Sri Lanka - Third Small and Medium Industries Project

Document of The World Bank FOR OmCAL USE ONLY Report No. 13199 PROJECT COMPLETION REPORT SRI LANKA THIRD SMALL AND MEDIUM INDUSTRIES PROJECT (SMI-III) (CREDIT 1860-CE) JUNE 27, 1994 Country Operations and Industry & Finance Division Country Department III South Asia Region 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 EQUIVALENT Currency Unit - Sri Lanka Rupee (annual average) Rs. Per US$1.00 US$ Per Rs.1.00 1986 28.017 0.0355 1987 29.445 0.0339 1988 31.807 0.0314 1989 36.047 0.0277 1990 40.063 0.0249 1991 41.372 0.0241 1992 43.830 0.0228 1993 48.322 0.0206 ABBREVIATIONS ADB - Asian Development Bank AWPR - Average Weighted Prime Rate BOC - Bank of Ceylon BMR - Balancing, Modernizing and Rehabilitation CBOC - Commercial Bank of Ceylon CBSL - Central Bank of Sri Lanka DFCC - Development Finance Corporation of Ceylon EDB - Export Development Board GOSL - Government of Sri Lanka HNB - Hatton National Bank IDB - Industrial Development Board IPS - Industrial Policy Statement NDB - National Development Bank of Sri Lanka PB - Peoples Bank PCI - Participating Credit Institutions PFDP - Private Finance Development Project PMEs - Public Manufacturing Enterprises SB - Sri Lanka Standardization Institute SMI - Small and Medium Industries TIPR - Trade and Industrial Policy Reform FISCAL YEARS GOSL = January 1 to December 31 Commercial Banks = January 1 to December 31 Development Finance Corporation of Ceylon (DFCC)= April 1 to March 31 National Development Bank of Sri Lanka (NDB) = January 1 to December 31  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation June 27, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Sri Lanka - Third SmaDl and Medium Industries Project (Credit 1860-CE) Attached is the Project Completion Report on Sri Lanka - Third Small and Medium Industries Project (Credit 1860-CE) prepared by the South Asia Regional Office. Part II was prepared by the Borrower. The project was the third in the series of credits that were extended to Sri Lanka, starting in 1979, for the development of small and medium scale industries. An additional US$15 million was cofinanced by ADB. The Credit (including the ADB funds) was committed to 2448 subprojects and created 28,630 jobs with incremental investment per job of around US$2000. The project had little impact on creation of employment in the rural areas, as most subprojects were implemented in the urban centers. There was also little demand for financing in the Northern and Eastern Provinces because of the ethnic disturbances. The technical assistance component of the project had a mixed result. It proved useful in training and upgrading the institutional capabilities of the participating credit institution, but the State Commercial Banks failed to fully implement the recommendation of the studies aimed at their restructuring. Thus the institutional development objectives of the project, especially the focus on improving the financial sector efficiency, were only partially realized. Given that most subprojects are operating and profitable, the project outcome is rated as satisfactory. However, it is not clear that a succession of three credits (followed by a fourth for US$45 million in 1991) has done much in enabling the commercial banks or other financial institutions to continue lending to the small and medium industries without the support of multilateral agencies. Consequently, the sustainability of this project is rated as uncertain. The PCR is of high quality. It provides a clear and candid account of the project experience. No audit is planned. Robert Picciotto by H. Eberhard Kopp Attachment Thi document has a rstricted distribution and may be used by recipients only in the perfornance of their official duties. ts contents may not otherwise be disclosed without World Bank authorization.   PROJIECT COMPLETION REPORT FOR OFFICIAL USE ONLY SRI LANKA THIRD SMALL AND MEDIUM INDUSTRIES PROJECT (Credit 1860-CE) TABLE OF CONTENTS Page No. PREFACE ..............................................i EVALUATION SUMMARY ...................................ii PART I: PROJECT REVIEW FROM THE IDA'S PERSPECTIVE .......... 1. Project Identity ................................... 1 2. Background ..................................... 1 The Industrial Sector ............................... 1 3. Project Objectives and Description ....................... 2 4. Project Design and Organization ......................... 2 5. Project Implementation and Results ....................... 3 Implementation and Results of Credit Component .... ......... 3 PCIs/NDB Performance in Project Implementation .... ........ 4 Credit Guarantee Scheme ............................ 5 Implementation and Results of Technical Assistance .... ........ 6 6. Project Sustainability ................................ 8 Sub-Projects's Performance ........................... 8 Participating Credit Institutions ......................... 8 7. IDA's Performance ................................. 9 8. Borrower's Performance ............................. 9 9. Project Relationships ................................ 9 10. Project Documentation and Data ........................ 9 11. Consulting Services ................................ 10 12. Conclusions and Lessons Learned ....................... 10 PART II: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE .... 12 PART III: STATISTICAL INFORMATION ......................... 14 1. Related Bank Loans ....... .......... ............... 14 2. Project Timetable ....... ......... .. ................ 15 3. Credit Disbursement ....... .......... ............... 15 4. Project Implementation .............. .. .............. 15 5. Project Costs and Financing ............ .. ............. 16 A. Project Costs .......... 16 B. Project Financing ............................... 16 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. |  6. Project Results .......... .. .. . . . ............... . .. . 16 7. Status of Covenants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 8. Use of Bank Resources ........ . . . .. . . . . . . . . . . . . . . . . . 17 A. Staff Inputs .......... . .. . . .. . .. . .. . . .. . .. . .. . 17 B. Missions .......... .. .. . .. .. . .. .. . .. .. . .. .. . . 17 ANNEXES I. Sub Loan Financing .................................... 18 II. Net Refinancing Approval of the Credit Components by PCIs ... ....... 19 III. Size Distribution of Sub Loans by PCIs ....................... 20 IV. Sectoral Distribution of Sub Loans .......................... 21 V. Geographical Distribution of Sub Loans ....................... 22 VI. Collection Performance of the PCIs .......................... 23 VII. Technical Assistance - Allocation and Utilization .................. 24 VIII. People's Bank and Bank of Ceylon Operational Review .25  - i - PROJECT COMPLETION REPORT SRI LANKA THIRD SMALL AND MEDIUM INDUSTRIES (CREDIT 1860-CE) PREFACE 1. This is the Project Completion Report (PCR) for the Third Small and Medium Industries Project (Credit 1860-CE) to the Government of Sri Lanka in the amount of US$ 20.0 million equivalent (inclusive of US$ 5.0 million technical assistance). The project was approved on December 15, 1987, signed on February 10, 1988, became effective on April 21, 1988, and closed as scheduled on June 30, 1993. The project supported the Government's objective of providing credit through the banking system to small and medium scale private manufacturing enterprises and contributed to policy reform and institutional strengthening in the areas of tariff administration, export promotion and financial sector operations. 2. The credit component was fully committed as of December 31, 1990 and fully disbursed as of May 1993. The TA component, however, was 92 percent committed of which only 88 percent was utilized at the time of the project closing; the unutilized fund was cancelled. The Asian Development Bank (ADB) provided, through co-financing, an additional US$ 14.5 million for sub-project financing and US$ 0.5 million for technical assistance. 3. The PCR (Preface, Evaluation Summary, Parts I and III) was prepared by the Country Operations and Industry and Finance Division, Country Department III, South Asia Region (SA3CI). Part II was prepared by the National Development Bank (NDB), the Apex Institution for the Small and Medium Industries Projects, on behalf of the Government. The NDB also provided part of the statistical information required for Part III. 4. This PCR is based, among other sources, on the Staff Appraisal Report (SAR); The Development Credit Agreement (DCA); supervision reports; internal Bank memoranda; NDB quarterly progress reports; correspondence between IDA and the borrower.   - ii - PROJECT COMPLETION REPORT SRI LANKA THIRD SMALL AND MEDIUM INDUSTRIES PROJECT (SMI-III) (CREDIT 1860-CE) EVALUATION SUMMARY 1. Obiectives and Description. The main project objective was to support and complement IDA's previous industrial sector operations by providing credit through the banking system to private small and medium manufacturing industries. The project also aimed at making a further contribution to policy reform and institutional strengthening in the areas of industrial and export promotion and financial sector operations. The project had two components: (i) a Credit component of US$ 15.0 million equivalent for term loans to private Small and Medium Industries (SMIs); and (ii) a US$ 5.0 million equivalent Technical Assistance (TA) component for improving financial sector efficiency, export development, and improvement of public manufacturing enterprises. An additional US$ 14.5 million for sub-project financing, and US$ 0.5 million for TA was co-financed by ADB. 2. Proiect Implementation and Results. SMI-III was signed on February 10, 1988, became effective on April 21, 1988 and closed as scheduled on June 30, 1993. The Credit component was fully committed as of December 31, 1990 and fully disbursed as of May 1993. The TA component was 92 percent committed of which 88 percent was utilized at the time of closing; the unutilized fund (about US$ 1.0 million) was cancelled. The Credit (including ADB fund) was committed to 2,448 small and medium industries and created about 28,630 jobs (appraisal estimated 31,400 new jobs) with incremental fixed investment per job of about US$ 2,032 (appraisal estimated US$ 1,750). The project contributed 50 percent of the cost of SMI fixed assets during the period of implementation. The People's Bank (PB) accounted for the largest number and amount of sub-loans approved followed by the Bank of Ceylon (BOC) and Development Finance Corporation of Ceylon (DFCC). The two state owned banks (PB and BOC), together, accounted for about 53 percent of the loan financing. 3. The implementation of Credit component was, in general, satisfactory. The Credit funds mainly financed enterprises in selected industries in a few urban areas. About 50 percent of sub-projects were concentrated in Colombo and Gampaha Districts in the Western province. Demand for sub-project financing in the Northern and Eastern Provinces was low mainly due to ethnic disturbances and lack of sufficient infrastructures in these areas. Southern Districts received only about 15 percent of the project financing, despite SMI-III's emphasis on employment generation in the South. The sectoral composition and geographical distribution of sub-projects financed under SMI-III were similar to those under SMI-I and II. The factors that contributed to this pattern were: (i) continued ethnic disturbances; (ii) differences in the availability of entrepreneurial development programs and supporting infrastructure between regions; and (iii) lack of adequate and effective marketing and technical assistance by public advisory agencies. 4. SMI-III had a limited impact on rural industrialization and employment, in particular, in the Northern, Eastern, and Southern provinces. Over 60 percent of the Credit, by number and amount, was allocated for expanding existing sub-projects as opposed to investment in new projects. Although, SMI-III emphasized the development of export-oriented ventures, relatively small number of sub-projects with direct export impact were financed. Under the Credit Guarantee Scheme (CGS), as of June 30, 1993, the  - iii - premium collected by the Central Bank of Sri Lanka (CBSL) amounted to Rs 31 million while the guarantee value of the loans in arrears was Rs 134 million. While it is too early at this point to determine the full cost of CGS, it is unlikely that CBSL would be able to cover the cost from this scheme. 5. The National Development Bank (NDB), the apex institution, continued satisfactory administration of project operations and used the TA funds to strengthen its capabilities in this area. NDB's monitoring system, however, lacks an in-built mechanism to collect ex-post data on the financed sub-projects with regard to their financial and economic benefits. Lack of adequate follow-up data on sub-project performance has made it difficult for NDB to carry out an analysis of the project benefits. At the request of IDA, a survey was conducted by NDB to assess the actual performance of a sample of sub-projects. The results of this survey of a sample of 100 sub-projects, financed under SMI-III, indicated that about one-third of the industries performed below the expected level, were financially unstable, and had poor sales performance. It should be noted that the performance of the sub- projects financed under SMI-III has not changed from the previous SMIs with respect to their profitability and output level. 6. The TA component focused mainly on improving financial sector efficiency and the fund was mostly allocated for: (i) restructuring the two state owned commercial banks (SCBs); (ii) institutional upgrading of the Participating Credit Institutions (PCIs) and NDB; and (iii) improving the supervision capabilities of the Central Bank. In line with the objective of promoting export-oriented SMIs, the project also supported the Export Development Board (EDB) in its efforts to introduce, develop, and promote priority export industries, as well as implement some of the elements of the Government 1987 trade and industrial policy, in particular, in the area of export promotion schemes. The TA funds proved to be useful for the PCIs in upgrading their institutional capabilities. The PCIs/NDB training programs were successful in helping these institutions to develop the capabilities of their staff in sub-project appraisal and supervision. However, some of the funds were used on an ad-hoc basis for training, consultancies, seminars, and studies, without determining strategy, and the results were ineffective. A major component of the TA fund was allocated for operational review of the SCBs with the objective of restructuring the banks in order to enable them to function as autonomous commercial organizations. While the overall use of TA funds by BOC has been relatively satisfactory, the impact of TA for PB has been low, especially in areas such as the management of human resource and information system. The SCBs implemented some of the recommendations of the studies, improved the functions of some units, and automated a limited number of branches. They, however, failed to implement a major restructuring which was the principal objective of the technical assistance. The result of the project was not as good as it was expected because of the slippage in the implementation of financial sector reform program. 7. Sustainabilit . Although the project contributed to the expansion of the SMI sector, the ability of the sub-projects to continue to generate employment, output, and profit is uncertain. The absence of long-term private investment funding in the financial sector and the high risk and high transaction costs, generally associated with SMI lending, suggest that the volume of SMI financing is likely to decline without the availability of multilateral funds which are the major source of term financing in Sri Lanka. In view of the fact that sustainability of the SMI sector is partly dependent on a sound capital structure of industries (which can be achieved through higher share of equity participation by the borrower), SMI-III and IV attempted to encourage the PCIs to become more independent in SMI financing by decreasing the IDA's financing share to 75 (from 80 percent under SMI-II) and 70 percent, respectively. Although, IDA approved four SMI lines of credit in the period 1979-1991, it is not clear that the financial institutions, particularly the commercial banks that participated in these credits, are able  - iv - to continue to finance SMI without further support from the multilateral agencies. IDA has been supporting the GOSL's efforts in improving the financial sector operations for some time. The recently approved ID project, Private Finance Development Project (PFDP), attempted to help the GOSL in changing and strengthening the role of financial institutions including the contractual savings institutions and strengthening the local bond market in order to mobilize domestic resources for long-term investment. GOSL's success in implementing the reform programs and addressing the weaknesses in the domestic resource mobilization will determine whether the commercial banks would continue to provide term lending to the SMI sector in the absence of support from multilateral agencies. The sustainability of SMI lending, in the long-term, depends on the success of the Government in its financial sector reform program. Unless the commercial banks can channel the domestically funded term savings to the private manufacturing sector, their term lending for SMI sector is likely to decline. 8. Effective implementation of the TA component is key to short-term success and long-term sustainability of the TA beneficiaries (PCIs, Government agencies, etc). The Government advisory agencies 1 that link the SMIs with the financial institutions have been ineffective despite continuous support from World Bank and ADB. Rationalizing the function of these institutions is a must for a sustainable SMI sector. 9. Conclusions and Lessons Learned. The SMI-III Credit component was implemented satisfactorily; the TA component, however, was implemented with mixed results. The project met its objectives in providing term finance for the development of a large number of small and medium scale industries which generated a significant number of jobs. The project played an important role in encouraging more private commercial banks, which accounted for about 47 percent of the market share, to participate in the credit and further increase their SMI lending activities. The training programs proved to be successful in capacity building of the staff of the PCIs/NDB and other institutions. Both NDB and PCIs benefited from the TA funds for expanding and upgrading their SMI units. The project supported the GOSL's trade and industrial policy reform programs, in particular, in removing constraints in export credits and duty drawbacks for indirect exporters in the SMI sector. The project also helped deepen IDA's dialogue with GOSL on issues affecting the efficiency of financial sector through an analysis of the operational and financial viability of the two state owned commercial banks and addressing issues in prudential regulations and banking supervision. 10. The project was unable to achieve a number of the objectives both in sub-project financing (due to prevailing circumstances in the country) and in TA programs. Like the previous SMI projects, SMI-III did not adequately contribute to promoting rural industrialization and employment. This was mainly due to limiting sub-project financing to selected industries in a few urban areas. Due to the absence of an in-built mechanism for collecting accurate follow-up data on sub-project performance, there is not much information available on the financial situation and economic impact of the sub-projects. Lack of a proper recording system that would automatically monitor the TA programs affected the satisfactory implementation of the TA. 11. Lessons learned include: (i) for more systematic and effective project monitoring (TA and sub-projects), the role of the Apex Institution needs to be redefined and its capabilities further improved to be able to carry out an in-depth evaluation of economic impact and financial performance 1/ Such as Ceylon Institute for Scientific and Industrial Research (CISIR), Industrial Development Board (IDB), National Engineering for Research and Development Center (NERDC), etc.  v of SMI sub-projects; (ii) broader geographical distribution of SMI sub- projects can only be achieved when adequate industrial infrastructure is in place, entrepreneur development programs benefit rural areas, and PCIs effectively utilize their branches for credit delivery; (iii) the SMI sector is still heavily dependent on external financing due to absence of long term investment funding in the financial sector, which implies that the volume of SMI financing is likely to decline without availability of multilateral funds; therefore, to ensure the sustainability of the SMI sector lending the financial sector needs to adopt lending to SMIs as part of their long-term strategy which, in turn, is dependent on the GOSL's success in mobilizing domestic resources for long-term investment; (iv) the design of the credit guarantee scheme needs to be improved in order to reduce the risk to the Central Bank and the need for such scheme should be reviewed further for future projects; (v) more attention must be given to the design and implementation of the TA component and to developing an effective monitoring system to ensure more efficient and successful project implementation; (vi) both IDA and GOSL should ensure that the PCIs obtain the project documents and that availability of the TA are widely publicized and disseminated to the concerned institutions and SMI sector; (vii) steps should be taken to improve the effectiveness of the experts through better selection process, evaluation of their performance, and establishing counterparts in the concerned institution; and (viii) for a more successful operation of an SMI, the role and effectiveness of the public advisory agencies need to be strengthened; therefore, to ensure quality services and effective delivery system by these agencies, efforts should be made at moving these institutions towards demand- driven support units and providing their services on a cost-sharing basis.  PROJECT COMPLETION REPORT SRI LANKA THIRD SMALL AND MEDIUM INDUSTRIES PROJECT (SMI-III) (CREDIT 1860-CE) PART I: PROJECT REVIEW FROM IDA'S PERSPECTIVE 1. Project Identitv Name Third Small and Medium Industries Project Credit Number : Credit 1860-CE RVP Unit South Asia Region Country Sri Lanka Sectors Finance and Industry 2. Background 1.01 Sri Lanka has had a long tradition of emphasizing social development and is characterized by levels of literacy, infant mortality and life expectancy that are comparable to those of more developed countries. But its success in generating growth in income and employment has not been commensurate with its social progress. Economic growth during the early 1970s was limited by the inefficiencies in the public sector and the insulation of the domestic market from foreign competition. In late 1977, the Government initiated a liberalization policy with the objective of increasing the growth rates of production and employment, saving and investment and exports. To achieve these objectives, the Government introduced policies which included liberalization of trade and industrial policies, lifting of foreign exchange restrictions, and relaxation of regulations on private domestic investment. The Government further adopted measures to promote exports through rationalizing the export incentive structure, encouraging non-traditional export products, streamlining the administrative arrangements for exporting, and creating of a Free Trade Zone to attract export-oriented foreign private investment. 1.02 The promotion of the industrial sector in Sri Lanka has long been part of the Bank's country assistance strategy. To assist the Government in its efforts to implement the policy reforms and industrial objectives (industrial growth, employment generation, and higher export earnings from industry), IDA initiated a series of industrial sector projects starting 1979. The objectives of these projects were: (i) to provide traditional lines of credit through the banking system to small, medium, and large scale enterprises; and (ii) to support the GOSL's policy reform in the areas of trade policy, industrial incentives, export promotion, public enterprise efficiency, and financial sector policies. 1.03 The Industrial Sector. Sri Lanka has a limited industrial base with the manufacturing sector accounts for about 15 percent of GDP and 45 percent of exports. At the time of the 1983 industrial census, there were about 100,000 industrial establishments in Sri Lanka. Over 86% of these employed fewer than five workers. Units with fewer than five employees accounted for 30% of employment and contributed about 8% of value added and output with a value added per person of only Rs 7,900 compared to Rs 40,000 for the larger private sector units. Since 1979, the GOSL and IDA have worked closely together through a number of industrial policy operations to develop and implement a phased program of trade and industrial policy reform. In 1986, with Technical Assistance (TA) funding from the first Industrial Development Project (IDP-I), GOSL prepared an Industrial Policy Statement (IPS) which gave high priority to private industrial development, with  -2- particular emphasis upon the role of SMIs. Recognizing the importance of the small scale industries' contribution to value added and employment and their potential for generating foreign exchange, the Government has taken a number of measures to remove impediments to their successful functioning. The Government policy has had a significant impact on industrial performance and contributed to strong manufacturing growth of 7 percent in 1988-1992. 3. Project Objectives and Description 1.04 The objectives of the project were to support and complement previous industrial operations, particularly the two SMI projects, by providing credit through the banking system to private manufacturing industries. It also aimed at making further contributions to policy reform and supporting institutional strengthening for industrial and export promotion and the financial sector. More specifically the project was to: o provide needed term financing for viable SMI projects, with an emphasis on export-oriented ventures and balancing, modernization, and rehabilitation of existing enterprises; reconstruction needs in the Northern and Eastern provinces; and employment generation in the high unemployment districts in the South; o continue the overall institutional development of the Participating Credit Institutions (PCIs); [ contribute to policy reform and institutional strengthening in the areas of tariff administration, export promotion and financial sector operations; and 3 continue the institutional development of SMI technical and marketing service agencies through the provision of technical and marketing assistance facilities to SMIs in order to upgrade their technology and product quality. 1.05 The project had two major components: (a) a Credit Component of US$ 15.0 million equivalent for term loans to private sector SMIs; and (b) a US$ 5.0 million equivalent technical assistance component for: (i) improving financial sector efficiency through continued training for the PCIs, providing technical assistance to the two state owned commercial banks to implement action programs to improve their overall operating/financial efficiency, and upgrading the evaluation system of the Central Bank's Supervision Department; (ii) export development through extension of export credit and duty drawback facilities to indirect exporters; and (iii) improvement of the public manufacturing enterprises (PMEs) through development of action programs developed under the IDA funded industrial development projects. An additional US$ 14.5 million for sub-project financing, and US$ 0.5 million for technical assistance was provided by ADB through co-financing. 4. Project Design and Oraanization 1.06 In view of successful IDP and SMI projects lending since 1979 which also supported the Government's broad policy reform program, in early 1987, the GOSL's requested further IDA funds to support small and medium scale industries. In response to the Government's request, IDA helped GOSL design the SMI-III project in order to provide needed credit for viable SMI projects through the financial system and to continue its support for the Government's trade and industrial sector reform. The GOSL's effective incentives to facilitate growth in industrial and export credit for SMI, NDB's provision of refinance for SMI lending, and SMI project promotion training resulted in substantial increases in the PCI term lending to SMIs. When the SMI-II credit  closed in December 1987 with a substantial number of pending loan applications for SMI being held with the PCIs, preparation of another SMI facility was considered appropriate. SMI-III was similar in design to the earlier two SMI projects. The project, however, initiated action to address some of the major issues identified in the 1986 joint IMF/IBRD Financial Sector Report with emphasis on measures to: (i) increase the role of market forces in determination of interest rates, (ii) improve credit discipline and reduce the level of the financial system's non-performing assets, and (iii) provide technical assistance to the state owned commercial banks in order to improve their overall operating and financial efficiency. 1.07 The size of the SMI-III lending component was geared to the SMI project lending capacities of the PCIs and met about 23 percent of the total demand for SMI funding. Together with the ADB co-financing facility, about 50 percent of the needed SMI funding was mobilized by these two multilateral agencies. The GOSL onlent the US$ 15.0 million project proceeds to NDB for 18 years, including a five year grace period, at an initial rate equivalent to 7 percentage points below the Average Weighted Prime Rate (AWPR) for the commercial banks for short-term lending operations. In turn, NDB financed eligible PCIs ' operations at AWPR, less 6 percentage points, giving NDB a spread of 1 percent to meet the cost of SMI operations. The PCIs relent the proceeds at 13%-16%, fixed or variable, which was estimated to be sufficient to cover their costs of operations and the project risks. Given the spread between the GOSL and IDA on-lending rates, GOSL bore the foreign exchange risk. 1.08 The PCIs were responsible for appraisal, supervision and collection of sub-loans of up to Rs 2.0 million and NDB was delegated authority to approve sub-project applications up to Rs 4.0 million (maximum sub-loan size). To encourage the PCIs to become more independent in SMI financing, the IDA's financing share of sub-loans was decreased from 80 percent under SMI-II to 75 percent. The PCIs were to encourage a minimum of 25 percent of sub-project costs be provided by the sponsor in the form of equity, although the actual equity requirement was determined on a case by case basis. 5. Prolect Implementation and Results 1.09 SMI-III for SDR 15.7 million (US$ 20.0 million) was signed on February 10, 1988, became effective on April 21, 1988 and closed as scheduled on June 30, 1993. The credit component was fully committed as of December 31, 1990 and fully disbursed as of May 1993 against the appraisal estimate of December 31, 1992. The TA component was 92 percent committed of which 88 percent was utilized at the time of the project closing; the unutilized fund (about US$ 1.0 million) was cancelled. In general, the SMI-III project was implemented satisfactorily; there were, however, issues with respect to the utilization and implementation of both the Credit and TA components which are addressed below. 1.10 Implementation and Results of Credit Component. Implementation of the Credit component was, in general, satisfactory. The credit component of a total of US$ 29.5 million (inclusive of ADB US$ 14.5 million credit) was fully committed to 2,448 small and medium industries. The employment impact was 1/ The eligible PCIs were two state owned commercial banks {People's Bank (PB) and Bank of Ceylon (BOC)}, a development finance institution {Development Finance Corporation of Ceylon (DFCC)}, four private banks {(Hatton National Bank (HNB), Commercial Bank of Ceylon (CBOC), Sampath Bank (SB), and Seylon Trust Bank (STB)}, and the Regional Rural Development Bank (RRDB) of Kurunegala.  -4- significant; SMI projects created about 28,630 jobs (appraisal estimated 31,400 new jobs) with incremental fixed investments per job of about US$ 2,032 (appraisal estimated US$ 1,750). The project contributed about 50 percent to the cost of SMI fixed assets during the period of implementation (Annex I). The PB accounted for the largest number and amount of loans approved followed by BOC, and DFCC (Annex II). According to the appraisal estimate, 60 percent of credit component was to be allocated for loans below Rs 1 million; the allocation was, actually, about 43 percent. Allocation to sub-loan size of Rs 1-2 million was about 26 percent as against the appraisal estimate of 20 percent. The remaining 31 percent was allocated to the sub-projects of Rs 2-4 million (Annex III). This indicates that the average sub-loan size (Rs 464,000) was slightly larger than the appraisal estimate of Rs 450,000. A larger proportion of Credit, by number and amount, was allocated for expansion projects as opposed to investments in new projects (Annex II). 1.11 The project mainly financed enterprises in food processing (23.8%) and garments (13.5%) (accounted for about 63 percent of the manufactured products during 1988-92), followed by construction material (7.8%), and metal products (7.0%) (Annex IV). About 65 percent of sub-projects, both in number and amount, were concentrated in a few districts in Central and Western provinces: Colombo (33.9%), Gampaha (15.5%), Galle (9.5%), and 4-7% in smaller towns of Kurunegala, Kandy, Kalutara (Annex V). Demand for sub-project financing in Northern and Eastern provinces was low mainly due to ethnic disturbances and lack of sufficient infrastructures in these areas. The sectoral composition and geographical distribution of the sub-projects financed under SMI-III project were similar to those under SMI-I & II. The factors that contributed to this pattern were: (i) continued ethnic disturbances; (ii) differences in the availability of entrepreneurial development programs and supporting infrastructure between regions; and (iii) lack of adequate and effective marketing and technical assistance by public advisory agencies (CISIR, IDB, NERDC, etc.) to promote investment in a wider geographical areas. SMI-III had limited impact on promoting rural industrialization and employment, in particular, in the Northern, Eastern and Southern provinces. 1.12 Some of the project risks were correctly identified in the Staff Appraisal Report. The project did not, however, take into consideration the risk due to ethnic disturbances and its possible escalation. Therefore, one of the project objectives of reconstruction of Northern and Eastern provinces could not materialize. In addition, SMI-III specifically aimed at employment generation in the high unemployment districts in the South. The Southern Districts received about 15 percent of the project financing for over 330 projects (13.5 percent of projects by number). Although, SMI-III emphasized the development of export-oriented ventures, relatively small number of sub- projects with direct export potential were financed; 149 sub-projects (excluding industries that supply to direct exporters) with an investment amount of Rs 172 million (15% of committed fund) contributed to exports (Annex II). 1.13 PCIs/NDB Performance in Project Implementation. SMI-III project was built on organizational capabilities developed under SMI-I & II with the expectation that: (i) the PCIs would further improve their credit, technical, and marketing services, (ii) NDB would monitor the project implementation more effectively, and (iii) IDA would provide more frequent supervision visits. While the PCIs increased the number of trained staff in order to maintain the standard of appraisal and supervision of sub-projects, they did not make any effort to provide technical and marketing services to SMIs. The PCIs' collection performance was improved under SMI-III as opposed to SMI-I & II mainly due to better sub-project appraisal and supervision. Because of the escalation of ethnic disturbances during the last quarter of 1988 and mid- 1989, the annual collection rates for DFCC, BOC and PB fell below the minimum collection rate of 70 percent. This was a transitory problem and by early  -5- 1990 all the PCIs met the minimum collection ratio required by IDA (Annex VI). NDB continued its satisfactory administration of project operations and by using the TA funds tried to strengthen its capabilities in this area. However, these efforts were blunted because NDB's monitoring system lacks an in-built mechanism to collect ex-post information on the financed sub-projects with regard to their financial and economic benefits. Although this shortcoming prevailed for the previous SMI projects as well and the importance of sub-projects' financial performance and economic impact was recognized by SMI-III, the project did not address the need for redefining the functions of NDB to conduct surveys of SMIs after the loans are made. The 1991 supervision mission, however, impressed upon NDB the importance of establishing an effective mechanism for collecting information on financial performance and economic impact of the SMI sub-projects. As mentioned in Part II, efforts are underway by NDB to improve the data on the economic and financial performance of the sub-project through conducting surveys of SMIs on a regular basis. 1.14 At the request of IDA, a survey was conducted by the NDB to assess the actual performance of a sample of sub-projects. A sample of 220 sub- projects was selected covering completed SMI-I, II, & III projects, and a few from SMI-IV. The performance of a sample of 100 sub-projects under SMI-III was separately assessed, the result of which can be summarized as follows: 65 projects were completed within or below the estimated cost; and 35 were completed with cost overruns mainly due to time taken by the banks to appraise the sub-projects and comply with legal formalities. In terms of project performance, out of 100 projects, 67 are operating above or at the expected level; and 33 are performing below the expected level or temporarily stopped. With regard to the financial performance, 51 sub-projects provided data of their sales for two years of operations and calculated their actual profits. Of these sub-projects, 31 percent achieved profits as planned, 33 percent achieved profits between 50% and 99% of the estimated profit and the remainder had profit below 50% of the original estimate. In addition, the survey concluded that some of the industries encountered difficulties with the high Business Turnover Tax and Defense Levy, high interest rates charged on SMI loans, finding adequate skilled labor and quality raw material, and complying with state formalities and regulation. The results of the survey further show: (i) that about one-third of the industries performed below the expected level, were financially unstable, and had poor sales performance, and (ii) that sub-project performance has not changed from the previous SMIs with respect to their profitability and output level. 1.15 To Coordinate the implementation of the IDA financed SMI activities, a Committee was established under SMI-II. This SMI Coordinating Committee consists of Secretaries of Ministry of Industry, Science and Technology, Ministry of Textiles, and Ministry of Tourism and Rural Industries; Director General of Department of National Planning; Chairman of IDB and EDB; and General Managers of PB and BOC. The role of the Committee was changed from that of sub-project coordination to policy review under the SMI-III project. It was responsible for reviewing the quarterly reports prepared by NDB and other implementing agencies, examining operational and coordination problems, and making recommendations to GOSL for improving efficiency in policy implementation. It was also responsible for evaluating and monitoring the TA programs. The committee has not been effective in complying with its mandate and had an insignificant role in project implementation. The committee does not have a proper recording system and relies basically on the NDB for information on the utilization of TA funds. 1.16 Credit Guarantee Scheme (CGS). Under SMI-III, the CBSL issued 3,129 guarantees (including those for SMI loans granted by NDB), between 1989 and 1993, with an aggregate value of Rs 1,312 million. As of June 30, 1993, of the total 2,380 loans outstanding with a balance of Rs 907 million and guarantee value of Rs 603 million, 638 with a guarantee value of Rs 134 million fell into arrears. The premium collected by the CBSL amounted to Rs  -6- 31 million. It is too early at this point to determine the full cost of the CGS under the SMI-III. If no more loans fall into arrears and PCIs can not collect their loans already in arrears, they would claim them from Central Bank and, Rs 134 million would represent the maximum liability of the CBSL under the scheme. Sub-loans with collateral for realization would reduce the CBSL' liability to the guarantee fund. The other problem that CBSL faced was the long delay in receiving its shares of recovered loans which, in turn, was caused by delays in post claims recoveries by the PCIs. 1.17 Implementation and Results of Technical Assistance. The TA program of US$ 5.5 million (inclusive of US$ 0.5 million provided by ADB) focused on improving financial sector efficiency (US$ 2.25 million),2 export promotion and marketing (US$ 0.65 million), short-term technical and productivity assistance to SMI units (US$ 0.5 million), and assisting the industrial sector through further trade and industrial policy reform (US$ 2.1 million). About 92 percent of the TA component was committed of which only 88 percent was disbursed at the time of project closing (Annex VII). The actual allocation was different from that in the appraisal report. About US$ 2.4 million or 47 percent (as against US$ 1.5 million original allocation) of the TA fund was allocated to PB and BOC for their restructuring with the objective of making their operations more efficient. These two banks dominate the banking system in terms of size and act as market leaders in establishing interest rates. About US$ 0.9 million was spent on systems improvements and training of the NDB and PCIs staff. In line with the SMI-III objective of promoting export- oriented SMIs, the project supported the Export Development Board (EDB) in its effort to introduce, develop, and promote priority export industries including leather products, shrimp and prawn culturing, gemstones, and rubber. EDB and Ministry of Finance also utilized the TA funds to implement some of the elements of the Government's 1987 trade and industrial policy; with the objective of extending export credit and duty drawback facilities to indirect exporters in addition to direct exporters, tariff reform and further improving the efficiency of the PMEs. The allocation for the Technical Assistance Facility (TAF), the training programs, and policy reform support had not been fully utilized, reflecting a reallocation of TA funds in favor of the financial sector. 1.18 The TA programs funded under the SMI-III were implemented with mixed results. TA funds were not managed efficiently and the TA activities, for the most part, were not monitored mainly due to the lack of a routine follow-up and proper recording system of the beneficiaries. Improvement of financial sector efficiency was one of the main objectives of the TA component under SMI-III. The TA to the PCIs/NDB for their training programs helped these institutions to develop the capabilities of their staff in sub-project appraisal and supervision. The success of the training program is attributed to factors such as: NDB's effective management of TA fund for training; design and content of the courses; clear objective and frequency of the courses; availability of funds to benefit a large number of staff; commitment of PCIs in improving the human resources; and staff motivation and career development objective. The TA fund proved to be useful for the PCIs in upgrading their SMI units and developing the institutions' technology. However, some of the funds which were used on an ad-hoc basis for foreign training, consultancies, seminars, and studies without setting strategy were ineffective. 1.19 A major component of the TA fund was allocated to the two state owned commercial Banks for their restructuring. The operational review of the banks (Annex VIII) that started in 1988 under SMI-III and IDP-II was undertaken by two international consultancy firms. These studies were continued and further funded under other IDA industrial sector operations (SMI-IV and IDP-II & III) and completed in 1992. While the use of TA funds by 2/ Figures in parenthesis represent allocation in the Staff Appraisal Report.  -7- BOC has been relatively satisfactory, the impact of TA for PB has been low. BOC has taken appropriate actions to implement the recommendations of the studies to improve the credit policy and procedures. It, however, failed to implement the recommendations in the area of technology (disregarding the consultants' recommendations on suitable computer software) and human resources (fearing adverse reactions from staff unions). BOC has reorganized the head office and some of the regional offices and is in the process of reorganizing the branches (out of 287 branches, 31 manual and 36 computerized branches were reorganized). BOC's Management Information System (MIS) has also been improved. As for the PB consultancy, the result was not satisfactory with respect to both the consultants performance and the project implementation by the bank. No progress has been made on the project implementation in the areas of the organization structure, human resource management, and MIS. Recommendations in the area of credit policy have been partially implemented and branch reorganization is proceeding slowly (out of 316 branches, 52 were reorganized). While SCBs implemented some of the recommendations of the studies, improved the functions of some units, and automated a limited number of their branches, they failed to implement a major restructuring which was the principal objective of the consultancies. The result was not as good as it was expected because of the slippage in the implementation of the financial sector reform program. The SCBs, which also participated in the SMI-IV credit as PCIs, incurred large operating losses and, as of December 31, 1991, were technically bankrupt. Therefore, both banks were suspended from utilizing the credit component of SMI-IV (IDA Credit 2250-CE) as they failed to maintain the capital adequacy requirements for eligibility as a PCI for that credit. These banks were recapitalized in April 1993. 1.20 To further support GOSL's reform program for improving the financial sector, in particular in the area of banking supervision and regulation and implementation of Banking Act (1988), TA was provided to the CBSL Banking Supervision Department. It was agreed with the GOSL/CBSL and the IMF that the IMF supervise the implementation of this component on behalf of IDA. The TA fund was utilized in 198-89 for five studies in the areas of prudential regulations and banking supervision. No significant improvement has yet taken place in increasing the effectiveness of central bank supervision, mainly due to lack of adequate manpower and technical resources required to implement some of the recommendations. However, the Banking Act established a number of regulations in the areas of loan classification, provisioning requirements, and capital adequacy standards which are now being followed by the banks. 1.21 TA allocated to the Export Development Board was mainly utilized for undertaking studies on export promotion schemes. Some of the recommendations derived from the studies were implemented by the GOSL, in particular those related to the Duty Rebate Scheme, Manufacture-in-Bond Scheme, and liberalization of exchange control. The ADB funded TA projects aimed at export development and promotion of export marketing in selected areas of high export potential. A number of producers and exporters benefitted from the program through participation in international fairs and seminars. However, the in-depth assessment of the impact of these programs (studies and seminars) still needs to be carried out by the EDB. Under the TAF, partial grants were provided to potential SMI entrepreneurs to finance eligible and viable "technology sub-projects" such as technical consulting, technical training programs, and small R&D projects which either introduce new technology or improve existing technologies in SMIs. NDB was responsible for appraising the eligibility and viability of each sub-project, submitting approved proposals for IDA approval, and supervising the project. The TAF has not been fully utilized, but has achieved its objective. Factors that led to the underutilization of the fund were: (i) lack of knowledge of how and where to obtain technology and technical and marketing knowledge, and (ii) lack of effective and sufficient publicity about the availability of TAF.  -8 - 6. Prolect Sustainability 1.22 Sub-Projects' Performance. Based on the limited information available on actual sub-project performance, about two third of industries are performing as expected with respect to profitability and output level. The rest are either below the expected level or completely abandoned. Although the project contributed to the expansion of the SMI sector, the ability of the sub-projects to continue to generate employment, output, and profit is uncertain. Project success needs to be evaluated on the basis of a combination of data before the loans are made and data on the financial performance and economic impact of the sub-projects after the loans are made. In the absence of an adequate information system, evaluation is difficult and less conclusive. PCIs and NDB need to develop their information system in order to collect necessary information on sub-projects for a more accurate and in depth evaluation of the SMI sector. 1.23 Participating Credit Institutions. The PCIs are, in general, financially viable institutions. To improve financial sector efficiency, a large amount of TA funds under all IDA industrial sector operations were allocated for restructuring the SCBs. The GOSL, however, has been slow to embrace the reorganization, citing political constraints to privatization. Absence of long term investment funding in the financial sector and high risk and high transaction costs of SMI lending suggest that the volume of SMI financing is likely to decline without availability of multilateral funds which are the major source of long-term financing in Sri Lanka. IDA approved four SMI lines of credit in the period 1979-1991; it is not, however, clear that the financial institutions, particularly the commercial banks that participated in these credits, are able to continue to finance SMIs from their own resources without having access to the multilateral credit lines. IDA has been supporting the GOSL's efforts in improving the financial sector operations for some time. The recent IDA's project, PFDP, was designed with the objectives of helping the GOSL in changing and strengthening the role of financial institutions including the contractual savings institutions and strengthening the local bond market in order to increase mobilization of domestic term savings for term investment. GOSL's success in implementing the reform program, addressing weaknesses in domestic resource mobilization and institutional constraints to profitable SMI lending will determine whether the commercial banks are likely to continue to provide term lending to the SMI sector in the absence of funds from multilateral sources. In view of the fact that long-term sustainability of the SMI sector is partly dependent on a sound capital structure of industries (which can partly be achieved by encouraging higher share of equity participation by the borrower), SMI-III and IV attempted to encourage the PCIs to become more independent in SMI financing by decreasing the IDA's refinancing share to 75 percent and 70 percent, respectively. Although SMI projects have attempted to achieve a resource additionality effect by pushing for significant contributions to investment financing by the sub-borrowers and PCIs, the SMI sector is still heavily dependent on the external financing. 1.24 Short-term success and long-term sustainability of the TA beneficiaries (PCIs, NDB, Government agencies) depend on their efficient use of TA funds and more importantly on the effective implementation of the projects. The TA beneficiaries need to focus on strengthening the local ability and be actively involved in the execution of the TA. An important element in building the capacity to achieve this objective is to provide qualified counterparts who are involved from the initiation stage of the TA. The Government agencies (such as IDB, CISIR, etc.) that provide technical and marketing services to the SMIs have been ineffective despite continuing support from the World Bank and ADB. Restructuring or strengthening these institutions is essential for a sustainable SMI sector.  - 9 - 7. IDA's Performance 1.25 IDA responded quickly to the need for additional financing in SMI sector in 1987/88. SMI-III followed closely the design of SMI-II and was prepared to support and complement the objectives of other industrial sector operations. Given the ambitious objectives and broad scope of the TA programs with emphasis on policy and institutional development and weak management and organizational capabilities of the Government agencies, IDA should have monitored and supervised the TA programs more effectively. Involvement of staff of the Resident Missions in the projects can contribute significantly to the effective and systematic monitoring of the TA component. IDA sent seven supervision missions from project effectiveness (May 1988) to closing with the last mission in June 1993 just before the project closed. 8. Borrower Performance 1.26 The performance of NDB and PCIs in the utilization of the credit component as well as upgrading their SMI units was satisfactory. NDB's management of the project and collaboration with the missions was also satisfactory. Although the credit component was committed on schedule and disbursements were satisfactory, the TA fund, for the most part, was not managed and utilized efficiently by the TA beneficiaries mainly due to inefficient organizational capabilities and lack of regular follow-up and proper recording system. This problem was also addressed in the SMI-II PCR (February 11, 1992). IDP/SMI Monitoring Unit in the Ministry of Finance and the SMI Coordinating Committee did not contribute to the effective implementation of TA under SMI-III. 1.27 The PCIs managed to maintain the eligibility criteria under the project and effectively channelled the credit component to viable and eligible SMIs. The TA fund proved to be useful for proper monitoring of their SMI portfolio. The restructuring of the SCBs in the context of the financial sector reform did not take place. The impact of the ADB/IDA funded TA programs for export development and export marketing promotion were not assessed by the GOSL and was not reported to IDA. In general, the GOSL did not attempt to assess the benefits and impacts of the TA programs provided under this project. 9. Project Relationships 1.28 The GOSL's contribution to the speedy and smooth effectiveness of the project and excellent co-operation with supervision missions were instruments to satisfactory implementation of the project. NDB was very receptive and helpful to the supervision missions and provided timely response to the IDA's requests throughout the life of the project. Regular reporting of the status of project implementation by NDB was useful. IDA supervision missions should have spent more time with the TA beneficiaries with regards to their utilization of TA funds and perhaps should have identified the weaknesses in this area from the beginning of the project implementation. 10. Project Documentation and Data 1.29 In general, the Development Credit Agreement (DCA) and Staff Appraisal Report (SAR) were adequate and provided a framework for IDA and NDB during project implementation. However, the SAR did not address some weaknesses noticed in earlier projects and was not used by some of the PCIs and TA beneficiaries as a guide for implementation of the project. Some of the PCIs that have been participating in the SMI schemes did not obtain the SAR and DCA and were not aware of the content of the project.  - 10 - 1.30 With respect to the credit component, data relevant to the preparation of the PCR was made available to the Association by NDB. However, information relating to the TA was not available in the format requested mainly due to the poor monitoring and documentation of the TA projects. 11. Consultina Services 1.31 The consultancies for the restructuring of the BOC and PB were completed as scheduled. Although the quality of work performed by the consultants selected by BOC was, in general, adequate and satisfactory, those for PB did not perform to the satisfaction of the bank. Lack of continuity in their teams, recommending USA practices without any consideration to local needs and culture, and overlapping the terms of reference for different stages of the studies were the main factors for their poor performance. The consultants provided their services to these banks for a period of four years. 12. Conclusions and Lessons Learned 1.32 The SMI-III Credit component was implemented satisfactorily; the TA component, however, was implemented with mixed results. The project was successful in providing needed credit for development of a large number of small and medium scale industries. The employment impace of the project was also very significant. The project played an important role in encouraging the private commercial banks, which accounted for about 47 percent of the market share, to increase their SMI lending activities. The PCIs, in turn, benefited from the TA fund for their institutional upgrading and staff training programs. The project was, to some extent, successful in supporting the GOSL's trade policy reforms, in particular, in removing the constraints to export financing and duty drawbacks for indirect exporters in the SMI sector. The project further deepened IDA's dialogue with GOSL on issues affecting the efficiency of the financial sector for industrial financing through an analysis of the operational and financial viability of the two state owned commercial banks and addressing issues in effective banking supervision and regulations. 1.33 The project was, however, not able to achieve a number of objectives both in sub-project financing (due to prevailing circumstances in the country) and technical assistance programs. The sub-projects financing was concentrated in selected industries with limited growth and/or export potential, and were concentrated in urban areas (about 50 percent in Colombo and Gampaha). Like the previous SMI projects, SMI-III did not adequately contribute to promoting rural industrialization and employment, in particular, in the Northern, Eastern, and Southern provinces. Due to the absence of an in-built mechanism for collecting accurate follow-up data on sub-project performance, it was difficult to carry out an objective analysis of the benefits of the project. The technical assistance component was not fully utilized and did not, in general, achieve its objectives except in few specific areas such as training and TAF. The SCBs' implementation of their operational review was far short of expectations. 1.34 The main lessons learned, some of which were brought out by the earlier PCRs for SMI I & II, are summarized as follows: i. The available records of the SMI projects in Sri Lanka indicate that these projects and associated TA, in general, have been successful in contributing to the SMI sector growth and efficient employment creation and could be an effective vehicle for poverty alleviation. An in-depth evaluation of SMI projects, however, requires better sectoral data on SMI output and employment through a systematic follow-up study of the sub-projects to determine the benefits of the project. For a more systematic and effective  - 11 - project monitoring and evaluation of the economic benefits of sub-projects, the role of the Apex Institution (NDB) needs to redefined and its capabilities need to be further strengthened. ii. SMI lending remained concentrated in selected industries and a few urban areas despite efforts at dispersion. It is appropriate for the GOSL to make efforts to remedy these problems through programs for entrepreneur development and decentralizing banking and technical services. Also, efforts should be made by the PCIs to utilize branches more effectively for credit delivery of such a project. iii. Recognizing the importance of the SMIs' contribution to value added and employment and their potential for generating foreign exchange, a sustainable expansion of the sector requires that the financial sector adopt lending to SMIs as part of their long-term strategy and not as a transient involvement due to the credit lines available from IDA and other multilateral donors. This can be achieved through improving the efficiency and deepening the financial system which would enable the commercial banks to channel the domestically funded savings to the entrepreneurs. iv. To reduce the CBSL's loss associated with the credit guarantee scheme, the design of the scheme needs to be improved and sub-project appraisal and supervision need to be further strengthened. The GOSL should further examine the need for such scheme for the future projects. v. The TA programs can only make significant contribution to increasing the overall efficiency when: (i) the program clearly defined for it to be effectively implemented; and (ii) both IDA and GOSL develop a more effective and systematic project monitoring and evaluation procedures. In the case of large TA programs (TA for BOC and PB), slippage in overall reform program greatly affect the effectiveness of TA. For the long-term, strengthening needs to begin from the project design and preparation stage. Alternative approaches (such as freestanding TA) for designing TA programs should also be examined by IDA and the GOSL. vi. Both IDA and GOSL should ensure that the PCIs obtain the project documents and that availability of the TA funds are widely publicized and disseminated to the concerned institutions and SMI sector. vii. Steps should be taken to improve the effectiveness of the consultants through better selection process, evaluation of their performance, strengthening the terms of reference, and establishing counter-part teams in the concerned institution. viii. Despite efforts at strengthening the role and developing the skills of the staff of the public sector organizations that are engaged in providing technical and marketing services to the SMIs, deficiencies still exist due to the quality of services and effectiveness of delivery system itself. In order to improve the public advisory organizations' efficiency and responsiveness to private sector demand, efforts should be made at moving these institutions towards demand-driven support agencies and providing their services on a cost-sharing basis.  - 12 - PROJECT COMPLETION REPORT SRI LANKA THIRD SMALL AND MEDIUM INDUSTRIES PROJECT (CREDIT 1860-CE) PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE The following comments on the Draft Project Completion Report (PCR), parts I and III, were submitted to the Association on January 24, 1994 by the National Development Bank which was the apex institution for SMI-III. We have examined the Project Completion Report for the SMI-III Project sent under cover of your letter dated December 14, 1993. Our observations on the salient issues identified in the report, based on NDB's experience as the apex agency, are set out below: (a) Post Sanction Follow-up The report highlights the fact that NDB's monitoring system lacks an in- built mechanism to collect ex-post data on the financed sub-projects with regard to their financial and economic benefits. You may recall that the reporting formats in the preceding two projects viz. SMI-I and II included separate reports on sub-project performance. These reports were to be submitted monthly/quarterly by the PCIs to the NDB. However, during the formulation stage of SMI-III the submission of all such reports by the PCIs to the NDB was dispensed with in order to ease the paper work which the PCIs submitted onerous. A corresponding effort to place the ex-post data collection responsibility on the Head Office units of the PCIs as well as on their regional and district offices has not had much success. Besides ex-post performance data, we even find it difficult to obtain loan collection data from the Head Office units resulting in the NDB directly collecting such data from all the branches of the Banks (numbering approximately 725) twice every year so as to ensure the timely computation of PCI collection ratios. The NDB attempted to address this shortcoming by conducting surveys of sub-loan performance on a sample basis. Accordingly, at the request of IDA a sample of 220 projects were surveyed obtaining most of the post sanction operational data in the process. Since the number of sub-loans under the SMI scheme have now increased to more than 12600 it would not be practical for NDB to collect ex-post data on all sub-projects on a regular basis as stated above, since we would have to collect them ourselves directly from the branches of the PCIs. It is therefore proposed that we conduct regular quarterly surveys of all SMI assisted sub-projects taking region by region. The SMI Department of the NDB can, in addition to conducting the bi-annual collection data survey, also carry out at least two more surveys every year. We are also exploring the possibility of making use of the Industrial Development Board (IDB) to conduct two more surveys on our behalf in two other regions every year since the IDB is represented in the regions. We believe they would not only be in a position to undertake the surveys but would also benefit from the experience since it would result in a closer interaction with industry.  - 13 - (b) Rural Industrialization The report points out that SMI III had a limited impact on rural industrialization and employment in particular in the Northern, Eastern and Southern provinces. This observation is correct. However, it must be pointed out that the above position is not the result of structural or other weaknesses in the SMI scheme but one resulting from the weak demand for credit in these regions. The report correctly points out that the low demand was in turn the result of continued ethnic disturbances in North and Eastern provinces, weak industrial infrastructure and the paucity of entrepreneurship in these regions. A further observation is that the commercial bank branches in these areas are located in the urban centers and tend to concentrate on the relatively larger projects generated in these areas. On the other hand, the level of entrepreneurship in these areas is best suited to undertake smaller projects which the Banks are reluctant to finance. Accordingly, the decision taken to introduce the Regional Rural Development Banks as PCIs in the SMI-III program is a step in the right direction since these banks with their field officers are in a better position to handle and service smaller projects. (c) Credit Guarantee Scheme The report highlights the gap between premia collection and the loan amount in arrears. Two matters need to be mentioned in this connection. (i) Premia were paid by the PCIs quarterly under SMI-TII. This has however been changed in SMI-IV. At present premia are paid initially upfront at the beginning of the year. This should improve the credit guarantee scheme cashflow with the Central Bank. (ii) The loan amount in arrears may not be the actual liability to the guarantee fund since each sub-loan would have its collateral for realization.  - 14 - PROJECT COMPLETION REPORT SRI LANKA THIRD SMALL AND MEDIUM INDUSTRIES PROJECT (CREDIT 1860-CE) PART III: STATISTICAL INFORMATION 1. Related Bank Loans Credit Title/Amount Purpose Year of Status Approval Small and Mediuim Industries, Promote private industrial development focusing on 1979 Closed on $ 16 m SMI's assistance through addressing constraints 6/30/85 (SMI 1) Credit 942-CE hindering their rapid growth in order to increase their contribution to employment generation, export expansion, and economic growth. Second Small and Medium To support and expand on the SMI I objectives. 1981 Closed on Industries Proj., $ 28 m (SMI II) 12/31/87 Credit 1182-CE Industrial Development Project Provide term credit through DFIs to assist in 1983 Closed on (IDP-I), $ 25 m strengthening the system of industrial financing, 9/30/88 Credit 1401-CE assist in trade and industrial reforms, and provide technical assistance to improve selected PMEs. Second Industrial Development Provide term credit to finance rehabilitation and 1986 Closing Project expansion (BMRE) of private firms and selected extended for S 20 m public manufacturing enterprises and assist in 6 more months; (IDP-II) Credit 1692-CE implementation of policy reforms to improve Closing sectoral performance and institutional development. 6/30/94 Third Industrial Development Complement previous and on-going industrial sector 1989 On-going Project operations to provide credit through banking system Closing S 43.8 to medium- and large-scale private industrial 6/30/95 (IDP-III) Credit 1948-CE enterprises, support implementation of GOSL's policy reforms and institutional strengthening. Fourth Small and Medium Complement previous and on-going industrial and 1991 on-going Industries Project financial sector intermediation; to provide credit Closing S 45 m for SMI development and to make further 10/31/96 (SMI-IV) Credit 2250-CE contributions to policy reform and institutional strengthening in the area of trade policy administration and financial sector operations. Private Finance Development Provide investment finance to the private sector; 1993 On-going Project; USS 60 m assist in domestic resource mobilization for long- Closing (PFDP) Credit 2484-CE term investment by stimulating the development of 6/30/99 local bond market; further assist the GOSL in improving the financiaL sector operations through strengthening the financial intermediation including the key players such as contractual savings institutions.  - 15- 2. Proiect Time Table Item Date Planned Actual Date Identification/EPS 3/1987 3/1987 Appraisal Mission 6/1987 6/1987 Credit Negotiations 10/1987 10/1987 Board Approval 12/1987 12/1987 Credit Signature 2/1988 2/1988 Credit Effectiveness 4/1988 4/1988 Credit Closing 6/1993 6/1993 Credit Account Closing 10/1993 10/1993 3. Credit Disbursement Cumulative Estimated and Actual Disbursements (US$ Millions) 1988 1989 1990 1991 1992 1993 Appraisal Estimate 3.0 6.0 10.0 15.0 19.0 20.0 Actual 1/ 0.0 5.9 9.9 17.7 19.5 19.8 Actual as % of 0.0 98.3 99.0 118.0 102.6 99.0 estimate 1/ Date of final disbursement was June 1993. 4. Project Implementation Indicator Appraisal Estimate Actual Number of sub-projects financed 3,850 2,448 Average Size of sub-loans Rs 450,000 Rs 464,000 (US$ 15,000) (US$ 15,500)  - 16- 5. Project Costs and Financing A. Project Costs (Incl. ADB Financing) US$ Millions Item SAR Estimate Actual Utilization a. Term loan for SMIs 29.4 29.4 b. Technical Assistance 5.5 4.6 Component Total 34.9 34.0 B. Project Financing (US$. million) Source Planned SAR Actual IDA 20.0 19.6 ADB 15.0 14.8 Government 2.0 2.0 PCIs 8.0 10.2 Sub-borrowers 10.0 16.9 Total 55.0 63.5 6. Project Results Indicator Aooraisal Estimate Actual a. Employment Generated 31,400 28,630 b. Incremental Cost/Job US$ 1,750 US$ 2,032 c. Economic/Financial NA NA Rates of Return 7. Status of Covenants The GOSL and executing agencies were generally in compliance with covenants.  - 17 - 8. Use of Bank Resources A. Staff Inputs (Staff Week) FY86 FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 Preparation 1.6 29.0 Appraisal 3.7 18.1 Negotiation 3.4 Supervision 5.9 12.3 10.3 10.5 12.7 12.2 0.9 Other 3.3 Total 1/ 2.2 33.7 27.4 12.3 10.3 10.5 12.7 12.2 4.2 1/ The total staff weeks were 125.5. B. Missions Stage of Month/ Date Rating Project Year of Cycle Return Preparation Appraisal Supervision I 7/1988 7/28/1988 - Supervision II 4/1989 5/11/1989 2 Supervision III 12/1989 12/22/1989 2 Supervision IV 12/1990 12/15/1990 2 Supervision V 8/1991 9/05/1991 2 Supervision VI 6/1992 7/10/1992 2 Supervision VIII 6/1993 6/18/1993 2  Annex I PROJECT COMPLETION REPORT THIRD SMALL AND MEDIUM INDUSTRIES PROJECT Sub-Loan Financing (Rs) IDA ADB PCls Sponsor's Equity Total PCls Financing BOC 123,900,093 173,856,002 100,587,367 285,233,528 683,576,990 PB 134,838,605 178,520,469 105,493,524 180,056,957 598,909,555 CBOC 29,188,500 39,897,750 23,028,750 71,766,400 163,881,400 HNB 36,962,632 45,084,750 27,639,127 69,346,846 179,033,355 DFCC 166,904,187 93,235,166 29,066,594 166,722,159 455,928,106 SBL 45,540,004 59,838,095 34,166,032 77,637,421 217,181,552 RRDB-Kur 2,726,250 908,750 5,368,829 9,003,829 Seylan 750,000 3,937,500 1,562,500 6,116,000 12,366,000 TOTAL 540,810,271 594,369,732 322,452,644 862,248,140 2,319,880,787 % of Total 23.3 25.6 13.9 37.2 100.0  Annex 11 PROJECT COMPLETION REPORT THIRD SMALL AND MEDIUM INDUSTRIES PROJECT Net Refinancing Approval of the Credit Component By the PCIs IDA AD8 IDA+ADB Projects Contributed PCls Net Approval Net Approval Net Approval To Exports No. Amount (Rs) No. Amount (Rs) No. Amount (Rs) % Share No. Amount (Rs) % of Total BOC 278 123,900,093 452 173,856,002 730 297,756,095 26.2 22 22,162,938 7.4 PB 383 134,838,605 556 178,520,469 939 313,359,074 27.6 67 78,903,543 25.2 CBOC 40 29,188,500 47 39,897,750 87 69,086,250 6.1 19 16,461,750 23.8 HNB 94 36,962.632 109 45,084,750 203 82,047,382 7.2 DFCC 160 166,904,187 104 93,235,166 264 260,139,353 22.9 24 31,791,855 12.2 SBL 61 45,540,004 99 59,838,095 160 105,378,099 9.3 13 17,774,000 16.9 RRDB-Kur 63 2,726.250 63 2,726,250 0.2 2 93,750 3.4 Seylan 1 750,000 1 3,937,500 2 4,687,500 0.4 2 4,687,500 100.0 TOTAL 1080 540,810,271 1,368 594,369,732 2,448 1,135,180,003 100.0 149 171,875,336 15.1 New Projects 327 204,200,425 468 237,493,038 795 441693463 38.9 62 70,299,723 40.9 Expansion Projects 753 336,609,846 900 356,876,694 1653 693486540 61.1 87 101,575,613 59.1  Annex III PROJECT COMPLETION REPORT THIRD SMALL AND MEDIUM INDUSTRIES PROJECT Size Distribution of Sub-Loans by PCIs PCls Rs 0-500,000 Rs 500,000 to Rs 1,000,000 to Rs 2,000,000 to TOTAL Rs 1,000.000 Rs 2,000,000 Rs 4,000,000 No. Amount (Rs) Y/ Share No. Amount (Rs) % Share No. Amount (Rs) % Share No. Amount (Rs) 0/0 Share No. Amount (Rs) O BOC 521 90,878,273 32.9 115 62,893,263 30.0 56 62.113,992 21.3 38 81,870,567 22.9 730 297,756,095 PB 754 113,478,370 41.1 75 43,193,632 20.6 71 75,309,459 25.8 39 81,377,613 22.8 939 313.359,074 CBOC 37 8.740,500 3.2 20 11,581,500 5.5 19 22,728,750 7.8 11 26,035,500 7.3 87 69,086,250 HNB 141 20,281,575 7.3 35 18,687,307 8.9 15 16,258,500 5.6 12 26,820,000 7.5 203 82,047,382 DFCC 88 22.033,931 8.0 77 51,907,953 24.8 64 86,659,504 29.7 35 99,537.965 27.8 264 260.139,353 SBL 77 17,858.045 6.5 39 20,666,618 9.9 27 28,941,252 9.9 17 37.912,184 10.6 160 105,378.099 RRDB-Kur 63 2.726,250 1.0 63 2,726,250 Seylan 1 750,000 0.4 0.0 1 3,937,500 1.1 2 4,687.500 TOTAL 1681 275,996,944 100.0 362 209,680.273 100 252 292,011,457 100 153 357,491,329 100 2,448 1.135,180,003 tt of Total 68.7 24.3 14.8 18.5 10.3 25.7 6.3 31.5  Annex IV PROJECT COMPLETION REPORT THIRD SMALL AND MEDIUM INDUSTRIES PROJECT Sectoral Distribution of Sub-Loans IDA ADB IDA + ADB Main Sectors No. Amount (Rs) No. Amount (Rs) No. Amount (Rs) % Share Food Processing, Beverage and Tobaco 218 121,700,362 305 147,963.463 523 269,663,825 23.8 Textiles 24 16,670,025 38 25,098,402 62 41,768,427 3.7 Garments 104 69,555,365 134 83,874,566 238 153,429,931 13.5 Metal Products 114 41,931,396 113 37,155,922 227 79,087,318 7.0 Construction Materials 92 34,509,607 164 54,387,464 256 88,897,071 7.8 Rubber Products 32 28,524,382 32 27,444,188 64 55,968,570 4.9 Wood and Wood Products 84 30,839,675 97 26,341,090 181 57,180,765 5.0 Leather and Allied Products 16 5,013,000 18 5,201,243 34 10,214,243 0.9 Plastic Products 10 7,230,000 25 19,990,702 35 27,220,702 2.4 Printing and Paper Products 49 36,931,465 44 19,392,199 93 56,323,664 5.0 Chemical and Chemical Products 17 11,680,050 17 7,392,288 34 19,072,338 1.7 Agro Industries 35 9,678,236 43 14,418,581 78 24,096,817 2.1 Fisheries 9 2,973,750 18 10,756,760 27 13,730,510 1.2 Animal Husbandry/Horticulture 70 22,090,647 77 20,186,796 147 42,277,443 3.7 Construction Contracting 11 6,432,000 8 3,685,275 19 10,117,275 0.9 Services 63 27,889,010 51 19,089,075 114 46,978,085 4.1 Miscellaneous 57 34,463,176 72 32,807,593 129 67,270,769 5.9 Transport/Storage and Communication 75 32,698,125 112 39,184,125 187 71,882,250 6.3 Total 1,080 540,810,271 1,368 594,369,732 2448 1,135,180,003 100.0  Annex V PROJECT COMPLETION REPORT THIRD SMALL AND MEDIUM INDUSTRIES PROJECT Geographical Distribution of Sub-Loans IDA ADB IDA + ADB Districts No. Amount (Rs) No. Amount (Rs) No. Amount (Rs)