India - Industrial Export (Engineering Products) Project

Document of The World Bank FOR OMCAL USE ONLY Repwt No. 12612 PROJECT COMPLETION REPORT INDIA INDUSTRIAL EXPORT PROJECT - ENGINEERING PRODUCTS (LOANS 2629-IN AND 2630-IN) DECEMBER 17, 1993 is iI C R GRgPH'-T F1 i I L, Report No i26'- Tvype: F'C Country Operations, Industry & Finance Division Country Department II South Asia This document has a restricted distribution and may be used b recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EOUIVALENTS (Annual Average) Currency Unit Indian Rupees (Rs.) 1986 $1.00 Rs. 12.6 Official Rate 1987 $1.00 Rs. 13.0 " 1988 $1.00 Rs. 13.9 1989 $J.00 Rs. 16.2 i 1990 $1.00 Rs. 17.5 " 1991 $1.00 Rs. 22.7 " 1992 $1.00 Rs. 26.0 " 1992 $1.00 = Rs. 30.4 Market Rate FISCAL YEAR April 1 to March 31 ABBREVIATIONS AND ACRONYMS USED AIEI: Association of Indian Engineering Industries BOB: Bank of Baroda DeCTA: Developing Countries Trade Agency EMF: Export Marketing Fund ERR: Economic Rate of Return EXIM: The Export Import Bank of India FRR: Financial Rate of Return GOI: Government of India ICICI: Industrial Credit and Investment Corporation of India Ltd. IEP: Industrial Export Project - Engineering Products PCB: Participating Commercial Bank PCR: Project Completion Report PF: Productivity Fund PNB: Punjab National Bank RBI: Reserve Bank of India SB}: State Bank of India  FOR OFFICIL USE ONLY THE WORLD BANK Washlngton, D.C. 20433 USA Office of Orectreneral Oper. ons Evaluaion December 17, 1993 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECr: Project Completion Report on India - Industrial Export Prolect, EXn&eering Produts (Loans 2629 and 2630-IN) Attached is the Project Completion Report on India - Industrial Export Project, Engineering Products (Loans 2629 and 2630-IN) prepared by the South Asia Regional Office. Part H was prepared by the Borrowers. The PCR is of satisfactory quality. It provides a clear account of the experience with a complex project. Although the project became effective by mid-1986 and two intermediary financial insttutions began committing fnds shortly thereafter, it took others some time to put in place procedures and staff to utilize the loan. The unfamiliarity of the sub-borrowers with the Bank's currency pool and their reluctance to accept the exchange risk eventually convinced the Government of India to take over this risk Once this decision was made and the interest rates were subsequently realigned with the market rates, both commitments and disbursements improved. The rate of commitment fauther improved by widening the use of fimds to cover all export projects, rather than focussing primarily on engineering products. When the loan closed on June 30, 1992, some $5.2 million of Loan 2630-N was undisbursed and hence cancelled. The project was followed by two subsequent loans for Export Development In 1989 (Loans 3058 and 3059-IN). It led to substantial investment in the export industries of India. It also helped two of the participating banks to derive valuable experience in promoting export markeedng and productivity enhancing activities. The outcome is rated as satisfactory and the project bas had partal institutional impact. Given the current economic environment in India, sustainability is rated as likely. An audit is planned in FY95. Attachment I doum ha rcted ltution ad may b used by recip only thyin perforance of their of ficidl duis. It contt y not odemws be disclosed wihou Wodd aOk authotizaon.  FOR OMCUIL USE ONLY PROJECTtCOMPLET9IO PRT INDIA XIDSTRIAL EXPORT PRojTzs - EaIEeN PmODUm (LOANS 2629-IN and 2630-IN) Table of Contents Paa2o. Preface . . . . . . . . . . . . . . . . . . . . . . . . ..... . Evaluation Summary . . . . . . . . . . . . . . ... ..iii PTIs PROJECT REVIEW FROM THE BANIK' S PERSPECT.IVE . ..... . 1 I. Proiect Identity . . . . . . . . . . . . . . . . . . . . . . . . . . 1 II. Project Background . . . . . . . . . . . . . . . . 1. . Link with Industrial and Trade Policy Reform . . . . . . . . . .1 _II. Proiect Obiectives and Description . . . . . . . . ........ . . 2 Project Objectives ... . . . . . . . . . . . . . . . *... . . 2 Project Description .. . . . . . o . . . 2 IV. Proiect Desian and Organization .. ... . . . . . . . . 2 The Participating Commercial Bank Sub-Loan Component . . . . . . . . 2 The ICICI Sub-Loan Component . . . . . . . . . . . . . . . . . . ... 2 The Export Marketing Fund .. . . . . . . . . . . . . . . . 3 The Productivity Fund . . . . . . . . . . 3. . ... 3 V. Proiect Imnlementation ..9. ....3 Effectiveness and Start Up . . . . . . . . . . . . . . . . . . . 3 Implementation *. . . . . . . . Pisbursements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Procurement . . . . . . . . . . . . . . . . . . . . . . . . . 5 Costs and Financing . . . . . . . . . . . . . . . . . . . . . . . . . 5 VI. Proiect Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Impact on Overall Export Performance . . . . . . . . . . . . . . . 5 Sub-Loan Performance . . . . . . . . . . . . . . . . . 5 Productivity and Export Marketing Funds Performance . . . . . . . . . 6 VII. Proiect Sustainabilitv ... . . . . . . . .. . 7 This document has a restricted distrtbution 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.  Pae no VIII. Performance of the Bank and Imolementing agencieg . . . . . . . . . . 8 World Bank Performance . . . . . . . . . . . . . . . . . . . . . . . 8 Implementing Agencies Performance . . . . .*.. . . . . . . . . . 8 IX. Lessons From the Proiect . .................... . 8 PARTIBfl PROJECT REVIXW PROM BORROWER'S PERSPBCTIVE . . . . . . . . . . . 9 Comments from ICICI .......... . . . . . . . . 10 Comments from the Export Import Bank of India . . . . . . . . . 12 Comments from Bank of Baroda ........ . ... ..... 15 Comments from Punjab National Bank . . . . . . . . . . . . . . 17 Comments from Canara Bank . . . . . . . . . .. . . . . . . 18 PART XII: STATISTICAL INFORMATION . . . . ... . . . . . . . . 21 Tables I Related Bank Loans . . . . . . . . .... .. . . . . . 22 2A. Planned Project Costs and Financing . . . . . . . . . . . . . . 23 25. Actual Project Costs and Financing . . . . . . . . . . . . . . . 23 3. Sub-loan Lending Terms . . . . . . . . . . . . . . . . . . . .24 4. Project Timetable . . . . . . . . . . . . . . . . . . . . . . . 25 5. Loan Commitments . . . . .... . ... . . . . . . . . . . . 26 6. Loan .isbursements . . . . . . . . . . . . . . ... . . . . . 27 7. Selected Indicators of India's Export Achievement . . . . . . . 28 8. Selected Indicators of Sub-loan Achievement . . . . . . . . . . 29 9A. Selected Indications of WM and PF Achievements . . . . . . . . 30 95. Activities Funded By Grant Programs . . . . . . . . . . . . . . 30 1OA. Performance of Implementing Agencies - 1987 . . . . . . . . . . 31 108. Performance of Implementing Agencies - 1992 . . . . . . . . . . 31 11. Use of Bank Resources . . .. . . . . . . . . . . .32 12. Mission Data . . . . . . 3 . . . . . . . . . . . 33 13. Status of Compliance with Covenants . . . . . . . . . . . . . . 34 14. List of ICICI Sub-loans . . . . . . . . . . . . . . . . . . . . 35 15. List of Commercial Bank Sub-loans . . . . . . . . . . . . * .38 16. List of EXIM PM Grants . . . . . . . . . . . . . . . . . . . . 44 17. List of ICCI PF Grants . .... . . . . . . . . . . 48  PROJECT COMPLETION REPORT INDIA INDUSTRIAL EXPORT PROJECT - ENGINEERING PRODUCTS (LOANS 2629-IN and 2630-IN) Preface This is the Project Completion Report (PCR) for the Industrial Export Project - Engineering Products in India for which Loan 2629-IN in the amount of US$90 million and Loan 2630-IN in the amount of US$160 million were approved on January 26, 19P6. Both loans closed on June 30, 1992, two and a half years later than originally envisaged. US$90 million was disbursed from Loan 2629-IN and US$154.8 million was disbursed from Loan 2630-IN. The balance of US$5.2 million of Loan 2630-1N was canceled. The PCR (Preface, Evaluation Summary, and Parts I and III) was prepared by the Country Operations, Industry and Finance Division of Country Department II, South Asia. Part II contains separate written comments on the project submitted by five of the six implementing agencies. In some cases, the implementing agencies submitted longer reports and more detailed statistics which are available for review in the project file. This PCR was prepared during FY93 and is based on the Staff Appraisal Report, the Loan and Project Agreements, supervision reports, correspondence between the World Bank and the Implementing Agencies and internal World Bank memoranda.  - iii - PROJECT COMPLEION RPORT INDIA INDUSTRIAL EXPORT PROJECT -- ENGINEERING PRODUCTS (LOINS 2629-ZN and 2630-IN) EVALUATION SRY Obiectives 1. The main objectives of the project were to support trade policy reform and boost Indian exports by: (a) increasing the availability of term export finance, particularly in the engineering sector; and (b) catalyzing increased spending by companies on export promotional activities (para. 3.1). The project built on a foundation of 14 previous loans to the Industrial Credit and Investment Corporation of India (ICICI) in providing term finance for industrial projects. However, it differed from past loans in targeting engineering and export projects, in providing matching grant support for pre- project export marketing and productivity activities, and in using multiple financial institutions (para. 2.2). 2. To achieve its objectives the project was designed to provide: (a) $160 million to ICICI to be relent to industrial companies for export projects primarily in the engineering field; (b) $70 million to four participating commercial banks (PCBs) to be relent to ancillary engineering companies for export or import substitution projects; (c) $10 million to be extended as matching grants by the Export Import Bank of India (EXIM) to engineering companies for marketing activities through the Export Marketing Fund (EMF); and (d) $10 million to be extended as matching grants by ICICI to engineering companies for productivity improvements through the Productivity Fund (PF) (para. 3.2). Imnlementation Exnerience 3. The loan was approved on January 21, 1985 and became effective on June 23 1986. ICICI and Bank of Baroda began to commit sub-loans shortly thereafter, in August of 1986. EXIM and Canara Bank began to commit funds in the first half of 1987 and the other PCBs in 1988. This delay was largely due to the need for the participating institutions to put into place appropriate procedures and staff (para. 5.1). An additional cause for delay in committing marketing and productivity grants was that many eligible activities (such as hiring overseas consultants) required governmental approval. To address this, in mid-1987 the Government of India (GOI) granted blanket government permits for approved activities under the EMF and PF programs (para. 5.2). 4. Loan funds were committed slowly through mid-1988 because sub-loan interest rates were proving unattractive to borrowing companies. In the case of ICICI, sub-loans were denominated in the Bank's currency pool which was not well understood by borrowing companies. To make these funds more attractive, in February 1988 the GOI agreed to take the exchange rate risk and for ICICI sub- loans to be denominated in Rupees at 14W. Interest rates on commercial sub- loans, initially set at 15%, were adjusted in November 1988 to a similar level (para. 5.3). The rate of commitments was further improved by widening the use of funds to cover all export projects, rather than focusing primarily on engineering projects (para. 5.4). These changes were successful and by 1990 most of the funds had been committed. The loans closed on June 30, 1992 and, in accordance with Bank policy, disbursements were permitted until October 1992.  - iv - Results S. The project funded 439 sub-loans amounting to $229 million and 359 grants under the PF and EMP amounting to $16 million. The average size of ICICI loans was about $1,000,000, of the PCB loans was about $250,000, and of the EMF and PF grants was about $50,000. 6. Overall, the loan seems to have helped promote export growth by enabling exporters to inport capital goods and make promotional investments during an economic period in which the availability of foreign exchange was constrained. Manufactured exoorts rose by about 10% in real terms per annum during tne covered five year period, greater than both overall export performance during the covered period and manufacturing export performance in the previous five years (para. 6.1). 7. The financial performance of sub-loans and grants appears to be strong. Average financial rates of return on sub-loans among the implementing agencies ranged from 20t to 40% and the number of cases of long term arrears or reschedulings was not excessive (para. 6.3). As regards EMP and PF grants, assisted companies achieved incremental exports over five years of, respectively, 60 times and 37 times the amount of the grant. lhis compares favorably to a minimum target of 10 times established in the follow-up project (para. 6.5). Of course, these figures need to be interpreted with caution since country and company export performance can only in part be attributable to Bank assistance. 8. The performance of the project goes beyond quantitative calculations. By helping six financial institutions to increase their export focus and by helping about 500 companies develop export strategies through either grants or sub-loans, the project was instrumental in helping to overcome India's supply- driven export orientation and in encouraging companies and financial institutions to focus instead on improving production efficiency and better meeting the needs of overseas customers. This change in thinking is likely to be of critical importance to the long term c.-npetitiveness of Indian companies (para. 6.7). Sustainability 9. Up until recently, both grant and sub-loan activities financed under the Industrial Export Prcject were being sustained through a follow-up project - the Export Development Project (Loans 3058-IN and 3059-IN) - approved in 1989. Due to recent reforms which included making the Rupee partially convertible, part of the sub-loan component under Loan 3059-IN was effectively stalled as sub- borrowers found the on-lending arrangements unattractive. In response, the affected part of the sub-loan component is being cancelled. In its stead, th.e Bank and the Government are considering a follow-up project that would make use of the single currency option recently approved by the Board for loans meeting certain criteria. A single currency option loan could be made more attractive to sub-borrowers in its on-lending arrangements and at the same time not require the Government to shoulder exchange risk. 10. Given the on-going reform program, the demand for export finance is likely to grow. ICICI and EXIM are considered to be generally well-run institutions that are likely to capitalize on their experience under this loan and maintain a strong ability to make term sub-loans in support of export projects (para. 7.2). Over the past few years, the commercial banks have significantly increased their short term export finance as well as their term lending, including for export projects. Nevertheless, in order to enhance their capability to respond to the emerging credit needs of exporters, both financial and institutional restructuring of commercial banks will be required. The Bank is currently discussing these issues with GOI within the context of a proposed financial sector adjustment operation (para. 7.3). The sustainability of the technical assistance component to exporters without further Bank assistance is  v limited. ICICI has obtained a small amount of additional funds from the British Agency DeCTA and EXIM has initiated a program that supports export promotional activities with Rupee based funds. Ultimately, however, to ensure sustainability, it would be necessary to shift from grant-based support to instruments which provide for the recycling of funds. Findinqs and Lessons Learned 11. A number of lessons can be learned from this project, some of which have been incorporated into the follow-up Export Development Project. These lessons (para. 9.1) include: (i) sectoral allocations in financial intermediation lending should be left to the market when possible. The initial allocation in the project for engineering goods was eased because, while engineering exports did perform well, many other non-traditional, chemical and computer software exports performed as well or better; (ii) on-lending arrangements in financial intermediation lending should be easily understood by sub-borrowers, market based, and easily adjustable. The initial interest rates on ICICI sub-loans were set using the Bank's currency pool but this proved unattractive to sub- borrc-ers who were not familiar with it. The initial interest rates on c' - aercial bank sub-loans was a Rupee rate but it was too high and * I to be reduced; and (iii) there should be more coordination across financial instruments provided for under the same project. While it was anticipated that thc use of loan and grant funds in supporting the same company would be coordinated, in fact they were approved independently by different implementing institutions, resulting in both duplication of effort and a weakening of an unified export strategy.  PROJECT COMPLETION REPORT INDIA INDUSTRIAL EXPORT PROJECT -- ENQINERRING PRODUCTS (LOANS 2629-IN and 2630-IN) PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE I. PROJECT IDENTITY Name Industrial Export Project -- Engineering Products Loan No. : 2629-IN and 2630-IN RVP Unit : South Asia Country : India Sector : Industry II. PROJECT BACKGROUND 2.1 The Bank's lending to industry in India has traditionally been dominated by general lines of credit to a few large development banks. From 1955 to 1985 the Bank financed 14 consecutive loans totalling over $1 billion to support Indian industry through the Industrial Credit and znvestment Corporation of India (ICICI). The "Industrial Export Project - Engineering Productsit (IEP) was, in part, a continuation of these loans. 2.2 At the same time, it differed from past loans in three important respects. First, in recognition of ICICI's ability to raise funds on its own, another general line of credit was not considered necessary. Rather, the main constraint in the financial sector was the allocation of credit, given the high set-asides for "priority" sectors and the rigid system of interest rates. Thus, the project targeted funds on engineering projects, in order to assure adequate funding for a sector considered to have a high level of export potential. Second, given the history of protection in India and the lack of exposure of Indian firms to international competition, it provided direct matching grant support, in addition to loan funds, to help catalyze export promotional investment by individual firms. Third, it reflected a shift in Bank policy towards using multiple financial institutions, for the first time including four participating commercial banks (PCBs) and the Indian Export-Import Bank (EXIM) as well as ICICI as implementing agencies. By using commercial banks, the Bank hoped to widen the expertise on exports in the financial sector and reach a wider nll-.er of downstream small and medium sized companies, including in particular ancillary companies. 2.3 Link With Industrial and Trade Policy Reform: The project's targeting of engineering products as having export potential grew out of three years of trade sector work, culminating in a paper on nLight Industrial Export Development and Finance" completed in 1983. Consistent with this sector work, the Government of India undertook a series of industrial and trade policy reforms in 1984 and 1985 including easing industrial license requirements, reducing tariffs on capital goods imports, and an exchange rate depreciation of about 25W. These reforms provided the foundation for the project by improving the potential international competitiveness of Indian industry. While the project did not directly containi policy conditionality, it was designed to support these reforms by ensuring that export projects, particularly in the engineering sector, had adequate financing in order to take full advantage of the policy changes.  III. PROJECT OB;RCTIVES AND DN8CRIPTION 3.1 Proiect oblectives: The main 3bjectives of the project were to support trade policy reforms and boost Indian exports by: (a) increasing the availability of term credit provided to companies with export potential, or companies producing efficient import substitutes with important linkages to exports; and (b) catalyzing increased investment and thinking by private companies on pre-project promotional activities in the areas of export marketing and production. Because it was felt that Indian engineering products had a comparative advantage for competing in the international marke;.?lace, a large part of the loan was initially targeted to this sector. 3.2 Prolect DescriPtion: The project comprised two loans, one to the Government of India and one to ICICI. The loan to the Government of India, Loan 2629-IN, was divided into three components: (a) US$70 milli - to be passed on to four government-owned participating commercial banks (PCBs) for financing term sub-loans to private ancillary engineering firms; (b) US$10 million to be provided as matching grants by EXIM to engineering companies for export marketing activities; and (c) US$10 million to be provided as matching grants by ICICI to engineering companies for productivity improvements necessary to compete internationally. Loan 2630-IN provided US$160 million directly to ICICI, guaranteed by the Government _f India, for financing term sub-loans for export- oriented projects, of which $100 million was restricted for engineering projects. The above Bank funding, totaling $250 million, was expected to catalyze an additional $750 million, thereby enabling total project costs of $1 billion to be funded. IV. PROJECT DESIGN AND ORGANIZATION 4.1 The project, as described above, had four components: (a) sub-loans administered by four participating commercial banks; (b) sub-loans administered by ICTCI; (c) the Export Marketing Fund (EMF) administered by EXIM; and (d) the Productivity Fund (PF) administered by ICICI. The design of each component is discussed below: 4.2 The Participatincq Commercial Bank Sub-Lcan Comoonent: This component was financed under Loan 2629-IN to the GOI. Under this component, the commercial banks extended sub-loans to eligible companies, which were refinanced by the GOI in the form of equity out of the proceeds of the Bank loan. In order to target ancillary firms, the maximum commercial bank sub-loans was set at $500,000 and the maximum size of supported companies was set at $2 million in fixed assets. Repayment was denominated in Rupees and the rate of interest on the sub-loans was initially set at 15%, the non-subsidized commercial bank term rate existing at the tirme of loan negotiations. This was reduced subsequently to 14* for medium sized companies and 13.5% for small companies. In order to be eligible for financing, the borrowing company had to finance 20% of the project internally, have a debt to equity ratio of not greater than 3:1, and have net revenues of at least 1.5 times total debt service. 4.3 The ICICI Sub-Loan Comionent: This component was financed under Loan 2630-IN, made directly to ICICI and guaranteed by the GOI. Out of a total amount of $160 million, $100 million was initially allocated for engineering companies (later reduced to $60 million). Initially, repayment by the sub-borrower was denominated using the World Bank's variable currency pool and interest rate plus a 2% intermediary spread to ICICI. However, this did not prove attractive to sub-borrower.s. Consequently, the GOI agreed to take the foreign exchange risk permitting thereby ICICI sub-loans to be denominated in Rupees with an interest rate of 14% (see para 5.3 below).  4.4 The Export Marketincr Fund (EMFI: The EMF was administered by EXIM. Under this component, EXIM extended grants to private companies covering up to 50% of eligible activities, with the remaining funds coming from internal company accruals. EXIM approved grants up to $50,000, with larger grants requiring approval from a steering committee composed of government and financial institution representat_ves. Initially, only 11 product groups were eligible for funding (composed of engineering products and computer software), but this restriction was subsequently eased. Eligible activities included desk and overseas research, travel to and from India, staff training, minor product adaptation, quality certification, and establishing overseas operations. EXIM was to es-i.ablish a separate cell to administer the EMF composed of a manager based at headquarters and four project officers, one each based in Delhi, Bombay, Calcutta, and1ladras. Currently, the EXIM marketing cell consists of a Manager, two project officers in Bombay, two project officers in Madras, and one project officer in New Delhi. 4.5 The Productivity Fund (PF): The PF was administered by ICICI. Like with the EMF, grants were extended under the PF to private companies covering up to 50% of eligible activities. ICICI could approve grants up to $100,000, in consultation with the Association of Indian Engineering Industries (AIEI). Larger grants required approval from a steering committee of government and financial institution representatives. PF financing per project was restricted to a maximum level of $200,000. Initially, only "thrust" engineering products were eligible for funding, although this restriction was subsequently eased. Eligible activities included hiring productivity consultants, staff training, product adaptation, and exposure trips to see foreign operating systems. This was later broadened to include all export marketing activities, as provided by the EMF. ICICI established a separate cell to administer the PF composed of a manager and four staff officers. V. PROJECT IMPLEMENTATION 5.1 Effectiveness and Start-an: The project was identified and appraised during 1984, negotiated during Aug-Sept 1985, approved by the Executive Directors in October 1985 and signed on Ja:uary 21, 1986. Both loans became effective on June 23, 1986, five months later. ICICI and Bank of Baroda began to commit sub- loans shortly thereafter, in August 1986. EXIM and Canara Bank began to commit funds in the first half of 1987 and the other PCBs in 1988 (see Table 5). For commercial banks, this delay was due primarily to the need for each bank to develop internal structures and procedures and to agree with the GOI on these procedures. Even after documentation was completed, these were new procedures for the commercial banks and it took time for them to adjust. 5.2 The EMF and the PF were new tasks which also took time for the implementing agencies, EXIM and ICICI, to gain experience. An additional factor contributing to the slow start up of the EMF and PF was that many of the eligible activities used foreign currency (such as hiring foreign consultants, overseas travel, and technology imports) and this required governmental clearances. To address this, in mid-1987 the GOI agreed to give the steering committees for both funds blanket government permits for approved activities. These blanket permits, not envisaged initially, eventually proved to be among the main attractions of the EMF and PF programs. 5.3 Implementation: Loan commitments were generally sluggish until 1989, after which commitments picked up considerably. The main reason for this sluggishness was that sub-loan interest rates, as initially set for both ICICI and commercial banks, were not attract&ife to borrowing companies. ICICI sub-loan interest rates, initially set at the World Bank currency pool rate plus 2*, proved unattractive to borrowing companies who were for the most part unfamiliar with the currency pool and reluctant to take on exchange rate risk at a time of rapid Rupee devaluation. To rectify this, the GOI agreed in February 1988 to  - 4 - take on the foreign exchange risk and redenominate ICICI sub-loans in Rupees carrying an interest rate of 14%. Unrder the revised scheme, for each sub-loan ICICI agreed to pay the GOI the difference between the World Bank interest rate and the sub-loan interest rate, less a retained spread. In return, the GOI agreed to compensate ICICI for the foreign exchange risk on servicing the World Bank loan. To bring commercial bank sub-loans into line with this, agreement was reached in November 1988 to adjust interest rates on commercial bank loans from 15% to 14*, and 13.5% for small scale enterprises (see Table 3). 5.4 Several other actions were taken in 1988 to speed commitments. First, it became apparent that several non-engineering products held significant export promise. As a result, in February 1988 at the request of the GOI, the Bank agreed to reduce the allocation for engineering products for ICICI loans from $100 million to $60 million, and eliminate restrictions for commercial bank sub-loans and the PF (restrictions were also informally eased for the EMF which had, in any case. been permitted at the outset to fund computer software projects). Second, the Bank agreed that ICICI sub-loans could cover Rupee costs as well as foreign currency costs, up to 80% of each sub-project. Third, for commercial barks, the Bank agreed to increase the maximum size of sub-loans from $500,000 to $1 million and the maximum asset size of eligible companies from $2 million to $4 million. An additional factor which acted to speed up commitments was that commercial banks obtained funds as equity. 5.5 Disbursements: At appraisal it was projected that disbursements would begin in January 1986 and be fully completed by December 1989, some four years later. In fact, for all components except ICICI sub-loans, disbursements did not start until 1988 and were not completed until October 1992. This delay in disbursements can be primarily attributed to the initial delay in commitments for reasons already discussed in paras. 5.1 - 5.4. An additional factor was that during 1990-91, the GOI imposed tight import controls which made the opening of new letters of credit difficult, even on projects already authorized for Bank funding. A third factor was that there was sometimes a time lag of up to six months between the submission of a claim for reimbursement by the implementing agencies to the GOI and the submission of those claims by the GOI to the World Bank. Finally, many PF and EMF grants remained partly undisbursed more than two years after commitment, in part because of the Rupee devaluation and in part because of changing market conditions. 5.6 In order to compensate for slow sub-loan disbursements and the possibility of eventual cancellation of some of these sub-loans, the Bank authorized sub-loan commitments on the commercial bank line of credit totalling an amount in excess of funds available by about $8 million. While at the time this was considered to be an effective way of ensuring maximum sub-loan utilization, claims were eventually submitted for most of the loan commitments. On the other hand, disbursements on the PF and EMF grants remained far below commitments due to reasons noted in para. 5.5. Because the underdisbursement on the EMF and PF largely compensated for overcommitments on sub-loans, the Bank was able to reimburse all claim applications submitted to it except for about $1.6 million. 5.7 The Closing Date was extended three times, first to December 1990, then to December 1991 and a third time to June 1992. In accordance with Bank practice, disbursements continued to be honored for four months after the closing date, i.e. until October 31, 1992. Underdisbursements of $5.2 million on the ICICI sub-loan component could not be used to meet claims from other components, since the ICICI sub-loan component was provided through a separate World Bank loan, and these funds were canceled.  - 5 - 5.8 Procurement: The implementing agencies were responsible for reviewing and approving procurement. The detailed reports submitted to the World Bank on purchasing plans were generally of a high quality and very few funding requests were denied. Given the limited size of the consultant contracts approved under the PF and EMF, the World Bank did not generally request competitive bidding documentation, although it encouraged participating agencies to follow Bank procurement guidelines. 5.9 Costs and Financinc: The project financing plan projected that World Bank funds of $250 million would be matched by $750 million from other sources for total project funding of $1 billion. A review of sub-loan data indicates that this target was more than achieved, with total project funding amounting to about $1.2 billion. However, projects did not obtain foreign commercial borrowings as anticipated. Rather, the bulk of the matching funds came through consortium lending from other Indian development banks (IDBI and IFCI) as well as a growing local equity market (see Tables 2A and 2B). VI. PROJFCT RESULTS 6.1 Imnact on Overall Exoort Performance: The goal of the loan was to boost Indian exports, particularly in the engineering sector. In fact, export performance in the manufacturing sector was very good during the five year period following 1987, increasing by about 10% in real terms per annum. This compared favorably with both the previous five year period and with total Indian export performance. Engineering exports rose even faster, by 11% in real terms per annum (see Table 7). How much of this export increase was due to the existence of the World Bank loan is uncertain, although surely some could be attributed both to the loan and to the improved export policies that it supported. Implementing agencies reported that the availability of the World Bank line of credit was particularly useful in enabling companies to import capital goods during 1990-91 when the GOI placed restrictions on such imports in order to protect its low foreign exchange reserves. In addition to facilitating exports, commercial banks reported that there was considerable foreign exchange savings stemming from enabling ancillary engineering firms to provide larger firms with import substitutes. 6.2 Sub-Loan Performance: In total, some $229 million was disbursed in Support of 439 loans. This was composed of $155 million on 144 loans from ICICI with an average size of about $1 million (including 4 loans of more than $5 million each) and $75 million on 295 loans from the four participating commercial banks with an average loan size of about $250,000. Most of these loans were used to import machinery necessary to make the manufacturing process more competitive. While most of these funds were reserved for engineering firms, ultimately these reservations were lifted and less than half went to such firms. While some of these loans were made to large exporting firms which might have had access to funding without this project, a large number also went for new firms or firms with new export projects. For example Bank of Baroda (BOB) reports that over half of its loans were for new export projects. 6.3 The financial results of sub-loans look good, although information on many sub-loans is spotty. The average financial rate of return varied across institutions, from about 20* reported by Bank of Baroda and PNB to 40% reported by Canara Bank. Repayment performance also appears to be reasonable, although not exceptional. About 13% of ICICI sub-loans extended under this project have been rescheduled or are in arrears by more than 1 year. This is roughly in line with ICICI's overall portfolio performance. At the same time, less than 5% of commercial bank sub-loans have been rescheduled or are in arrears by more than 1 year (see Table 8).  - 6 - 6.4 Productivity and Exrort Marketing Funds Performance: The Productivity and Marketing Funds together committed $22 million ;nd disbursed $15.6 million consisting of 359 grants to 292 companies (63 companies obtained more than one grant). The average grant size was about $50,000 although grant levels varied considerably with 45 companies obtaining grants of less than $10,000 and 17 companies obtaining grants of more than $200,000. The EXIM reports that about half of EMF funded companies had no previous export experience. While initially reserved primarily for engineering companies, these reservations were lifted and, in the end, only about 71% of the EMF and 61% of the PF went to engineering companies, the largest share of which was for Autos and Auto parts. Of the activities funded, about half of the funds were spent on consulting or research and on business promotional travel abroad (see Table 9B). The blanket permit extended by the GOI on activities funded under the EMF and PF proved to be a particularly attractive feature of the funds. 6.5 While the Staff Appraisal Report for this loan does not include a performance measure for these funds, the follow-up project suggests that they should generate incremental exports of 10 times the grant support over a five year period. Using this measure, the EXIM and ICICI report that the companies supported by the EMF and the PF generated incremental exports over the past five year period of, respectively, 60 times and 37 times the amount of grant support in real or US Dollar terms. A second measure reported was that on average assisted companies under the EMF and PF achieved annual export growth rates of 68% and 48% respectively (see Table 9A). ICICI also reported that companies which received PF grants have shown a significant improvement in their plant capacity utilization and a reduction in their rejection rates. While these measures are impressive, they should, of course, be interpreted with caution, since company export performance can only in part be attributable to obtaining grant funding. Other factors at work include government policies, company management, and product competitiveness. 6.6 Operationally, one performance issue was that a significant number of grants were disbursed either on activities which took place well before or far after the original approval date. Under the EMF, 20 grants financed activities which had taken place more than six months before authorization and 32 grants continued to disburse more than two years after authorization. For the PF the average time lag from sanction to final disbursement has also been long, about 18 months. While no provision was made in the project to restrict ex-post EMF & Pr funding, such funding brings into doubt the catalytic nature of the grants. In the case of long-delayed disbursements, activities sometimes differed from those originally authorized. A related problem was that no formal cancellations were requested for PF grants, despite the fact that a significant amount of committed funds was never disbursed. To address these concerns, ex-post funding over 3 months should be restricted in future projects and funds not disbursed more than two years from approval should be automatically cancelled unless an explicit reauthorization is obtained. 6.7 The performance of both the sub-loan components and the PF and EMF grant components goes beyond strictly numerical calculations. Each loan or grant request was accompanied by an in-depth review of the company's export strategy. The implementing agencies have reported that these reviews have been useful in helping overcome the supply-driven nature of Indian exports and encouraging companies to focus rather on improving production efficiency and on better meeting the needs of overseas clients. Regardless of the effects of particular grants or loans, supported companies should be more likely to recognize the importance of export investments in maintaining their competitiveness in the future.  VII. PROJICT SUSTAINAfBIITY Sustainabilitv 7.1 Up until recently, both grant and sub-loan activities financed under the Industrial Export Project were being sustained through a follow-up project - the Export Development Project (Loans 3058-IN and 3059-IN) - approved in 1989. This follow-up project provided funding for the same activities and included, in addition, a technical assistance component for improving the managerial capabilities of the implementing agencies in evaluating export projects. Due to recent economic reforms which included measures to make the Rupee partially convertible, part of the sub-loan component under Loan 3059-IN was effectively stalled as sub-borrowers found the on-lending arrangements no longer attractive. In response, the affected part of the sub-loan component is being cancelled. In its stead, the Bank and the Government are considering a follow-up project, involving the single currency option recently approved by the Board, that could be made more attractive to sub-borrowers in its on-lending arrangements, and at the same time not require the Government or the RBI to shoulder exchange risk. 7.2 Lending for export projects by EXIM and ICICI should be sustainable in the medium to long term. ICICI is a mature and generally well-run institution which is in the process of reforming its operations to be more market oriented and to mobilize increasing amounts of resources on its own. In a freer policy environment it is likely to be able to capitalize on its experience under this project and continue to make loans to bankable export projects. The EXIM is a newer institution, but given its export orientation it is also likely to internalize significant gains from this project. The EXIM has also begun to supplement its public support through raising funds from fixed deposits and borrowing from public institutions and this should further enhance its lending capacity. 7.3 Commercial banks are also expected to play an expanding role in financing export activity. A major goal of the on-going reforms is to increase India's export performance which, if successful, should result in increasing demand for both short term export finance and term finance, including for export projects, from commercial banks. From 1991 to 1993 commercial banks have, in fact, increased their short term export finance by about 50t in Rupee terms. This has been facilitated in part by the reduction in mandatory reserve requirements of commercial banks, generous refinance facilities provided by the RBI, and a government directive to the commercial banks to provide at least 10% of their credit for export finance, a level that some of the banks have already surpassed (for example 12.7% of Canara Bank's total credit is now export oriented). The commercial banks have also begun to increase their term lending for both export and domestic oriented clients. Term loans now account for about 30t of credit provided by Canara Bank, Bank of Baroda, and Punjab National Bank. However, assuming the demand for export finance continues to increase, additional steps will be needed to enhance the capability of banks to respond to the emerging credit needs of exporters, including financial and institutional restructuring. The Bank is currently discussing these issues with GOl within the context of a proposed financial sector adjustment operation. 7.4 The ability of ICICI and EXIM to sustain their grant-financed programs for export promotion without World Bank support is more limited. ICICI has obtained a small amount of additional grant funds for export promotion from the British agency DeCTA and EXIM has initiated a program that supports export marketing activities with Rupee based funds. However, these programs are much more limited than that supported by the Bank. Since grant funding does not involve any reflow of funds, continual funding would be needed to sustain this component as currently structured. Of course, even without external assistance, supported companies should be better able to see the value of making export investment on their own.  - V *III. PEFORMANCE OF THE BANK AND THE niPLEmEng AGENCIES 8.1 World Bank Performance: During the seven years between the Loan's approval and its completion, the World Bank supervised the project 12 times (see table 12). Missions met with implementing agencies, exporters, and GOI officials and were directed largely at monitoring commitments and disbursements and resolving bottlenecks. In addition to mission work, the Task Manager reviewed and approved each individual sub-loan and grant from headquarters. These approvals were largely done in a timely manner, usually within two weeks. 8.2 ImDlementinc Aaencies Performance: A number of the implementing agencies were slow to start, which is understandable since the activities were relatively new. As they staffed units and gained expertise, commitments picked up. All parties worked well together in 1987 and 1988 to resolve interest rate and eligibility bottlenecks. The implementing agencies have generally been diligent in seeking out and working with potential exporters in reviewing projects and preparing detailed export strategies. Each agency generally provided comprehensive documentation on each sub-loan and grant to the World Bank Task Manager for review and prepared a useful general implementation review during each supervision mission (normally about twice yearly). However, these reports were prepared using different formats and did not generally include planned vs. actual performance measures. A more systematic approach towards data collection would have facilitated monitoring during supervision, preparing the PCR, and designing future operations. IX. LESSONS FROM TUN PROJECT 9.1 While the project was generally a success, several lessons can be learned from this project, some of which have been incorporated into the follow- up Export Development Project. These lessons include: (i) Let the market decide where exnorts will be: The initial focus of the project on engineering goods as having the most export potential was well documented. However, while engineering exports did outperform overall exports, there were other products such as non- traditional goods, chemicals, and computer software which performed as well or better. In recognition of this, the allocation in this project for engineering goods were eased and in the follow-up project funds were made available to all manufacturing and computer software firms. (ii) Let the market decide the on-lendincr rate: Slow initial commitments could be directly traced to the fact that the initial rates established for sub-loans were uncompetitive, unattractive to borrowers, and had to be revised. Towards the end of the project rates were somewhat below-market, which may have also distorted the demand for loans. In future projects, on-lending arrangements used in financial intermediation should be easily understood by sub- borrowers, market based, and easily adjustable. A related point is that passing loan funds to commercial banks in the form of equity introduced demand pressures unrelated to sub-project quality. For this reason, the issues of recapitalizing the financial sector and loan funding should be dealt with separately when feasible. (iii) Coordinate the Productivity Fund. the Marketing Fund and sub-loans more: While it was anticipated that the use of loan and grant funds would be coordinated, in fact sub-loans and grants for the same company were largely approved independently by different implementing institutions, resulting in duplication in both documentation and government clearances as well as not developing a coordinated approach to a company's overall export strategy. The follow-up project attempts to address this by combining the activities provided under the Productivity and Marketing Funds and permitting each participating institution to provide both grant funds and sub-loans.  9- PART IIs 2ROJBCT RXVIEW FROM TIM BORROWUR' PZRBPECTXV Written comments on the project have been received from five of the six implementing agencies. The comments listed below are executive summaries prepared by these agencies. In some cases, the implementing agencies submitted longer reports and more detailed statistics which are available for review in the project file.  Page 1 of 2 - 10 - PROJECT REVIEW FROM ICICI'S PERSPECTIVB (Productivity Fund) OBJECTIVE 1. The Productivity Fund of US$10 million was operated by ICICI for providing financial assistance by way of grant to Indian exporters undertaking Export Development programmes. The purpose of the fund when it was conceived, was to improve the exports and export capability of India companies. Initially the fund was for providing assistance for Productivity Improvement programmes for export mlarkets. Subsequently, export market related activities were also included within the framework of the fund. TJTLIZATION 2. A total of 152 schemes for 112 companies were assisted under the Productivity Fund during the period 1986-1992. The details of the utilization of the funds are: o Total fund was US$10 million * Total sanctions US$11.67 million * Total Disbursements US$9.463 million IMPACT OP TRE FMND 3. The following indicators have been used to analyze the impact of the fund: Exvort Related: * Total exports for the group of Productivity Fund assisted companies was US$61.77 million in 1986-87. This had increased to exports of US$208.68 million by 1991-1992. e Incremental exports for the companies over the 5 year period 1986-1992 has been US$434.42 million. This is 37.2 times the sanctioned amount. e The actual incremental exports is about 9t lower than the planned incremental exports. The Electronic/Software and Consumer Product groups performed below expectations, while Chemicals and Textiles performed above planned. o In 1987-88, 47 companies had exports more than 5% of the total sales whereas, by 1991-92, 70 companies had export more than 5% of the total sales. * 16 companies had exports more than 25% of the total sales in 1987-88 while 34 companies had exports of more than 25% of the total sales by 1991-92. Productivity Related o In 1987-88 40 companies had capacity utilization lower than 50% while in 1991-92 only 16 companies had capacity utilization lower than 50%. o The entire group has shown substantial improvement in rejection and rework rates. * Most of the PF assisted companies have improved their profitability through increased exports, lower inventories, and better planning.  Pace 2 of 2 Other oualitative Imorovement e By 1991-92, most of the companies are seeing exports as a profitable business opportunity. * The companies have been becoming more dynamic with exports emerging as a main line activity for the companies. o The export awareness has increased all throughout the organizations of the assisted companies. * Most of the companies feel that their international orientation and quality standards have improved drastically. PATTERN OF ASSISTANCE e The average amount per assistance was around US$76,775. * Maximum assistance of 61% to Engsncering Industries (includes Auto- ancillaries, Electricals, Electronics, Heavy a'd Light Engg). * Average time lag from sanction to disbursement has been around 18 months - mainly because of the developmental nature of most programmes. o Maximum assistance of 20* was sanctioned for consultancy followed by 12.8% for export promotion visits.