMPLEMENTATION AND RESULTS ........................... 2 Outcome .................................................. 2 P roject C ost ... .. . .. .. ...... ... .. ... ... .... ... ... ........ .. 3 4. ISSUES ................................................... 4 RE Schem es Design .......................................... 4 M aintenance .............................................. 4 Demand Side Management and Quality of Service ....................... 5 L essons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Financial Condition of the SEBs ................................. 5 Lessons .................................................. 7 Recom m endations .......................................... 7 The Role of REC in Supporting Financial Reform in the SEBs ............ 8 L esson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Institutional Reform in Time Slice Projects .......................... 9 Recommendation .......................................... 9 The Role of REC as an Apex Onlending Institution .................... 9 Bank Performance . ................ .......... 10 Lesson and Recommendation .................................. .11 Economic Rate of Return ...................................... 11 Producer Surplus ............................................ 12 M icroeconomic Analysis ...................................... 12 5. PROJECT OUTCOME AND SUSTAINABILITY .................. 14 Annex 1: Comments from the Ministry of Power ......................... 15 This report was prepared by Richard Berney, Task Manager with input by Alain Barbu (Finance) and Alvaro Covarrubias (Technical), and secretarial assistance provided by Mrs. Lorna Sibblies. 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.  1 PERFORMANCE AUDIT REPORT INDIA THIRD RURAL ELECTRIFICATION PROJECT (LOAN 2165-IN) PREFACE This is a Performance Audit Report (PAR) on the Third Rural Electrification Project in India, Loan 2165-IN, for US$304.5 million. The Bank approved the loan on June 1. 1982. The loan closed on June 30, 1988, two years behind schedule. The Bank disbursed 97% of the loan, canceling the US$9.0 million undisbursed portion. The PAR was prepared by the Operations Evaluation Department (OED). It is based on the President's Report, sector and economic reports, Country Strategy and Policy Framework Papers, the loan documents, study of the program files, and discussions with Bank staff. A Project Completion Report (PCR) was prepared by the East Asia and Pacific Regional Office of the Bank (Report No. 10780) and was distributed to the Board on June 26, 1992. An OED mission visited India in May, 1993 and discussed the effectiveness of the Bank's assistance with Government officials, and management of the Rural Electrification Corporation (REC), the State Electricity Boards (SEBs) of the states of Haryana, Maharashtra and Rajastan. Their kind cooperation and invaluable assistance in the preparation of this report is gratefully acknowledged. The mission also discussed the impact of India's rural electrification with many electricity users in rural areas of these states. The PCR provides a satisfactory account of the experience of the project with respect to implementation, achievements and sustainability. It concludes that the project was financially and economically unjustified and, with the continuation of current tariffs, unsustainable. The PAR reviews the assumptions underlying the PCR economic analysis and finds that many important external effects of the RE program have not been fully accounted for. In particular, the PAR finds that RE in India has played a central role in shifting the country's agriculture from monsoon dependent dry farming, to irrigated farming, and, in this way, has been a major contributor to a successful national food security program that has eliminated the risks of widespread distress that had accompanied failures of the monsoon rains in the past. The project has, therefore, made a significant contribution to improving the lives of the poorest of the poor. Copies of the draft PAR were sent to the relevant Government officials and agencies concerned for their review and comments received have been attached as an Annex.  ii PERFORMANCE AUDIT REPORT INDIA RURAL ELECTRIFICATION III PROJECT (LOAN 2165-IN) BASIC DATA SHEET KEY PROJECT DATA Appraisal Expectation Actual Total Project Cost (US$ million) 795.3 1813.3 Loan Amount (US$ million) 300.0 291.0 Economic Rate of Return 53 % CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS 1983 1984 1985 1986 1987 1988 1989 Appraisal Estimate (US$ million) 7 141 267 300 Actual (US$ million) 1.7 33.9 103.8 176.6 255.0 287.1 291.0 Actual/Appraisal, % 24.3 24.0 38.9 58.9 85.5 95.7 97.0 PROJECT TIMETABLE Original Plan Actual Appraisal Mission 5/3/81 5/3/81 Negotiations 4/16/82 4/16/82 Board Approval 6/1/82 6/1/82 Loan Signature 6/22/82 6/22/82 Loan Effectiveness 10/21/82 10/21/82 Closing Date 6/30/86 6/30/88 Date of Final Disbursement 2/3/89 iii STAFF INPUTS (Staff Weeks) Through Appraisal 74.3 Through Board Approval 24.8 Supervision 90.2 Total 189.3 MISSION DATA Activity Date Duration Composition Overall Probect Rating (mo/yr) (days) Appraisal 5/81 15* Eng, Ec n.a. Supervision 1 11/82 20 Eng, FA 1, Problem free 2 12/83 15 Eng, FA 2, Moderate problems 3 /85 n.s. Eng, FA 4, Serious problems a/ 4 2/86 n.s. Eng, FA 3, Major problems 5 1-2/87 10* FA, Cons Not rated. b/ 6 2/88 n.s. Eng, FA 3, Major problems n.a.: Not applicable. n.s.. Not specified in supervision reports. a/ Project not likely to meet objectives. b/ Consultant review of SEB finances. Ec: Economist. Eng: Engineer. FA: Financial analyst. Cons: Consultant. Estimate. OTHER PROJECT DATA Borrower: Government of India Beneficiary: Rural Electrification Corporation (REC). Related Bank Prooects Purpose Year Amount Status of Approval (US$ million) First Rural Electrification Rural Electrification 1975 57.0 Completed Project (Credit No.: 572-IN) Second Rural Rural Electrification 1979 175.0 Completed Electrification Project (Credit No.: 911-IN) iv PERFORMANCE AUDIT REPORT INDIA THIRD RURAL ELECTRIFICATION PROJECT (LOAN 2165-IN) EVALUATION SUMMARY Background expansion and improvement investments in 14 States. During this time period, the RE 1. This project was the third in a series of program was projected to electrify 18,760 closely linked Bank lending operations in villages, 543,000 irrigation pumps, 25,000 small support of time slices of India's Rural industries and 600,000 commercial consumers. Electrification (RE) program. The primary Bank funds would support about 3500 eligible objective of the RE program has been to distribution schemes and 110 system provide reliable and economical power for improvement schemes (para. 2.1). irrigation pumps, as the foundation of an agricultural policy which would eliminate the 4. The main sector issues was the weak risk of large scale crop failures (from the failure financial condition of many SEBs and the of monsoon rains) and establish the conditions financial burden that RE programs were for the growth of agricultural productivity creating for them. The Bank focused much of (para. 1.1). its efforts on improving the financial performance of the SEB. Each SEB had to 2. All three RE loans were administered agree to meet minimum financial criteria to by the Rural Electrification Corporation (REC), participate in the loan. Since each State a Central Government institution established in government had authority over its SEBs, 1969. Since that time, REC has been including the authority to set tariffs, they also responsible for administering various RE had to agree to implement the policies programs created under subsequent Five-Year necessary for meeting these criteria (paras. 2.3, Plans and onlending borrowed resources to the 2.4). SEBs in support of these RE programs. In addition to appraising and approving the RE Implementation and Results projects proposed by the SEBs before onlending funds, REC has developed and produced 5. Although disbursements met with standards, guidelines and manuals covering significant delays, the project was implemented most aspects of RE, all of which have been successfully. The project implementation adopted by the SEBs (para. 1.2). period was extended by two years to enable the executing agencies to utilize all the allocated Project Description funds. The time slice of the RE investment program supported by the project was also 3. The loan was designed to support the increased, from two years to five years. During GOI's RE strategy in accordance with India's the five-year period, 83,000 villages were Sixth Five-Year Plan (FY81-FY85). It was to electrified and 2 million pump sets were provide funds for the import component of a energized, about four times the number two-year time slice (FY84 and FY85) of the RE estimated in the appraisal for the original two- V year time period. However, villages' utilization performance with the Central Government of electricity has not grown rapidly. On (paras. 4.8-4.11). average in India, 85% of electricity usage in rural areas is for irrigation pumping (paras. 3.1, 9. However, the Central Government 3.2). could not dictate tariff policies to the States, and agricultural tariff policies were the central 6. However, the project failed to meet the element in the improving SEB demand side sector reform objectives of improving the management policies and financial performance. financial health of participating SEBs. In fact, Average tariffs were below US$ .005 per KWh, India's RE programs had a profound negative less than 10% of the LRMC of electricity impact on the financial condition of all SEBs. production. In addition, some SEBs provided By 1992, demand from irrigation and power for agricultural use at a fixed monthly agriculture accounted for 31% of India's charge, so that the marginal cost to the user was electricity consumption (in GWh), but zero. Rationing, rather than pricing was used in accounted for only 4% of SEBs' gross states with power shortages, to shift away from revenues. The resulting financial losses were at peak periods. Because the Bank did not least $2.4 billion for 1992, equivalent to about dialogue with the States, there could be no 30% of the gross revenue of all the SEBs. agreement on the basic issue of tariffs and These losses were a major contributing factor to without such an agreement there could be no the financial problems that most of India's plausible plan to improve the SEB's financial SEBs are facing today (para. 3.3). performance. The entities responsible for making the difficult political decisions about Project Cost tariffs never accepted ownership of the problem or its solution (paras. 4.12-4.14). 7. The total project cost increased substantially more than did the number of Lessons and Recommendations schemes. The cost overrun of the average on scheme was about 110%. Since the average * In India, the Bank needs to work cost of equipment and materials financed by the closely with individual SEBs and the Bank was lower than estimated at appraisal, the State governments to ensure that they audit attributes the increase to higher than have accepted ownership of the reforms anticipated costs for goods and services needed to establish a financially sound contracted by the SEBs (paras. 3.4, 3.5). electric power sector. The Bank has, in fact, already incorporated this lesson Issues into its lending strategy. It has begun to lend significant sums directly to Tariff Policies individual SEBs where agreement can be reached on action needed to restore 8. The Bank was operating under the financial viability. assumption that it could reform the financial practices of the individual State Governments * For a RE program to be sustainable, by getting Central Government institutions to RE consumers must be charged a tariff obtain the desired policy changes from the that, at minimum, is sufficient to cover individual SEBs. It therefore did not work the cost of supplying electricity to the directly with the SEBs and their Governments system. That is, RE needs to cover its to come to an agreement on how to resolve the operating costs without cross-subsidies. SEB financial problems. Instead, it negotiated If tariffs are below this level, the the financial covenants related to SEB Government will be committing itself vi to a long-term subsidy program which rather than increasing it. They might, therefore, will grow in proportion to the increase provide a platform for reopening in rural energy consumption. In communications with recalcitrant SEBs without addition, it is desirable that tariffs are increasing distortions in the energy sector high enough to also finance a (paras. 4.22-4.25). substantial part of the RE distribution investment costs. Bank Performance Where a government decides that 12. Bank performance during project subsidization of RE is politically or preparation, appraisal and supervision was socially desirable, the method of weak. Project design relied on Central subsidization should be an up-front and Government institutions to achieve State level transparent subsidy covering a portion reforms. Conceptually, this design should have of the distribution investment cost. allowed the project to achieve widespread reforms, without Bank staff having to undertake The Role of REC as an Apex Institution detailed State by State analysis and negotiations. However, this approach has 10. REC has played a central role in the proved to be unworkable in practice. Given implementation of India's RE program, acting India's federal political system, there is no as both a financial intermediary and a source of substitute for entering into a policy dialogue technical assistance expertise. It was the directly with those responsible for making State appropriate institution for supporting the Bank's policy (para. 4.26). onlending program in RE. REC effectively supervised the formulation of the individual Compliance with Covenants subproject RE schemes, ensuring that the subprojects conformed with pre-agreed 13. Under the Loan Covenants, only those parameters and disbursing against actual SEBs who satisfied and continued to satisfy completions (paras. 4.17-4.18). specific financial eligibility criteria were eligible to receive project funding. Covenant 11. With the maturation of India's RE 2.03 gave REC the responsibility for monitoring program, the decline in funding for new the compliance and withholding loans for expansion schemes, and the increasing cost for noncompliance. Most SEBs were seldom in connecting the more remote, last ten percent of compliance with these covenants. The Bank villages with the national grid, RECs will need took no action when the SEBs failed, for four to shift more of its efforts to improving the years, to provide plans for meeting their efficiency of existing systems. Even without financial commitments and REC, taking its cue significant extensions of their RE systems, from the Bank's lax attitude, neglected to SEBs will need to make substantial investments review the financial statements of the SEBs on in system strengthening to keep system losses an annual basis. Using an apex financial within tolerable limits. RECs' studies have institution also was supposed to allow the Bank shown that in most cases energy saving to avoid supervising the SEBs. However, the investments have a high economic return to the Bank should not have expected REC, which SEBs. In the future, the Bank should consider lacked any in-depth financial expertise, to take supporting a shift in focus of India's RE on a new responsibility of supervising the investment program from further expansion to financial performance of the SEBs. Without energy saving, system improvement programs. this supervision it proved difficult to enforce the Such programs might also be of greater interest financial covenants. There were no other to SEBs, since they would reduce energy use, vii significant failures to meet loan covenants tranche conditions to put the borrower on notice (para. 4.27). that certain measurable practices need to be implemented and to require the Bank to certify 14. The Bank also failed to resolve many of that these conditions have been met before the the other institutional and operational problems entire loan can be disbursed. The lack of such identified during appraisal, including excessive conditionality made it easier for all parties to energy losses, poor operating performance by neglect the implementation of the covenanted some SEBs, the lack of metering of electricity conditions (para. 4.20). use at all levels of the operating and low efficiency end use patterns, for irrigation pumps Lesson with high power factor. However, the project failed to include any specific programs to 17. The sector policy conditionality of time ensure that these problems would be addressed slice projects such as this one would, and during project implementation, and the audit should, be strengthened by the inclusion of found no evidence that they were systematically specific tranche conditions, as practiced in dealt with by the executing agencies (para. adjustment loans. Such tranching would have 4.28). strengthened the Bank's hand in its subsequent dealings with individual SEBs (para. 4.21). Lesson and Recommendation Outcome and Sustainability 15. While Apex institutions are useful to ensure effective physical implementation of RE Economic Rate Of Return programs spread over many executing agencies, they cannot resolve the specific operating 18. The PCR update of the SAR's problems of the executing agencies. Greater economic calculations found that benefits were efforts are needed at appraisal to analyze and lower and costs were higher. The PCR resolve problems at the operating level. More therefore concluded that India's RE was work at the local level is also important to economically unjustified. It is the audit's ensure that the implementing agencies have strong presumption that this conclusion is accepted ownership of the project and are incorrect, because the marginal analysis on willing to take the steps needed to ameliorate which it is based captures neither the full the worst of the identified operating problems. benefit stream that has come from India's RE Future RE programs should place special program, nor the full cost and institutional emphasis on programs designed to improve end difficulties of implementing the diesel use efficiency, particularly for non-household alternative, when applied on a nationwide scale. electricity usage (para. 4.29). The audit presents a detailed analysis of assumptions underlying the calculations (para. Sector Conditionality in Time Slice Projects 4.30-4.38). 16. The projects' time slice disbursement 19. The analysis failed to capture: (i) the profile was much shorter than the agreed timing producer surplus that farmers derived from for reaching sector policy objectives related to irrigation; (ii) many of the broader benefits that the actual improvement in the financial came with the transformation of agriculture performance of the SEBs, and disbursement from monsoon dependent dry farming to year conditions were only loosely tied to these round irrigated farming; and (iii) many of the improvements. Unlike in a SAL loan, which costs that would have been associated with a also disburses against time slices of imports, in national irrigation program based on the this time slice loan there were no specific viii introduction of 8 million diesel pumps in the 22. However, the pricing policies that past two decades (para. 4.33). accompanied the RE program have led to a substantial misallocation of resources and to an 20. With the introduction of over eight intolerable financial burden on all the country's million electric pumps over the period 1974 to electricity producing and distributing 1993, irrigated agriculture has completely institutions. The audit agrees that both the changed the nature of India's agricultural sector, economic and financial cost of RE could have on which some 70% of the India population still been reduced by eliminating agricultural rely, either directly or indirectly, for their pumping during hours of peak system demand livelihood. The result has been that when the providing price incentives to encourage farmers monsoon rains failed in 1987, and India to use more efficient pumps. If pumping were experienced the worst drought in fifty years, restricted to off peak nighttime hours, economic agricultural production fell only marginally, and and financial costs would be sharply lower. the famine and the concomitant massive However, electricity for pumping has only been population dislocations of the rural poor that, in restricted by the competing demand by other the past, had accompanied failures in the users in a greatly overloaded system, and there monsoon rains did not occur (paras. 4.39-4.40). have been no successful efforts to get farmers to use efficient pumps. Thus, the project failed 21. Could India have introduced irrigation to achieve its covenanted energy sector policy on the scale that it did through the use of diesel goal (paras. 5.1, 5.2). powered pumps? A definitive answer may not be possible, but there are many reasons to 23. The audit concludes, as the PCR also believe that this alternative path was not did, that the project outcome was unsatisfactory. economically superior, or even feasible Given the continued financial problems of most (paras. 4.41-4.44). of India's SEBs and the growing burden of their RE programs, the audit also confirms the PCR's conclusion that the project's benefits are unlikely to be sustainable. The audit rates the Institutional Development impact on the SEBs, the ultimate borrowers, as negligible and not partial as concluded in the PCR (para. 5.3).  PERFORMANCE AUDIT REPORT INDIA THIRD RURAL ELECTRIFICATION PROJECT (LOAN 2165-IN) 1. BACKGROUND 1.1 This project was the third in a series of closely linked Bank lending operations in support of time slices of India's national Rural Electrification (RE) program.' The primary objective of India's RE program has been to provide reliable and economical power for irrigation pumps, as the foundation of policy aimed at eliminating the risk of large-scale crop failures (from the failure of monsoon rains) and establishing the conditions for increased agricultural productivity. 1.2 All three loans were administered by the Rural Electrification Corporation (REC), an institution established in 1969 to support the Government of India's (GOI) greatly expanded, country-wide RE strategy. REC's role has been to help State Electricity Boards (SEBs) formulate effective RE schemes, and to provide financing for these scheme, in line with India's Five-Year Plans. In addition to appraising and approving the RE projects proposed by the SEBs, REC has developed and produced standards, guidelines and manuals covering most aspects of RE, all of which have been adopted by the SEBs. 2. PROJECT DESCRIPTION 2.1 The project was designed to support the GOI's RE strategy in accordance with India's Sixth Five-Year Plan (FY81-FY85). It was to provide funds for the import component of a two-year time slice (FY84 and FY85) of the RE expansion and improvement investments in 14 States. During this time period, the RE program was projected to electrify 18,760 villages, 543,000 irrigation pumps, 25,000 small industries and 600,000 commercial consumers. The loan was to be used to support the following specific project components: * 1780 RE schemes from the 1890 RE schemes approved by REC between April 1, 1978 and January 31, 1982; * About 1700 new RE schemes approved or to be approved by REC during the period February 1, 1982 to March 31, 1984; * About I10 distribution system improvement schemes; * Buildings, furniture and equipment for REC's Central Training Institute for RE. 2.2 As in preceding two RE projects, the Bank delegated responsibility for approving individual RE schemes to REC, based on agreed criteria for evaluation and monitoring. The same five categories of I The first (Credit 572-IN), for $57 million, was approved by the Board in June 1975 and was closed in June 1978; the second (Credit 911-IN), for $175 million, was approved by the Board in May 1979, and was closed in March 1984. A Project Performance Audit Report for both projects (OED Report No. 6307) was distributed on June 16, 1986. At the time that RE III was appraised (May 1981) RE I had been completed and RE II had already committed the bulk of it's resources. 2 scheme that were supported in RE I and RE II were also supported by RE III. They were: Ordinary Advanced (OA) schemes, which electrify villages in economically advanced areas; Ordinary Backward (OB) schemes, which electrify villages in less advanced areas; Special Agricultural (SPA) schemes to connect irrigation pumps; Special Industry (SPI) schemes, which connected industry in rural industrial estates were to be co-financed by other institutions; and System Improvement (SI) schemes for rural distribution system reinforcement, rather than for new connections. All schemes financed under the project were to be individually evaluated by REC, and all were to have an economic rate of return well above the opportunity cost of capital. The Loan Agreement stipulated monetary allocations for specific types of project materials (conductors, transformers, switch gear, insulators and galvanized iron wire) which were to be used in the approved schemes. 2.3 The main sector issues were the weak financial condition of many SEBs and the financial burden that RE programs were creating for them. The Bank focused much of its efforts at the appraisal stage on improving the financial performance of the SEB. Each SEB had to agree to meet minimum financial criteria to participate in the loan. Since the State governments (SG) had authority over their SEBs, including the authority to set tariffs, they also had to agree to implement the policies necessary for meeting these criteria. 2.4 Under the two previous RE projects, the Bank had sought assurances from the State Governments (SG) that they would provide direct financial subsidies to compensate for RE losses, either fully or until the SEB's return on historical assets reached 9.5%. Under RE III, it was agreed that the criteria defining SEB's financial stability would be changed from reaching 9.5% financial return on net historical assets to one related to cash generation for self-financing of investments (at least 20% of investment). 3. IMPLEMENTATION AND RESULTS 3.1 Disbursement met with significant delays, and the project implementation period was extended by two years to enable the executing agencies to utilize all the allocated funds. During this extension, disbursement rose from 59% to 97%. In the process of extending the period over which materials were procured, the time slice of the RE investment program that the project supported was also extended from two years (FY84 and FY85) to five years (FY84 through FY88). Outcome 3.2 The national RE program met or exceeded its original physical implementation targets. During the extended five year period, 83,000 villages were electrified and 2 million pump sets were energized, about four times the numbers estimated in the appraisal for the original two-year time period. Program cost for this larger time slice was about US$1.8 billion, compared with the appraisal estimate of US$800 million for the shorter time slice. Cost per unit declined, partly as a result of the rapid devaluation of the Rupiah during this period. The declining unit cost may also have been partially related to a decline in the average number of households connected in newly connected villages. Only one connection was needed to qualify a village as being electrified, and Government pressure during this period to increase the target for the number of villages electrified may well have led to a strategy of minimizing the number of connections. Village consumption has always been a far less important part of rural consumption. Agriculture has, on average accounted for 85% of all electricity consumption in rural 3 areas. There is, therefore, no doubt that India's RE program met its physical objectives over the period of project implementation, and that the Bank loan helped to support the implementation of these objectives. 3.3 The project failed to meet the sector reform objectives of improving the financial stability of participating SEBs. In fact, India's RE programs had a profound negative impact on the financial condition of all SEBs. While it is not possible to measure precisely the impact of RE on the finances of SEBs, because RE is not carried as a separate account, it is possible to make some gross estimates of its impact. By the beginning of the 1990s, RE was already a very important element in the total demand for power. In 1992, demand from irrigation and agriculture accounted for 31% of India's electricity consumption (in GWh), but contributed very little to SEB income generation, accounting for only 4% of SEBs' gross revenues. The resulting financial losses were very large, at least $2.4 billion for 1992, equivalent to about 30% of gross revenue, and this is likely to be an underestimate, since it assumes RE costs were equal to the average system costs. Since actual costs of RE are considerably higher than the average system, the real financial losses from the RE program are, undoubtedly also much greater. These losses were a large contributing factor to the financial problems that most of India's SEBs are facing today. Project Cost 3.4 The cumulative credit disbursement by the original closing date of about US$168 million (or about 56% of the total credit) provided about Rp3,600 million, which was more than the Rp2,200 million estimated in the SAR. By the revised closing date (1988) the credit provided about Rp3,800 million. This was a consequence of the devaluation of the Rupee. The appraisal used an exchange rate of 8.3 Rp/US$. At the start of the project in 1983 it was 10.1 Rp/US$ and by the revised closing date in 1988 it had risen to 16.2 Rp/US$. 3.5 There is an indication that the credit would have bought more equipment and material than estimated by the original credit closing date, because the unit cost of the principal goods of the RE schemes, i.e. conductors, transformers and steel wires (and poles), in average, appear about 60% of that estimated at time of appraisal. This would explain the fact that by the original credit closing date about 4,600 RE eligible schemes were added, 32% more than the total 3,480 estimated at appraisal. On a range, they had about 70% more pump sets connected than originally planned. 3.6 How large was the cost overrun per scheme and what was its probable source? The number of schemes increased by 70%, while the total project cost increased by 260%. The cost per scheme therefore increased by 112%. The increase in total costs was unlikely to be a consequence of RE design features, which the audit found simple and appropriate for RE purposes. Nor was it in the cost of equipment, conductors and materials, financed by the Bank, which actually decreased in unit terms by about 40%. The audit therefore concludes that the increase was due to higher than anticipated number of pumps per scheme - which increased by 28% - and to higher than anticipated equipment, construction and engineering services, provided by, or contracted by the SEBs. The latter would suggest the possibility of some major deficiencies in project implementation at the local level. 4 4. ISSUES RE Schemes Design 4.1 REC has standardized the design and construction of RE schemes. The standardized schemes entail an up-stream link of electric pump sets, small industries and households with 3-phase distribution transformers, 3-phase 33 kV or 11 kV spurs and radial lines from 33 kV and 11 kV switching and transformer substations. These, in turn, are connected to the major transformer substation of the high voltage power grid by 66 or 154 kV. These scheme designs are simple, based on similar ones for urban distribution system and appropriate for rural service. 4.2 The use of 3-phase system was appropriate in the Indian RE context. The principal objective of RE in India is to supply electric motor pump sets for irrigation and secondarily, to electrify the villages located near the distribution lines. The use of 1-phase distribution lines as a cost effective alternative to the 3-phase 33kV or 11kV rural lines should be considered only when electrifying several villages where demand density is low, more or less uniformly distributed along the lines, and where there is no need for massive water pumping for irrigation--which was not the case of the RE III project in India. 4.3 Rural distribution lines use concrete poles, metallic cross arms and pin insulators for supporting conductors. Pole mounted distribution transformers manufactured in India are used to step down distribution voltage to final user's low voltage. Design parameters for mechanical strength of these structures is based on local wind and temperature conditions. Electric motor pump sets of sizes ranging from 6 to 10 HP, 3-phase, 400-Volt are of widespread use for irrigation purposes. The technology used is appropriate. These electric motor pumps are designed and fabricated locally and conform to Indian standards. The alternative of using smaller electric pumps for 1-phase 220 Volts would be an inferior choice for the rural irrigation conditions in India. However, the design and fabrication are of a relatively low standard and, as a result the motors are not very efficient. Maintenance 4.4 The audit mission reviewed RE construction and maintenance practices in several rural areas of three SEBs. The RE schemes appeared to be built in conformance to REC standards, but there was significant physical deterioration evident for the older facilities. The visited schemes practice corrective maintenance rather than preventive maintenance for the large areas covered by the RE schemes. Most of the maintenance is carried out by SEBs and little is contracted out. Physical conditions of maintenance shops visited are no more than fair. As an indicative example of decreasing effectiveness of maintenance done to aging distribution transformers in one state: from 1981 to 1992 the number of installed distribution transformers increased from about 32,000 to 75,000 and the number of damaged distribution transformers increased from 3,700 to 17,000. Thus the rate of failure increased steadily from 12% to 23%. The efficiency of the maintenance shops decreased as measured by the percentage of transformers repaired to the total failed, which decreased steadily from 100% in 1981 to 70% in 1992. 5 Demand Side Management and Quality of Service 4.5 The only demand management tool used in RE is programmed load shedding of entire lines in a cyclical schedule during capacity shortage at peak hours. Load shedding is done manually. Voltage and power factor control by transformer taps and capacitor banks at the distribution substations are practiced in the SEBs visited by the audit. This allows the system to maintain power factors close to 80% and to limit voltage drops close to the permissible maximum 10%. Since electric pump sets are highly inductive loads, further improvement of system power factor could be achieved by installing capacitors in all the electric motors of the irrigation pump sets. 4.6 Total system losses in the order of 20-22% are high but difficult to reduce, particularly the non- technical losses related to un-metered connections (legal and illegal ones). Reduction of technical losses are addressed in India by upgrading overloaded lines, as well as voltage and power factor corrections, as indicated above. Lessons 4.7 The REC could help to improve user efficiency and decrease future investment costs by: (i) promoting standardization of electric motors with built-in capacitors as a practical measure to improve power factor of rural lines and reduce technical losses; (ii) undertaking studies to reassess the potential for using single phase lines to expand village connections; and (iii) promoting the privatization of construction, engineering and equipment repair services for RE schemes through competitive bidding. Financial Condition of the SEBs 4.8 The financial issues raised during negotiations on RE III were an integral part of the Bank's broader energy sector policy dialogue with Central Government (GOI) and State Government (SG) entities. Most SGs were reluctant to impose significant electricity tariffs on their rural sectors. As a result, the rapidly expanding RE programs were creating growing financial strains on the SEBs. To resolve this problem without having to confront SGs on their tariff policies for the agricultural sector, the Bank had agreed in the two previous RE projects that the States would make direct subsidy payments to their SEBs to cover the financial losses of the RE programs, and that these subsidy payments would be considered as income by the SEB for the purposes of achieving the Bank's financial covenants. The idea behind this approach was that the cost of subsidizing agricultural tariffs would be identified and acknowledged by the SGs, which would, in turn, create some pressure to reduce these subsidies by rationalizing tariffs. 4.9 In the previous RE loans, financial covenants for participating SEBs had focused on obtaining a 9.5% return to historically valued assets, in line with the covenants on other power projects. But by the late 1970s it was apparent that a covenant on the rate of return on historically valued assets would not be sufficient to maintain SEB's financial viability. Therefore, in 1981, during negotiations for the Second Korba Thermal Power Project, the Government agreed to establishing revised sector financial objectives. Rather than introducing a process for revaluing assets, the GOI and the Bank decided to change their strategy to one focusing on getting the SEBs to provide a "reasonable contribution to investment", from internal cash generation. The definition of "reasonable contribution" was somewhat complex, but was based on a premise of providing about 20% of capital expenditure from internally generated funds, after all other costs have been accounted for. 6 4.10 In addition, it was discovered that in many States the need for large and growing RE subsidies had failed to provoke any changes in electricity tariffs to the rural sector. In some states, RE subsidies had grown to where they were more than 25% of the SEB's total electricity sales revenue. A covenant was, therefore, added to RE III to reduce the maximum subsidy that a State could provide, on a sliding scale, with a goal of limiting it to no more than 10% of sales revenue over a three-year period. As a condition of receiving project funds, each SEB was to submit to the Bank, by December 31, 1982, an action plan outlining the steps to be taken and the improvements to be made to ensure that the SEBs could meet the required levels of contribution to investment even with the smaller subsidies. REC was supposed to monitor these conditions and inform the Bank when a SEB was not in compliance. 4.11 The basic financial problem was the extremely low levels of electricity tariffs for agricultural use. Average tariffs were below US$ .005 per KWh, which was less than 10% of the LRMC of electricity production for the urban power systems. Even this low price overestimated the actual payments by the agricultural sector. In some States, power was provided on a fixed monthly charge, related to pump size, so that the marginal cost to the user was zero. In addition, the collection system was often so erratic that the average price paid was effectively zero. Although low tariffs were a central issue there is no mention of it in the project documents. 4.12 The Bank negotiated the loan with the (GOI), but the GOI had no authority over the tariff policies of the SEBs. Since the GOI had no authority in the area of tariffs, it could only agree to insisting that the SGs had to make their own policy pronouncements to qualify their SEBs for funding under the loan. Yet, because the negotiations were between the Bank and the Central Government (GOI), and did not include the State governments, who had the authority to decide tariff policies within their states, it was not possible to reach an agreement on policies for rationalizing tariffs, and without an agreement on this basic issue, there could be no plausible plan to improve the SEB's financial performance. The entities responsible for making the difficult political decisions to increase tariffs simply never undertook to accept ownership of the problem or its solution. 4.13 By the end of 1983 -- one year after the dated covenant -- all the SEBs had registered their commitment to improve financial performance. However, none had presented a plan of action, approved by their SG's to meet the targets for contribution to investment and reduction in subsidies. Eventually, the action plans were sent to the Bank during the first quarter of 1986, some three years after the dated covenant. Although this was a dated covenant, the Bank never declared any of the delinquent SEBs ineligible for onlending, even when the action plans they provided were clearly and obviously unimplementable. 4.14 Less than a third of the action plans were plausible, since they lacked any plan for increasing tariffs. In fact, the tariff situation deteriorated during the period of project implementation. Between 1982 through 1987, India's average irrigation tariff increased by only 2%, while the average cost of providing electricity for the system as a whole more than doubled. Not surprisingly, rather than improving, the financial condition of most SEBs deteriorated throughout the period. During this whole period, up-to-date financial data on SEBs was largely unavailable and the Bank was unable (or possibly unwilling) to carry out any meaningful financial supervision of SEB's finances. It is only in early 1987 that, subsequent to a request by GOI to change back the loan's financial covenant from the 20% contribution to investment (CTI) to a modified 3% r.o.r. on fixed assets - so as to make it consistent with recently enacted changes to the Electricity Act - the Bank carried out its first detailed review of SEB's finances since project appraisal. The results of the review were predictably negative: it showed that half the SEBs could be considered insolvent, in that their total liabilities exceeded their total assets. 7 A majority of the rest had serious imbalances in their financial structures, including negative working capital. Furthermore, the mission's report concluded that "...both the existing [CTI] covenant and the proposed [R.O.R.] covenant are not appropriate measures of financial performance and therefore little would be achieved by changing the covenant. There is a need for the Bank to return to basics.." (i.e. thorough financial restructuring of most SEBs). There is no trace in the files of the mission's findings having been discussed at managerial level, in spite of the importance of its findings.2 Indeed, in March 1987, the Bank agreed to change the covenant via a corresponding amendment to the Loan Agreement. The audit could not find any evidence of further work on this topic having been carried out by the Bank before loan closing in June 1988. Lesson: 4.15 The PCR identified some important lessons from the weak financial performance of the SEBs under REC III. The audit is in full agreement with them, and therefore repeats them here: * Setting financial targets without a viable enforcement mechanism is unlikely to be successful, particularly when the borrower has little impact on the policies of the sub-borrowers. * Financial covenants have to be realistic and enforceable. * Having set realistic financial targets, the Bank should be prepared to cancel funds if the beneficiaries are unable to comply with them. * RE, as currently practiced in India, is unsustainable because it imposes too heavy a financial burden. 4.16 The audit also believes that the failure to achieve the sector's financial goals was a result of a lack of understanding of the dynamics of the sector's institutions. The Bank was operating under the belief that it could reform the financial practices of the individual SGs through pressure from Central Government institutions. It therefore failed to work directly with the individual State SEBs and their SGs to reach agreement on how to resolve the SEB's financial problems. Instead, it negotiated the loan covenant conditions with the GOT and expected that the GOT would, in turn, be able to impose the agreed conditions on the SGs, who controlled their own electricity tariff policies. The Bank had hoped that the leverage of a large project loan would be sufficient to ensure compliance. However, this leverage never materialized; while large in aggregate, the loan provided relatively small sums to each of the fourteen participating SEBs and represented only a small share of total power lending to India, and the entities responsible for making the difficult political decisions to increase tariffs never accepted ownership of the agreed solutions. Recommendations: 1. In India, the Bank needs to work closely with individual SEBs and the State governments to ensure that they have accepted ownership of the reforms needed to establish a financially sound electric power sector. The Bank has, in fact, already incorporated this lesson into its lending strategy. It has begun to lend significant sums directly to individual SEBs where agreement can be reached on action needed to restore It is perhaps not totally coincidental that the mission's report was issued at about the same time the Bank reorganization was starting to mobilize much of staff time and energy (Spring 1987). 8 financial viability. The first such loan to incorporate a RE component was the Second Maharashtra Power Project, which was approved by the Board in June 1992. 2. For a RE program to be sustainable, RE consumers must be charged a tariff that, at a minimum, is sufficient to cover the cost of supplying electricity to the system. That is, RE needs to cover its operating costs without cross-subsidies. If tariffs are below this level, the Government will be committing itself to a long-term subsidy program which will grow in proportion to the increase in rural energy consumption. In addition, it is desirable that tariffs are high enough to also finance a substantial part of the RE distribution investment costs. 3. Where a government decides that subsidization of RE is politically or socially desirable, the method of subsidization should be an up-front subsidy covering a portion of the distribution investment cost. Tariffs should then be set so that the utility can earn a reasonable rate of return on the non-subsidized portion of the investment, including the full cost for the electricity consumed. Such a systematized approach to subsidized support for RE programs ensures that subsidies correspond to the resources the government has available in its current budget, and that the RE program will not create a continued obstacle to the establishment of a financially viable electric power sector. The Role of REC in Supporting Financial Reform in the SEBs 4.17 The Bank thought that it could leverage its own supervision resources by using the financial intermediary, REC, to administer the Loan Agreement's financial covenants. However, REC was somewhat ambivalent about taking an active part in this process. REC's official mandate was to ensure that the funds provided for RE were properly spent and that the national RE development goals were met. To accomplish this goal, it needed to ensure that funds kept on flowing to all the SEBs. But to police the financial covenant and to declare a SEB ineligible under the Bank loan would run counter to its original mandate. REC therefore took its cue from the Bank, overlooking or making excuses for SEBs failure to meet their commitments, and continued to onlend to all SEBs, without concern for financial viability issues. 4.18 REC was, of course, concerned with maintaining its own financial health. In its role as a financial intermediary, it borrowed from domestic and international sources, and onlent them to SEBs. However, like the World Bank, it ensured that its loans to SEBs were counter guaranteed by the State Governments. And as long as its loans were guaranteed, it felt no need to discriminate among its borrowers in terms of their financial strength and ability to repay their loans. This strategy appears to have been quite successful, at least until recently. REC has maintained a strong financial position throughout the 1970s and 1980. It is only since the early 1990's that it has begun to experience some significant arrears in repayments from some of the most financially distressed SEBs. Lesson: 4.19 The Bank should not have expected REC to take the initiative on financial reform issues. The Bank could, however, have used REC more effectively to monitor SEB's financial performance, if it had structured the loan to insure SEB compliance and if it had been willing to stop disbursements to those SEBs who failed to meet their commitments. After reviewing the disappointing experience with the performance of some SEBs during the implementation of a number of projects in the 1980s, the 9 Region has substantially revised its policies. In recent loans -- such as the 1992 loan to the Power Finance Corporation -- the Bank has taken a strong position with SEBs that fail to maintain full compliance with all financial covenants.' Institutional Reform in Time Slice Projects 4.20 Lacking agreement on immediate steps to increase agricultural tariffs, the Bank took the position that improving the SEB's financial condition could only be implemented over a substantial time period. In fact, for the SEBs with the greatest financial problems, the Bank agreed to a four-year improvement program. But since the project was supposed to finance only a two-year time slice of the RE program, many of the improvements that were to be made would come only after most of the disbursement program was scheduled to be completed. One must therefore conclude that there was a mismatch between the type of loan and the implementation of the sector policy agreements. Recommendation: 4.21 When a loan is designed to finance a time slice of a sector investment program, it takes on many of the attributes of a generalized sector policy loan. In a time slice project, implementation and procurement are divisible, so that temporary suspension of the disbursements for the second time slice would not jeopardize the completion of the first time slice investment program. In this sense, procurement for a time slice project is similar to the general import procurement of an adjustment loan. Therefore, it is possible, and advisable, for such time slice projects to have their sector policy conditionality strengthened by inclusion of specific tranche conditions, as practiced in adjustment loans. Such tranches might have strengthened the Bank's hand in its subsequent dealings with individual SEBs. The Role of REC as an Apex Onlending Institution 4.22 REC has played a central role in the implementation of India's RE program, acting as both a financial intermediary and a source of technical assistance expertise. It was created to supervise Government lending programs aimed at expanding RE in the States, and thereby to ensure that the funds were used for the designated purposes and that the RE development targets were actually met. 4.23 REC was, therefore, the appropriate institution for supporting the Bank's onlending program in RE. REC effectively supervised the formulation of the individual subproject RE schemes, ensuring that they conformed with pre-agreed parameters for each scheme category, including availability of sufficient water, number of potential customers per line kilometer, and potential for load growth. REC then checked on the implementation progress, disbursing against actual completions. Given the financial constraints on all SEBs and the relatively low priority that many SEBs placed on their RE programs, it is unlikely that India's RE program could have been expanded without the assistance of a specialized agency dedicated exclusively to carrying out the Government's mandate. REC's supervision was, therefore, an important element in the successful implementation of India's RE program. Given the fungibility of the materials purchased for electricity distribution, the active participation of REC was essential to ensure that the SEBs used their project funds for the agreed purpose. 3The Government notes that the number of SEBs that have met the statutory minimum financial performance on an accrual basis has risen from two in FY90 to nine in FY93. (See Annex 1, para. 7). 10 4.24 It can be expected that REC will, in the future, focus more on improving the efficiency of existing systems. Even without significant extensions of their RE systems, SEBs will need to make substantial investments in system strengthening to keep system losses within tolerable limits. In the future, the Bank should consider supporting a shift in focus of India's RE investment program from further expansion to energy saving, system improvement programs. Such programs might also be of greater interest to SEBs, since they would reduce energy use, rather than increasing it, and might, therefore, provide a platform for reopening communications with recalcitrant SEBs without increasing distortions in the energy sector. 4.25 One criticism of the use of such an apex institution might be that it limited the Bank's influence over the choice of subprojects. Of the 9300 subprojects implemented during the project period, many were undoubtedly of marginal value. For instance, it is questionable whether the capital costs of electrification can be justified when irrigation pumps are used less than three hours a day. However, it is doubtful that the program as a whole would have improved if the Bank had chosen the specific subprojects that it wanted to finance, even if it had the resources to undertake the necessary analysis. Since the materials purchased under the project were, in effect, totally fungible between various RE schemes, it made no difference whether the Bank chose to finance a small percentage of a large number of schemes or a large percentage of a smaller number of schemes. Thus, the Bank's influence on the types of schemes implemented comes entirely through its policy dialogue with the government on the overall envelope for the country's aggregated RE investment program, and through its support for the introduction of economically based procedures for evaluating the schemes to be chosen. Bank Performance 4.26 Bank performance during project preparation, appraisal and supervision was weak. Project design relied on Central Government institutions to achieve State level reforms and help the Bank minimize resource inputs, by working exclusively with one apex financial institution. In theory, this design would allow the project to achieve widespread reforms, without Bank staff having to undertake detailed State by State analysis and negotiations. In practice, this approach proved unworkable. It explains the failure to see the need for a forecast of RE losses for each SEB. In the absence of separate RE accounts at the SEB level, an arbitrary loss allocation formula was used. Nor was there an analysis of the implicit and explicit subsidy mix for each SEB. Without this analysis there could be no assessment of the future long-term impact on SEB finances. In the end, the Bank's expectation that it could improve the overall financial performance of most of India's SEBs, through a lending operation for an activity that was peripheral to the SEBs' main areas of interest, proved to be totally unrealistic. 4.27 Using an apex financial institution also allowed the Bank to minimize the resources used in supervision. No more than a quarter of the SEBs who were implementing the projects were ever visited during supervision missions in six years of project implementation. However, the result was that there was essentially no financial supervision of the SEBs for the first four years after Board approval, and there was little concern about enforcement of the financial covenants. The Bank should not have expected a local institution, which lacked any in-depth financial appraisal expertise, to supervise the financial performance of the SEBs.4 'The Government points out that the Bank was kept informed of the financial health of the SEBs for which REC did not report directly, through reports from the Central Electricity Authority (CEA) (See Annex 1, para. 4(d). 11 4.28 The Bank also failed to resolve many of the other institutional and operational problems identified during appraisal, including excessive energy losses, poor operating performance by some SEBs, the lack of metering of electricity use and low efficiency end use patterns for irrigation pumps with high power factor. The project failed to include any specific programs to ensure that these problems would be addressed during project implementation, and the audit found no evidence that they were systematically dealt with by the executing agencies. Lesson and Recommendation. 4.29 While Apex institutions are useful to ensure effective physical implementation of RE programs spread over many executing agencies, they cannot resolve the specific operating problems of the executing agencies. Greater efforts are needed at appraisal to analyze and resolve problems at the operating level. More work at the local level is also important to ensure that the implementing agencies have accepted ownership of the project and are willing to take the steps needed to ameliorate the worst of the identified operating problems. Future RE programs should place special emphasis on programs designed to improve end use efficiency, particularly for non-household electricity usage. Economic Rate Of Return 4.30 The SAR undertook a detailed economic analysis of fifteen "typical" RE schemes, covering all 14 States. This analysis showed that all the schemes had acceptable economic rates of return, unless one made what was, at that time, extremely conservative assumptions on the valuation of benefits. The primary benefits in the SAR methodology were the cost savings to the economy obtained through using electricity rather than alternative conventional energy sources (kerosene, diesel, etc). This approach avoided allocating incremental agricultural output specifically to electricity. 4.31 When the PCR updated these calculations, it found substantially less positive results. It found that economic costs had been greatly underestimated because of a methodological error in evaluating the cost of electricity for the time pattern of use of irrigation pumping. The costs were equal to the LRMC of supplying peak power to the system because agricultural usage, which accounted for 85% of demand, was almost exclusively during daylight peak power demand periods, in seasons where hydropower was least likely to be available. The result was that, for the central base case, six of the fifteen schemes yielded negative rates of return, five yielded low positive rates of return (below 10%) and four yielded attractive returns of over 20%. The PCR also found that economic benefits, while properly estimated, had declined since appraisal in 1981, in parallel with the decline in the cost of petroleum fuels. 4.32 The PCR concluded, therefore, that RE was economically unjustified in India, as long as power is provided for agricultural pumping during hours of peak system demand and there are no price incentives to encourage farmers to use more efficient pumps. This is a rather startling conclusion, since the India RE program has, arguably, had the highest utilization of electricity for productive purposes, and the highest impact on the level and structure of production in rural economies of any RE program supported by the Bank. It is argued below that the analysis used in the PCR failed to consider the strategic importance of rural electrification to India's economic well-being. 4.33 The analysis failed to capture: (i) the producer surplus that farmers derived from irrigation; (ii) many of the broader benefits that came with the transformation of agriculture from monsoon dependent dry farming to year round irrigated farming; and (iii) many of the costs that would have been associated 12 with a national irrigation program based on the introduction of 8 million diesel pumps in the past two decades. Producer Surplus 4.34 The analysis of the PCR evaluates the willingness to pay for electricity as the area under the downward sloping demand curve. The curve is defined by two points: the upper one is the value of electricity in the displacement of a diesel pump set, for those pumps displaced; the lower one is the very low actual price charged and amount of electricity consumed. This methodology assumes that some of the demand for electricity is replacement demand, where electric pumps replace diesel pumps and some is induced demand, where the lower price will induce farmers to purchase new electric pumps, even though they would not have purchased diesel pumps. It therefore assumes that the induced demand pumps have a lower economic value to the farmers and to the economy as a whole than the replacement demand pumps. Microeconomic Analysis: 4.35 The shape of the demand curve is based on some important implicit assumptions about farmers' decision-making behavior that are highly questionable. First, it assumes farmers make rational economic decisions on production technologies, based on perfect knowledge of the benefits that will flow from these technologies. This assumption requires that farmers are fully aware of how much they can increase their agricultural output by shifting to irrigated farming, including the additional benefits of being able to use new high yield seeds which can take advantage of higher fertilizer inputs. Second, it assumes that farmers are not highly risk adverse, so that they are willing to introduce these new technologies until the marginal costs of introducing them is equal to the expected, average, marginal benefits. Third, it assumes that farmers can take advantage of perfect financial markets, so that they can borrow to pay for the cost of diesel fuel in the beginning of the season, when the pumping requirements are highest, along with borrowing for seed, fertilizer and other input requirements, at rates that are in line with India's marginal cost of capital. 4.36 These assumptions run counter to what we know about small scale farmers, in India or in any other developing country. Farmers are risk adverse, and are therefore slow to accept new agricultural practices until they are sure that the benefits at the margin will be worth the risks of giving up technology that has worked for generations. Even when extension services help promote new technologies, most farmers discount a large part of the potential benefits, and only begin to accept the new technology when they see it practiced by one of their neighbors. Thus, new technology and best practices spread slowly, as more farmers are convinced of the potential benefits and willing to take the risks of early acceptance. This is the reason that farmers may have to be induced to introduce these new techniques by being offered subsidized seed and fertilizer until they can judge the benefits for themselves. Subsidized electricity played much the same role for the introduction of irrigated farming in India. In addition to the slow spread of information and the aversion to risk taking, farmers are often unable to take advantage of market opportunities because they do not have access to additional credit needed for implementing the new technology at reasonable rates. Policy for electricity bill collection also substitute, to a large degree, for some short-term credit requirements. 4.37 If the demand curve for electricity in agriculture were an accurate measure of farmer's derived demand for electric pumps, it would, most probably, be almost completely price elastic (horizontal at the level of diesel alternative costs), over a wide range of demand, since this derived demand would 13 be based on the actual marginal productivity of irrigating more land. The failure of most of the assumptions used in the SAR/PCR to be substantiated by studies of agricultural institutions and farmer behavior together with the very low tariff used as the end point for the approximation by the demand curve suggests that the demand curve used in these papers to estimate producer surplus for electricity in agriculture may greatly underestimate the economic value of electricity in agricultural production. 4.38 The assumption that the lower point on the demand may be calculated by observed prices and levels of consumption is also questionable in the Indian context. Point on the actual demand curve must be a market clearing price; that is, all farmers who want to connect electric pumps are connected and electricity is available when it is wanted. But in many states in India, severe electricity shortages necessitated formal and informal rationing. The audit found that in all states it visited, there were long queues for connecting new irrigation pumps, and electricity was rationed. SEBs were in no hurry to reduce backlogs for new connections, since, at the existing tariff levels, electricity sales to agriculture did not cover even the current costs of purchasing it from the central authorities. Some states were considering the introduction of connecting fees to allocate new connections. 4.39 There is another fundamental problem with using a microeconomic demand analysis to evaluate RE. It would still be unable to incorporate the nationwide benefits that derived from the successful transformation of agriculture through implementation of its national RE program. The benefits of India's RE program should be viewed from a global, macroeconomic perspective. 4.40 In the 1960's, the GOI made food security a national development priority. To reach this goal it needed to transform the agricultural sector from monsoon fed, dry farming to irrigated farming, because this was the only way to eliminate the social and economic costs of widespread famines and social disruption that had, historically accompanied periodic failures of the monsoon rains. In addition, irrigated farming was to lay the foundation for the adaptation of new, emerging technologies for enhancing agricultural output, which was the only way to obtain the sustained growth in agricultural output needed for long-run food security. 4.41 Could the objectives have been met through the use of diesel pumps instead of electric ones? There can be no definitive answer to this question, but there are many reasons to believe that this alternative path was not economically superior, or even institutionally feasible. It is not at all clear that India could have successfully implemented a program for installing 8 million more diesel pumps and supplying them with imported fuel. First, the cost of the fuel would be much higher than it is today. The additional fuel imports, required even in periods of exchange crisis, would have impacted India's terms of trade, and the economic cost of distributing the fuel would have been substantially higher than it is today. New institutions would have had to be created to store and distribute the fuel in close proximity to the users, since, according to studies done in India, diesel pump usage declines rapidly when farmers have to travel more than eight to ten miles to purchase fuel. These facilities would have to be large enough to meet planting period peak demand requirements, and the distribution chain would have to be efficient, to avoid shortages at critical times. 4.42 Second, to reach similar growth rates as electrified pumps have, diesel pumps would also have had to be heavily subsidized. If the subsidy was for diesel fuel, it too would have put a heavy strain on government budgets. It would, however, be much more difficult to restrict the subsidy to users of agricultural pumps. The leakage to other users, for rural and urban transport, might have made the cost of such a subsidy prohibitive. 14 4.43 India has been successful in meeting its strategic goals. By the early 1990s India's RE program had energized over 8 million pumps. The failure of the monsoon rains in 1987/88 resulted in some of the worst drought conditions in fifty years, but the drought did not result in the widespread starvation and massive population dislocation that had accompanied previous monsoon failures. In fact, in this period agricultural production declined by less than 15%, and additional food imports were minimal. 4.44 The audit therefore concludes that the development objectives of the project, specifically, the support for a country-wide program of electrified irrigation in support of a program to eliminate widespread intermittent droughts and to increase agricultural output, were appropriate and were successfully met in a cost effective fashion. 5. PROJECT OUTCOME AND SUSTAINABILITY 5.1 The pricing policies that accompanied the RE program have led to substantial resource waste, and to an intolerable financial burden on all the country's electricity producing and distributing institutions. This sector policy issue needed to be effectively addressed after it had been identified in two previous RE loans. The Bank's main efforts were directed to resolving these problems, and all the project specific covenanted conditions were focused on this aspect of project implementation, but the project failed to achieve its sector objectives. 5.2 The audit agrees with the PCR that both the economic and financial cost of RE could have been reduced by eliminating agricultural pumping during hours of peak system demand providing price incentives to encourage farmers to use more efficient pumps. If pumping were restricted to off peak nighttime hours, economic and financial costs would be sharply lower. However, electricity for pumping has only been restricted by the competing demand by other users in a greatly overloaded system, and there have been no successful efforts to get farmers to use efficient pumps. Thus, the project failed to achieve its covenanted energy sector policy goal. 5.3 The audit therefore concludes that the project outcome was unsatisfactory. Given the continued financial problems of most of India's SEBs and the growing burden of their RE programs, the audit also concludes that the project's benefits are unlikely to be sustainable if policies are unchanged. Although the project was instrumental in getting the SEBs to adopt modem accounting standards, overall, the project's Institutional Development impact on the SEBs, the ultimate borrower, was negligible. 15 ANNEX 1 Page 1 of 7 GOV9RNMENT OF INDIA T. SETHUMADHAVAN k% kt V ftgm MINISTRY OF POWER JOINT SECRETARY ww dftImw, xt 414, ttwnt on 1001oa D.O.No.44/5/93-D(RE) New 0*IN-1100M 17.6.1994, Dear Mr. Jozsef, Please refer to your letter dated 28th April, 1994 and FAX meagw dated 2nd June, 1994. regarding Draft PAR on Third Rural Electrification Project- (Loan No.2165). 2. Please find enclosed our comments on the draft Performance Audit Report on Third RE Project. With regards, Yours sincerely, End; As above. [ T. SET DHAVAN ] Mr. Jozsef B. Buky, Acting Chief, Infrastructure & Energy Division, Operations Evaluation Department, World Bank, 1818, H Street, N,W. Washington, D.C. 20433, U.S.A. 16 ANNEX 1 Page 2 of 7 PERFORMANCE AUDIT REPORT (PAR) ON THE THIRD RURAL ELECTRIFICATION PROJECT IN INDIA, LOAN 2165-IN This is in response to the Performance Audit Report (PAR) on the Third Rural Electrification Project in India, Loan 2165-IN, for US $304.5 Million. The Bank approved the loan on June 1, 1982. The loan closed on June 30, 1988, two years behind schedule. The Bank disbursed 97% of the loan, canceling the US$9.0 Million undisbursed portion. 2. The Ministry endorses the basic thrust of the PAR that Rural Electrification has played a central role in shifting the country's agriculture from monsoon dependent dry farming to irrigated farming and implementation of India's rural electrification programme and has been responsible for the larger achievements in village electrification & pumpset energisation. The PAR's conclusion that, the project has, therefore made a significant contribution to improving the lives of the poorest of the poor is also correct. 3. The Ministry also agrees with the PAR's view that the conclusion of the Project Completion Report (PCR) is not correct in suggesting that India's RE was economically unjustified. This is mainly because, the PCR has not taken into account the full benefit stream that has come from Indian rural electrification programme nor the full cost and institutional difficulties of implementing diesel alternative when applied on a nation wide scale. The PAR's that, the rural electrification programme has greatly contributed to irrigated agriculture and, when India experienced its worst drought in 50 years (1987), agricultural production fell only marginally also stands substantiated. 17 ANNEX 1 Page 3 of 7 RO4. With regard to the performance and Role of REC in RE4 the programme the following issues are relevant :- (a) During the 5 year period uqder consideration, REC went well beyond fulfilling targets set by the World Bank, mainly, of 18,760 villages to be electrified and 543,000 pumpsets to be energised. In fact, it was able to electrify 83,000 villages and energised 2 Million pumpsets. (b) As regards the role of the REC as an Apex Institution, as already stated, it has undoubtedly played a central role in the implementation of India's RE programme, acting as both a financial inter-mediary and source of technical assistance expertise. The Audit Report also endorses this view that REC was the appropriate institution to chanelise the flow of funds and that it effectively supervised the formulation of the individual RE schemes ensuring that the projects were within the technical parameters laid down and moneys were disbursed against actual completion of the sub-projects. (c) Having electrified 85% of India's villages, the emphasis of the REC has already shifted from mere village electrification to load intensification and improving the efficiency of the existing systems to make the rural electrification programme more effective and also more economically and financially viable to the SEBs. 18 ANNEX 1 Page 4 of 7 (d) With regard to the criticism that the REC lacked financial expertise to take on the responsibility of supervising financial performance of the SEBs, the Ministry is of the view that achieving financial stability for an SEB through an RE loan is not an easy proposition keeping in view the socio-political compulsions of the State Governments. Any shortcomings in achieving the financial objectives of the Project cannot be attributed to the suggested lack of Indepth financial expertise of REC, but due to the compulsions of a developing State under trying conditions. Moreover, all through the loan period, REC either directly or through Reports sent by Central Electricity Authority (CEA) had kept the Bank informed of the financial health of the Electricity Boards which received the loan. The utilisation of the loan amount was initially as per the Project Agreement. However, some gaps started appearing in a few Boards, and the Bank was duly kept informed. RECARDING 5. The other main thrust of the PAR Report centres STIU Or on the financial stability of the SEBs. It suggests that rather than improving the financial condition of the SEBs through the chanelisation of the World Bank RE loans, it, in fact, had an adverse impact on financial healthof the SEB's. The Ministry would like to point out here that in order to ensure as much financial discipline as possible, the World Bank funds were routed to those SEBs where State Governments provided subsidy for the RE loans of the SEBe as per agreed norms. Further, more in association with the Government of India, the implementing agency 19 ANNEX 1 Page 5 of 7 commissioned reputed Consultancy firms to introduce commercial accounting in all SEBs. The PAR has also commented that this has been an achievement of some note. 5.1 As regards the sector reform objective of improving the financial health of SEBs through the RE loan, it has to be remembered that this can not be seen in isolation. The financial viability of the SEBs has been a matter of sustained interest and continued concern to the Government of India. Some of the major issues concerned have been analysed in the study on Long-terms perspectiveA of the Power Sector, commissioned in 1991. The problem identified include Sub-optimal capacity utilisation, high system losses and low level productivity, low tariff structure etc. It will not be correct to envisage that all these could be remedied by intervention through RE loan alone. On a more comprehensive level, the need for reform in the power sector has now been accepted and some SEBs are subjecting themselves for detailed studies for re-structuring, with the assistance of international consultants and Bank assistance. The RE-III loan was sanctioned/implemented prior to the liberalisation of economy in India and long before the need for restructuring of the public sector was accepted at the national level. Even though the revenue from RE has remained more or less constant, it is relevant that the Central Government has been able to persuade about 16 State Governments to fix a minimum agricultural tariff of 0.50 P/kwh. It has also been feasible in the recent years to persuade most SEBS to revise their tariff periodically. 20 ANNEX 1 Page 6 of 7 elL OF 6, The in-ract of tie effort In iirFrovinr the FOWER FINANCE financial health of the SEBs through the aegis of the FoLer linance Corporaticn (FFC) I-as been effcctive. SEEs have acknow%ledged the role of OTTEB s FFC in helpino then to inrrove their financial Ferforn-ance through constant mcnitoring as also by intervention through the State Governments concerned. However, REC -as not conceived in the sarre -culd and the obleclives of REC vere not so much inatitutional develoywent of SEBs. L-ut expension of FE yrotrarmre In the country. Nevertheless, the ccncluecn in the FAR that RE consumers r-ust be charged a reasonable cost and that any suleidy sbculd be urfront and translarent is endorsed. Ihis is presently being atterrFted through the rreclanism cf FFC's OFAF. Ferhars, from this angle, the colenants attached to tI-c RE III 'vere nct adeouately transparent. 7. As 2egards the financial heeltb of SEBe, it nay not be out of place to mention here that as against only 2 SEBs in 1989-90, the number of Boards v6bich have reccrded the statutory n-lninutr zurrlus on accural basis has gone uF to 9 in 1992-93 and this wes possible through the FersusiEicn of Central Government agency, lending to SEBs, %alth a vie to bring abcut overall financial iuyprovetrent and insisting on ccvenents attached to loans. Since State Governments control SEBE, direct inter-ection b principal lender %ith State Gcvernrente way be useful, but thiE could be achieved more effectively through tle Inetrument cf a Central funding agency, vJo ai]] have constant inter- aclicn vith State Governmrents/SEBs and adeauste leverage. 21 ANNEX 1 Page 7 of 7 8. As regards reform in tariff policy, the Central Government has decided to set up a National Power Tariff Board at the Centre and Five Regional Tariff Boards with a view to streamlining the tariff structure in various SEBs. This will help in improving the revenue realisation including from the Rural Sector. 9. Finally, it will not be correct to conclude that the project outcome was unsatisfactory. The loan was implemented at a tiime when the main objective of REC/recipient SEBs was to reach the benefits of electricity to the majority of villages. spread over the length and breadth of the country. The implementation of the project has also assisted in improving the food production substantially and has improved the quality of life of the poorest of the poor, which indeed is truly gratifying. It is true that action for improving the financial health of the borrowers did not receive equal attention while implementing the project. Fortunately, the situation has improved since then, and the subsequent efforts are showing results. The finding that sectoral reform can better be achieved through direct inter- action with the State Governments is to an extent, valid, but not wholly correct. Any assistance, granted as per RE III would indeed be sustainable, provided it is matched by necessary follow up measures on a holistic basis, as is being attempted now by Central Govt. through concerted measures.