Nepal - Second Cottage and Small Industries Project

Document of The World Bank FOR OmCLAL USE ONLY Report No. 14049 PROJECT COMPLETION REPORT NEPAL SECOND COTTAGE AND SMALL INDUSTRIES PROJECT (CREDIT 1696-NEP) MARCH 13, 1995 Country Operations, Industry and Finance Country Iepartment II South Asia Regional Office 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 EOUIVALENT Currency Unit = Nepalese Rupees (Rs) Appraisal Year (FY86): SDR 1 = US$ 1.16 US$ 1 = Rs. 20.8 Rs. 1 = US$ 0.05 Completion Year (FY92): SDR 1 = US$ 1.40 US$ 1 = Rs. 43.5 Rs. 1 = US$ 0.02 ABBREVIATIONS AND ACRONYMS ADB/N - Agricultural Development Bank of Nepal CIDB - Cottage Industry Development Board CIHE - Cottage Industry Handicrafts Emporium CGC - Credit Guarantee Corporation CGS - Credit Guarantee Scheme CSI - Cottage and Small Industries CSIDC - Cottage and Small Industry Development Center ESC - Export Services Center HMG - His Majesty's Government of Nepal JVB - Joint Venture Bank MOI - Ministry of Industry NBL - Nepal Bank Limited NMTC - Nepal Metal Trading Company NRB - Nepal Rastra Bank NWTC - Nepal Wool Trading Company PCC - Project Coordination Committee PCIs - Participating Credit Institutions PCR - Project Completion Report PCU - Project Coordination Unit PDC - Product Development Center PON - Products of Nepal office in New York RBB - Rastriya Banijya Bank SACs - Structural Adjustment Credits TPC - Trade Promotion Center UNDP - United Nations Development Programme WCDB - Wool and Carpet Development Board FISCAL YEAR July 16 - July 15  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation March 13, 1995 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Nepal - Second Cottage and Small Industries Project (Credit 1696-NEP) Attached is the Project Completion Report on Nepal - Second Cottage and Small Industries Project (Credit 1696-NEP), prepared by the South Asia Regional Office, with Part II contributed by the Borrower. This SDR 8.7 million credit was approved in May 1986. Following a major restructuring in September 1991, project activities were halted in June 1992. The credit was finally closed in March 1994, with only about hall of the original amount disbursed. On the heels of a perceived highly successful first operation, the project was launched to continue promoting cottage and small industries (CSIs) by facilitating their access to unsecured long- term credit; by strengthening public and private agencies; and by improving policies and procedures. A large technical assistance and training component (TTA) was included, financed in roughly equal parts by the credit and by UNDP. Unfortunately, the firsl CSI credit had not been an unqualified success. It had weaknesses which were magnified in the second, including complex and ambitious components, beyond the capabilities of local institutions. In the Borrower's words, the greatest drawback of the second project was that IDA "designed it in Washington and tried to implement it in Nepalese environment with almost no flexibility in its policy." Implementation of the credit was delayed by cofinancing difficulties with UNDP, by the project's complexity and by mounting deraults on the subloans. The credit restructuring was not successful, and by closing, an undisbursed balance of SDR 4.2 million was cancelled. But some of the objectives were partially achieved. Almost 20,000 jobs were created; over one-third of the artisans traincd wcnt on to start thcir own businesses; and CSI exports expanded, assisted by the TTA program. Dcspite these achievemcnts. CSIs have not gained access to unsecured credit. None of the implementing agencies crcaLed undcr the project remain alive, the recommendations of TTA studies were not implemented, and sectoral policies were not improved. For these reasons, the project oulcome is rated as marginally unsatisfactory, and its institutional development impact as negligibic. Sustainahility of the meager benefits of the project is considered unlikely. The quality of the PCR is very good. The Borrower's contribution is particularly frank and perceptive. No audiL is planned. Attachment This document has a resiricted distribution and may he used by recipients only in the performance of their ofticial duties, Its contents miiay not otherwise hc disclosed without World Bank authorization.   FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT NEPAL SECOND COTTAGE AND SMALL INDUSTRIES PROJECT (Credit 1696-NEP) TABLE OF CONTENTS PaEe No. Preface .................................................. i Evaluation Summary ........................................ ii PART I: PROJECT REVIEW FROM IDA'S PERSPECTIVE .... ........ 1 1. Project Identity .................................... 1 2. Project Background .................................. 1 3. Project Objectives and Description ........................ 2 4. Project Design and Organization .......................... 4 5. Implementation .................................... 5 6. Project Results .................................... 10 7. Project Sustainability and Impact ......................... 10 8. IDA Performance ................................... 11 9. Borrower Performance ................................ 12 10. Project Relationships ................................. 12 11. Consultant Services ................................. 12 12. Project Documentation and Data ......................... 12 13. Lessons from the Project .............................. 13 PART 1I: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE . . 15 1. Overview ........................................ 15 2. Project Design .................................... 15 3. Lending Operations .................................. 16 4. Achievements ...................................... 16 5. Prime Cause of Failure ................................ 16 6. IDA Performance ................................... 17 7. Borrower Performance ................................ 18 8. Relationship Between IDA and HMG ....................... 18 9. UNDP Performance .................................. 18 10. Services of Credit Advisor ............................. 19 PART III: STATISTICAL SUMMARY ........................... 20 1. Related IDA Credits ................................. 20 2. Project Timetable ................................... 21 3. Cumulative Credit Disbursements ......................... 21 4. Project Costs and Financing .23 5. Project Results.24 6. Status of Covenants.26 7. Use of IDA Resources .29 This document has a restricted distribution and mnay be used by recipients only in the performance of their | official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Table of Contents (cont'd) ANNEX I: Table 1. Credit Performance of PCIs ......................... 30 Table 2. Subsectoral Distribution of Subloans .................... 30 Table 3. Geographic Distribution of Subloans .................... 31 Table 4. Composition of Overseas Exports ..................... 32 Table 5. Growth Trends in Key CSI Exports .................... 33 ANNEX I: List of Consultants Provided under CSI-Il ................... 34 ANNEX III: Comments from the Cofinancier (UNDP/Nepal) ............... 35  PROJECT COMPLETION REPORT NEPAL SECOND COTTAGE AND SMALL INDUSTRIES PROJECT (Credit 1696-NEP) PREFACE This is the Project Completion Report (PCR) for the Second Cottage and Small Industries Project in Nepal. for which Credit 1696-NEP in the amount of US$10.0 million (SDR 8.7 million) was approved on May 13. 1986. The project activities were halted on June 30, 1992, three years prior to the official closing date of June 30, 1995. The credit account was closed on March 29, 1994, with total disbursements of US$6.0 million (SDR 4.5 million), The amount of SDR 4.2 million (49% of the original credit amount) was cancelled and the last disbuL-sement was made on August 19, 1993. The PCR was prepared jointly by the Country Operations, Industry and Finance Division of the (then) Country Department I of South Asia Regional Office (Preface, Evaluation Summary, Parts I and 111), and the Borrower (Part 11). Preparation of this PCR was based on IDA's project completion mission for the project in February 1994 and information from the Staff Appraisal Report, the Development Credit and Project Agreements. supervision reports. correspondence between IDA and the Borrower, internal IDA correspondence. UNDP Project Terminal Report, documentation provided by the implementing agencies, field visits to selected project sites and discussions with project, UNDP and IDA staff associated with project implementation.   - ii - PROJECT COMPLETION REPORT NEPAL SECOND COTTAGE AND SMALL INDUSTRIES PROJECT (Credit 1696-NEP) EVALUATION SUMMARY Project Objectives i. As a follow-up operation to the first Cottage and Small Industries Project (CSI-I, Cr. 1 191-NEP), the basic objective of the project was to continue to support the Government's objectives of developing efficient private sector cottage and small industries (CSIs), enhancing exports of CSI products and increasing employment in these industries. To this end, the more specific project objectives were to: (a) finance sub-projects in cottage and small industries through the financial system; (b) strengthen the term-lending capabilities of participating credit institutions (PCIs) and upgrade overall operations of two commercial banks; (c) strengthen public and private sector agencies providing marketing, raw material supply and training services to CSls; and (d) support improvements in industrial and export policy and procedures to foster efficient industrial investment. Under the project, the geographic coverage was increased to 27 districts from 9 under CSI-I. Implementation Experience ii. Although the initial processing of the project proceeded satisfactorily, effectiveness was considerably delayed because of a change in IDA's policy on executing UNDP funded cofinancing. As a consequence, UNDP had to renegotiate cofinancing and implementation arrangements with His Majesty's Government of Nepal (HMG). This delayed credit effectiveness by almost 14 months and slowed the early impetus that had been built up during processing (para 5.1). iii. After effectiveness, implementation proceeded slowly. The project was very complex and over-optimistic in its scope and its design did not give enough attention to the country's limited trained manpower. Soon after the project was launched, earlier loans made under CSI-I became due for repayment and sub-borrowers began to default on payments at a rapid rate. Alarmed by the poor collection performance, IDA tightened the recovery requirements which, in turn, caused two out of three PCIs (and many of their branches) to become ineligible to make new subloans under CSI-1I. The poor recovery level led the PCIs to revert to traditional collateral-based lending to avoid making further bad loans which they believed were caused by the new project-based lending approach of the CSIs. As a result, few new subloans were sanctioned and disbursement of the credit component was extremely slow. IDA's rigid stance on not lowering recovery rate requirements was not well received by HMG/NRB which wanted to increase the pace of lending. The technical assistance component also suffered from the various project design and coordination problems and only two (export development and extension and training) of the five TA components were implemented with reasonable success (paras. 5.2, 5.6). iv. In an attempt to improve project implementation, HMG agreed to restructure both the credit and the technical assistance components in 1991. The main features of the restructuring were: (a) to allow the private (foreign) joint venture banks (JVBs) to participate in the credit component;  - iii - (b) to remove the bankwide loan recovery requirement but keep the branchwise qualifying criteria only to permit better performing individual branches to resume lending; and (c) to allow the PCIs to charge market interest rates. On the TA side, to avoid problems associated with joint-financing of TA components, it was agreed that UNDP would exclusively finance and execute the Cottage Industry and Development Board (CIDB) entrepreneurial promotion and training components, while IDA concentrated on assisting the Export Services Center (ESC). However, the restructuring was not successful as none of the JVBs ultimately participated in the project despite their earlier interest and, therefore, disbursements failed to pickup. Additionally, HMG failed to decide on the future role and status of CIDB and ESC as envisaged at the time of restructuring. Against this background, it was agreed that the project activities should be halted as of June 30, 1992, some three years prior to the official closing date (paras. 5.13-14). The credit account, which had been kept open to complete disbursement of approved commitments under the subloan component and to recover the credit balance in the special account, was closed on March 29, 1994. The last disbursement was made on August 19, 1993, and the recovery of the balance in the special account took place on March 29, 1994. Of the SDR 8.7 million original credit amount, SDR 4.5 million (or 51 %) was actually disbursed and about SDR 4.2 million (49%) was cancelled. Project Results v. Achievement of project objectives has been mixed. Only 44% of expected number of subloans were approved and only 49% of the line of credit was utilized. The recovery rate of about 63 % ' was significantly below the targeted level of 95% or more. Additionally, the project failed to achieve its objectives of establishing sustainable financial intermediaries for lending to CSIs. Of the TA components, the upgrading of two commercial banks' operations were unsuccessful, two components were never implemented (upgrading of Cottage Industry Handicraft Emporium and supporting HMG's industrial and trade policy improvements); only the expansion of export promotion operations and strengthening extension services were implemented with reasonable success. While this gives a clear picture of a failed project, the CSI sector, as a whole, continued to expand in line with the quantitative targets anticipated at appraisal and some of its growth could be attributed to the project as indicated below. vi. According to the data provided on approved subloans, the subprojects were expected to generate about 19,347 new employment. This, combined with about 500 jobs created as a result of CIDB's entrepreneurial development program, nearly fulfilled the appraisal target. In addition, the project provided skills and entrepreneurship development training to about 3,000 artisans and entrepreneurs (fully meeting the quantitative target). According to CIDB's monitoring data of FY89-90 trainees, about 36% of CIDB-trained artisans started business and another 12% were employed as a result of the training. Through the export development TA component, the project played an important supporting role in the significant growth of CSI exports. CSI exports over the project period grew an average 33% per annum in US dollar terms (or 55% per annum in Rs. against 50% aimed at appraisal) and accounted for about 86% of Nepal's total exports (Part III: 5. Project Results and Annex I: Tables 4 and 5). These achievements suggest that the project did help promote Nepal's major exports and increase employment. However, since the existing data base is poor, it cannot be clearly determined how much of these achievements, especially the export growth-related, were the direct result of the project. It is also difficult to assess the quality and the sustainability of businesses and employment I Provided by NRB Refinance Unit.  - iv - created and the training provided. Nevertheless, the fact that about 63% of the CSI units established under the project are currently repaying the loan indicates that they are commercially successful and contributing to increasing output and employment. Also, as the majority of CSI exporters (especially carpets) have benefitted from the project -- less from the line of credit but mostly from the technical assistance component which succeeded in upgrading the quality of Nepalese carpets to international standards -- some credit is clearly due to the project for the recent rapid growth of CSI exports. Sustainability and Impact vii. Prospects for project sustainability are poor. Because of the poor recovery rates and the high cost of project-based lending, the credit component is not sustainable. The aim to provide CSIs with easier access to loans through introduction of project-based lending was not successful as all PCIs basically reverted to collateral-based lending. The positive impact of technical assistance for institutional development was negligible and none of the implementing agencies created under the project, including the CSI units established in PCIs and in NRB, outlived the project closing date. Also, recommendations of studies financed by the project were often not implemented, nor did they result in improved sectoral policies. viii. However, taking the loan recovery rate as a proxy, the majority of the CSIs established under the project appear to be operating successfully and continuing to contribute to increasing output and employment. As an innovation, the project introduced the concept of project-based lending to Nepal and since the closure of the project the PCIs have begun to experiment with modified non-collateral lending. In addition, before the ESC was dissolved, it helped create the Wool and Carpet Development Board (WCDB) to continue the technical consultancy and research functions for carpet and wool quality improvement. WCDB, headed by a senior ex-ESC official who received overseas training under the project, is actively seeking to improve and maintain Nepal's carpet quality and export market share (para.5 .7). Lessons Learned ix. The most important lessons learned from this project are that: (a) successor projects should not be launched before proper assessment of on-going similar projects. Most of the problems encountered with CSI-II were those already discernable under the first project. If this project had been launched after a proper assessment of the performance of the first project, many of the design flaws could have been minimized (para. 4.1). (b) the design of investment credits for CSI, especially in countries like Nepal with underdeveloped financial and industrial sectors and limited implementation capacity, should be kept simple with clear and focussed objectives. Components should be well- linked and supportive of each other and the implementation of the project should not involve an excessive number of institutions. At appraisal, IDA must assess more carefully the institutional capacity of the Borrower to implement complex projects. Components to be included in the project should be selected in view of their importance to achieve project objectives and not to solve all sector problems. Similarly, successor projects should not be expanded too rapidly based on insufficient evaluation of the pilot project results. Expanding CSI-II, from 9 under the pilot project to 27 districts, was  - v - overly ambitious and beyond the implementation capacity of the agencies concerned (para 4.1). (c) creating new agencies to bypass existing inefficient units often causes duplication and overlapping of activities. Since these entities are created for the project, they often cease to exist as soon as project ends and funding stops. Due to uncertainties about their future resulting from their temporary status, these entities often have difficulty recruiting and retaining qualified staff and their performance is adversely affected. Furthermore, the impact of training provided to staff of these entities under the project was limited because all of the agencies created under it were disbanded or subsumed by different units in the relevant Ministries or PCIs before or after project closing. Similarly, rather than creating a separate credit guarantee scheme for the project, the proposed scheme should have been administered by the already existing Credit Guarantee Corporation (CGC) after strengthening it. In all these cases institutional strengthening would have been the byproduct (paras. 5.4, 5.12, 7.1). (d) innovative concepts like "project-based lending", should be introduced in a project only after ensuring that the essential ingredients required for successful implementation (sound appraisal, monitoring and supervision capacities and effective legal system for debt enforcement) are in place. Thus, strengthening the appraisal, monitoring and supervision skills and procedures of the PCIs should have been the pre-requisites for such an initiative (para. 5.3). (e) the risks of CSI lending in a poor credit environment could have been minimized through an effective management information system in the PCIs which could track and monitor each sub-loan on a regular basis and respond to problems as they emerged (para. 5.10). (f) key actions should have clear completion schedules and their compliance within the deadline should be strictly monitored (para. 5.9). (g) Relying on studies alone in a small project such as CSI-II was not an effective way to encourage policy reform. Important policy reforms should be proposed as part of a structural or sectoral adjustment project (para. 5.8).  NEPAL SECOND COTTAGE AND SMALL INDUSTRIES PROJECT (Credit 1696-NEP) PROJECT COMPLETION REPORT PART I: PROJECT REVIEW FROM IDA'S PERSPECTIVE 1. Proiect Identity Name Second Cottage and Small Industries Project Credit No. Cr. 1696-NEP RVP Unit South Asia Region Country Nepal Sector Industry 2. Project Background 2.1 With an average per capita income of only US$170 (1992), Nepal is one of the poorest countries in the world. Reflecting the fact that about 90% of Nepal's people live in rural areas, Nepal's economy is based primarily on traditional rainfed agriculture, which accounts for about 56% of GDP and 80% of employment. Agricultural productivity, however, has been stagnant resulting from rapid environmental degradation caused by inappropriate policies and strategies, and population pressure combined with excessive grazing and deforestation. Real GDP growth over the past three decades has been virtually offset by population growth (2.1 % in 1992) over this period. In a normal rainfall year, Nepal produces barely enough foodgrains to feed its population. While agricultural growth is important to improve the population's low income levels, it alone will not be sufficient to improve Nepal's prospects for increasing employment and income and improving the balance of trade. Increases in industrial production and exports are needed to supplement agricultural income. 2.2 Nepal's industrial sector is small (16% of GDP in 1992) and underdeveloped, with limited prospects for import substitution or exports of most modern industrial products. Industrial development is constrained by Nepal's limited natural resource base, small domestic demand, lack of skilled labor, landlocked position, and poor transport systems. Because of these constraints, Nepal needs to adopt a selective approach to industrial development. Promotion of cottage and small'industries (CSIs) is a key part of Nepal's approach for industrial development because CSIs I constitute about 90% of industrial establishments in Nepal, and provide about 95 % of industrial employment and about 90% of public revenues from industry. About 87% of Nepal's manufacturing value-added in the private sector is concentrated in cottage and small industries and they are the major source of export earnings (83% of total exports in 1991/92). 2.3 In view of the above, CSI development has been a priority of the Government's industrial development program under the Fifth, Sixth and Seventh Five-Year Plans. The strategy to supplement I Cottage industries are defined in the Industrial Enterprises Act (October 1982) as those with fixed investment, including land and buildings, of up to Rs. 800,000 in urban areas and Rs. 500,000 in rural areas. Small industries are defined in the Industrial Enterprises Act, 1992, as those with fixed investment not exceeding Rs. 10 million (increased from Rs. 2 million previously).  - 2 - also been an important element of IDA's industrial strategy for Nepal. In 1978, when HMG requested assistance from IDA to prepare a CSI project, IDA provided funds from the Technical Assistance Credit (Cr. 659-NEP) for an in-depth study of the CSI sector and project preparation in five product groups (carpets, cotton handloom products, metal utensils and curios, wood, bamboo and reed products, and certain agro-industries). The results of this assistance provided the basis for IDA's first Cottage and Small Industries Project (CSI-I; Cr. 1191-NEP). 2.4 CSI-I for US$6.5 million covering nine pilot districts was approved in November 1981 and closed on December 31, 1986. The project was designed to help achieve the Government's objectives of increasing incomes and generating employment and foreign exchange through expanded production and export of CSI products by: (a) building programs and capacities of key commercial, public and banking institutions to provide effective services to viable CSIs; (b) implementing innovative, product-specific extension services in two zones (Bagmati and Gandaki), which could be replicated in other areas; and (c) providing improved, simplified incentives to encourage efficient production and expanded sales of CSI products. A major objective was to tap marketing knowhow and organizing skills of private exporters and other commercial traders to provide training and commercial services to cottage industries. 2.5 CSI-I, which achieved most of its quantitative targets for the lending component and nearly full disbursement, was initially considered a success. In fact, its disbursement rate was the best of all IDA projects in Nepal at the time. According to its PCR, CSI-I had succeeded in developing the Nepal Rastra Bank (NRB) as an apex institution to administer project-based lending and in inducing all three PCIs to promote, appraise and supervise CSI subprojects and in setting "a solid framework" for successive projects. However, in retrospect it is now apparent that this assessment was premature. While NRB did play an active role in administering the project based lending and enhancing the project appraisal capabilities of the PCIs during the implementation of CSI-1, it reduced this operational role in the CSI-II; at the time of CSI-I1 appraisal, it was prematurely believed that the PCIs had developed the necessary capacity to carry on much of the project appraisal and loan sanctioning on their own. The speed of CSI-I's commitments and disbursements was misconstrued as indications that there was ample demand for CSI lending and that the implementing institutions had acquired adequate skills and experience. In fact, as is now evident, the first project disbursed quickly simply because PCIs were not concerned with the repayment of subloans (most subloans were not due for repayment before that project closed) and processed loans without proper screening and appraisal, under pressure to disburse the project funds as quickly as possible. It is also now clear that many CSI-I achievements claimed in the PCR, including "the solid framework for successor projects," were too optimistic an assessment and did not exist when CSI-II, which was designed based on this optimistic assessment, was launched in a greater scale and scope. 3. Project Objectives and Description 3.1 Proiect Objectives: As a follow-on operation to CSI-1, the project's basic objective was, as in CSI-I, to support HMG's objectives of developing efficient cottage and small industries, enhancing exports of CSI products and increasing employment in these industries. The project proposed to expand the scope and scale of CSI operations from nine district to 27 and strengthening public and private managerial, technical, marketing and other services to CSIs through provision of staff, training, equipment, vehicles, facilities and advisory services.  - 3 - 3.2 Project Description: The principal components and costs of the project were: (i) subloans to eligible cottage and small industries, market agents, and raw material supply companies (US$7.6 million, IDA); (ii) a program to strengthen the term-lending capability of the participating commercial institutions (PCIs), upgrading overall operations of the two commercial banks, local and foreign training for CSI officers of the PCIs, and strengthening the Bankers Training Institute (BTI) and NRB/CSI Unit (US$0.6 million, UNDP); (iii) an export development program, in which the Export Service Center (ESC) and a Products of Nepal (PON) office in New York would carry out market and product development, including improvements in design, production techniques, cost reduction and quality control; and strengthening the manufacturer/exporter associations to improve information flows and communication between industry and HMG (US$0.7 million IDA and US$0.8 million UNDP); (iv) a program to support HMG in formulating and implementing improved industrial and export policies (US$0.1 million, UNDP); (v) extension services for skills training and entrepreneurial development, including a Raw Material Procurement Fund and technical services to facilitate raw material supply and marketing of rural artisans' products (US$1.2 million IDA and US$0.36 million UNDP); and (vi) a program to provide management support and monitor and evaluate project components (a Chief Technical Advisor financed by IDA, US$0.3 million). 3.3 Total project costs, phased over seven years, were estimated at US$17.3 million and an IDA credit of SDR 8.7 million (US$10.0 million equivalent at the time of appraisal) was approved to fund 58% of costs, while UNDP, HMG, PCIs and CSIs were to contribute respectively 10.4%, 6%, 11.6% and 14%. Under the lending component, the same three PCIs as in the first CSI project -- two local commercial banks, Nepal Bank Limited (NBL) and Rastriya Banijya Bank (RBB), and the Agricultural Development Bank of Nepal (ADB/N) -- were responsible for channeling the subloan component of about US$7.6 million as term loans to CSIs, related market agents, and input supply companies. The PCIs could apply to the NRB (the central bank), for refinancing up to 80% of the eligible sub-loans from the project funds. The refinanced subloans of the PCIs were eligible for guarantees, under the CSI Credit Guarantee Scheme (CGS), up to 75 % of the principal amount (para. 5.4). PCIs bore the credit risks associated with the sub-loans, while the foreign exchange risk was borne by HMG. 3.4 The technical and export development component was financed through US$ 2.4 million of IDA funding and a US$ 1.8 million (later increased to approximately US$ 2.0 million) UNDP grant on a parallel basis. It was implemented by the Ministries of Industry and Commerce and NRB and included: (a) upgrading of two commercial banks' overall operations to enable them to expand term- lending operations; (b) expansion of operations of ESC in Kathmandu and the Products of Nepal (PON) office in New York; (c) strengthening of CIDB's extension services, skills training, and entrepreneurship development; (d) upgrading of the Cottage Industry Handicraft Emporium (CIHE)  operations; and (e) support for HMG's industrial and trade policy improvements by providing technical assistance to implement duty-free access to imported inputs for export production and to improve import and export administration and raw material supply. 4. Project Design and Organiza&on 4.1 The project was prepared by a committee of representative from key implementing agencies with assistance from UNDP-financed consultants and IDA. Recognizing that credit alone was not enough to promote CSIs and exports, the designers of the project tried to deliver an integrated package of services (easier access to credit, training for small entrepreneurs, improvement of raw material supplies, easier marketing and export promotion, technical assistance for strengthening relevant lending and implementing agencies, and assistance for the introduction of appropriate sectoral policy changes). While this approach to address the sector's problems through a comprehensive and coordinated package of services was conceptually sound, it presented a challenging tasks of implementation. This showed that: (a) the project contained too many components, involved too many agencies and was too complicated to coordinate and monitor. The project involved three line ministries (Finance, Industry and Commerce), seven independent agencies (NRB, three PCIs, ESC, CIDB and CIHE) and two trading companies (NWTC and NMTC), each with different objectives and policies. Combining them all in one project made the project complicated and unmanageable. Coordination and monitoring was further complicated because of cofinancing, which introduced its own set of problems; (b) the project was launched, and substantially enlarged in scope and complexity, before the final results of the pilot project were in; implementation weaknesses and poor credit recovery which began to surface towards the end of the first project were not adequately addressed in the design of the second project. The overly ambitious expansion severely overstretched the Borrower's implementing capacity, leading to the inevitable failure of the project (paras. 5.5, 5.9); and (c) the project objectives were defined too broadly and the activities under each component were not well coordinated to support each other to achieve the common objectives. Rather, each component was serving different beneficiaries and achieving different objectives as in separate projects (para. 5.9). For instance, the skills and entrepreneurship development training was not linked to CSI credit, and training was given to anyone on an open basis. As there was no assurance for easier loan approvals after the successful acquisition of skills for business, only the small number (less than 14%) of trainees with collateral were able to secure CSI loans for their business start-up. The training could have been more effective and conducive to promoting cottage and small industries if it were provided selectively to people with. genuine interest and potential for business. By linking it formally with the CSI subloan process and allowing trainees an easier access to CSI loans for the seed money required for starting a business after successful completion of the training, the project objectives could have been better met. This would also have helped the PCIs in identifying good loans, discouraged willful defaulters from accessing these loans and supported eager CSI entrepreneurs with better potential for success thereby resulting in better recovery rates. Direct linkage to business and employment would have enabled the training component to be more  effective in achieving the project objective of developing efficient industries and increasing employment. 5. Implementation Project Start-up 5.1 The Credit became effective on July 31, 1987, fourteen months after Board approval. The delay was mainly in processing the UNDP project document for technical assistance components which were to be financed by UNDP. The UNDP related effectiveness conditions were delayed because, after the project was appraised and the design agreed by HMG, IDA and UNDP, IDA's policy regarding its execution role on cofinanced projects changed and IDA decided not to execute the UNDP- financed portions of the project. This required a reappraisal of the UNDP component and the introduction of significant time consuming changes to the original project design and financing. Implementation of the Credit Component 5.2 Under the credit component, only 2,058 subloans (or 44%) of the appraisal target of 4,720 were sanctioned and SDR 3.2 million (or 49%) of the total line of credit were disbursed. The major performance problem under the credit component was the poor recovery rate (Annex 1: Table 1) which was below the SAR targets. Originally, the qualifying criteria for lending for PCIs was that each bank should achieve an aggregate recovery rate of at least 60% and each participating branch should collect at least 70% of amounts due. These rates were subsequently increased to 70% bankwide and 79% branchwise by the time project restructuring (para 5.13) was being discussed. Unable to meet the aggregate bankwide target, two of the three PCIs (and their high performing branches), became ineligible to participate. To avoid penalizing good performing branches and their clients, IDA agreed to waive the bankwide recovery requirement as part of the project restructuring attempt in July 1990. However, performance continued to deteriorate and by June 1992, over 57% of the PCI branches were below the qualifying criteria and remained ineligible to sanction new loans under the line of credit. IDA's rigid stance on not lowering the recovery rate requirements was not well received by the HMG/NRB which wanted to increase the pace of lending. 5.3 Some of the reasons for the poor performance of the credit component include: (a) PCIs' lack of skills and experience crucial to project-based lending (sound appraisal, monitoring and supervision capacities); (b) general lack of financial discipline and high incidence of willful defaults of the subloans sanctioned under CSI-I partly due to widespread perception earlier among some of the sub- borrowers that the CSI loans were really a grant and partly due to wrong identification of clients and subprojects; (c) extension of the project from 9 to 27 districts including remote hill districts with low industrial base; and (d) to some extent, the adverse business environment resulting from the trade and transit impasse with India, followed by Nepal's political instability and change of government. 5.4 The poor repayment performance of subloans led the PCIs to abandon the concept of project-based lending and to revert to traditional collateral-based lending. The failure of the CSI Credit Guarantee Scheme (CGS) may have also played a part in the demise of the innovative but riskier project-based lending (para. 2.5). When CSI-I was launched, a new Credit Guarantee Scheme specifically for the CSI refinancing program was established, despite the existence of the Credit Guarantee Corporation (Private) Limited (CGC) which automatically guaranteed all priority sector (including CSI sector) loans. At the time, because CGC was not effective due to a severe shortage of  - 6 - funds, it was thought crucial to create a better performing guarantee scheme to give confidence to the PCIs in lending on the merit of a project instead of collateral. The CGS, which was primarily funded by a 1 % premium on refinanced loans, was still not adequately capitalized and was not effective. Under CSI-II, the premium was thus increased to 3% and the Government took responsibility for maintaining an adequate level of funds in the Scheme. However, the Scheme failed to work under CSI-II also due to the Government failure to maintain the required level of funding. Of an estimated total guaranteed subloans of about Rs. 152 million under CSI-I, PCIs registered claims of Rs. 32 million, against which the Scheme paid out only Rs. 10 million (31 % of claims) due partly to lack of funds and partly to PCIs' failure to submit required documents. The performance of CGS under the CSI-II was even less satisfactory. As of project closing, no claims had been paid. 5.5 The geographical and subsectoral distribution of subloans are shown in Annex I: Tables 2 and 3. Reflecting the rapid growth of the carpet sector during the project period, the highest number of CSI units apd amount financed were in the wool subsector (33% and 43%, respectively). Also, approximately 58% (compared to 40% in CSI-I) of lending has been for export-oriented industries, such as wool and metal products) and 42% (vs. 60% in CSI-I) for import-substitution activities, including handloom and agro-based enterprises. The average amount of subloans was Rs. 76,356 (about US$1,561) and the average amount sanctioned per worker was Rs. 8,122 (US$166). About 73% of the total amount sanctioned went to the two zones originally included in the pilot project (Bagmati and Gandaki), while only 27% went to the new five zones added under this project. The main reason for the marked imbalance in the zonal distribution of subloans sanctioned was due to the presence of the better prospects and potentiality in the two zones originally included in the pilot project. The low volume of lending in the new areas and the indications of low repayment rates would suggest that the expansion strategy may have led to additional losses to the PCIs. Data on the current status of CSI units established under the project was unavailable because there were no subloan monitoring system at the PCIs (para. 5.10). Implementation of TA Components 5.6 Implementation of the technical assistance component was also problematic and of the five TA components included in the project, only two (export development and extension and training) were implemented with reasonable success; while implementation of the component to upgrade the overall operation of two commercial banks was very limited. Of the two remaining, CIHE component was never implemented because the CIHE later decided not to participate in the project and the component relating to industrial and export policy improvement was superseded by initiatives introduced under IDA's First and Second Structural Adjustment Credits (SACs in FY87 and FY89). 5.7 The export development component, which may be considered the most successful part of the project, was implemented by ESC. Focusing on products with proven export potential (carpets, readymade garments and knitwear), ESC provided valuable marketing and technical consultancy services to exporter/manufacturers of CSI products and conducted a variety of training programs to improve product quality. A particularly noteworthy contribution by ESC was by its wool laboratory, which succeeded in developing color-and light-fast vegetable dyes and formulas for over 300 color matchings to enable both carpet and garment manufacturers to improve the quality of Nepalese products. ESC succeeded in obtaining the International Wool Secretariat authorization to act as an approved test house for "woolmark" certification of Nepalese carpets and recommended HMG to initiate the voluntary certification scheme; however, no action was taken by HMG. ESC also published rosters of exporters/manufacturers of CSI products for potential foreign buyers and provided technical advisory  services and training in wool scouring, color matching, dyeing and color and design forecasting services. Seven ESC officials received overseas training to increase their in-country training capabilities and they in turn provided a variety of training to about 240 persons. ESC's efforts for exploring US market through the PON office in New York, however, were not cost-effective. PON was later handed over to a private enterprise, which also failed to make a profit and eventually closed it down. ESC was dissolved when the project activities were terminated in June 1992 and part of its export promotion functions have since been subsumed by the Trade Promotion Center, while the Wool and Carpet Development Board (WCDB), established with the help of ESC, took over the technical and research functions. 5.8 To help facilitate the project to achieve its export development objectives, technical assistance was provided to HMG. for formulating and implementing improved industrial and export policies. A study was carried out on specific policy and procedural issues affecting the export performance of CSI products (including the allocation of US garment quotas, overcoming of problems of carpet wool shortages, import licensing for dyes and chemicals and other inputs). Relying on studies alone in a small project was not an effective way to encourage policy reform, however. While the general findings of this report were well received, no direct action was taken by HMG on the recommendations before project closing. Although it is possible that the studies may have contributed somewhat, important measures to change the structure of trade incentives to promote exports and efficient import substitution were, however, introduced under the two SACs, e.g. liberalization of Nepal's trade and industrial licensing system, removal of quantitative restrictions, rationalization of the tariff structure on third country imports, and increase in export incentives. The elimination of import duties on raw wool in FY90 has played a critical role in the increase of carpet exports and the duty drawback system introduced under SAC has contributed to a similar growth in garment exports. 5.9 CIDB implemented the extension services and skills training component which included provision of extension services (technical skills training and entrepreneurship development), management of 13 product development centers (PDCs), local marketing support and provision of access to raw materials and equipment to rural area and implementation of the handmade paper production component. Implementation of this component was limited and not as successful as the export development component because CIDB was asked to carry out too many activities in too many areas (27 districts) without the commensurate increase in the manpower or financial support to implement the activities effectively. Partly due to poor communication between the Project Coordination Unit (PCU) and its supporting consultants and UNDP and partly due to lack of action planning and a clear timetable for completion, many key actions (e.g., training of trainers, preparation of training manuals, recruitment of product development experts, overseas supervisory management training) were completed so late in the project cycle that achievements under this component were modest and ineffective. The project provided overseas training for 10 middle level managers and five field staff (out of total 20 targeted), in-country training for 42 EDP and industrial extension trainers (vs. 112 targeted), and product specific training (bamboo product, fabric design, textile dyeing) for another 78 technicians. Although CIDB succeeded in achieving the quantitative targets for training some 3,000 artisans and entrepreneurs, it was unable to provide comprehensive and sustained services to CSI units due to lack of qualified technical staff and delays in many essential activities. The aim to upgrade the levels of training in CSI-II, rather than continuing on basic training given under CSI-I to a large number of people, was not achieved mainly because the substantive training for the trainers never materialized. Middle-level managerial and technical capabilities of CIDB improved in a limited way. Also, due to lack of coordination, CIDB's activities sometimes overlapped and duplicated with the Department of Village and Cottage Industries, further worsening its already limited resource situation. The combined UNDP and IDA financing of  - 8 - the TA components and subcontracting arrangement for project implementation through an international management team, created further complication of coordination and resulted in the loss of the sense of accountability and leadership. CIDB was disbanded when UNDP funding ran out in 1990 and its functions have since been absorbed by the Cottage and Small Industry Development Center (CSIDC) and the Department of Cottage Industries of MOI. 5.10 Technical assistance in the area of institutional and financial capacity building within the government agencies and the financial institutions providing services to the CSI sector was not implemented effectively. Critical TA aimed at upgrading "all aspects" of the banking operations of NBL and RBB and computerizing their loan monitoring system for "all loan data" could not be implemented because of a lack of commitment on the part of the PCIs. Eventually this component was dropped and a new IDA/UNDP initiative, the Commercial Bank Problem Analysis and Strategy Study (CBPASS), was developed. However, changes made in the two banks were largely cosmetic due to the relatively low priority accorded by the Government to the banking sector problems. This component was later superseded by the SAC-I1 which included several financial sector reforms. Due to lack of commitment by the PCIs and NRB/CSI, the information monitoring system to computerize subloan data, monitor subloan performance and detect problems at an early stage, was not developed and implemented. Also, to strengthen the institutional capability of MOI and thereby help minimize the project implementation risks, the project provided a full-time chief technical advisor and supporting experts to the PCU. However, the PCU did not have over-riding authority to lead and effectively coordinate the activities of all of the implementing agencies. The project failed to effectively institutionalize a coordinating and monitoring role through the Project Coordination Committee and the PCU in the MOI. Consequently, the PCU was disbanded when the UNDP program ended in July 1990. 5.11 At appraisal, it was identified that the major risk affecting manufacturing projects in rural areas was the weak raw material supply and market system. To minimize this problem, the project included: (a) provision of credits to rural-based market agents; (b) improvement of the operations of CIHE, which was active during CSI-1, in supplying raw material to rural areas and marketing handicrafts produced by the rural artisans; and (c) expansion of the distribution and marketing assistance provided through CIDB. None of these were implemented successfully: only an insignificant number of market agents were identified and very few operated effectively; CIHE did not participate in the project because it did not wish to expand its operations to remote areas with little prospect for success; and rural marketing support by the CIDB was abandoned eventually because of the constraints created by the small volume of activity and the distances covered in the rural areas and because of non- participation of CIHE, which was to purchase the local production collected by CIDB. Further, HMG decided against utilization of the raw material procurement fund (US$100,000) which was set aside to finance the purchase of raw materials by CIDB for sale to rural artisans. As discussed in para. 5.8 above, the major breakthrough on the raw materials for carpets and garments was made as a result of the elimination of import duties on raw wool and the introduction of the duty drawback system under the SACs. The raw material supply problems in the rural areas may have diminished somewhat by the abolition of import licensing requirements. 5.12 Implementation of TA components suffered from a variety of problems including the complexity of the project design, limited technical resources, lack of permanent funding, lack of adequate Government support, excessive bureaucratic controls and insufficient involvement of the private sector. Also, since both ESC and CIDB had been created only for this project, their efficiency suffered from their temporary status, which made it difficult for these agencies to recruit and retain qualified staff and make operational decisions without excessive government intervention.  - 9 - Restructurin2 and Accelerated Closiny 5.13 Recognizing that the project contained many design flaws, IDA attempted to radically restructure the project to improve project implementation, taking into account lessons learned until then. This restructuring also provided IDA the opportunity to introduce new Bank policies (e.g. to bring in the private commercial banks and promote private sector participation, moving away from the earlier focus which relied exclusively on public sector). To streamline the project and to speed up disbursements, it was agreed in September 1991 to restructure both the credit and the technical assistance components. The main features of the restructuring were to: allow the private foreign JVBs to participate in the credit component; change the qualifying loan recovery criteria to permit better performing individual branches to resume lending so as not to penalize them and allow PCIs to charge market interest rates. Also, to reduce the coordination problems caused by the joint UNDP and IDA financing of the technical assistance components, it was agreed that UNDP would be exclusively responsible for funding and executing the entrepreneurial promotion and training entity (CIDB), while IDA would concentrate on export development (ESC). Also included in the TA restructuring was a study to explore the possibility for the private sector to pick up the ESC. At the same time, to address the problems concerning TA, HMG planned a number of institutional reforms affecting CIDB and ESC. The closing date was extended to June 30, 1995 in connection with the project restructuring in 1991 to allow more time for the new JVBs to set up CSI lending. 5.14 However, notwithstanding the restructuring, the project completely failed to achieve the expected results. More than a year after the restructuring was agreed, disbursements had not improved as none of the JVBs participated in the project; HMG did not carry out the institutional reforms concerning CIDB and ESC and was still ambivalent about the future of the two agencies which implemented TA components. The ESC study found no potential for privatization of the agency either. In the meantime, obviating the need for the CSI project as structured, HMG announced plans to establish new financial institutions to carry out priority sector lending and to assume the PCIs' role of term-lending to small borrowers under the project. Under these circumstances, IDA was unwilling to agree to the amendments to the DCA which would have been required to continue the project (extension of the deadline for subloan submissions and transfer of funds into the depleted TA categories). HMG accepted this position and agreed that continuation was unlikely to meet the project objectives. Hence, despite an undisbursed balance of more than SDR 3.4 million (or 51 % of the total credit amount), IDA and HMG agreed to halt the project activities as of June 30, 1992. Procurement 5.15 There were no particular procurement problems identified in the supervision reports. Disbursement and Project Cost 5.16 The estimated and actual disbursements of the project are given in Tables 3A and 3B in Part III. Originally, the Credit was expected to be committed in three-and-a-half years and disbursed in seven years. Disbursements under the line of credit were significantly slower than planned owing to problems discussed above, which eventually resulted in a cancellation of SDR 4.2 million, or about 49% of project proceeds. In all, SDR 4.5 million (51 % of total project proceeds) were actually disbursed with the final disbursement on August 19, 1993. The estimated cost of the project at appraisal was about US$17.3 million. The actual cost was about US$11 million (37% decrease) as shown in Table 4A in Part III. The lower actual cost was due to slow and incomplete implementation,  - 10 - depreciation of the Nepalese currency (by over 50%) and termination of the project activities three years before the official closing date. 6. Project Results 6.1 The attainment of project objectives has been mixed. The credit component, with the cancellation of more than half the line of credit due to slow commitment and poor recovery performance, failed to achieve its project objective of establishing sustainable financial intermediaries for lending to CSIs. Also, only two out of the five TA components were implemented with reasonable success. However, while this clearly gives a picture of a failed project, the CSI sector, as a whole, continued to expand in line with the quantitative targets anticipated at appraisal and some of its growth could be attributed to the project as indicated below. 6.2 According to the data provided on approved subloans, the subprojects were expected to generate about 19,347 new employment. This, combined with another about 500 employment created as a result of CIDB's entrepreneurial development program, nearly fulfills the appraisal target. In addition, the project provided skills and entrepreneurship development training to about 3,000 artisans and entrepreneurs (fully meeting the quantitative target). According to CIDB's monitoring data of FY89-90 trainees, about 36% of CIDB-assisted artisans started business (22% through self-financing and 14% CSI loans) and another 12% were employed as a result of the training. Through the export development TA component, the project played an important supporting role in the significant growth of CSI exports. CSI exports over the project period grew an average 33% per annum in US dollar terms (or 55% per annum in Rs. against 50% aimed at appraisal) and accounted for about 86% of Nepal's total exports (Part III: 5. Project Results and Annex I: Tables 4 and 5). These achievements indicate that the project achieved its objectives of promoting Nepal's CSI product exports and enhancing employment. However, due to lack of adequate data, it cannot be clearly determined how much of these achievement, especially the export growth-related, were the direct result of the project. It is also difficult to assess the quality and the sustainability of businesses and employments created and the training provided. Nevertheless, the fact that some 63% of the CSI units established under the project are currently repaying the loan indicates that they are commercially successful and contributing to increasing output and employment. Also, as the majority of CSI exporters (especially carpet, which showed the strongest growth) have benefitted from the project -- not so much from the line of credit as from the technical assistance component which succeeded in upgrading the quality of Nepalese carpets to international standards -- some credit is clearly due to the project for the recent rapid growth of CSI exports. 6.3 While the project was unable to induce PCIs to adopt the project-based lending approach during implementation, the two CSI projects introduced the new lending concept to Nepal and provided the necessary trial and exposure for better results in the future. It also made the banks aware of the importance of recovery rates and proper appraisal and supervision of subloans and improved, although in a limited way, the operating capability of CSI banking. Indeed, as summed up in UNDP's completion report, "it was an innovative project and helped Nepal move closer to meeting its development objectives." 7. Project Sustainability and Impact 7.1 Prospects for project sustainability are poor. Because of the poor recovery rates and the high cost of project-based lending, the credit component is not sustainable. The aim to provide CSIs  - 11 - easier access to loans through introducing the project-based lending concept was not successful as all PCIs basically reverted to collateral-based lending. The impact of technical assistance for institutional development was negligible and none of the implementing agencies created under the project, including CSI units established in PCIs and in NRB, outlived the project closing date. Also, recommendations of studies financed by the project were often not implemented, nor did they result in improved sectoral policies. 7.2 However, taking loan recovery rate as a proxy, the majority of the cottage and small industries established under the project appear to be operating successfully and continuing to contribute to increasing output and employment. Also, the project introduced the concept of project-based lending in Nepal. Since the closure of the project, PCIs, having learned the pros and cons of the new concept from the two CSI projects have, now on their own, begun to experiment with modified forms of non- collateral lending.' In addition, before ESC was dissolved, it helped create WCDB to continue the technical consultancy and research functions for carpet and wool quality improvement (para. 5.7). WCDB, headed by a senior ex-ESC official who received overseas training under the project, is actively seeking to improve and maintain Nepal's carpet quality and export market share. 8. IDA Performance 8.1 IDA proceeded with the preparation of CSI-II before any tangible results could be ascertained from the implementation of CSI-I. As a result, it failed to design the successor project to deal more effectively with the risks. Instead, IDA optimistically viewed CSI-I performance satisfactory for a pilot project and believed the implementing agencies involved had adequately acquired the necessary skills and experiences. As a result, the project design was kept essentially the same as in CSI-I while its scope was expanded to cover three times more areas. This magnified project problems and caused serious delays in implementation, limited its impact and eventually led to a premature closing of the project (paras. 2.5, 4.1, 5.14). 8.2 IDA only fielded seven supervision missions during the five years between project effectiveness and termination (Table 7B, Part III). Compared to CSI-I, which was heavily supervised (13 supervision missions in six years, including five missions in 1983) and was accorded the personal attention of a task manager who designed and nurtured the project through most of its implementation period, CSI-II was very unevenly supervised despite its more acute need for closer supervision, partly due to the Bank's reorganization (twice). The infrequency of supervision missions, combined with frequent change of task managers (at least six task managers were involved), undoubtedly contributed to the Government's loss of interest and the conflict with UNDP. Also, most IDA supervision missions were so concerned with low recovery rates and unfulfilled collection targets that all attention was diverted to those two issues rather than on effective supervision of the entire project. Consequently, very little information on progress under the technical assistance component could be found in the mission reports. This was available at a later stage when the need for TA restructuring was raised upon completion of UNDP-financed components. However, for 18 months between October 1991 and May 1993, there were no supervision missions. RBB's 'Banidng with the Poor is a business" program, and operation of NRB's two new regional banks, Sudur Paschimanchal Grameen Bikash Bank and Purbanchal Grameen Bikash Bank.  - 12 - 9. Borrower Performance 9.1 The Borrower complied with most of the credit covenants and organizational requirements but did not give adequate priority to the project. Thus, although the Project Coordination Committee (PCC) and the PCU were set up, neither PCC nor PCU took complete charge of the project and did not coordinate the project effectively or resolve the implementation problems that were referred to them. The lack of coordination among the various implementing agencies involved in the project caused operational inefficiencies and delays. It is apparent that the project failed to generate a sense of ownership by the Borrower and suffered from a lack of leadership. 9.2 The Borrower failed to maintain the capitalization of the Credit Guarantee Scheme (CGS) of the project at 10% of the guarantees outstanding in any given fiscal year. As it also failed to capitalize the Credit Guarantee Corporation (CGC) adequately, the intended transfer of the project's credit guarantee scheme to the CGC could not be effected. In addition, HMG was slow or often took no actions to implement recommendations of studies financed under the project (paras. 5.4, 5.8). 10. Project Relationships 10.1 Relationships under the project were generally cordial, although there were coordination problems between IDA and UNDP in implementing the jointly-financed TA component. Also, UNDP's subcontracting arrangement with an international management firm (SGV & Co.) created additional coordination delay and minor irritation between UNDP and TA implementing agencies. 11. Consultant Services 11.1 In all, 10 consultants (representing about 87 manmonths) were provided to CSI Unit of NRB, ESC, CIDB and PCU (see Annex I Table 4 for details). Most of the consultancy months were effectively utilized and all the agencies were generally pleased with the performance of the consultants, except in NRB/CSI and NBL where computerization could not be made operational' and was not cost effective. Several studies were also carried out by consultants (feasibility study of privatization of ESC, two technical studies on wool yarn processing and vegetable dye, an evaluation of and feasibility study of privatization of PON, Nepal's export policies and procedures). While the findings of these studies were well received by the Borrower, their recommendations were often not implemented and did not result in improved sectoral policies (para. 5.8) (Table 6B, Part III). 12. Project Documentation and Data 12.1 The Appraisal Report provided a useful framework for the review of project implementation. However, project documents held in IDA files on the earlier implementation period were not complete and little data on technical assistance performance was available. Language in the project documents was not always clear and caused some confusion regarding the project objectives. For instance, there was no mention of the future of the TA components after UNDP completed its portion of the technical assistance component half way into the project implementation period. I Later, a local consultant company provided new program to computerize the CSI loan data.  - 13 - 12.2 Also, due to lack of an information monitoring system at the CSI units of PCIs and NRB (para. 5.10), much of the key data relevant for preparing the PCR was not available. For instance, the NRB Refinance Unit was not able to provide detailed data relevant for calculating the recovery rate of the subloans made by the PCIs, and the PCI head offices could not generate the detailed data on the CSI subloans. I Actual project cost information was not readily available to allow comparison with appraisal estimates. The situation was worsened by the fact that the project activities had been halted and all implementing agencies dissolved almost two years before the actual closing of the credit. As most of the institutional memory was lost, a meaningful assessment of this project may be difficult. The lesson learned from this experience is that a PCR for projects being terminated prematurely should be initiated immediately after a decision to halt its implementation is made. 13. Lessons from the Project 13.1 The most important lessons learned from the experience of this project are that: (a) successor projects should not be launched before proper assessment of on-going similar projects. Most of the problems encountered with CSI-II were those already discernable under the first project. If this project had been launched after a proper assessment of the performance of the first project, many of the design flaws could have been minimized (para. 4.1). (b) the design of investment credits for CSI, especially in countries like Nepal with underdeveloped financial and industrial sectors and limited implementation capacity, should be kept simple with clear and focussed objectives. Components should be well- linked and supportive of each other and the implementation of the project should not involve an excessive number of institutions. At appraisal, IDA must assess more carefully the institutional capacity of the Borrower to implement complex projects. Components to be included should be selected in view of their importance to achieve project objective and not to solve all sector problems. Similarly, successor projects should not be expanded too rapidly based on insufficient evaluation of the pilot project results. Expanding CSI-II, from 9 under the pilot project to 27 districts, was overly ambitious and beyond the implementation capacity of the agencies concerned (para 4.1). (c) creating new agencies to bypass existing inefficient units often causes duplication and overlapping of activities. Since these entities are created for the project, they often cease to exist as soon as project ends and funding stops. Due to uncertainties about their future resulting from their temporary status, these entities often have difficulty recruiting and retaining qualified staff and their performance is adversely affected. Furthermore, the impact of training provided to staff of these entities under the project is limited because all of the agencies created under the project were disbanded or subsumed by different units in the relevant Ministries or PCIs before or after project closing. Similarly, rather than creating a separate credit guarantee scheme for the project, it should have been administered by the already existing Credit Guarantee Corporation (CGC) after strengthening it prior to effectiveness. In all these cases institutional strengthening would have been the byproduct (paras. 5.4, 5.12, 7.1). NRB Refinance Unit was only able to provide the mission the aggregate data on recovery rates.  - 14 - (d) innovative concepts like "Project-based lending", should be introduced in a project only after ensuring that the essential ingredients required for successful implementation (sound appraisal, monitoring and supervision capacities and effective legal system for debt enforcement) are in place. Thus strengthening the appraisal, monitoring and supervision skills and procedures of the PCIs should have been the pre-requisites for such an initiative (para. 5.3). (e) the risks of CSI lending in a poor credit environment could have been minimized through an effective management information system in the PCIs which could track and monitor each sub-loan on a regular basis and respond to problems as they emerged (para. 5.10). (f) key actions should have clear completion schedule and their compliance within the deadline should be strictly monitored (para. 5.9). (g) Relying on studies alone in a small project such as CSI-11 was not an effective way to encourage policy reform. Important policy reforms should be proposed as part of a structural or sectoral adjustment project (para. 5.8).  - 15 - PART II PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ' 1. Overview 1.1 The project implementation was delayed at the outset because of not being effective in time. The project was very complex and over-optimistic in its scope. The objectives laid out on this project, were very much attractive and high-sounding but turned out to be very difficult in translating them into action. We agree with the PCR Mission's views that the project contained too many components, involved too many agencies and was too complicated to coordinate and monitor. Based on the near full disbursement rate of CSI-I, the scope, scale and the objectives of the CSI-II were considerably widened without giving much attention to the recovery rate and capacity improvements of the participating banks. Possible difficulties on the part of implementing agencies were either overlooked or forced to ignore while attaching several conditions. The reappraisal of the UNDP component because of change of IDA's policy on co- financing, pushed project fourteen-month behind the schedule. 1.2 One of the reasons for poor repayment performance of subloans was the use of the concept of project-based lending. The concept of project-based lending helped the loan amount disburse faster, but failed to consider the default risk. 2. Proiect Design 2.1 The CSI-II was designed by a committee with assistance from UNDP and IDA. The design includes the following credit related components: * Continuation of project based lending; PCIs to sanction subloans of up to US$13,000; * Enhancement of staff's appraisal, supervision and monitoring capabilities through appropriate training; * NRB to provide refinance to the PCIs and administer the Credit Guarantee Scheme (CGS) with initial contribution of Rs. two million from the government; * Maintenance of high recovery rate by PCIs; 70 percent bankwise and 79 percent branchwise; * Extension of the project area from 9 to 27 districts; * Maintenance of Project Coordination Committee (PCC) and project Coordination Unit for overall coordination, monitoring, evaluation and supervision of project implementation; and * Strengthening of CIDB's extension services, skills training, and entrepreneurship development. This is an assessment made by the Borrower and its contents have not been edited and are not attributable to IDA.  - 16 - 3. Lending Operations 3.1 During 1987/88 - Mid July 1992 period, a total of Rs. 157.14 million was sanctioned to 2,058 CSI units. This amount was about 49 percent of the total IDA line of credit. Of the total CSI units, 58 percent were export oriented for which almost 60 percent of the loan was sanctioned., Subsectoral distribution of subloans reveals woolen based industries having received the highest proportion of CSI loan (43 percent) followed by agro-based industries (21 percent). The cotton, forest, metal based and other industries combined together received the remaining 35 percent loan. The number of employment created during the project period was estimated at 19,347. The share of the three PCIs, namely the RBB, ADB/N and NBL in the total loan amount was 49.0 percent, 41.7 percent and 9.3 percent respectively. 4. Achievements 4.1 The greatest achievement of the CSI project was that it brought about greater awareness and created favorable environment to establish cottage and small scale industrial units in the project areas in general and a remarkable growth of export oriented woolen carpet industry in particular. Besides, it also provided employment to over 19,000 persons, training to over 3,000 artisans and entrepreneurs and an exposer of project based lending to the PCIs. But, on the whole, the project may be considered as a failure case to achieve its objective of establishing sustainable financial intermediaries for lending to CSIs. 5. Prime Cause of Failure 5.1 The single most important cause of the failure of Credit component of the Project was the poor recovery performance. Poor recovery of sub-loans also gave an excuse for the PCIs to give up project based lending and return to the traditional collateral based lending. Some of the factors responsible for poor recovery performance include: * Role of NRB not well defined; * Willful defaults resulting from poor follow-ups and wrong identification of clients; * Inadequate working capital loans caused by deficient appraisals and defective policy of one- time lending for working capital; * Diversion of funds by borrowers partly caused by weak supervision and partly due to preferential rate of interest on CSI loan; * Weak and inefficient institutional support services; * PCIs growing unwillingness to participate in the project; * IDA's rigidity on branchwise recovery target. It made most of the branches ineligible for loan disbursement. Non-disbursement of further loans encouraged the borrowers to default repayments; * Frequent transfer of trained manpower to non-CSI branches and replacement by untrained staff;  - 17 - * Lack of trained staff in a number of CSI branches; and * Adverse business environment due to trade and transit impasse with India. 6. IDA Performance 6.1 The CSI-II designed by IDA was over-optimistic in its scope and its design did not give enough attention to the country's reality, e.g., underdeveloped infrastructure, primitive technology, low level of entrepreneurship, and limited trained manpower and implementation capabilities of the PCIs. The greatest drawback of the IDA was that it designed the project in Washington and tried to implement in Nepalese environment with almost no flexibility in its policy. 6.2 Defective design and too much rigidity on the part of IDA were the major causes of CSI-II failure. Further, IDA's assumption that all the pre-requisites for CSI-II implementation were in place was another mistake. The three PCIs which had to carry out the project based lending function were reluctant partners. IDA's policy to maintain recovery rate of 70 percent bankwise and 79 percent branchwise as qualifying criteria for CSI lending automatically forced two of the three PCIs to become ineligible, which they had thought as unnecessary burden imposed on them. 6.3 Creation and maintenance of PCC and PCU with functions such as coordination, monitoring, evaluation and supervision of project implementation was another flaw in the project design. Coordination aspect under the CSI-II project was a new approach. The general practices of monitoring and coordination in the public sector were ignored while designing the coordination activities. PCU was to coordinate the overall functions of the different components such as credit, extension, and export. It was difficult to implement different functions of different nature by three different HMG/N's line agencies. IDA should have made appropriate institutions responsible individually. Another shortcoming of the design was to extend project area from 9 to 27 districts. It was far beyond the managerial capabilities of the PCIs. The effectiveness and impact of the coordination cell, during the first phase of the CSI project, was not properly analyzed before implementing the CSI-II phase. 6.4 The thrust of the project was mainly on the credit component. Activities of other components (such as the extension component, which was responsible for entrepreneurship training) were not seriously taken into account by the appraisal missions. The major focus of appraisal has been on the credit component of the project. But, the other two components that were responsible for the development of the economic decision making capability of the people should have been more of importance to the country. 6.5 The IDA also failed to supervise the project activities to the extent it was needed. There were only seven supervision missions during the whole project period compared to 13 in CSI-I period. Even those missions lacked to develop and suggest appropriate project monitoring indicators and flexible mid-stream correction mechanism. 6.6 The questions as to what way the project was defective and why the project failed are already answered in Part-I, para 2.5 and para (iii) of Evaluation Summary (Draft). Our view on IDA performance matches exactly with what have been expressed in those two paragraphs.  - 18 - 7. Borrower Performance 7.1 Borrower's perception and performance were also partly responsible for the failure of the project. The weaknesses on the part of the Borrowers are highlighted in the paragraphs below: - The roles given to the PCUs were not in line with the bureaucratic practices in Nepal. Because of this, the PCC and PCU which were established in the Ministry of Industry and in the Ministry of Commerce could not discharge their roles of bringing coordination and resolving implementation problems effectively. - The Government which was supposed to capitalize the CGS to the extent of 10 percent of the guarantees outstanding in any given fiscal year, did not meet that commitment. - The PCIs did not give sufficient priority to the CSI activities. The reasons were several, but the most important of them was that of low interest spread (7 to 8 percent) which brought them small income but had to incur high lending cost. The PCIs either did not depute adequate number of trained manpower in the field offices or made frequent transfer of trained manpower from CSI-offices to non-CSI offices and replaced by untrained staff which caused deficient sub-loan appraisals, wrong identification of clients and weak supervision and follow-ups. - NRB which had taken quite active role in CSI-I did not take equally active role in coordination, supervision and follow-ups of the credit component. This was because the role of NRB was not well specified in CSI-I1. - Neither the NRB nor the Government could get the services of the consultants effectively who were to advise the authority in Project related policies and procedures. Besides, the government also did not take effective measure to implement the recommendations of studies financed under the Project. 8. Relationship Between IDA and HMG Relationships between IDA and His Majesty's Government of Nepal was cordial and friendly. 9. UNDP Performance 9.1 The technical component of the project also included multiple programs of complex dimensions which posed problems while implementing the project. Some of the programs were abandoned, some of them never implemented. The objectives of the technical component were also overly emphasized without considering the institutional capabilities of the concerned agencies. We, however, do not agree with the comments made in Part I para 5.12 that no adequate government support was given and excessive bureaucratic controls were responsible for the project failure. Several conditionalities and recommendations put forward by the consultants or the project designer were in most cases not very much appropriate and not commensurable by the borrowers. The timing of the project was also not favored by the political upheaval and trade embargo by India for about 18 months.  - 19 - 9.2 The integrated plan of operations was not ready until the PCU was disbanded. There was a Project Management Team comprising of a team leader and one expert each in banking, export and extension component. The primary function of the Project Management Team was to produce an integrated plan of operations for the duration of the project. A workshop was also organized for the purpose. 9.3 The Project Management Team was functioning as a separate entity isolated from the PCU. Decisions were taken by either side without prior consultation to each other and with the Chief Technical Advisor of the Project. 9.4 Inadequate monitoring and evaluation of the field level project implementation are a result of the inefficiency in the coordination aspects. The entire problem of the field level activities was not seen and therefore ignored by the joint management team at the center. 9.5 UNDP mainly provided technical assistance to the Project. Such assistance included staff training on project appraisal, supervision and monitoring, trainer's training; physical facilities such as computer, photo copy machines, motor bikes, and consultant services. Training component helped the NRB and PCI staff to enhance their appraisal capabilities to some extent. The quality however, was not as good as it was in CSI-I. The physical facilities provided under the TA were helpful and also useful to carry out the respective project activities. The consultancy services of the Credit Advisor was not up to the expectation. 10. Services of Credit Advisor 10.1 The performance of the credit advisor was less than satisfactory and was not fruitful. While the NRB virtually failed to make minimum use of their expertise, presence of bureaucracies and lack of commitment on the part of the PCIs, NRB and HMG/N on a number of issues seriously affected the performance of the Credit Advisor. The observations and suggestions made were either overlooked or undermined by the authorities and, at the same time, the advisor virtually failed to convince the authorities on a number of critical issues. Failure to develop and implement Management Information System in PCIs through the computerization of the subloan data to monitor subloan performance and to detect problems early through flexible midstream correction mechanisms is one of the instances.  - 20 - PART III. STATISTICAL SUMMARY 1. Related IDA Credits Loan/Credit Purpose/Comments Year of Status l________________________________________ A pproval Cr. 659-NEP Extensive study for exportable CSI's and 1978 completed Technical preparation of the CSI I project were carried Assistance out. Cr. 705-NEP Assist NIDC to become a stronger 1978 completed First Industrial development financial institution, provide Development foreign exchange for private sector. Project Cr. 1191-NEP Employment creation, small-scale industry 1981 completed Cottage and promotion, institutional strengthening through Small Industries technical and financial assistance. Project Project objectives mostly achieved. Project under review was a follow-on to this project. l Cr. 1535-NEP Support to finance medium and large-scale 1984 completed Second agro-based, light engineering and service Industrial industries, utilizing NIDC as a financial Development intermediary. Project Follow-on Project None. The project in review was not achieving the project objectives and was halted before its completion. IDA decided to review the sector before launching on a similar type of project.  - 21 - 2. Project Timetable Item Date Planned Date Revised Date Actual Identification Preparation Appraisal August 1985 Negotiations April 1986 Board Approval May 1986 Credit Signature September 1986 Credit Effectiveness June 1986 July 1987 Project Completion June 1992 June 1995 June 30, 1992 Credit Closing June 30, 1993 per DCA June 30, 1995 March 29, 1994 3.A. Cumulative Credit Disbursements (US$ Millions) FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 Appraisal Estimate 1.5 3.0 5.0 7.5 8.5 9.5 10.0 Actual 0.0 1.3 3.3 4.2 5.1 5.7 6.2 6.0 Actual as % of Estimate 0% 43% 66% 56% 60% 60% 62% SDR US$ Original Amount: 8,700,000 10,000,000 2 Amount Disbursed: 4,471,537 3 6,046,309 Amount Cancelled: 4,228,463 5,983,951 Date of Final Disbursement: August 19, 1993 Due to recovery of balance in the Special Account, disbursements in FY94 was negative (-US$ 241,264). 2 The revised Credit amount in US dollars equivalent was substantially higher due to depreciation of US dollars against the SDR. 3 Including an exchange loss of SDR 12,166.20 (or US$17,217) incurred due to fluctuations in the exchange rate associated with transactions in the Special Account.  - 22 - 3.B. Disbursements bv Category (in '000 SDR equivalents) Amount Actual Disbursed as % Allocated at Disbursements of Original Category Appraisal i 12/31/93 Allocation I Refinancing of Sub-loans, Part A 6,580.0 3,204.6 49% 2 Works under Part C (Extension & Training) 104.0 9.8 9% 3-A Vehicles, equipment, furniture, raw materials 34.0 12.6 37% under Part A 3-B Vehicles, equipment, furniture, raw materials 26.0 48.4 186% under Part B (Export Development) 3-C Vehicles, equipment, furniture, raw materials 315.0 79.4 25% under Part C 3-D Vehicles, equipment, furniture, raw materials 16.0 12.4 78% under Part D (Program to improve procurement, inventory, TA) 4-A Consultancy services under Part B 22.0 0.7 3% 4-B Consultancy services under Part C 9.0 0.0 - 4-C Consultancy services under Part P (Management 85.0 21.1 25% support and TA, PCU) . 5 Raw material procurement fund, Part C 87.0 0.0 - 6-A Salaries and allowances of staff, Part B 44.0 120.7 274% 6-B Salaries and allowances of staff, Part C 554.0 491.9 89% 6-C Salaries and allowances of staff, Part D 9.0 0.0 - 6-D Salaries and allowances of staff, Part F 5.0 10.3 206% 7 Equipment, consultancies and technical services 520.0 447.5 89% and operating costs, Part B 8 Unallocated 290.0 0.0 TOTAL 8,700.0 4,471.5 2 51% As revised at effectiveness, reflecting new UNDP financing for TA, 6/87. 2 Including an exchange loss of SDR 12,166.20 (or US$17,217) incurred due to fluctuations in the exchange rate associated with transactions in the Special Account.  - 23 - 4. Project Costs and Financing A. Project Costs (US$ million) Item Appraisal Estimate Actual Subloan Component 11.9 6.72 Technical Assistance 5.4 4.24 (est.) Total 17.3 10.96 B. Project Financing (US$ million) Source Appraisal Estimate Actual US$ million % US$ million % IDA 10.0 58.0% 6.04 55% UNDP 1.8 10.4% 1.97 18% HMG 1.1 6.0% 0.80 (est.) 7% CSIs/Associations 2.4 14.0% 1.06 (est.) 10% PCIs 2.0 11.6% 1.09 10% Total: 17.3 100.0% 10.96 100%  - 24 - 5. Project Results A. Direct Results Anpraisal Actual at Closing 1. Number of subloans sanctioned 4,720 2,058 2. Permanent Jobs Created 21,000 19,847' 3. Cost/Job (Rs.) $400 $350 4. Export and Product Development2 - Incremental Increase of Handicrafts Export 5 - 10% p.a. 32% p.a. - Incremental Increase of 20% p.a. 56% p.a. Carpet Export - Incremental Increase of Maintain the Readymade Garments Export growth rate 45% p.a. (26% average during FY82-84) - Export Growth of Total CSI products 50% p.a. 55% p.a. - Incremental Foreign Exchange Earned $ 150 m. $ 312 m. 5. Trainings - Total # of Artisans Trained: by ESC 240 by CIDB 3000 - # of Officials Trained: ESC 7 CIDB 130 (of which 15 trained overseas) - # trained for CSI lending by NRB/Bankers Training Center: NRB 59 RBB 218 NBL 144 ADB/N 123 Other 6 Including about 500 created as a result of CIDB's EDP training program (about 250 trainees opened business by self- financing). 2 As project-specific data were not available, total CSI export-related data (in Rs.) are presented here. See ANNEX 1: Table 5B for the same data in US dollars terms.  - 25 - B. Studies Title Purpose Status Impact "Feasibility Study Feasibility Study Completed July The study recommended an autonomous on Privatization of 1990 by status for ESC in the form of "Export Export Service Methodical Service Association" or "Export Service Center" Research Center, Council." However, it found no scope Kathmandu for self-sufficiency and privatization of ESC ownership. "Commercial Wool Feasibility study Completed The report proved useful as a reference Yarn Processing November 1989 by material to ESC clients. Services" DECORE, Kathmandu "Commnercial Feasibility study Completed This report was not thorough and did Applicability of December 1989 by not prove useful. Vegetables" Laxmi Malla "An Evaluation of Evaluation of Completed in The report recommended closure of the PON" effectiveness and March 1989 by office, with its efforts replaced by one- feasibility study for Andrew Singer to-one linkages between Nepalese privatization exporters and US businessmen. HMG handed over the PON office to Nepal Handicrafts Company. The "Export Market Development Fund" suggested by the report did not materialize. "Nepal's Export Studies on specific Completed in The study proposed strengthening TPC Policies and policy and procedural November 1989 by and ESC, more clearly defining their Procedures" issues affecting eh Frederick Glover. functions, reforming the regulatory export performance of environment and forceful leadership in CSI products. the export sector. While the general findings of this report were well received, HMG did not take action while the project was in operation. "An Assessment of Review and develop Completed by The report recommended that CIDB be CSI Sector and an strategy relating to UNDP consultants restructured and made an autonomous Evaluation of training and extension (Claire E. Cottage and Small Industry CIDB" activities for rural Humphrey and Development Center (CSIDC) to carry industrialization Govind R. out promotional activities for CSI sector carried out by CIDB. Agarwal) in August from a single window. 1990.  - 26 - 6. Status of Covenants Cove- Relevant Covenants: Status of Compliance nant Development Credit Agreement DCA HMG/N shall maintain the following units/entities Complied with. 3.04 with such function, powers, staffing, funds ... to enable each to achieve its purpose: a) Project Coordination Unit (PCU) b) Project Coordination Committee (PCC) c) Export Service Center (ESC) d) Product of Nepal (PON) DCA Adoption and application of Statement of Operating Policies and Procedures by: 3:05 NWTC NWTC: Complied with. 3:06 NMTC NMTC: Complied with. 3:07 ESC ESC: Complied with. 3:08 CIDB - for management of a raw material and CIDB: HMG did not wish to utilize the allocation for marketing fund "raw material and marketing fund." DCA Use of detailed action plans by ESC and CIDB in Complied with. carrying out their respective activities under the Project. Commencing 3/1/88 and every 3/1 thereafter, each entity shall review the action plan, and if required, prepare and furnish IDA for its review and comments a proposal for updating the action plan for the forthcoming fiscal year. 3.07 ESC 3.08 CIDB DCA ESC to prepare and furnish IDA by 12/31/88 a Complied with. 3.09 program for market research, product development and export sales, including eligibility criteria to be implemented by exporters on a cost-sharing basis with ESC. DCA CIHE to develop and thereafter implement an CIHE did not participate in the project. 3.10 inventory management program to reduce costs and monitor inventory levels. 3.12 NRB and the PCIs shall: (1) apply the interest rates Under the restructuring program reflected in the legal set forth in the schedule to the PA; (2) implement document amendment signed between HMG and IDA and changes in the interest rates for spreads as on 6/25/91, (1) PCIs became free to fix their lending warranted. rate; (2) CSI's refinancing rate were as per NRB's general refinancing rate. DCA Use of detailed action plans by CIHE in carrying CIHE did not participate in the project. 3.13 out their respective activities under the project. DCA HMG/N shall prepare and furnish IDA by 12/31/87 Superseded by SAL I and II. 3.14 a draft action program of import and export policies and procedures. DCA Units/entities responsible for carrying such parts of Complied with. 4.01 the Project shall maintain separate accounts (including the Special Account) audited annually, with certified copy of the audit furnished to IDA no later than 12 months after the end of such year.  - 27 - DCA HMG/N to contribute Rs. 2,000,000 to the CSI HMG/N contributed the required amount in 12/87 and 4.02 Credit Guarantee Scheme (CGS) and, as and when recapitalized during the first half of 1989 to maintain required, pay additional amounts into the CSI CGS the 10% level. However, this was not kept up and so that the CGS capitalization is maintained at no currently, CGS is undercapitalized. less than 10% of the guarantee's outstanding. Transfer of the resources and liabilities of the CSI CGS to CGC shall take place once CGC is No longer applicable. Under the newly agreed adequately capitalized and has adopted appropriate reforms, PCIs were given the option to negotiate their operating policies and procedures. own credit guarantee arrangements, if any, with CGC. NRB will continue to administer the CGS for the CSI loans made prior to the new agreements. DCA PCU to prepare and furnish to IDA, commencing Complied with. 4.04 on 1/1/87 and thereafter, quarterly progress reports within 30 days after the end of each reporting period. Sch. 2 HMG/N to employ and assign a consultant to PCU Complied with. Sec. Ilb to assist PCU in the overall coordination, monitoring and evaluation of project implementation.  - 28 - PROJECT AGREEMENT PA NRB shall maintain the CSI Refinance Unit headed Complied with. 2.07 by a qualified and experienced officer. PA NRB to issue operating instructions to each PCI Complied with. 2.08 including sub-project eligibility criteria, appraisal standards, monitoring and evaluation, interest rates, spreads, terms and conditions of sub-loans, and minimum staffing and training requirements. PA NRB to adopt and apply a Statement of Operating Complied with. Policies and Procedures for: 2.09 CSI Refinance Unit 2.10 CSI Credit Guarantee Scheme. PA NRB to transfer the liabilities and resources to the Refer to remarks in DCA Sec. 4.02 2.10 CSI Credit Guarantee Scheme to CGC promptly after HMG/N has complied with DCA Sec. 4.02(b). PA NRB to furnish IDA quarterly progress reports Complied with. 2.15 within 30 days after the end of reporting period commencing 1/1/87 and every 1/1 thereafter. PA NRB will review, in consultation with IDA, and if Complied with. 2.16 required, revise the staffing requirement of PCIs and PCI branches for purposes of the eligibility criteria set forth in Part 2 of the Schedule to PA commencing 5/1/87 and every 5/1 thereafter. PA NRB to maintain records and accounts which shall Complied with. 3.01 be audited annually, with certified copy of the audit furnished to IDA no later than 9 months at the end of each such year.  - 29 - 7. Use of IDA Resources A. Staff Inputs Stage of Project Cycle Staff Weeks Through Appraisal 44.8 Appraisal through Board Approval 1.8 Board Approval through Effectiveness 12.8 Supervision 112.9 TOTAL 172.3 B. Supervision Missions Month/ear No. of Specialization' Days in Performance Types of Problem3 Persons Field Rating 2 Nov. 87 3 FA, TA 19 1 Nov. 88 2 FA, TA 11 2 M Apr. 89 3 FA, ID, TA Dec. 89 2 FA, TA 14 3 F, M, T Apr. 90 3 FA, TA 16 3 F, M, T Dec. 1 - 17, 90 2 FA, TA 16 3 F, M, T Sept.26- Oct. 8, 91 2 IE, FA 12 3 F, M, T May/June, 93 2 FA, TA 10 3 F, M, T Jan.27 - Feb. 22, 2 FA, TA 27 4 - Project Completion mission. FA = Financial Analyst, Credit Line; TA = TA programming; IE = Industrial Economist. 2 1 = Problem free or minor problems; 2 = Moderate problems; 3 = Major problems. F = Financial; M = Management; T = Technical. 4 The PCR mission was for two projects (Nepal IDP-II).  - 30 - ANNEX I Page I TABLE 1: CREDIT PERFORMANCE OF PCIS (As of mid July 1992) NBL RBB ADB/N TOTAL No. of Units Sanctioned 279 788 991 2,058 Subloans Sanctioned (Rs. M) 14.6 77.0 65.6 157.2 Subloans Disbursed (Rs. M) 13.9 71.3 58.4 143.6 Principal and Interest Receivables 16.0 42.5 42.5 101.0 (Rs. M) Principal and Interest Repaid (Rs. 8.3 26.2 29.6 64.1 M) Percentage of Receivables Repaid 51.8% 61.6% 69.7% 63.4% (Recovery Rate) TABLE 2: SUBSECTORAL DISTRIBUTION OF SUB-LOANS (As of mid July 1992) No. of % No. of % Amount % Sub-sector Units Employment Sanctioned (Rs. '000) Wool 684 33 10435 54 67990 43 Agro-based 354 17 1735 9 32975 21 Cotton 346 17 2075 11 14381 9 Forest-based 216 11 1257 6 8565 5 Metal 171 8 1024 5 11248 7 Others 287 14 2821 15 21981 14 Total 2058 100 19347 100 157140 100 of which export oriented 1205 58 13534 70 93619 60 industries Source: NRB.  TABLE 3: GEOGRAPHIC DISTRIBUTION OF SUBLOANS (As of mid-July 1992) RBB NBL ADB/N Percent Total Total of Zone No of Amount No. of Amount No. of Amount Unit Amount Total Units Sanctioned Units Sanctioned Units Sanctioned Bagmati 524 59469 85 4985 269 26457 878 90911 58% Gandaki III 9885 53 3549 221 9940 385 23374 15% Narayani 57 3072 44 2831 95 10739 196 16642 11% Rapti 46 3107 58 1767 174 9638 278 14512 9% Lunmbini 29 1059 13 647 128 3584 170 5290 3% Janakpur 0 0 0 0 61 3344 61 3344 2 % Dhaulagiri 21 391 26 826 43 1850 90 3067 2% Total 788 76983 279 14605 991 65552 2058 157140 100% sc m  - 32 - ANNEX I Page 3 TABLE 4: COMPOSITION OF OVERSEAS EXPORTS 1986/87 1991/92 (Rs. M) (US$ M) % of Total (Rs. M) (US$ M) % of Total Total Overseas Exports 1,659 76 100% 12,185 271 100% CSI Exports 1,267 58 76% 10,373 230 85% Carpets 628 29 38% 7,131 159 59% Garments 611 28 37% 3,112 69 25% Handicrafts 28 I I % 94 2 1% Silver Jewelry, etc. 0% 36 0% Source: Trade Promotion Center.  - 33 - ANNEX I Page 4 TABLE 5A: GROWTH TRENDS IN KEY CS! EXPORTS (Rs. Million) TOTAL EXPORTS CARPET GARMENTS HANDICRAFTS FY Change from Change from Change from Change from Rs. (m) Prev. Yr. Rs. (in) Prev. Yr. Rs. (m) Prev. Yr. Rs. (m) Prev. Yr. 1986/87 1659 .. 628 .. 611 .. 28 1987/88 2489 50.0% 1212 93.0% 897 46.8% 27 -3.6% 1988/89 3041 22.2% 1589 31.1% 1108 23.5% 31 14.8% 1989/90 4389 44.3% 2294 44.4% 1409 27.2% 48 54.8 % 1990/91 5671 29.2% 3702 61.4% 1343 -4.7% 49 2.1% 1991/92 12185 114.9% 7131 92.6% 3112 131.7% 94 91.8% Average Growth: 52.1% p.a. 64.5% p.a. 44.9% p.a. 32.0% p.a. TABLE 5B: GROWTH TRENDS IN KEY CS! EXPORTS (US$ Million) TOTAL EXPORTS CARPET GARMENTS HANDICRAFTS FY Change from Change from Change from Change from US$ (m) Prev. Yr. US$ (m) Prev. Yr. US$ (m) Prev. Yr. US$ (m) Prev. Yr. 1986/87 76.45 .. 28.94 .. 28.16 .. 1.29 1987/88 112.62 47.3% 54.84 89.5% 40.59 44.2% 1.22 -5.3% 1988/89 119.25 5.9% 62.31 13.6% 43.45 7.1% 1.22 -0.5% 1989/90 153.46 28.7% 80.21 28.7 % 49.27 13.4% 1.68 38.1% 1990/91 179.46 16.9% 117.15 46.1% 42.50 -13.7% 1.55 -7.6% 1991/92 271.38 51.2% 158.82 35.6% 69.31 63.1% 2.09 35.0% Average Growth: 30.0% p.a. 42.7% p.a. 22.8% p.a. 11.9% p.a. Source: TPC/NRB.  - 34 - ANNEX II LIST OF CONSULTANTS PROVIDED UNDER CSI-II Project Management Team (SGV & Co. Philippines) Jan. 1988 - Dec. 1989 1. Dr. P.V. Viloria, Project Team Leader, PCU 2. Mr. F.T. Chio, Credit Advisor, NRB-CSI 3. Mr. A.L. Cagalingan, Export Marketing Advisor, ESC United Nations Volunteer 4. Ms. Christiane Ken, Silversmith Oct. 5, 1988 - Aug. 5, 1990 Consultants 5. Mr. Alfredo L. Carlos, Philippines Jan. 1, 1980 - June. 1988 Entrepreneurship and Extension Training Dec. 16, 1989 - Mar. 15, 1990 6. Mr. Rick Kersey, Computers (NRB) Flanan Associates, Ottawa, Canada Mar. 2, 1989 - Dec. 1989 7. Mr. Andrew Singer, England Effectiveness of PON, New York Jan. - Mar. 1989 8. Mr. Frederick Gover, USA Export Development July 16 - Nov. 25, 1989 9. Mr. M.M. Macasiling, Philippines Bamboo Products Specialist July 28 - Oct. 5, 1989 10. Ms. C. Thenuwara, Sri Lanka Wearing and Fabric Design Specialist Nov. 15, 1990 - Feb. 14, 1991  - 35 - ANNEX III Page I UNDP/NEPAL COMMENTS ON THE PROJECT COMPLETION REPORT ON COTTAGE AND SMALL INDUSTRIES (CSI-II) PROJECT (CREDIT 1696-NEP), NEPAL 1 A. General Comments on the CSI Project 1. The CSI-II Project aimed to assist the Government to achieve its development objectives of increasing employment opportunities, foreign exchange earnings, and domestic production through the promotion of the CSI sector. Since this sector was defined in terms of the magnitude of capital investment, it can cover any industrial enterprise within the definition under the maximum limit of capital investment of Rs. 10 million. Besides this, the CSI project envisaged to cover 27 districts and produce thirty outputs and seven immediate objectives. 2. The UNDP TA focused on the strengthening institutional capabilities of organizations providing specialized support services which mainly include credit, skill development training, entrepreneurship development, marketing to the CSI sector, etc. 3. The direct beneficiaries comprised the Nepal Rastra Bank together with its Banking Training Center, the three Participating Credit Institutions (PCIs), Cottage and Small Industry Development Board (CSIDB), Export Promotion Center, Cottage Industries and Handicrafts Emporium (CIHE). It implied the need for coordination amongst the sectoral line ministries (Ministry of Industry, Ministry of Commerce, Ministry of Labour, and Ministry of Finance) and a number of NGOs. 4. The project thus laid special emphasis on the coordination and monitoring of the project's numerous activities. The strategy adopted for this was to strengthen the supervisory and monitoring capability of the Ministry of Industry working through different layers of supervisory authorities constituted in committees namely, Project Coordination Committee, Project Coordination Unit, Project Management Team, etc. The Project Coordination Committee was the apex body constituted under the chairmanship of the MOI Secretary, with representation of senior officials from concerned line ministries, other national parties and donor agencies. 5. The above brief description of the project clearly indicates that the project was indeed overambitious. It attempted to link independent organizations with different constituencies with contradictory goals. By the same token, the project lacked the necessary focus and failed to clarify the responsibilities of the implementing parties and to define the target beneficiaries in clear terms. 6. The UNDP's Evaluation Mission fielded in December 1989 summarized its findings in the following words: "Although some of the project's immediate objectives have been met and many of the planned activities have been completed, CSI-II has had little impact in meeting the nation's development objectives or stimulating the CSI sector." This is an assessment made by the cofinancier (UNDP) and its contents have not been edited and are not attributable to IDA.  - 36 - ANNEX III Page 2 7. The major areas of the project's failure relate to project coordination and monitoring, both at the policy level and at operation level, the administration of the credit component, and training and extension activities. Due to the weak project coordination and monitoring, the linkage among the implementing parties did not function effectively. The Project Coordination Committee (PCC), for example, failed to draw the Government's attention to solve the current fiscal and monetary problems faced by the specific CSI industries. The Export Services Center served a different group of beneficiaries (for example, carpet and ready-made garments) who were not interested in the extension training and credit scheme offered by the project. The skilled entrepreneurs/craft persons produced by CSIDB failed to receive credit from the CSI financial scheme. 8. Another very weak component of the project was the marketing of CSI products. One important reason why the CSI clients failed to repay loans was that the project b    I  I