Document of The World Bank FOR OFFICIAL USE ONLY STAFF APPRAISAL REPORT NEPAL COTTAGE AND SMALL INDUSTRIES PROJECT

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Document of The World Bank FOR OFFICIAL USE ONLY STAFF APPRAISAL REPORT NEPAL COTTAGE AND SMALL INDUSTRIES PROJECT October 29, 1981 Report No NEP Industrial Development and Finance Division Projects Department I 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.

2 CURRENCY EQUIVALENTS Currency Unit = Nepalese Rupee (Rs) US$ 1.00 = Rs 13.2 Rs 1.00 = US$ FISCAL YEARS HMG - July 16 - July 15 IDA - July 1 - June 30 UNDP - January 1 - December 31 PRINCIPAL ABBREVIATIONS ADB - Asian Development Bank ADBN - Agricultural Development Bank of Nepal CIDB - Cottage Industry Development Board CIED - Cottage Industry Export Development Division - Trade Promotion Centre CGC - Credit Guarantee Corporation, Private Limited CSI - Cottage and Small Industries DCVI - Department of Cottage and Village Industries Emporium - Cottage and Handicraft Sales Emporium HMG - His Majesty's Government of Nepal ISC - Industrial Services Centre NBL - Nepal Bank Limited NIDC - Nepal Industrial Development Corporation NRB - Nepal Rastra Bank RBB - Rastriya Banijya Bank SBI - State Bank of India TPC - Trade Promotion Centre UNDP - United Nations Development Programme

3 NEPAL FOR OFFICIAL USE ONLY COTTAGE AND SMALL INDUSTRIES PROJECT Table of Contents Page No. 1. INTRODUCTION AND SUMMARY... 1 II. SECTORAL SETTING *... *.*...*.*.*.* 2 A. Economic Setting... 2 B. Industrial Structure and Performance... 3 C. Role of Cottage and Small Industries... 3 D. Industrial Policy Framework... 5 E. Financing of CSis... 7 F. Commercial and Technical Services... 9 III. SCOPE AND STRATEGY IN SELECTED SUBSECTORS A. Woolen Carpets and Garments B. Cotton Handloom Products C. Metal Crafts D. Forest Based Products E. Selected Agroindustries IV. THE PROPOSED PROJECT A. Objectives, Strategy and Institutional Arrangements * B. Subloan Component CSI Refinance Fund - Rastra Bank Eligibility, Terms and Conditions Credit Guarantee Scheme The Participating Banks C. Commercial and Technical Service Components Export Development - Trade Promotion Centre Extension Services and Development Centres - CIDB Raw Material Arrangements - Companies and the Emporium Coordination and Monitoring and Preparation for Expansion UNDP Financing for Training and Technical Assistance..40 D. Policies Affecting Cottage Industries 41 This report is based on the findings of an appraisal mission which visited Nepal in June/July The mission comprised N. Barry, C. Bam, S. Kandel (IBRD/IDA) and J. Starkey (Consultant). This document has a restricted distribution and may be used by recipients only in the performance of 1 their official duties. Its contents may not otherwise be disclosed without World Bank authorization.

4 - ii - TABLE OF CONTENTS (continued) Page No. V. THE PROPOSED CREDIT VI. PROJECT BENEFITS AND RISKS VII. RECOMMENDATIONS ANNEX 1: Supporting Tables Table 1 Indicators of Performance in CSIs Table 2 Gandaki Zone - Commercial Bank Branches and Cottage Industry Concentrations Table 3 Probable Locations of Performance Contracts in Gandaki Zone - Year 1 Table 4 Projected Project Benefits - Export Earnings, Output, Employment Table 5 Optimistic, Pessimistic and Expected Results with and without the Project Table 6 Estimated Credit Requirements to be Met under the Project Table 7 Interest Rate Structure of Credit Institutions Table 8 Costs and Sources of Finance for Commercial and Technical Service Components Table 9 Summary of UNDP Financing for Training and Consultancy Table 10 CSI Fund Financial Projections Table 11 Estimated Distribution of Subloans by Size Table 12 Estimated Disbursement Schedules - IDA and UNTDP ANNEX 2: ANNEX 3: Map 1 Map 2 Supporting Documents Available in Project File Maps of Project Area Nepal, Cottage and Small Industries Project, Project Area (IBRD No ) Nepal, Cottage and Small Industries Project, Craft Concentrations in Gandaki Zone (IBRD No ).

5 NEPAL APPRAISAL OF A COTTAGE AND SMALL INDUSTRIES PROJECT I. INTRODUCTION AND SUMMARY 1.01 In the Sixth Five Year Plan ( ), His Majesty's Government of Nepal (HMG) is giving priority to cottage industry development, to supplement rural earnings and expand non-traditional exports. At HMG's request, a portion of an IDA Technical Assistance Credit (659-NEP) was used for an intensive two year project preparation study. The proposed project, an outgrowth of the study and successive IDA missions to Nepal, would involve an IDA credit of SDR 5.7 million (US$6.5 million equivalent) with a US$4.5 million subloan component and US$2.0 million for commercial and technical services in promising product groups. The project also would incorporate a US$2.0 million UNDP grant, to be executed by the Bank and implemented by the appropriate agencies of HMG, for most training, consultancy and marketing inputs of the project The project area would be limited to the Kathmandu Valley 1/ and Gandaki Zone, with the objectives of strengthening services at the center and implementing product-specific schemes which could be replicated in other hill areas where cottage industries are concentrated. Kathmandu Valley has the majority of export-oriented cottage units and the headquarters of the major commercial, banking and public agencies. Gandaki Zone typifies the rural hill areas but has relatively large craft clusters and reasonable road links with Kathmandu. Commercial and technical services would focus on those major product groups with strong market prospects and problems which can be addressed within a project: carpets, woolen and cotton handloom garments, metal crafts, forest products and selected agroindustries Where feasible, private commercial organizations would be used for input supply, marketing and training, since exporters and agents have marketing knowhow and have demonstrated the ability to organize cottage industry producers in Kathmandu Valley. Under the project, the Trade Promotion Centre (TPC) would provide exporters with assistance in market contacts and product adaptation, and would engage selected exporters to extend their commercial networks to craft clusters in the hills. Capabilities of the three banks dealing with cottage and small industry (CSI) loans would be increased, with partial refinance and credit guarantees from the central bank, attractive spreads, and training of appraisal and supervision officers. These banks would make loans to individual cottage and small enterprises, related market agents, and input supply companies developed under the project. While commercial agents would be the major organizers, public institutions dealing with CSIs would have responsibilities in training, filling some gaps in commercial services, and creating a positive and stable policy environment for industrial growth. The Cottage Industry Development Board would build its extension services and provide integrated assistance to more remote cottage industry clusters; the Emporium would procure and distribute cotton yarn and organize institutional marketing of handloom products; and the Industrial Services Centre would do monitoring and evaluation. Coordination of project inputs would be provided by the Ministry of Industries through a CSI Coordinating 1/ Kathmandu, Lalitpur, and Bhaktapur Districts.

6 -2 - Committee chaired by the Secretary, Ministry of Industries and supported by full-time staff To date, IDA's involvement in industrial development in Nepal has been limited. IDA Credit 705-NEP has provided US$4 million to the Nepal Industrial Development Corporation (NIDC) for foreign exchange portions of industrial subloans. Subloans have averaged about US$155,000; commitments have been slow, reflecting problems in promotion, limited prospects for medium and large industry, and NIDC's inadequate local currency resources. Two rural development projects have included small experimental cottage industry components (Credits 617-NEP and 939-NEP). However, these deal with out-stations of agencies which require improvements in strategy, organization and incentives at the center. Under the IDA Technical Assistance Credit (659-NEP), preparation of the proposed cottage industry project has been financed and some industrial subsector feasibility studies have been undertaken. The proposed cottage industry project provides IDA with the opportunity to play a significant role in shaping assistance to one of the more promising productive sectors in Nepal. Analyses and discussions during project preparation already have resulted in improvements in overall strategy, institutional arrangements and industrial policies affecting CSI development. Training and consultancy to prepare the implementing agencies for the project have been undertaken, with UNDP financing. HMIG views the proposed cottage and small industries (CSI) project as the lead project in its program to expand CSI production and exports during the Sixth Five Year Plan period. By building institutional capabilities and replicable programs, the proposed project should have significant immediate and long-term impact on earnings, output and exports from cottage and small industries. II. SECTORAL BACKGROUND A. Economic Setting 2.01 Nepal remains one of the least developed countries, with annual per capita incomes averaging about US$140. Agriculture, accounting for more than 60% of GDP and 75% of exports, provides a livelihood to over 90% of the population. Food production has not kept pace with population growth; over the period , growth of foodgrain output averaged 1.5% p.a., while population grew at 2.6%. Stagnation in agriculture limited growth in other sectors, and GDP grew by only 2.4% p.a. in real terms during the Fifth Development Plan period ( ). The economic situation deteriorated further in 1979/80: GDP decreased by about 1% and, with major crop losses due to poor monsoons, crop output probably declined by roughly 4%. Industrial growth continued to be curbed by supply bottlenecks, delayed preparation of public sector projects, and limited potential for large industry. This weak performance led to accelerated inflation and a widening trade deficit, largely due to reduced rice exports and a surging import bill for oil and foodgrains. The situation improved somewhat during 1980/81, due to a good monsoon and consequent increases in agricultural production Agricultural growth will not be sufficient to meet Nepal's needs. In the hills and mountains, which contain two-thirds of the population but only one-third of the arable land, population pressures have pushed cultiva-

7 tion onto marginal land and average yields have declined; food production meets only two-thirds of minimum subsistence needs. While hydropower and tourism offer opportunities to increase foreign exchange earnings, these sectors do not generate the needed employment. In industry, the Sixth Five Year Plan ( ) gives priority to promotion of CSIs, to generate earnings for the hill population and strengthen the trade balance by increasing exports of non-traditional goods. B. Industrial Structure and Performance 2.03 Role of Industry. Although small and at an early stage, Nepal's industrial sector represents the largest non-agricultural income source, accounting for over 5% of GDP, as officially calculated. 1/ About 70% of industrial output involves agroprocessing; textiles, apparel and leather account for about 14%; and forest products represent about 8%. About half of industrial output is produced in 12 public enterprises; most public firms face difficulties due to low productivity, inappropriate pricing policies, and technical and managerial weaknesses. About 20% of the value added in private industry is produced in 3,500 registered, mainly small scale firms, employing 35,000; the remaining 80% comes from unregistered, mainly rural CSIs. The skill base in CSIs is large, with over one million working full or part time in about 400,000 mainly household enterprises The low level of industrial development reflects the paucity of obvious opportunities. Constraints to industrial development include: limited natural resources; a small domestic market with low purchasing power 2/; a landlocked position which increases costs of raw material imports and exports; competition from India; and absence of entrepreneurial and technical skills for modern industry, making import-substitution difficult and exports of most modern industrial products unlikely. In view of these problems, Nepal needs to adopt a highly selective approach to industrial development. Performance of public enterprises needs to be improved. Promotion of CSIs can yield substantial employment and foreign exchange benefits at relatively low cost. Prospects for increasing exports of carpets, metal crafts and wool and cotton handloom garments are good. Development of CSIs is crucial in view of their predominance, labor intensity, export potential and ability to provide substantial supplementary * earnings to the hill population. To realize this potential, CSIs need improved access to credit and effective technical and commercial services. 1/ In Nepal, GDP calculations do not include many cottage industry activities which are conducted at the household level. 2/ Cash incomes of the majority of inhabitants are less than US$50 p.a.; in hill regions most transactions are on a barter basis.

8 - 4 - C. Role of Cottage and Small Industries 2.05 Under the new Industrial Enterprise Act, the definition of cottage industries for purposes of fiscal incentives and technical services includes those with fixed investment, including land and buildings, of up to Rs 500,000 in rural areas and Rs 800,000 in urban areas. 1/ Small scale industries would be defined as those with fixed investment not exceeding Rs 2 million. 2/ In view of past inflation, these changes seem appropriate. Over 80% of the 3,500 registered industries have fewer than 10 employees; these small firms mill grain, extract oil, and produce a limited range of consumer goods. Most cottage industries are household units; fixed assets range from Rs 100 to Rs 15,000 and annual output per worker varies from Rs 100 to Rs 10,000 (Annex 1, Table 1) Subsectoral Distribution. Major cottage industry subsectors are woolen and handloom items, agro-based operations and forest products. In exports, carpets, metal crafts, woolens and cotton garments account for about 95% of sales. Five major subsectors have identified export or local market potential: woolen carpets and garments; cotton handloom fabrics and finished products; metal curios and utensils; food processing; and forest-based products Geographical Distribution. Bagmati and Gandaki Zones account for roughly 60% of cottage industries in Nepal (Annex 3, Map 1). Kathmandu Valley holds sizeable concentrations of wool, cotton and metal CSIs. In the six districts the Gandaki, almost all units are operated by poor hill families who provide goods and services for the rural subsistence economy. Agroindustries are widely dispersed, but in wool, cotton and forestry products, Gandaki holds over 40 clusters 3/ of 100 to 200 cottage units each (Annex 3, Map 2) Constraints. In most hill areas, purchasing power, already below subsistence levels, is stagnant; local markets offer limited scope for expansion in most CSI products. Seasonal demands from agriculture mean that rural workers are available for cottage industry production for about two-thirds of the year. In exports manufactured by CSIs, Nepal's landlocked position results in high transport costs which add a sizeable margin to export prices of the lower value items. Most craftsmen are not entrepreneurs; their tradition, scale, and knowledge hamper growth beyond the household level and necessitate services by market agents to provide inputs and sell finished products. Institutional constraints are serious. Public programs to assist CSIs have 1/ Previously, cottage industries were defined as those with fixed investment of up to Rs 200,000, including land and buildings; small industries were considered as those with fixed investment not exceeding Rs 500,000. 2/ Credit under the proposed project would be available for cottage and "modern" small firms with fixed assets of up to Rs 800,000 after the subloan is made, as NIDC is established to meet the credit needs of the larger end of the small industry sector. 3/ Locations of the main craft clusters in Gandaki are depicted in Annex 1, Tables 2 and 3; and in Annex 3, Map 2.

9 - 5 - been weak. Private agents, who play a major role in input supply, quality control and marketing in the Kathmandu Valley, have not had sufficient incentives to extend commercial networks to craft clusters in the hills. Exporters have inadequate market information, face delays in licensing and shipment, and have insufficient credit access. If the more serious of these constraints are addressed, CSI potential could be tapped, particularly for exports Export Potential. CSI exports have increased by about 35% p.a. since 1972/73, from a low base, and already constitute about 15% of cottage industry output and the majority of non-traditional manufactured exports. In 1980/81, exporters marketed over Rs 140 million in woolen, handicraft and handloom items; sales to tourists generated another Rs 50 million. Studies of export potential in Europe and the United States 1/ indicate prospects for handknotted Nepalese carpets, woolen knitwear, cotton handloom garments and metalware. Except in carpets, for which existing markets are relatively strong, Nepal will need to combine its exotic image with competitive price, quality, and responsiveness to western fashion trends. Concentrations of traditional skills at relatively low wages can be utilized as a strong production base. The 80 to IOU exporters of cottage industry products are among Nepal's chief assets; it is they who have increased exports, despite the substantial difficulties. While at present, these exporters work mainly with cottage units in Kathmandu Valley, their market contacts and organizing knowhow can be tapped in mobilizing production from the hills. In Gandaki about 30 craft clusters involving about 5,000 producers have reasonable access to the main roads and can be competitive, if products are adapted and production geared to exports (Annex 3, Map 2); transport costs to Kathmandu are marginal and can be outweighed by lower wages in rural areas. However, exporters would need direct incentives to help cover the initial costs of extending their networks to rural CSI clusters. In addition to these direct incentives to promote rural employment, handicraft exporters need assistance in adapting products to meet concrete market prospects If systematic marketing, technical and credit services are provided, expanded production to meet export demand would be expected to generate about US$30 million in incremental foreign exchange over four years, providing significant additional earnings to CSIs, equivalent to roughly 40,000 manyears of additional employment (paras ). While the majority of production for export would continue to come from Kathmandu, rural Gandaki Zone could generate as much as 30% of this growth Local Potential. Local market prospects in most product groups are more limited. Analysis of HMG's statistics on legal industrial imports indicates that less than 10% of the value could be replaced by reasonably competitive products from domestic CSIs. Not only is local purchasing power small, but in several product lines demand is price inelastic. There is, however, significant import-substitution potential in handloom products, and in agroprocessing if crop output can be expanded; some scope for expansion also exists in building materials, implements and footwear. Public sector and tourist markets are promising, particularly in processed foods and textiles. 1/ These two marketing surveys of Nepalese CSI export potential were prepared as part of the project preparation study.

10 - 6 - D. Industrial Policy Framework 2.12 In the Sixth Five Year Development Plan ( ), HMG recognizes the need to promote private industrial investment by reducing bureaucratic controls and providing effective fiscal and export incentives; to achieve these objectives, HMG has revised its industrial policy. idmg used consultancy services, financed by UNDP and executed by the Bank under preparatory assistance for the proposed project, in helping to finalize a revised Industrial Enterprise Act. The new Act, incorporating some recommended improvements, was passed in August 1981 and is with His Majesty for approval and issuance. While the Act includes improvements in income tax and export incentives, the mission's recommendations on more liberal industrial approval procedures and a streamlined fiscal incentive system have not been fully incorporated. Development of rules and procedures, needed to ensure effective implementation of the Act, could incorporate some of these recommendations. The following section summarizes the mission's recommendations and modifications planned by HMG on policies with direct effects on CSI development (para 4.53) Fiscal Incentives. The Act introduces product lists and size categories for CSIs, with different licensing requirements and fiscal incentives for various groups. Such scaling is cumbersome and unlikely to have significant impact on profits or economic benefits of the enterprises. The mission recommended a simple fiscal system for all CSIs; any differential incentives should be based on in-depth subsector analyses to ensure that efficient CSIs are promoted Registration and Licensing. The Act envisages registration of CSIs which seek incentives. While this is reasonable, the mission recommended that registration procedures do not inhibit CSIs from receiving credit and technical services; rules and procedures for the Act should clarify this aspect. Also, the Act specifies a number of organized small industry product lines which would require licensing in addition to registration. While this licensing would not affect most cottage enterprises it would represent an increase in controls and delays for some, in contrast to H4GN's stated objectives of reducing controls during the Sixth Five Year Plan period Joint Ventures. Under the Act, foreign investment would be allowed only in medium and large industries, with priority to enterprises employing sophisticated technology. The mission recommended that foreign partners also be encouraged in export-oriented light industries to help with organization of production and marketing channels; foreign participation by both manufacturing concerns and trading companies dealing in CSI products should be considered. The mission also recommended establishment of a "one stop office" to assist potential foreign investors in procedural matters Export Incentives and Procedures. HMG intends to introduce temporary income tax holidays and import duty drawbacks for exporting firms. Most CSIs, which would pay only nominal duty on imports of equipment and raw materials, are too small to import directly. The majority purchase from traders who pay regular import duty, pass on the cascading effect of a sales

11 tax, and charge high prices during periods of scarcity. Under the Act HMG extends import concessions to private limited companies importing inputs for CSIs, provided that CSI producers and exporters are shareholders (paras ). This measure is expected to provide effective incentives in the formation of companies importing key CSI inputs in bulk Exporters often do not obtain the full benefit of exchange rate incentives due to faulty operation of the Export Valuation Committee, which is to set standard prices for most export products. Standard prices for carpets and garments cannot incorporate differences in product quality or markets. Unit prices set by the Valuation Committee often are undervalued, discouraging product improvement. The Committee meets irregularly and often does not adjust prices to reflect cost increases. As such, the Valuation Committee represents a serious impediment to the development of non-traditional exports in Nepal. It has been recommended that the Valuation Committee be abolished. With the recent removal of the dual exchange rate system, the original rationale for the Export Valuation Committee no longer applies, and HMG is considering abolishing it Transportation Costs. A major obstacle to exports is the transport cost arising from Nepal's landlocked position. Transport costs can outweigh the advantage of lower labor costs in Nepal, pricing low value, large volume items out of the market. It has been recommended that HMG develop a transport sector policy and take concrete steps to address the issue. Short-term improvements in transport arrangements for non-traditional exports which merit early consideration include: low air freight rates to Bangladesh, Bangkok, and Colombo by the Royal Nepal Airlines Corporation to enable sea-shipment from ports other than highly congested Calcutta 1/; negotiating with India to extend to Nepal its rebates on international freight; organizing air cargo charter flights with international airlines consolidating export shipments; and improving cargo handling and warehousing facilities at Tribhuvan Airport. Based on the appraisal mission's recommendation, UNDP plans to finance a consultant to help organize the cargo facilities under a transport project Export Financing. Preshipment export financing, important in export growth, is difficult to obtain in Nepal. The appraisal mission recommended that HMG consider establishing an export credit guarantee scheme, which could start in Nepal Rastra Bank (NRB). 2/ At a minimum, NRB should reiterate its earlier guidelines, advising commercial banks on conditions under which they should extend preshipment finance. 1/ Net incremental foreign exchange earnings through increased exports would far exceed revenue losses from this measure according to a team which worked in Nepal during 1981 on a UNCTAD-ESCAP project for "Assistance to Least Developed Landlocked Countries." 2/ Detailed recommendations regarding export financing are provided in a report by the International Trade Centre in January 1980.

12 -8- E. Financing of CSIs 2.20 During the latter part of the 1970s, total demand for credit increased rapidly. In 1977/78, credit by the banking system rose by 36%, outstripping the increase in deposits. While there was some expansion in lending to industry and agriculture, most bank lending to the private sector flowed to trade, particularly for imports. Rapid credit expansion generated demand pressures which contributed to the deteriorating balance of payments and the accelerating inflation rate. In 1978/79, growth of domestic credit decreased to 21%; lending to the private sector fell, largely due to tighter credit controls from the NRB. Total industrial credit, which nearly doubled in 1977/78, increased by less than 5% in 1978/79, with banks' reduced resources lent mainly for commerce The banking system encompasses two government-owned commercial banks (Nepal Bank Ltd. and Rastrya Banijya Bank), the Agricultural Development Bank (ADBN), the Nepal Industrial Development Corporation (NIDC), and the Provident Fund Corporation. Since 1970, the commercial banks and ADBN have expanded their share in the total assets of the financial sector. Nepal Rastra Bank (NRB), the central bank, has instructed the commercial banks to lend 7% of total deposits to small agricultural, industrial and services enterprises; however, these targets have not been accompanied by training in project evaluation, margins to make small loan operations profitable, or adequate credit guarantee provisions. As a result, lending under the "priority sector scheme" has averaged only about 5% of total deposits and most CSI loans still are made on the basis of clients' collateral rather than evaluation of project viability. ADBN, which focusses on agricultural lending, also finances small agro-based and cottage industries, based on project evaluation. Industrial credit to larger industries is provided by NIDC. During the Sixth Five Year Plan period, HMG intends to strengthen the banks' programs and capabilities in meeting credit needs of CSIs and related commercial concerns. Training of over 20 key staff of NRB and the banks has been provided in India and the Philippines and local training of about 100 CSI officers has been arranged with State Bank of India during November and December The 300,000 cottage and small scale units in Gandaki and the Kathmandu Valley have a potentially large demand for credit; however, market links, raw material supply arrangements, and production methods would have to be improved before large amounts could be absorbed fruitfully by cottage industries. Also, the banks need to upgrade their CSI staff and credit operations. The project would require about Rs 67 million (US$5.1 million) in institutional term credit over a three year commitment period, for cottage units, related commercial agents, organized small industries and raw material supply companies (Annex 1, Table 6). This would represent an annual increase of about 20% in CSI lending by the two commercial banks and ADBN, which is reasonable if training and incentives in CSI lending are improved Interest Rates and Inflation. Until recently, commercial banks made short term loans at 13-18% p.a; NIDC provided long term finance for

13 - 9 - larger industries at 11-14% 1/; ADBN lent to CSIs at 8-12%; and commercial banks offered medium term credit to CSIs at 10% under the priority sector scheme (Annex 1, Table 7). The tnepal Rastra Bank recently decided to rationalize interest rates and eliminate the gap between regular commercial lending rates and rates for CSIs, with all rates to reflect inflationary trends and enable the banks and ADBN to increase the spreads between deposits and lending rates. Medium term inflation projections, previously estimated to be 10-13%, have now dropped; the anticipated annual rate for 1981/82 is about 8%, reflected the good mansoon and resulting improvements in agricultural production. Medium and long term loans to cottage industry, including those under the proposed project, would be made at rates which are positive in relation to medium term inflation projections and consistent with term rates for all industry in Nepal. The initial rate for CSI loans would be 11%; this rate for CSI would be reviewed at least annually and adjusted to maintain positive real interest rates and consistency with commercial bank rates for term loans to large industry (para 4.11). F. Commercial and Technical Services 2.24 HMG recognizes that public technical and commercial services for CSIs have been inadequate. The final draft Sixth Five Year Plan contains measures to modify and improve public and private services to CSIs; these build on and are consistent with the institutional arrangements under the proposed project. The Department of Cottage and Village Industries (DCVI), the government agency which has been responsible for CSI development, has provided some formal training and technical extension through regional and district offices. tiowever, the majority of DCVI officers have no technical background; tight controls over departmental salaries and financial management preclude performance incentives and flexible operations; and DCVI has concentrated on administrative tasks such as registration and licensing rather than promoting development of CSIs. A Cottage Industry Development Board exists, which has wide powers for commercial and developmental functions, but CIDB never has been made truly operative; during the Sixth Five Year Plan period ( ), HMG intends to shift extension services and development functions from DCVI to the CIDB, beginning with the project area Public agencies' commercial services for cottage industries are particularly weak. National Trading Ltd. has imported metal and wool yarn, but supplies have been sporadic, prices have been high to cover costs of losing items, and NTL never has been a significant factor in imports of these two comrmodities, which are vital for export-oriented cottage industries. The Cottage Industries and Handicraft Sales Eraporium imports about 15% of the total cotton yarn supply; about 10 large and 100 small cotton yarn importers account for the majority of yarn operations. The Emporium's 1/ Commercial banks' interest rates vary depending upon the type of enterprise and the collateral provided by clients. NIDC's interest rates vary by location and type of industry.

14 efforts at cottage industry retail sales have not been successful, but the Emporium has served as a stabilizing factor in cotton yarn imports and has a comparative advantage in marketing to the public sector; the Emporium's services are to be strengthened in cotton-related areas The Trade Promotion Centre (TPC) is perceived as the government agency most concerned with the problems of private exporters. However, TPC has had limited resources and technical staff, and market information and sales promotion activities have been ad hoc and sporadic. HMG intends to have TPC play a more active and systematic role in promoting non-traditional exports, particularly in cottage industry products. TPC recently has established a Cottage Industry Export Development Division to provide exporters with assistance in product adaptation in response to specific market requirements, and to mobilize exporters to extend their commercial networks to rural cottage industry clusters. Exporters provide the majority of marketing, input supply and technical services to cottage industries in Nepal. Private commercial agents have built-in performance incentives; higher salaries enable them to attract, retain and motivate qualified staff to do raw material supply, quality control and purchasing. HMG's strategy, as reflected in the Sixth Five Year Plan and the proposed project, would focus on encouraging the private commercial sector to expand their services to CSIs. III. SCOPE AND STRATEGY IN SELECTED SUBSECTORS 3.01 Project preparation involved detailed analyses of problems, potential and assistance needs in five major CSI subsectors: carpets and other woolen products; handloom fabric and garments; metal crafts; forestry-based products; and selected agroindustries. RMG and IDA chose these subsectors due to their importance, market prospects and problems which could be addressed within a project. The project preparation study, sponsored by DCVI and performed by ISC staff with 30 manmonths of outside specialists, consisted of two phases. Phase I involved a survey of cottage industries in Gandaki and Bagmati Zones. 1/ Through the survey, clusters of cottage industries were identified, CSI operations analyzed, and public and private services observed. In Phase II, product specialists did detailed analyses of the key product groups and made proposals on institutional arrangements, subsector components and costs. Also, export potential studies for Nepalese woolen, cotton and handicraft products were performed in the United States and selected European countries. 2/ IDA missions were active in this project preparation and performed additional analyses, particularly on needed reforms in institutional services and policies affecting CSIs. The prospects, problems and proposed strategies for developing the major cottage industry subsectors are outlined below. 1/ Aggregated statistics from this survey are presented in Annex 1, Table 1. 2/ Detailed reports on the survey, subsector analyses, export potential studies and final recommendations are available in the project file.

15 A. Wool Carpets and Garments 3.02 Characteristics. In Nepal, about 97,000 spinners and weavers in 46,500 units make woolen carpets, garments, and rugs; fixed investment in the industry is about Rs 135 million; annual output totals roughly Rs 240 million. Kathmandu Valley and Gandaki Zone account for about 20% of woolen product output and employment. About 70% of the Rs 75 million in carpets and wool garments exported in 1980/81 were made in Kathinandu, with the remainder coming from Gandaki Zone and other rural areas. Net foreign exchange earnings constitute about 60% of export value. Most of the 13,200 workers in 6,900 household units in the project area make traditional blankets, rugs and jackets for rural markets; due to limited and stagnant local demand, these weavers work an average of days, earning about Rs 360 annually from their work in wool. In contrast, annual earnings of carpet weavers in Kathmandu average Rs 5, Potential. There is strong immediate potential for expanding production and exports of Tibetan hand knotted carpets, 1/ due to good Nepalese designs and established contacts with importers. The carpets produced in Nepal are not direct substitutes for Persian-style carpets; however, Iran's exodus has resulted in buyers seeking new sources for a variety of ethnic carpets. While Nepal's carpet exports have been growing at 30-55% annually for the last five years and now constitute the second largest manufactured export, Nepal still represents less than 0.5% of the world carpet trade. Market surveys indicate that a 35% annual increase in Nepalese carpet exports can be maintained if the production base is expanded, wool imports secured, and market contacts broadened; without these inputs, growth would be likely to slow to about 20%. These inputs would be likely to generate incremental foreign exchange earnings of about Rs 160 million (US$12.1 million) over four years, with incremental local value added of about US$7 million equivalent (Annex 1, Tables 4 and 5). Since about 90% of the existing carpet loom capacity is utilized, additional looms would be required to meet the production increase of about 35,000 meters annually. With each meter requiring about 23 days of spinning, weaving and finishing, a 35% annual expansion represents about 12,000 additional man years by the end of 1984 assuming a 200 day working year. About 40% of the workers required are unskilled, for carding, spinning, and finishing The high value added of carpets makes it possible for rural units to be competitive; transport of wool to and finished carpets from weaving concentrations in Gandaki adds only Rs 21 to the costs of a Rs 1,200 carpet, more than outweighed by lower labor costs in the hills. 2/ In Gandaki, 1/ While carpet production began with Tibetan refugees, the success has resulted in spreading of production to Nepalese units. Now, Nepalese workers account for over 70% of production of carpets. 2/ Daily earnings average Rs 13 in Kathmandu Valley; with wages representing about 54% of production costs, rural wages would need to be 3,/ lower to make products competitive.

16 about 1,500 of the weaving families are located in 12 rural clusters; these families have skills in spinning and weaving from making blankets on traditional looms. With introduction of carpet looms and adaptation of skills from rugs to Tibetan carpets, several of these concentrations could be mobilized for export markets. With initial training and organization of input supply, orders and quality control, Gandaki could contribute about 20-30% to Nepal's growth in carpet exports over the next three years, providing additional earnings of up to Rs 2,300 (US$175) annually to rural weaver families who now live below subsistence levels The other product lines with strong export prospects, particularly in US markets, are woolen jackets and hand knits. The current trend in western markets is for products designed specifically to suit US and European fashions rather than for ethnic outerwear. To meet competition, Nepalese exporters would need close relationships with buyers and designers to respond quickly to fashion trends. While prices are competitive and buyers appreciate the quality of wool from Nepal, improvements in sizing and finish are needed The attraction of expanding jacket and knitwear production in the hills is that existing equipment and methods would be used; piece rate wages are attractive, yielding about Rs 10 per day for knitters, Rs 12 for weavers, and Rs 25 for tailors. If the necessary improvements in design and marketing are made, traditional household production methods could yield a competitive product. Given the small present export base, project inputs could generate a 30% annual increase in exports of this product, lifting exports from about Rs 5 million in 1978/79 to about Rs 15 million by 1984, creating about 5,000 additional man years of employment, with annual incremental earnings averaging Rs 860 (US$65) per worker. 1/ In the absence of specific project inputs, export earnings probably would not grow by more than 10% annually, since the rough quality of the products available has limited appeal in export markets (Annex 1, Tables 4 and 5). While knitting and weaving operations could be performed in traditional craft concentrations in the hills, tailoring and ready-made garment operations are mainly in Kathmandu; skills and economies of scale in garment production would make it advantageous to weave woolen yardage for jackets in the hills but do sewing operations in Kathmandu Needs and Strategy. Major constraints in expanding carpet exports are wool supply and production capacity. Insufficient imports of Bhanglung wool have caused price increases, 2/ and inadequacy of supplies have limited output, foreign exchange earnings and ability to meet orders. At present, National Training Ltd. imports only about 60 tons of Bhanglung wool from China under a barter agreement; prices are high and operations inefficient. The 1/ Assuming a 100 day work year which is reasonable, given seasonality of demand for woolen garments. 2/ Wool prices doubled from Rs 22 to Rs 42 per kilogram from 1978 to 1979; with about 6 kg of wool required per meter of carpet, this increased production costs by about Rs 120 per meter, reducing profitability from about 40% to about 30%.

17 remaining 340 tons are imported, mainly from Tibet, often unofficially, by private traders. Although Tibet produces large quantities of Bhanglung wool, China is expanding its own carpet exports, aided by its Most Favored Nation status. Other sources also may have to be tapped; New Zealand appears promising, since it produces a variety similar to Bhanglung wool at comparable prices. No public sector organization is judged sufficiently competent to arrange bulk wool imports; a private company, structured to assure fair prices to the producers, is being formed to help ensure stable availability of wool for the carpet and garment industry. About Rs 2.5 million in financing would be required to launch this company, enabling it to purchase in bulk and hold the required stocks of wools and dyes (paras ) Another problem is the high cost of air freighting carpets; due to unreliability and delays in sea shipment from Calcutta, Nepalese exporters air freight carpets, adding 20% to the CIF cost of a carpet shipped to Europe and 40% to the US versus 5% and 10% for sea shipment. With air freight, goods are received in one to two weeks vs. one to three months by sea. Advantages of air could be reduced by cutting manufacturing and handling time and by air freighting to Colombo, Chittagong or Bangkok and sending goods by sea from there (para 2.18) Modifications in design and sizing, coupled with improved market links, are prerequisites for sizeable market penetration in woven or knitted woolen garments. Production costs would be reduced in both carpets and garments by introducing some simple equipment at a few key points in the process (e.g., carding and milling) and by improving speed and quality through short upgrading courses and in-house extension services. Credit requirements of the project for units making carpets and other woolen products are estimated to be Rs 11.0 million from 1981 to Most CSI subborrowers are expected to be household units employing an average of two family members; however, about 20%o of the loans to wool units are expected to be to units employing about 10 workers (Annex 1, Table 6). Exporters who supply raw materials, organize decentralized producers and market products would need about Rs 7.0 million in term credit. The raw material supply company importing wool yarn would need about Rs 3-4 million in credit over the commitment period (para 4.42). B. Cotton Handloom Products 3.10 Characteristics. The project area contains about 20,500 handloom units employing 25,500; about 85% of the looms are in clusters in Kathmandu Valley with the remaining 15% in Gandaki. Fixed investment per job averages Rs 690 and annual output per worker, Rs 2,900. The product range is extremely limited; most units produce simple shirting, saree material and bedcoverings. Prices of Nepalese handloom products are less than Indian imports by 5-10% in simple shirting and bedcover material but more for sarees and towelling; productivity on flyshuttle looms is comparable, averaging one meter per hour for simple weaves About 44% of the looms are traditional waist looms, 55% are flyshuttle hlandlooms, and 1% are pedal looms. Waist looms, used mainly in rural

18 Gandaki, are only 20% as efficient as fly shuttle looms, and the narrow loom width limits product possibilities. Pedal looms, which do shedding, picking and beating automatically, are still more efficient with one-shift capabilities of 15 meters vs. 6 to 8 meters on fly shuttle looms. However, the higher costs of pedal looms are warranted only for simple weaves which do not require frequent stoppages for design formation. Scarcity of power, market uncertainty, and the desirability of focussing on products with higher value added make the Rs 1,000 flyshuttle loom a viable alternative to the Rs 4,500 pedal loom In rural Gandaki, weavers in household units work only about 30 days a year, due to inadequate yarn supply, local marketing difficulties, inferiority of waistloom technology, and requirements from agriculture during one-third of the year. In Kathmandu Valley most weavers work about 180 days and in concentrations such as Kirtipur are employed for up to 270 days a year; however, daily earnings often are low with weavers engaged in less profitable lines for the low end of the local market. Kathmandu Valley has a stronger handloom tradition, easier access to yarn and markets, and produces better quality; however, productivity on the flyshuttle loom is similar in Gandaki and Bagmati Zones. With improved equipment and methods, rural Gandaki could be competitive, since transport costs to bring yarn in and take cloth out of rural clusters would add only about 0.3% to 4.0% to the cost of a meter, which would be outweighed by lower rural wages Local Market Potential. Nepal produced only 15% of the cloth which it purchased in 1979/80 1/, the remainder having been imported. Some imports were in synthetics; however, a sizeable portion was simple Indian handloom sarees, shirting, bed sheets, furnishing fabrics and towels. With labor representing about 40% of production costs at current Nepali wage rates, it is worthwhile for Nepal to buy yarn and gain the value added and employment. If Nepal could increase import substitution of cotton cloth by 10% per year, an additional 3,300 people would be employed annually. 2/ The most promising products are shirting, dhakka cloth, and furnishing fabric. Daily earnings average Rs 10 in these lines. In Gandaki, where weavers need to upgrade looms and methods to achieve competitive price and quality, institutional demand can be tapped. Government departments, hospitals, schools and private hotels import at least Rs 20 million annually in bed sheets, uniforms and furnishing fabrics; hill weavers can gain efficiency and steady earnings by producing standard products, gradually moving into more competitive consumer markets Export Potential. India exported US$77 million in handloom cloth in 1979; Nepal exported about US$150,000. In the same year, India exported US$450 million in garments, over 50% being of handloom cloth; Nepal's garment exports in 1979 totalled US$666,000. With the quotas affecting Indian and other exporters in US and European markets, there is scope for Nepalese 1/ The project area supplies about 7% of this effective demand. 2/ Assuming six meters per day per worker with a 200 day working year, which is reasonable given seasonal labor requirements from agriculture; if the working year were 150 days, the number employed would be 4,000.

19 exports, particularly if foreign collaboration can be arranged. 1/ The most likely areas for export expansion are shirts, casual wear, furnishing fabric, and table linens. In the short term, only a small portion of Gandaki's looms could be geared for export, making bolts of cloth with finishing done in Kathmandu; however, the thousands of skilled weavers in Kirtipur, Luvu and Bhaktapur of Kathmandu Valley could be organized for export-oriented handloom cloth production if marginal improvements in looms, methods, and designs are made and if production from these concentrations is organized Needs and Strategy. Handloom units, not in a position to import yarn in bulk, need to rely on traders. Fragmented purchasing precludes volume discounts and systematic yarn supply arrangements; during periods of scarcily, prices increase by 20-30% and the yarn available often is not of the varieties in demand. The nature of producers and traders in cotton handloom products and the limited profitability of the industry at present make it unlikely that the private sector will solve the cotton yarn supply problem. Services by the Emporium need to be improved significantly with better purchasing, inventory control and personnel management; bulk procurement from Indian and Chinese mills would reduce yarn prices by 5-10% In addition to yarn supply constraints, the narrow range of Nepali handloom products limits their competitiveness with Indian varieties in local and export markets. Assistance is needed in product adaptation and design development, in response to concrete market opportunities (para 4.27 and 4.37). Credit to upgrade equipment and finance working capital requirements is needed; about Rs 5.8 million in credit for the CSI handloom units and about Rs 6.0 million in term credit for the organizers would be needed from CY (Annex 1, Table 6). The Emporium would need about Rs 5.0 million in credit over the three year commitment period for import and distribution of cotton yarn and handloom accessories and for wholesale operations to meet public sector cotton handloom needs (para 4.44). C. Metal Crafts 3.17 Characteristics. The main center of metal crafts is Patan City in the Kathmandu Valley, which accounts for about 90% of total production (Annex 3, Map 1). Metal curios and utensils represent a small but highly profitable product group. About 340 units employ 1,000 workers who produce about Rs 77 million worth of artistic and utility items; metal craft exports reached Rs 37 million in 1978/79, constituting about 35% of cottage industry exports; annual wages average about Rs 3,600. The existing products include artistic statues and ritual objects, filigree art works and domestic metal utensils. Production technology is traditional: lost wax casting is used in manufacturing metal statues and bronze utensils, and simple beating and soldering is employed for copper and brass utensils and studded filigree artwork. 1/ It is unlikely that Nepal would face garment import quotas until annual exports exceed US$30 million, as was the case in Sri Lanka. Furthermore, garments of handloom cloth are in a category which is less sensitive to protectionism in Western markets.

20 Potential. While export growth has been rapid during the past five years, maintenance of this pace will require product diversification, increased metal supplies, and selective improvements in technology to compete with India and other metal craft exporters. Some market scope exists for religious brass and bronze work, particularly to Europe and Asia, both as art objects and lower priced curios. Growth prospects are good in filigree work and for hammered copper and brass culinary ware, with some design modification. If production is expanded, raw materials are secured, and design modifications responding to market requirements are made, exports of metal crafts could provide incremental foreign exchange earnings of Rs 145 million (US$11 million) over the period, relative to export earnings without project inputs (Annex 1, Table 4). Since most metal craft units are in Lalitpur, credit, technical and raw material services could be provided with relative ease and efficiency Needs and Strategy. The sporadic supply and high price of metal and other inputs are major constraints; the need for volume imports of metal is increasing with local scrap becoming scarce. At present, craftsmen depend upon private traders who often charge three times their import prices, particularly in periods of scarcity. These importers pay import duties and sales taxes; craftsmen, who are exempt from both, usually are too small to import directly. Members of the Nepal Handicrafts Association are establishing a private limited company to import required raw materials in bulk for supply to manufacturers at reasonable prices (para ). Under the Industrial Enterprises Act, HMG has made provisions to exempt such companies from import duties and sales taxes (para 2.16). With bulk purchasing and reasonable profits to such a company, metal prices to CSIs could be reduced by up to 200%. Metal craft producers also have difficulty obtaining credit. Finished metal handicrafts are not accepted by the banks as collateral since value on these items is difficult to assess; no pre-export financing is available to execute orders against letters of credit. About Rs 5 million in term credit would be required in the project area for fixed investment and permanent working capital in the commitment period, for metal craft units, related agents and the input company. Training and consultancy on selective improvements in traditional methods, tools and designs would be provided (para 4.36), and exporters would be assisted in product modification, quality control, and increased export contacts (para 4.28). D. Forest Based Products 3.20 Characteristics. About 125,000 people make forest-based products in the project area; most work during slack periods of the agricultural cycle. Raw materials used are indigenous bamboo, reeds, grass and timber. Main items are utility containers and floor matting. Most furniture making and wood carving skills are found in Kathmandu Valley. Production of bamboo products and reedware is concentrated in Gandaki Zone (Map 2); fixed investments in these lines average Rs 100, generally in very simple tools. Annual output value of forest based products in the project area is an estimated Rs 20 million.

21 Potential. Raw material available would be sufficient to support even a threefold expansion in production. Sizeable producer concentrations exist which would facilitate cost effective services and marketing. In Gandaki, 7,500 units in 9 clusters are within 20 miles of district headquarters with road connections to Kathmandu; Kathmandu Valley holds about 3,000 units in 13 clusters. The existence of large producer concentrations, the ample supply of local raw materials, and the extremely low fixed investments make bamboo, reed and straw products an attractive area for expansion. However, local markets are saturated. If improvements are made in sizing and workmanship, tourist markets for traditional products could be tapped. There is some scope for exports of basketry and knocked-down furniture. In both U.S. and European markets, there is always demand for a small quantity of "typical" baskets to provide variety. However, to realize sizeable growth, Nepal would need to be competitive with the Philippines and China, the main new entrant; attractive air freight rates and more efficient sea shipping arrangements would be necessary for volume exports of baskets. Packable designs are needed. Thus, bamboo and reed items, while appealing from an employment viewpoint, are likely to grow slowly relative to the other major product groups. Sizeable immediate export potential exists for expanded wood carving in Kathmandu Valley Needs and Strategy. Production is constrained by: lack of commercial orientation and resistance to change among craftsmen who now produce bamboo and reed goods mainly for household needs during free time; use of traditional tools and methods which often limit productivity and quality output; lack of effective credit or marketing services; and the need to make improvements in workmanship, seasoning, sizing and design of existing products. Services of public institutions would be improved and private agents motivated to address these needs (paras 4.32 and 4.37). Credit requirements in the project area, mainly for working capital, have been estimated at Rs 2 million over the three year period (Annex 1, Table 6). With improvements in methods and tools and adequate credit during commitment period, production value from forest-based products could increase by roughly Rs 3 million over what it would have been without the project; additional employment by the end of 1984 would be about 144,000 mandays. E. Selected Agroindustries 3.23 Characteristics. The project area contains about 103,000 agrobased units employing 140,000, processing about Rs 181 million 1/ worth of goods, with fixed investment of about Rs 18 million. The food subsector includes a wide range of activities. Rice beating, paddy pounding, ghee and khuwa making, and beekeeping are rural household activities pursued on a seasonal, part time basis to supply household needs and provide marginal income supplements; these activities are dispersed throughout rural Gandaki. 1/ This includes about Rs 118 million in service units milling rice, wheat and oil. The production figures in Annex 1, Table 1 do not include these service units.

22 Commercial units, concentrated in Kathmandu Valley and towns of Gandaki, include milling, bakeries, and units making noodles and sweets. Milling operations often are of the integrated type, encompassing rice milling, flour making and oil expelling. If the unit is located properly, it gets business throughout the year, after harvesting of paddy, maize, wheat and mustard seed Potential. Intensive agricultural development efforts, planned for Gandaki Zone, are expected to result in increased output of agricultural commodities, which would create requirements for milling operations and other agro-processing facilities. Also, there are lines in which the unit itself "makes" the raw materials; beekeeping, such an activity, is simple, requires limited capital investment in improved hives and honey extractors, and demands little time; parts of Syangya, Gorkha, and Lamjung Districts of Gandaki have vegetation suitable for beekeeping. Potential for adding value to existing agricultural commodities also exists in the project area; fruits, herbs and ginger available in Gandaki could be dried; some of the estimated 50% of household's surplus milk can be processed into cheese; and processing costs of milling operations can be reduced through improved water wheels and upgraded metal grinders. Honey, marmalade, dried fruits, and dairy products have strong demand from trekkers and hotels, with prospects for import substitution. Some relatively organized small units have been established, mainly in Kathmandu Valley, to provide jams, juices, and butter and other processed foods for the tourist markets; with improved quality control and some modernization, some of these units could produce competitive products. Prospects for effective institutional assistance to household agro-enterprises are more limited since many "units" pursue these activities mainly to meet household requirements; local rural purchasing power is limited; and the dispersal of these units hinders commercial organization Needs and Strategy. Since most agroindustries process locally available inputs for regional markets, commercial service needs would be limited, particularly for standard operations such as grain and oil milling. For honey, dairy, fruit and herb products catering to tourist or export markets, services to improve quality control and commercial contacts would be required. ADBN has done detailed preparation for agroindustrial projects, particularly beekeeping; ADBN provides some commercial and technical services in addition to credit. The CIDB would have a small unit to provide technical services to agroindustries. While at the outset, the primary market for processed foods would be local and tourist markets, the Trade Promotion Centre plans to include honey, cheese, dried ginger and herbs in its promotion of forestry-based products. Credit requirements in the project area for food processing and other agroindustrial CSIs are estimated to be Rs million 1/ over the three year 1/ The Asian Development Bank (ADB) and IFAD are financing two agricultural projects, with small agroindustrial components, through ADBN. However, overlap between the proposed IDA Credit and theqe projects is minimal: ADBN plans only about 30 milling operations, us kyd'opov-er. in the project area and the cottage industry component o. the IFai project would not be executed until CY83, to provide time to learn from experience under the IDA-financed CSI project.

23 commitment period. ADBN projects about Rs 3 million for beekeeping operations including hives, extractors, packing facilities and working capital. Average subproject costs for agroindustrial CSIs, including fixed investments and permanent working capital, are estimated to be Rs 30,000 for grain and oil milling, Rs 65,000 for beaten rice, Rs 52,000 for bakeries, Rs 85,000 for noodle making, Rs 25,000 for confectionary operations, and Rs 20,000 to Rs 130,000 for solar drying of ginger, fruits and vegetables, depending upon the scale of the enterprise. 1/ In addition to credit, for those more organized CSIs, the project would provide finance for viable household agroenterprises; ADBN anticipates lending to beekeeping cooperatives for which each household enterprise would require Rs 500-Rs 3,000 depending upon the number of hives. IV. THE PROPOSED PROJECT A. Objectives, Strategy and Institutional Arrangements 4.01 Objectives. The proposed project is designed to help achieve HMG's objectives of increasing earnings and generating foreign exchange through expanded CSI activity by: (a) building programs and capabilities of key commercial, public and banking institutions to provide effective services to viable CSIs; (b) implementing innovative, product-specific schemes in Bagmati and Gandaki Zones 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 would be to tap marketing knowhow and organizing skills of private exporters and other commercial agents in providing training and commerical services to cottage industries Scope. The first project would focus on Kathmandu Valley and the Gandaki Zone, which account for over half of the cottage industry output in Nepal (Map 1). The geographical scope could be broadened in subsequent projects, once services are strengthened and schemes tested. Since Kathmandu Valley holds the headquarters of banking, public service and private commercial agencies, efforts to upgrade institutions would be facilitated. Gandaki Zone typifies the rural hill areas but has relatively large craft clusters and road links with Kathmandu. Commercial and technical service components would focus on product groups which have: strong growth potential once methods and services are upgraded; artisan concentrations to enable efficient services; and high present or potential value added. Carpets, woolen and cotton handloom garments, metal crafts, forest products and selected agroindustries would be the major groups. 1/ The average size of agroindustries under the proposed project is expected to be higher than for existing operations, due to inflation and the emphasis on commercial operations.

24 Institutional Arrangements. The proposed project would: build capabilities of the banks for CSI credit; mobilize private organizations for most commercial functions; and help public sector organizations to focus on promotion, extension, and filling some gaps in private commercial services. Strengthening of institutional programs and staff training would be major elements. The two commercial banks and ADBN would be responsible for term credit to CSIs, market agents involved directly in cottage industry production, input supply companies for metal and wool, and the Emporium. The Cottage Industry Export Development Division of TPC would assist exporters in improving products, output and exports; TPC would engage selected exporters as "performance contractors" to extend their service networks to selected rural cottage industry clusters with reasonable access to roads. The CIDB would provide extension and design facilities in Kathmandu Valley and commercial services in more remote areas of Gandaki Zone. The Emporium would concentrate on bulk imports of cotton yarn and handloom equipment, and on institutional marketing. The Industrial Services Centre would be responsible for monitoring and evaluation, and for preparation of subsequent projects, building on the experience gained in the Kathmandu Valley and Gandaki Zone. The Ministry of Industry would provide project coordination. These institutional arrangements are consistent with those outlined for all of Nepal under the Sixth Five Year Plan; by providing sizeable training and consultancy inputs under this first project, private, public and banking capabilities would be built, enabling more rapid development of the CSI sector in the future Components. A cottage and small industries project of about US$12.0 million equivalent is recommended with IDA contributing about US$6.5 million and UNDP about US$2.0 million. Under the subloan component, the two local commercial banks (Nepal Bank Ltd. and Rastriya Banijya Bank) and the Agricultural Development Bank would be responsible for term lending to CSIs, related market agents, and input supply companies. These lending activities would be encouraged and strengthened by: (a) supplying US$4.5 million equivalent to help fill the gap between term lending requirements of CSIs and related commercial agents in the project area and credit resources available for this sector; (b) establishing a CSI Refinance Unit within Nepal Rastra Bank to do rapid subproject review and provide partial refinance of eligible subloans made by the participating banks; and (c) encouraging the banks to increase lending to viable CSIs and related commercial concerns by providing the banks with attractive spreads, effective guarantees, and trained CSI officers at headquarters and by key branches. Under the commercial and technical service components, private and public services to CSIs in key product groups would be improved by:

25 (a) supporting the Cottage Industry Export and Product Development (CIED) Division within TPC to assist exporters with market contacts, product development, and strengthening of the supply base. Product specialists, sales and exposure trips, performance contracts, and two foreign promotional offices would be financed; (b) reactivating the CI Board to assume responsibility for public development services to CSIs, with product specific extension services in Kathmandu Valley, an improved Handicraft Design and Promotion Centre, and integrated services to selected cottage industry clusters in areas of Gandaki Zone which are not covered by private commercial agents; (c) strengthening the Emporium's handloom operations by providing an advisor and seed capital for procurement and distribution of cotton yarn, handloom accessories, and textiles in the project area; (d) providing advisory services to the Rastra Bank's CSI Refinance Unit and training for CSI staff of NRB and the participating banks; and (e) funding ISC staff and consultancy for monitoring, and evaluation and preparation of subsequent projects Financing Plan. Total costs for the CSI project would be about US$12.0 million equivalent of which IDA would finance US$6.5 million or 54%, and UNDP 17%. For the subloan component, the IDA US$4.5 million would cover 100% of the portion refinanced by the Nepal Rastra Bank, equivalent to 80% of the subloan amounts made by the participating credit institutions, and averaging 64% of subproject costs. Under the commercial and technical service elements, IDA would provide US$2.0 million to cover the costs of the IMPACT centres, performance contract payments, 50% of incremental staffing costs and related expenditures, equipment for subsector development centres, vehicles, and seed capital for the Emporium (Annex 1, Table 8). The commercial and technical service component also would include about US$2.0 million in UWIDP grant funds for training and consultancy (Annex 1, Table 9); HMG would contribute Rs 6.6 million (US$500,000) to cover half of incremental staff and related expenditures, as well as land and buildings for these components; Rastra Bank would contribute Rs 5 million (US$379,000) to the CSI Refinance Fund, mainly to cover the lag between refinance by NRB and its withdrawals from IDA. H1MG would provide 20% of the CSI subloan amounts made by ADBN which are not eligible for IDA disbursements, since ADBN does not have sufficient deposits (para 4.22); these requirements are estimated to be Rs 6.6 million (US$500,000) over the disbursement period. During negotiations, agreement was reached that HMG and Rastra Bank would make the necessary annual allocations to execute the project. The following table summarizes project costs and sources of finance:

26 Project Financing Plan (in US$ million) Subloan Component (a) Term Credit for HMG/NRB/ IDA UNDP Agencies Banks CSIs Total CSIs and Agents (b) NRB - Administrative and Revolving Fund Subtotal, Subloan Component Commercial and Technical Services (a) TPC - Export Development (b) CIDB (c) Emporium (d) ISC (e) Training and Consultancy for NRB and Banks (f) Unallocated and other Subtotal, Services TOTAL B. Subloan Component 1. CSI Refinance Fund - Rastra Bank 4.06 The Nepal Rastra Bank (NRB) 1/ recognizes that target setting for CSI lending has not been effective, and that to develop sound CSI lending operations in the banks, specialized staff, attractive spreads, effective guarantees, and adequate refinance are needed. The proposed project would help meet these requirements. A CSI Refinance Fund, with its own policies, procedures, staff, and financial resources, is being established in NRB, to channel IDA funds. Staff of the Fund would: review appraisal reports and refinance CSI subloans submitted by participating banks; monitor appraisal, supervision and collection standards; collect repayments of refinanced amounts; and report regularly to the CSI Coordinating Committee and IDA Organization and Staffing. The CSI Refinance Unit to administer the Fund would be within the Banking and Credit Division of the Nepal Rastra Bank. The CSI Fund would be headed by a manager, assisted by an outside advisor financed under the project. Other professional staff would total 1/ Nepal Rastra Bank, the Central Bank, was established and operates pursuant to the Nepal Rastra Bank Act of 1955.

27 five at the outset 1/: two for analysis and review; one to administer the credit guarantee scheme; one for monitoring and reporting; and one for accounting and disbursements. Four of these officers already have received intensive training in CSI lending and refinance operations, in India and the Philippines, financed under the UNDP-financed, Bank-executed preparatory project; training of the remaining officers has been arranged for November-December 1981 (para 4.51). 2/ The manager and professional staff appointed to the Fund would have qualifications, training, and experience acceptable to IDA. The manager would need to be experienced in industrial appraisal, management, and development of procedures for project review and supervision. For the other officers, experience in industrial refinance and bank supervision would be satisfactory. Prior to credit effectiveness, management and staff for the CSI Fund, acceptable to IDA, would be in place Policies and Procedures. A draft statement of policies and operating procedures, which would provide a satisfactory basis for the CSI Fund's operations, has been prepared and discussed during appraisal; the Board of Directors has agreed with the substance of the Statement (Annex 2). Prior to credit effectiveness, Rastra Bank would need to adopt such a statement and accounting procedures satisfactory to IDA; any subsequent modifications would require IDA agreement. The statement would outline procedures for: review of subprojects; monitoring of bank activities; and methods and conditions of refinance. For subloans of over Rs 20,000 (US$1,500), the Fund would review appraisals prepared by the banks prior to granting refinance; for smaller subloans, refinance would be granted on the basis of the participating bank's appraisal showing the subproject to be eligible, financially viable and having appropriate lending terms. However, a 20% random sample of these smaller subloans would be subject to a more detailed post-approval review of technical, financial and marketing aspects. To ensure expeditious review of subloan applications of above Rs 20,000, the Fund would need to complete its procedures within three weeks of receiving the application. The review procedures are designed to ensure that about 30% by number and 86% by amount of subloans receive preapproval review by the CSI Fund (Annex 1, Table 11). Appraisal standards would be checked through this process; the CSI Fund would monitor the banks' supervision standards through review of follow-up reports and field visits to a sample of subprojects. Banks would submit quarterly reports to the Fund for review, aggregation and submission to IDA and the CSI Coordinating Committee (para 4.48). 1/ If the volume of work justifies an increase in this number, expansion of staff would take place during project implementation. 2/ In addition, two staff members of NRB's training institute have completed training with the State Bank of India, to enable NRB to provide continuous local training on CSI appraisal and supervision for officers of the banks.

28 Eligibility, Terms and Conditions 4.09 Eligibility for Refinance. Three banks are expected to participate in the CSI project: the two commercial banks, Nepal Bank Ltd. and Rastriya Banijya Bank, and the Agricultural Development Bank. Other financial institutions also would be eligible if HMG, IDA and NRB agree that they have sound overall credit operations and have complied with the conditions of participation in the project. To be eligible for refinance, each participating bank would need to have complied with minimum staffing and training requirements and have adopted standards and procedures acceptable to IDA (para 4.16). Enterprises eligible for subloans under the project would be: manufacturing and marketing concerns engaged directly in the production of cottage industry products; and CSIs with fixed investment, after the subloan, of up to Rs 800,000 (US$60,600). Only CSIs with fixed costs per job not exceeding Rs 12,000 (US$900) after the subloan would be eligible for credit. Eligible investment enterprises could be owned by private individuals, partnerships, incorporated bodies, or cooperatives. The maximum subloan size would be Rs 800,000, 1/ including fixed investment and permanent working capital financing. Subloans would be to help finance foreign and local costs of fixed assets and permanent working capital 2/ in either new or expansion subprojects. The credit would not be allocated among the participating banks. Exporter organizations engaged by the Trade Promotion Centre as "performance contractors" in Gandaki Zone would be eligible, provided the credit institution's standards of viability are met. 3/ Refinance to projects of market agents for production facilities located in Kathmandu Valley would be limited to no more than 30% of the total refinance granted under the project In most cases, banks would provide no more than 80% of the subproject cost, leaving 20% to be financed by sponsors. Exceptions could be made for existing artisans with established markets requiring subproject amounts of under Rs 5,000 (US$380); for these, the banks could recommend waiving the 20% minimum and provide up to 100% of the total subproject amount. 4/ In all cases, the CSI Fund would refinance a maximum of 80% of the bank's subloan amount. Each bank would assume full repayment risks on refinance; amounts due would be repayable to the Fund on a fixed schedule, corresponding to the 1/ Except in the case of the raw material supply companies and the Emporium (paras 4.42 and 4.45). 2/ This would not include fluctuating or seasonal working capital, which would be financed through separate loans by the commercial banks. 3/ "Performance contractors" (para 4.33) would be eligible for a second loan; normally, these second loans would be made only after the performance contractor had demonstrated regular repayment for one year after the grace period had expired. 4/ For exporters serving as performance contractors, TPC would pay up to 20% of subproject costs and the borrower would contribute at least 10% as equity; commercial banks would provide the remainder provided that the subloan portion does not exceed 80% of total subproject costs.

29 subloan amortization. IDA would reimburse the CSI Fund of NRB for 100% of refinanced amounts for eligible subloans, on the basis of monthly or quarterly requests Terms and Conditions. The initial lending rate by the participating banks to eligible subborrowers would be 11%, consistent with minimum commercial rates for term loans to industry in Nepal. The onlending rates and spreads would be reviewed at least annually with adjustments made as necessary to ensure that: (a) CSI term lending rates remain consistent with rates for term loans to all industry; (b) these rates are positive in relation to medium terms inflation projections; and (c) spreads are adequate to cover the banks' administrative costs and to induce them to expand CSI lending. The CSI lending rate established for the project would be applied to all CSI loans made by the participating credit institutions, except in a few areas defined as remote, for which the Industrial Enterprises Act specifies that rates of 2% less than the prevailing rate would be applied. The list of remote areas would need to be satisfactory to IDA, to ensure that only mountainous areas not served by substantial cottage industry development programs are included as remote areas. Maturities on subloans refinanced under the project would be from 18 months to 7 years, including grace periods ranging from 3 months to 2 years; banks would determine appropriate maturities and grace periods for each subloan based on the subproject's debt service ability. To help ensure credit access by smaller enterprises, the refinance rate would be adjusted to provide banks with a spread of 6% for subloans of up to Rs 20,000 and 4% for subloans of over Rs 20,000. Larger spreads for smaller subloans would compensate the banks for the higher administrative costs as a percentage of subloan size CSI Fund Resources. The CSI Fund's resources would comprise the US$4.5 million IDA credit component plus an initial contribution by NRB of at least Rs 5 million (US$379,000). A small portion of the Rs 5 million would be used to meet the Fund's initial operating costs; the majority would provide a revolving fund to cover the time lag between the Fund's refinance of the banks' subloans and its withdrawals from IDA. Rastra Bank would provide Rs 5 million prior to credit effectiveness and, thereafter, make available additional funds as needed to operate the CSI Fund. The Fund's spread would be set at 2% p.a., which is appropriate in view of the anticipated costs and the initial contribution. It is expected that after three years of operation, the Fund would begin accumulating operating cash surpluses; initially, the low level of average outstandings would result in operating cash losses, partially offset by income from NDB's investment contribution (Annex 1, Table 10) Institutional and Training Needs. Prior to credit effectiveness, the CSI Refinance Fund would need to have been established in a manner acceptable to IDA, including: suitably qualified management and staff (para 4.07); an approved statement of policies, operating procedures, and accounting practices (para 4.08); and Rs 5 million provided by NRB as initial capitalization of the CSI Refinance Fund. Outside training in CSI lending operations, financed under the UNDP preproject facility, has been completed for the review officers and bank liaison officer of the CSI Fund. A suitable outside advisor, with experience in CSI lending and refinance

30 operations, would be recruited to assist the manager of the Fund. This post, financed by UNDP under the project, would be for 24 man-months (US$170,000). 3. Credit Guarantee Scheme 4.14 To encourage the banks to relax their collateral requirements for subloans refinanced by the Fund, NRB would establish a CSI credit guarantee scheme. The NRB guarantee scheme, to be available to the participating banks only for CSI subloans refinanced by NRB under the project, would be administered by the staff of the CSI Fund. Coverage would be provided for up to 75% of the subloan amount and a premium of 1% p.a. on the guaranteed amount outstanding would be charged; this 1% premium would be added to NRB's 2% spread. During the initial three years, participation in the scheme would be mandatory for all subloans refinanced under the project and would be automatic once refinance is granted from NRB. The scheme would have reasonably liberal claim procedures; the banks would be able to claim once their auditors certify a debt as bad or irrecoverable. HMG would pay into the scheme an initial amount of Rs 2 million and would pay in further amounts on an annual basis to maintain the capitalization of the scheme at no less than 10% of the guarantees outstanding. Suggested operating procedures were outlined during appraisal and discussed during negotiations, with agreement in substance reached (Annex 2). Prior to credit effectiveness, the credit guarantee scheme would have to be established with policies and procedures satisfactory to IDA, and the initial capital contribution paid At present, commercial bank loans to the priority sector are covered by the Credit Guarantee Corporation Private Limited (CGC), in which the Rastra Bank holds 50% of the shares and the two banks, 25% each. However, this company, which relies entirely on share capital and an inadequate guarantee premium for resources, is seriously undercapitalized. Furthermore, the banks bear 50% of the guarantee schemes' credit risk on their own loans. To ensure that subloans under the proposed project have adequate guarantee coverage, a separate scheme would be established in NRB; once the CGC is suitably strengthened and is in a position to undertake further guarantee operations, the scheme could be transferred, if mutually agreed by HMG, Rastra Bank, and IDA. During negotiations, agreement was reached on minimum capitalization, equity structure, premium level and procedural improvements which would need to take place prior to transferring to the CGC the guarantee function for CSI subloans refinanced under the proposed project; these conditions were outlined in a side letter. 4. The Participating Banks 4.16 NRB would issue a set of operating instructions to each of the three banks; these instructions would outline procedures and respective obligations of the CSI Fund and the banks under the project. 1/ NRB's instructions I/ These instructions would take the place of participation agreements between the refinance institution and the banks. Management of Nepal Rastra Bank is opposed to participation agreements since NRB's position vis a vis the banks makes instructions binding. While not ideal, these instructions would be adequate, provided that the contents were acceptable to the banks and IDA.

31 would stipulate each bank's duty to: (a) assign and train an agreed minimum number of qualified CSI staff for head office units and key branches; (b) prepare appraisals of a satisfactory standard for each subloan; (c) supervise subprojects to ensure proper utilization of funds and monitor progress; (d) adhere to the terms of lending and repayment of refinance; and (e) submit quarterly reports on activities, subproject portfolio and arrears in a format prescribed by the CSI Fund, with all satisfactory to IDA. The operating procedures also would emphasize means to address two major lending problems faced by banks in Nepal: diversion of funds from the stated purpose and difficulties in enforcing repayments by instalment. Measures to control abuses, some of which are already in practice in Nepal, include: direct payments to suppliers; disbursements in kind; a system of coupons redeemable against approved raw material supplies; and partial disbursements, closely controlled by frequent supervision visits. The refinance terms would require repayments by instalment and banks would be expected to exert all possible pressure on subborrowers to honor this arrangement. The principal features of the operating instructions were discussed with the management of the Rastra Bank and the banks during appraisal (Annex 2); agreement was reached on the substance of the instructions during negotiations. Prior to credit effectiveness, NRB would have issued instructions, satisfactory to IDA to at least two of the participating banks, with these banks having complied with the agreed minimum staffing and training requirements; the banks would need to confirm that necessary actions had been taken and policies confirmed to implement NRB's instructions The Commercial Banks. The larger and more established of the two commercial banks is Nepal Bank Ltd. (NBL), which is 51% owned by HMG and 49% held by private Nepali citizens. It accounts for approximately 65% of deposits and 61% of loans and advances of the commercial banking system; Rastriya Banijya Bank (RBB), which is wholly owned by HMG, consitutes the balance. Both banks have extensive branch networks; NBL has a total of five regional offices and 133 branches, while RBB has four regional offices and 98 branches. The majority of commercial bank loans are short term, 1/ usually without rollovers. Collateral is the basis for most lending decisions. Project appraisal capabilities are limited due to the shortage of suitably trained staff. Both banks are instructed to lend at least 7% of their deposit liabilities to "priority sector clients. However, the total advances to this sector constitute less than 6% of deposits. 2/ The proposed project would assist the banks in developing project-based lending by ensuring that trained CSI officers were assigned to headquarters and key branches in the project area and by helping to establish sound appraisal, disbursement, supervision and collection practices for CSI lending operations. Also, since most CSI subloans would be made to units in identified concentrations of activity in Gandaki Zone and Kathmandu Valley, logistical problems and administrative costs of appraisal and supervision would be reduced. 1/ Usually from 3 months to 2 years; maturities rarely exceed 3 years. 2/ Including indirect financing by the purchase of bonds and loans to the Agricultural Development Bank. Excluding this, advances to the sector were 4.8% of deposits.

32 Nepal Bank Limited. NBL has 24 offices in Gandaki and 22 in the Kathmandu Valley. Total lending to the priority sector has totalled Rs 24.9 million, of which Rs 11.1 million (45%) was made to 596 clients in the cottage industry and service 1/ subsector; 62% of this was to 339 clients in the central development region, encompassing the proposed project area. The outstanding balance of CSI loans as of June 15, 1979 was Rs 9.5 million or 0.7% of total outstandings. In addition, NBL had advanced Rs 30 million to ADBN and held Rs 20 million of its debentures, bringing NBL's total direct and indirect loans to the priority sector to 5.2% of all loans and investments and 3.9% of deposit liabilities, well short of the mandatory 7%. No proper arrears data are available, but total amounts in arrears as a percentage of the portfolio are thought to be as much as 20% for CSI loans. The bulk of priority sector loans are processed by the branches and approved by regional offices. Disbursements are usually in instalments against proof of purchase, to avoid diversion of funds; Nepal Bank attempts to supervise borrowers during disbursement and throughout the period of the loan. However, proper supervision has been made difficult by lack of trained staff 2/ and logistical problems in rural areas To participate in the proposed project, NBL would establish a head office CSI unit, reporting through the credit manager to the general manager and staffed by a CSI unit chief and at least four officers. The head office unit would appraise larger projects, supervise branch performance and review subloan proposals. At the zonal level, each office would have two CSI officers. Twenty key branches have been identified in the project area; each branch would have at least one trained officer assigned full time for CSI work. Four CSI officers for the headquarters unit have received training in India, financed under the UNDP preparatory project; local training prior to project implementation, to be provided for about 30 CSI officers and their immediate supervisors, has been arranged for November-December 1981 (para 4.25); trained officers would be assigned to CSI work for at least two years. 3/ Approval of loans of up to Rs 10,000 would be made by the branch manager; zonal managers would approve loans of up to Rs 20,000; and larger loans would be approved at head office. It is anticipated that roughly 950 subloans would be approved by NBL under the proposed project; this would represent an average of one approval per month by each CSI officer over the three year commitment period. 1/ Loans to services mainly involving transport, trade and repair enterprises, represent a small proportion of the total. 2/ Nepal Bank has 22 officers who have received formal training in small agricultural and industry lending in India. However, several of these officers are not directly involved in priority sector lending operations. 3/ Seven officers of NBL, to be assigned to CSI units at headquarters and zonal offices, have undergone training with the State Bank of India (SBI) during December-January 1980, with financing under the UNDP project preparation facility. Local courses on CSI appraisal and supervision will be conducted in November-December 1981, with participation of SBI training staff (para 4.51).

33 Rastriya Banijya Bank. Twenty-eight of RBB'S 98 branch offices are in the proposed project area; 23 are in Bagmati and only 5 are in Gandaki Zone. Priority sector lending by RBB over the period totalled Rs 47.7 million, including Rs 24.1 million in 2,098 loans for cottage industry; 816 of them have been in the central development region covering the proposed project area. Outstandings at June 15, 1979 totalled Rs 18.6 million for 1,700 cottage industry loans. In addition, RBB has loaned Rs 10 million to ADBN, bringing the total direct and indirect lending to the priority sector to 5.2% of loans and advances and 4.4% of deposit liabilities, below the mandatory 7%. Arrears data are not available, but are thought to be somewhat higher than Nepal Bank's figures. RBB has a special section with five officers in its head office Loan and Investment Department, reviewing applications for priority sector loans in excess of Rs 20,000 (US$1,520). The department head has authority to approve loans up to Rs 50,000; the deputy general manager, up to Rs 75,000; and the general manager up to Rs 200,000. Disbursements of loan proceeds rarely are done on an instalment basis and RBB does not undertake regular supervision of its clients unless the loans are overdue at full maturity RBB would establish a separate CSI unit under the direct supervision of the general manager and staffed by a unit manager and four officers. At the zonal level, each office would have a unit manager responsible to the zonal manager, and two officers. In 20 key branches, officers would be appointed full time for CSI operations for a minimum of two years. Prior to participation in the project, the CSI officers and their immediate supervisors would undergo training (para 4.25). 1/ The present delegation of authority would be amended to allow zonal managers to approve loans of up to Rs 20,000 and the general manager up to the maximum under the project. With this delegation, it is anticipated that branch offices would process some 400 loans and the head office about 100. The head office unit would coordinate GSI operations and supervise CSI lending operations of zonal and branch offices Agricultural Development Bank of Nepal. ADBN, the main institution for agricultural credit, has 173 offices including 107 small depots, 61 branches and sub-branches, four regional offices, and the head office in Kathmandu. The proposed project area holds 35 offices, 13 in Gandaki and 22 in Bagmati. Resources are provided by share capital and borrowings, including four lines of credit from the Asian Development Bank (ADB). 2/ Domestic borrowings include refinance from the NRB, and direct borrowings from and debentures held by the commercial banks. ADBN does not have the advantages of sizeable deposit mobilization and, due to shortages of refinance funds at NRB, ADBN is experiencing difficultues in raising sufficient local resources. Also, lending to risky agricultural areas, high administrative costs and a decline 1/ RBB and ADBN each have sent seven officers for training in CSI appraisal and supervision with SBI; local training for CSI branch officers, financed by UNDP, will take place in February-March. 2/ ADB's fourth credit to ADBN includes some financing of agroindustries using mini-hydroplants. About Tk 2.6 million (US$200,000) of this is expected to be lent in Gandaki Zone, which has been taken into account in the credit projections for the proposed IDA project (Annex 1, Table 6).

34 in spreads resulted small net losses by ADBN in FY79 and FY80. Nevertheless, ADBN's lending and operating standards and procedures are more advanced that those of the commercial banks. Loans are made largely on the basis of project appraisals. In addition, ADBN has a policy of regular and close supervision of projects, to the extent that field staff availability makes this possible. Disbursements are made by instalment after inspections, and ADBN has pioneered the use of disbursements by coupon or in kind to curtail diversion of funds. Given these skills and procedures, and ADBN's capabilities in technical services to agroindustries, ADBN's participation in the proposed project would be beneficial At June 15, 1979, ADBN's outstandings totalled Rs million, of which 17% was long term, 53% medium and 30% short term. Branch managers have authority to approve loans up to Rs 10,000; regional offices have approval authority for up to Rs 200,000; the general manager has full authority for CSI loan approvals. To participate in the proposed project, ADBN would create a separate CSI unit at head office with a manager and four officers. Twenty key branches, expected to account for the bulk of CSI lending, have been identified and at least 20 full-time CSI officers would be assigned to them for a period of no less than three years. No change in the present delegation of authority is proposed. Branches are expected to appraise about 600 projects over the three year commitment period. The head office is expected to review and appraise a further 200 projects To meet its domestic currency requirements for the 20% of subloan amounts not covered by refinance, ADBN would require an estimated Rs 6 million over the project period with at least Rs 2 million transferred by HMG prior to ADBN's participation in the project. In addition, funds would be required to help ADBN finance the cost of any arrears as the project does not provide for rescheduling of refinance obligations to the Rastra Bank. During negotiations agreement was reached that HMG would make the necessary funds available to ADBN, or suitable terms Training and Institutional Needs. Six weeks of training in India and the Philippines for 21 officers of NRB and the participating credit institutions, financed under the UNDP preparatory facility, have been completed. Local training courses of three to four weeks for about 110 CSI officers from the participating banks has been arranged with the State Bank of India for November-December The completion of training for the agreed level of CSI personnel would be a condition of each bank's participation under the Credit. In addition, shorter 1 to 2 week appreciation courses for about 20 managers and supervisory personnel of the banks would be provided in December 1981 to ensure that branch managers are aware of the objectives and procedures of the project. The project also would provide for local courses and about 12 foreign training slots for NRB and the participating banks each year (US$70,000). A major institutional benefit of the subloan component would be to help the commercial banks introduce project appraisal as the basis of credit decisions, which should have a direct influence on overall credit policies. Also, since the project would include commercial as well as credit services for identified concentrations of artisans under the project, the banks' coverage of CSIs can be expanded, while reducing risks and administrative costs. Also, procedures for disbursement, supervision and amortization by instalment would influence the banks' operations in other areas.

35 C. Commercial and Technical Service Components 1. Export Development - Trade Promotion Centre 4.26 Development of cottage industry exports has to respond to specific market opportunities. Recognizing this need, HMG approved and provided initial budgetary funding for the establishment of a Cottage Industry Export Development (CIED) Division in the Trade Promotion Centre. This Division would provide exporters with assistance in making market contacts, adapting products, and developing the export-oriented supply base. Product groups for immediate attention would be carpets, woolen jackets and knitwear, cotton handloom garments and household articles, metal crafts, and wood and fiber items Functions. The CIED Division would assist exporters in expanding production and sales in cottage industry lines by: (a) providing exporters with practical consultancy in product adaptation, technical improvements in production, and organization of the decentralized production base; (b) improving links between exporters and major importing countries by establishing IMPACT offices 1/ in New York and Frankfurt; (c) organizing and financing trade missions to importing countries and trips to neighboring countries which are leading exporters of similar products; and (d) engaging exporters as performance contractors to establish service networks for selected clusters of artisans in rural areas, concentrating in Gandaki Zone Organization and Staffing. In Kathmandu, CIED would consist of: a general manager; product managers for carpets, woolen and cotton garments, metal crafts, and forestry products; and officers responsible for information, monitoring, and accounts. One assistant would be provided in each section. Four district officers would be assigned to Tanahu, Syangja, Kaski, and Gorkha in Gandaki Zone to monitor the performance contracts (para 4.32). Attractive salaries would be provided to product managers and district officers to attract highly motivated staff who can learn quickly. TPC has transferred four TPC officers to CIED-Kathmandu. The other nine officers, recruited from other institutions involved in industry, would be hired or deputed for a minimum of two years. TPC plans to hire these officers and send them for outside training during November Costs for establishment, staffing, vehicles and overhead costs for CIED-Kathmandu during the three year implementation period are estimated to be Rs 4.5 million (US$340,000) Foreign Offices. The extension of CIED-Kathmandu would be small IMPACT offices of the TPC in New York and Frankfurt. These offices are of critical importance in assuring that production activities in the project

36 respond to market realities. The major functions of these offices would be to: (a) establish and update information on products, prices, and supply capabilities of Nepalese exporters and producers in the priority cottage lines, using this information to promote CSI products among potential importers; (b) provide TPC and Nepalese exporters with current information on market possibilities and fashion trends, attending the permanent and temporary trade fairs, tapping knowledge of importers and getting information on competitors' products and prices; (c) prepare and follow up on Nepalese participation in trade fairs and individual exporter's sales trips; (d) organize consultancy on design, using information available on raw materials, existing designs, and production capabilities; (e) provide feedback from importers on designs, sizing, and prices without necessitating their taking trips to Nepal; and (f) perform necessary trouble shooting, followup and liaison on cottage export orders from Nepal, linked by telex to TPC, helping build Nepal's reputation for reliability Each TPC IMPACT office would have one Nepalese staff member and two from the importing country. Since skills required to design systems and establish operations for the centres are different from the sales capabilities for ongoing management, a firm would be hired to establish the New York IMPACT Centre, at a cost of about US$50,000. The New York IMPACT office would be established by July 31, 1982; operations of the Frankfurt office would be launched by July 31, 1983, with systems adapted from the New York office. UNDP would cover full costs of staff, travel, rent and overhead expenditures. Estimated staff costs for the New York office would be US$150,000 for the project period, plus US$40,000 for short term consultancy. Rent, furniture, telex, and overhead expenditures would be roughly US$86,000 for the three year period. For Frankfurt, staff costs for two years of initial operations are estimated to be US$120,000 equivalent; consultancy, US$40,000; and establishment and operating expenditures, US$49,000. A detailed action program and cost estimates for launching and operating these offices have been prepared (Annex 2) Consultancy. Product specialist consultants would be attached to each of the four product units of the CIED Division of TPC, Kathmandu, and one general consultant would be hired to assist the CIED Director in overall operations. These five consultants would be hired for one year each at a total cost of US$420,000; TPC would have an option to extend the contracts. The product specialists, with their Nepalese counterparts, would provide concrete assistance to exporters on improving designs, quality, and production

37 methods, and in organizing the decentralized supply base in Gandaki Zone and Kathmandu Valley. Consultants would have strong knowledge and successful experience in introducing production and quality improvements in the relevant product lines. In addition, short term consultants in more specialized design technology and export promotion operations would be financed (US$190,000). Measures to promote exports, product adaptation and development of supply capabilities probably would include financing of a firm specialized in intensive preparation for and organization of single country trade fairs, working closely with the New York IMPACT office. The project would incorporate about US$150,000 to sponsor this integrated approach with consultants assessing the most promising cottage lines and exporters, working with them on product adaptation, and then arranging single country fairs. This approach, which has proved effective in Bangladesh and Sri Lanka, would be introduced in about month 9, once initial work has been done through product specialists in Kathmandu and the IMPACT Centre in New York Performance Contracts. Skills in sales, input supply and organization of cottage production are concentrated among the growing number of medium scale exporters. The project would assist these agents and induce them to extend commercial services beyond Kathmandu Valley by engaging the most qualified as "performance contractors to develop selected cottage industry clusters in rural Gandaki and Bagmati. Contracts would specify the exporter organization's phased responsibilities for training, input supply, orders, quality control, and processing facilities over a two to three year period. Contracts would provide grant payments to cover specified initial costs in establishing rural operations. Prospective performance contractors would need to prepare their proposed programs for selected rural craft clusters; these proposals would be evaluated by TPC. 1/ Pilot phases would be built into most contracts. Performance contract payments could be utilized for: staff to provide training to upgrade artisan skills and introduce modified product lines; raw material expenditures during the pilot production phase when skills are being developed to meet export standards; and a portion of costs of establishing quality control agents and depots in Gandaki Zone. Performance contract payments would constitute no more than 20% of total subproject costs, or Rs 200,000 (US$15,200) 2/ whichever is lower. The remainder would be financed from sponsor's equity and bank term credit; minimum equity contributions would be 10% of total project costs. The performance contractors would require term credit from the banks to cover the majority of their financing needs for processing and storage facilities, centralized production in certain processes, and permanent working capital. Subloans for performance contracts and related CSIs would be eligible for NRB refinancing under the proposed project (para 4.10); in most cases, the contractors would seek credit for the permanent 1/ About 15 exporters have visited Gandaki Zone and are preparing their proposals. TPC has arranged trips to Gandaki to visit the rural concentrations of woolen cotton and fiber producers; about 30 exporters have expressed active interest in participating. 2/ These maximum amounts exclude sales and exposure trips to be financed by TPC under the project.

38 working capital needs of the artisans in their networks while individual CSIs would borrow directly from the banks for their equipment needs (Annex 1, Table 6). Since each performance contract would cover rural CSI concentrations, credit and commercial services would be facilitated Most contracts would be established in Syangja, Tanahu, Kaski and Gorkha of Gandaki, since these districts hold sizeable craft concentrations within reasonable access of trunk roads (Annex 1, Table 3 and Annex 3, Map 2). Up to five contracts would be available for rural areas of Kathmandu Valley. About 25 contracts would be financed from the US$310,000 project allocation, with each contract covering 1 to 3 panchayat clusters of 100 to 150 units each. 1/ Payments by TPC would be made on a quarterly basis against completion of agreed tasks, checked by field-level TPC staff. Flexibility would be maintained; contracts could be revised, upon mutual agreement by the contractor, TPC, and IDA as experience is gained. A statement of policies and procedures for the CIED Division, including norms for operating the contracts, prepared during appraisal, has been ratified by the Board of Directors of TPC (Annex 2) Sales and Exposure Trips. About US$225,000 would be provided for 25 to 35 sales trips for performance contractors and about 30 to 40 trips for exporters and manufacturers to expose them to marketing and production methods neighbouring countries with successful experience exporting similar items. The CIED Division of TPC would be responsible for organizing these trips Funding and Procedures. The Trade Promotion Centre would establish a Cottage Industries Export and Product Development Fund to finance the above activities. The Fund would be managed by the CIED Division under the general direction of TPC's Executive Chairman and Board of Directors. Prior to January 1, 1982, the Fund would need to be established with an agreed statement of policies and operating procedures, an action program for the second six months of FY 81/82, and adequate staff and long-term consultants in place in the CIED Division. The New York IMPACT office would be established by July 31, 1981 and operations of the Frankfurt office would be launched by July 31, Annual action programs, prepared by CIED- Kathmandu, would be sent to IDA by each May 31 for the second and third fiscal years. IDA would need to approve selection, action programs and contractual terms performance contracts. IDA and UNDP would disburse against regular statements of actual expenditures, with documentation provided on consultancy payments and expenditures for trips; 1MG would cover half of the costs of staff and operating expenditures and provide TPC with budgetary allocations to cover the project expenses, including the IDA and UNDP financed amounts. 1/ Most contracts are expected to be in carpets, woolen fabrics, and cotton handloom products (Annex 1, Table 3); however, a few contracts in reed and bamboo products are expected in Gandaki, metal craft networks in rural Kathmandu Valley would be eligible and export-oriented agroindustrial CSIs such as beekeeping could be covered by performance contracts in years 2 and 3 of the project.

39 Extension Services and Development Centres - CIDB 4.36 Strategy. The Cottage Industry Development Board would be reactivated, taking over responsibility for development activities from DCVI in Kathmandu and in Gandaki Zone. 1/ The CIDB would be responsible for: (a) technical extension services in Kathmandu Valley; (b) extension and management of subsector development centres in areas of Gandaki zone which are too remote to interest private performance contractors; and (c) operations of an improved Handicraft Design and Promotion Centre. During appraisal, drafts of the following key statements were prepared: policies and procedures for the CIDB; an organizational chart and staffing requirements; and a year I action program (Annex 2). These were discussed during negotiations with agreement in substance reached. Adoption of a statement of policies and operating procedures, financial rules, reorganization along subsector lines, and hiring and training of minimum staff would need to be completed by January 1, The CIDB would provide annual action programs to IDA, for concurrence, by May 31 of each year Operations and Functions. The CIDB would take over the training, production, and industrial service sections of DCVI headquarters, organizing into subsector divisions: cotton textile, woolen, metal, forestry-based products, agroindustries and others. These divisions would design and oversee subsector-specific extension programs and development centres, train CIDB staff and leading village craftspeople as trainers, and provide direct extension services and short-term training in Kathmandu Valley. In addition to subsector divisions, the CIDB would have a monitoring, reporting, and coordination division for regional projects and an accounting and administration division. Local staff costs and related expenditures for headquarters would cost Rs 2.8 million (US$210,000) over three years; about 50% of this cost would be incremental since a portion of staff would be shifted from DCVI. Specialist consultants would be connected with the four major subsector cells for six months to help train the CI Board trainers and launch extension services and development centre programs (US$190,000). Costs for short-term training and consultancy for CI Board staff and artisans would total about US$290,000. A parallel but smaller subsector organization in an office in Gandaki would be organized to provide backup services to the cotton handloom, wool and forest products development centres in Gandaki. The CIDB would launch and manage about six wool, six handloom, and four forestry products development centres in Gandaki over the three year commitment period to service household artisans in selected clusters, mainly of Lamjung, Upper Kaski, Gorkha, and possibly Manang. 2/ Services of these subsector development centres would be similar to those of private performance contractors, but the centres would cover more remote 1/ The Sixth Five Year Plan envisages that the CI Board would be charged with responsibility for executing cottage industry components of this lead project as well as small CSI components of integrated rural development projects which receive outside financial assistance. 2/ These are areas too remote to be of immediate interest to private performance contractors.

40 areas, have longer training periods, and concentrate on local markets. Services would include upgrading skills, purchasing raw materials, offering pre- and post-processing facilities, procuring finished products and supplying them to the Emporium and private traders. Costs of equipment and working capital for these centres would total about Rs 4.8 million (US$360,000); about 1,300 units would be serviced. The Handicraft Design and Promotion Centre in Kathmandu would be revamped with qualified management and resultoriented staff to provide prototype designs and samples to exporters and CI Board centres, and to develop appropriate technologies for use in CI Board centres and extension programs. Costs for these improvements would total about Rs 2.5 million (US$190,000). A breakdown of costs for the CI Board elements is provided in Annex 1, Table Organization. To reactivate the CI Board, the original legislation would not need to be changed. The composition of the Board of Directors has been modified to include representatives of the Ministries of Industries and Finance, TPC, ISC, the Rastra Bank and at least two private sector representatives engaged in manufacturing and marketing of cottage industry products. The Director General of DCVI would be the chief executive during the transitional period, assisted by a Chief Coordinator of CIDB; subsequent selection of a full time executive is planned Staffing. Only the most qualified and motivated of DCVI's staff would be transferred to the CIDB. Except in cases of extremely qualified technical persons, DCVI staff of over 45 years of age would not be transferred to the CIDB, since the work would be heavily field-oriented. Total staffing costs for the project would be about Rs 5.4 million (US$410,000), including Kathmandu, Gandaki, Design Centre and subsector development centre staff. Staff would be hired to the Board, deputed from othiei organizations for at least two years, or hired on a fixed contract basis. The financial rules for the Board would need to grant the Cottage Industry Board autonomy in personnel hiring and payment of performance incentives, within agreed limits. The Cottage Industry Board would provide sufficient salaries and incentives to attract and retain qualified, motivated staff for cottage industry work particularly in rural areas. 1/ These aspects are covered in the draft statement of policies and operating procedures, upon which agreement in substance was reached during negotiations; such a statement would need to be ratified by the CIDB by January 1, Raw Material Arrangements - Companies and Emporium 4.40 Sporadic availability and resultant price volatility of wool, cotton yarn and metal represent significant problems in the priority cottage subsectors identified under this project. Services by the National Trading Ltd. (NTL) and the Emporium have not been satisfactory (paras 3.07, 3.19). Import duty and sales tax rebates are granted only to cottage manufacturers, 1/ The statutes of the CIDB enable it to pay performance bonuses; these would be paid based upon staff within the subsector cells, the Design Centre, and subsector development centres exceeding established performance targets.

41 Private exporters and manufacturers intend to establish two private limited companies to import wool and metal in bulk, to assure stable prices, availability and adequacy of types of raw materials. Under the Industrial Enterprises Act, the same reduced import duties and sales tax rebates which apply to CI producers would be applied to such companies importing on behalf of member producers and exporters (para 2.16). In the case of wool, it also would be useful if the company could be authorized by HMG to negotiate with China, although bulk purchasing from New Zealand and Australia also is viable (para 3.07). Membership would consist mainly of exporters and manufacturers in wool or metal products respectively; minority shares of not more than 10% could be held by the Emporium. Importers, who could not hold more than 20% collectively, could provide procurement knowhow and contacts; the Emporium could help monitor procurement and sales to see that raw materials were sold at reasonable prices. The Handicraft Association would assure legitimacy of members in the metal company; a similar association is being organized for woolen products. Each company would have a board of directors consisting of a general manager, and representatives of exporters and manufacturers holding shares, and, if requested, the Emporium The companies would start on a limited scale. If initial operations are successful, the scope would be expanded. Estimated financing requirements of the two companies, including term credit portions eligible for IDA financing, are as follows: for the metal importing company, raw material stocks for six months (Rs 2 million) 1/ and office, storage and staffing expenses for one year (Rs 500,000) 2/; and for the wool importing company, raw material stocks for six months (Rs 2.5 million), and overheads by Rs 400,000. Additional funds could be made available, once successful operations are demonstrated At least twenty percent of financing requirements would come from equity of the shareholders. The remaining 80% would be provided by the commercial banks. Top executives of Nepal Bank Ltd. have indicated that NBL would be willing to be the banker for these companies. Maximum total lending limits would be enforced during year 1 of operation to ensure the companies are established on a strong footing before attempting to supply the majority of input requirements of the industry. Medium term credit requirements would be covered under the project, with partial refinance of bank loans by the CSI Unit of Rastra Bank (para 4.10). Short term credit needs would be supplied through normal loans, probably from NBL. As a condition of disbursement, the respective company would have to be legally established, with articles of association, membership composition, paid in capital, a financing plan and a year I action program satisfactory to IDA. Regular reports on operations would be submitted to IDA, on a quarterly basis. 1/ Calculated on the basis of 30% of annual metal requirements for metal working industry in Kathmandu Valley. 2/ Under the assumption that overheads for subsequent years will be financed out of company profits.

42 Cotton Yarn, Equipment and Cloth--Emporium. Under the project, the Emporium would concentrate on improving its procurement and distribution of cotton yarn and critical equipment needed by cottage industries in the project area. The Emporium also would provide an outlet for CSI handloom products from the CI Board development centres by taking public institutional orders and subcontracting these to CI Board centres. During the three year implementation period, the Emporium would make available its services in bulk procurement and distribution of cotton yarn and handloom accessories to: about six performance contractors for clusters in Gandaki and about three in rural Kathmandu Valley; and about six handloom development centers of the Board in Gandaki. Each of these agents would be expected to service 50 to 150 looms, with each loom requiring the Emporium to hold an equivalent of Rs 2,500 of yarn, representing three months in input requirements. By year 3, it is expected that the Emporium would need about Rs 5.0 million to meet these needs of the project. The Emporium would stock yarn and equipment and collect textiles for the project in Kathmandu and Pokhara; performance contractors and the CI Board would be responsible for other transport and storage During year 1, about Rs 2.2 million (US$167,000), or three months worth for about 400 to 500 units, would be needed. Under the project, the Emporium would be eligible for medium term bank credit to meet these needs, under the same terms and refinancing arrangements as for other eligible subprojects. The project could cover the 20% equity requirement of the Emporium which would be made as a seed capital grant to help enable Nepalese handloom producers to upgrade quality and improve price. 1/ The 20% equity is expected to equal about Rs 1.2 million by year 3, assuming a gradual buildup of stocks to Rs 5 million. HMG would make these seed capital funds available; the equivalent in cotton yarn or handloom accessories procured by the Emporium would be eligible for IDA disbursements. This assistance to the Emporium on commercial operations for the handloom sector is considered necessary because of fragmentation of private commercial agents and the need to upgrade quality before major expansion scope can be tapped. If HMG and IDA agree that the Emporium has performed well in these cotton-related activities and that diversification would be useful, during the initial period loans for expansion in other lines could be made eligible for project, the Emporium The cotton yarn operations under the project would represent a sizeable expansion in the Emporium's importing and sales activities. 2/ The 1/ The Rs 1.2 million equity is roughly equivalent to paying the interest rate charges if the Emporium were to borrow for their total financing requirements. This seed capital is judged necessary to provide a development period for improving price and quality of handloom products; during this period normal interest charges on yarn and accessory purchases would not be passed on to handloom weavers. 2/ From 1976/77 to 1978/79, the Emporium's annual turnover in yarn ranged from Rs 1.5 million to Rs 6.2 million; assuming three months worth were stocked, average stocks ranged from Rs 300,000 to Rs 1.5 million. Assuming that sales to project agents would be incremental, the Emporium would be expected to double its yarn stocks by the end of year 1 of the project.

43 project would provide a procurement advisor for six months, with the option by the Emporium to extend the contract for six additional months if required. The advisor would assist the Emporium in improving its procurement procedures, with particular attention to bulk procurement, more systematic information gathering on producers' requirements, and inventory management methods. In addition to this advisor's assistance, the International Trade Centre intends to provide the Emporium with regular information on price trends and alternative sources of cotton yarn The Emporium would maintain its procurement procedures, obtaining at least three quotations from different suppliers for each order of cotton yarn. The Emporium would maintain separate accounts for this yarn fund, including amounts utilized by each project agent. Prior to disbursements for this component, the Emporium would: ratify a statement of policies and procedures for this fund; establish separate accounts; and have hired the advisor. The Emporium would provide IDA, by May 31 of each year, annual action programs for the subsequent fiscal year. 4. Coordination, Monitoring and Preparation for Expansion 4.48 Coordination. The implementing agencies would be responsible for carrying out their specific activities and responsibilities within the broad framework of objectives, guidelines and procedures which would be agreed between HMG and IDA in the credit, subsidiary loan, and project agreements, and in the implementation plans. In view of the complementary nature of the activities of the various agencies, mechanisms would be established to help ensure adequate coordination at headquarters and district levels. At the district level, the TPC and CIDB would be operating in different areas of Gandaki; therefore, meetings between the banks and one technical assistance agency would be needed. Local government officials would be consulted as required. At headquarters, a CSI Coordinating Committee would be established to review progress of the project, facilitate coordination among the implementing agencies, and ensure that remedial measures are taken if necessary. The Coordinating Committee would be chaired by the Secretary, Ministry of Industry; key members would include representatives of the National Planning Commission, Ministry of Finance, Ministry of Commerce and Supplies, Rastra Bank, Cottage Industry Board, Trade Promotion Centre, Emporium, the Industrial Services Centre and representative private sector delegates from cottage industries and exporter organizations. 1/ Monthly meetings would be held during the initial year of operations, after which the coordinating committee would meet at least quarterly. The Joint Secretary of the Ministry of Industry would be a key contact point and decision maker, to help resolve problems without having to elevate them to the CSI Coordinating Committee; one senior officer would be assigned full-time to provide day-to-day coordination. ISC, as the technical arm of the Committee, would do informal liaison Monitoring and Evaluation. Each implementing agency would install its own monitoring and reporting system, summarizing findings in the quarterly 1/ A Steering Committee for the CI Development Study, with similar composition, has been meeting regularly for the last year.

44 reports. The Industrial Services Centre would collect and collate these reports as well as do independent monitoring of inputs, impact, and integration of the various components. The ISC monitoring staff, working closely with the agencies to maintain a good understanding or operations of each component, would submit the quarterly report to the members of the CSI committee. In addition to this regular monitoring and reporting, a formal review would be conducted by HMG and IDA after eighteen months of credit effectiveness; this would enable formal modifications to be made, if necessary, and would facilitate preparation of other projects. Four man-months of consultancy services, completed in February 1981, were provided to assist ISC in developing the overall monitoring and evaluation system for the project, and to work with each implementing agency on its own reporting system (para 4.51). The CSI project would cover four man-months of consultancy on monitoring over the project implementation period (US$30,000) Preparation for Next Projects. It is envisaged that HMG, IDA and other donors would be prepared to make sizeable financing available to CSIs once the implementing agencies are strengthened and the most successful types of assistance are determined through experience under the first project. ISC would do surveys of 20 to 30 districts which are either representative or which have particularly strong potential for CSI development; most districts would be from hilly areas where cottage industries are concentrated. On the basis of district surveys, discussions with the implementing agencies and consultants, and evaluation of the first project, ISC will make proposals for strategies and components for area and product schemes. The project would finance six full time ISC staff to work on monitoring and project preparation tasks; in addition, funds would be allocated for survey staff and expenditures (US$90,000 equivalent). Outside consultancy to ISC in project preparation is not envisaged; ISC is experienced in preparation of cottage industry projects and can tap knowhow of specialists assigned to the implementing agencies. 5. UNDP Financing for Training and Technical Assistance 4.51 UNDP project preparation financing of about US$224,000 has been provided. The Bank is the executing agency; NRB and the Ministry of Industry are the implementing agencies. Most elements of this preparatory assistance are expected to be completed by January Main components have included: (a) foreign and local training in CSI lending for staff, the banking institutions: NRB, NBL, RBB and ADBN; (b) outside training and exposure trips for management and specialists of the Cottage Industry Board, Trade Promotion Centre, Emporium, and private exporters; and (c) consultancy services to help finalize the Industrial Enterprise Act including foreign investment aspects (para 4.53); develop the monitoring and evaluation system for the project (para 4.49); and help prepare TPC and exporters to implement activities to be financed from the Cottage Industry Export and Product Development Fund. 1/ 1/ Consultances in monitoring and evaluation and on the Industrial Enterprises Act have been completed. Orientation and training in Nepal are scheduled to begin in January 1982.

45 In addition to this project preparation facility, about US$2.0 million in UNDP grant funds would be provided for technical assistance, training and marketing inputs under the proposed project. These components are of critical importance in building capabilities of private and public institutions. The elements and costs are outlined in Annex 1, Table 8 reflect understandings reached with representatives of the relevant ministries, Planning Commission, Nepal Rastra Bank, other implementing agencies 1/, and the UNDP. These training, consultancy, and marketing inputs would be part of the proposed Cottage and Small Industries Project; NRB, the Ministry of Industry and the Ministry of Commerce and Supplies would be the implementing agencies; and the Bank would be the executing agency. As the executing agency, the Bank would make the final selection and contractual arrangements for consultants financed by UNDP under the project. About US$110,000 million would be needed in CY 1981, US$1.06 million in 1982, and about US$680,000 in 1983 (Annex 1, Table 12). During negotiations, HMG presented a copy of HMG's request that UNDP make this funding available for the proposed Cottage Industry Project. Prior to credit effectiveness, the UNDP Project document would need to have been signed. D. Policies Affecting Cottage Industries 4.53 The new Industrial Enterprises Act, passed by the Panchayat in August 1981 and awaiting Royal assent and issuance, contains several measures which improve incentives for CSIs and related commercial agents. However, further improvements in industrial policies and procedures, outlined in paras , would increase profitability and incentives in CSI development. The new Industrial Policy and the Industrial Enterprises Act were discussed during negotiations. The IDA delegation expressed the need for further improvements, through rules and other measures, in the following areas: (a) simplified fiscal incentives for CSIs; (b) provisions for foreign collaboration; (c) extension of import concessions to input supply companies which are composed of private exporters and CSIs in the relevant product group; (d) streamlining of export procedures; and (e) planned steps for improving transport arrangements and export financing (paras ). If necessary, a portion of the contingency fund of the segments of the project financed by UNDP or IDA could be utilized for short-term advisors to address key policy and procedural needs of the sector. The project contains provisions for a continued dialogue between HMG and IDA on policies affecting CSIs. 1/ Consultancy to be financed by UNDP under the proposed CSI project is outlined in the relevant sections of this report: paras 4.07, 4.30, 4.31, 4.37, 4.46, 4.47, 4.49 and 4.52.

46 V. THE PROPOSED CREDIT 5.01 Lending Arrangements. The proposed IDA credit of SDR 5.7 million (US$6.5 million equivalent), consisting of a US$4.5 million equivalent subloan component and a US$2.0 million technical and commercial service component, would be made available to HMG on standard IDA terms and conditions. In addition, a US$2.0 million UNDP grant for training, consultancy and marketing inputs would be executed by the Bank, implemented by the Ministry of Industry and NRB. The credit is expected to be committed in about three years and disbursed in about 3.5 years from the date of credit effectiveness. Annex 1, Table 9 provides a breakdown of the portion to be financed by UNDP. For the subloan component, HMG would onlend the proceeds to NRB for the account of CSI Fund to cover disbursements to refinance eligible SCI subloans made by the three participating banks. The proceeds would be onlent tq NRB in rupees with HMG bearing the foreign exchange risk on repayments to IDA. Repayment by NRB to HMG would be on a fixed amortization schedule. During negotiations, it was agreed that the CSI Fund of NRB would repay HMG after fourteen years, to enable funds to be revolved at least twice. During the fourteen years, the funds would be relent by the CSI Fund to the participating banks for refinance of CSI loans. These arrangements are specified in the draft Project Agreement, and will be incorporated a Subsidiary Loan Agreement between HMG and NRB. The commercial and technical service component, including the portion financed by UNDP and executed by IDA, would cover specific items of training, consultancy, equipment and staffing expenses for the participating agencies. The CSI Fund would be responsible for the portion of the technical service component allocated to NRB and the participating banks. The other service elements would be under the respective implementing agencies; each would have separate action plans, withdrawal, and reporting requirements. HMGN would need to provide full budget allocations for the technical and marketing service components regardless of whether the project element is to be financed by HMGN, IDA or UNDP. The proceeds of the technical and commercial components would not be recovered directly; these components are of critical importance in building institutional capabilities as well as addressing key marketing, input supply and technical upgrading needs Onlending Terms. The participating banks would charge CSI subborrowers a standard annual interest rate of not less than 11% with rates reviewed at least annually and revised to assure positive, commercial interest rates and consistency with rates for term loans to industry in Nepal. The banks would maintain spreads of at least 4% for subloans above Rs 20,000 and 6% for subloans of up to Rs 20,000. The CSI Fund would receive a minimum spread of 2% to cover administrative costs, and 1% as a premium for the credit guarantee scheme. The CSI Fund of NRB would pay HMG an interest charge of 4-6% p.a. on the refinance amounts outstanding in its account. Agreement on minimum interest rates, spreads and review and adjustment procedures was reached during negotiations (para 4.11).

47 PROPOSED ANNUAL LENDING RATES AND SPREADS Size of Banks' Subloan Up to Rs 20,000 Above Rs 20,000 (%) (%) (a) Initial final onlending rate (b) Minimum spread to participating bank 5 4 (c) Minimum spread to CSI Fund, NRB 2 2 (d) Credit guarantee premium, NRB 1/ 1 1 (e) Minimum onlending rates to NRB by HMG 2 4 No commitment fee would be charged on the subloan component due to the multiple institutions involved and the snmall scale nature of the subborrowers Initial Contribution to the CSI Fund. Prior to credit effectiveness, NRB would pay into the CSI Fund the initial contribution of at least Rs 5.0 million. This contribution would be used to cover initial operating costs as well as the time lag between NRB's refinancing and reimbursement from IDA (para 4.12). NRB would make additional contributions to the CSI Fund as needed. Surpluses generated by the Fund and any benefit derived from a longer repayment term to HMG than that granted to the banks would be ploughed back into the Fund and used to increase the amount available for refinancing eligible CSI prospects Subloan Size and IDA Review. The maximum subloan size would be Rs 800,000 (US$60,600), except for the raw material supply companies and the Emporium. Refinance by the CSI Fund would cover 80% of the subloan amounts. The average size of subloans is expected to be about Rs 31,700 (US$240) resulting in a total of about 2,000 subprojects; including organized SSIs, market agents, and raw material supply companies, the project would involve about 2,200 subloans averaging Rs 311,000 (S$23,560). In view of the large number and small sizes of subloans, simplified review and approval procedures would be used to keep administrative costs and processing times within reasonable bounds. In refinancing subloans of below Rs 20,000, the CSI Fund would do a rapid check on the eligibility of the subproject, its financial viability and the adequacy of proposed subloan terms. Periodically, the CSI Fund would undertake a detailed post-approval review of a 20% sample of appraisal reports submitted to check that appropriate appraisal standards are being maintained. 2/ For subloans of over Rs 20,000, a detailed review by the CSI Fund would be required prior to NRB's approval of refinance and reimbursement applications to IDA. To ensure expeditious review of these subloans, the Fund would complete its procedures within three weeks of receiving completed applications from the banks. In claiming reimbursement from IDA, the Fund would certify 1/ Credit guarantee premiums would be 1% on amount guaranteed which is outstanding, not on the full subloan amount. 2/ In the initial period, the CSI Fund staff would pay closer attention to these smaller projects at the time of approval to provide the participating banks with immediate feedback on appraisal standards. Once these standards have been firmly established, full implementation of post approval sampling can take place.

48 that subprojects meet eligibility criteria, are financially viable, and that subloan terms are appropriate. IDA's prior review and approval would be required for refinance applications for all performance contracts, market agents, raw material supply companies and CSIs requiring subloans in excess of Rs 330,000 (US$25,000). This procedure is expected to result in detailed pre-approval review by the CSI Fund of some 370 subprojects, 20% by number and 84% by amount. IDA would review, prior to authorization, about 90 subprojects, 5% by number and 60% by amount of all subprojects (Annex 1, Table 11). IDA supervision missions would review appraisal standards on smaller subprojects Procurement. Participating banks would be responsible for ensuring that items procured for subprojects are suitable, reasonably priced and that sponsors have canvassed the main sources of supply; the CSI Fund would monitor these procurement procedures. Contracts for goods or services procured outside Nepal costing the equivalent of US$5,000 or more per item, or US$20,000 or more per contract, would be on the basis of international shopping with at least three quotations. Since it would be difficult and costly to seek quotations for smaller contracts, these provisions would not apply, but participating banks would maintain appropriate records of the method of procurement for post-review by the CSI Fund staff and IDA supervision missions. Procurement of goods and services under the technical and commercial service component would be made according to procedures of the CIDB, TPC and other implementing agencies (Annex 2). Procurement of vehicles and equipment would be fully dbcumented Disbursement. Estimated schedules of disbursements for the IDAand UNDP-financed portions of the project are presented in Annex 1. Table 12. All subloan disbursements for eligible foreign or local expenditures would qualify for 80% refinancing by the CSI Fund. IDA would reimburse 100% of the refinance granted by the CSI Fund, resulting in IDA funds covering about 64% of total subproject costs. 1/ The average foreign exchange content of subprojects is estimated to be about 30%. Given the large number of subprojects involve6 and :he small subloan sizes, normal DFC procedure of disbursing against expenditures on individual subproject accounts would be inappropriate. Accordingly, for refinance of subloans of Rs 330,000 (US$25,000) or less, the CSI Fund would claim reimbursement from IDA on the basis of periodic certified statements of expenditure. For these subprojects, documentation would not be submitted to IDA for review but suitable documentation, 2/ evidencing expenditure would be provided by the bank and retained by the CSI Fund. The state- 1/ The ratio would be somewhat higher depending on the number of subprojects where the sponso-'s contribution is waived in whole or in part (para 4.iO). As this is likely to be confined largely to smaller loans the aggregate of subp-oject costs covered by IDA financing is likely to be only marginally nigher than o4%. 2/ Excluding expenditure on permanent working capital.

49 ments of expenditure and the supporting documentation 1/ would be audited annually and an audit certificate would be presented to IDA. For refinancing of subloans of over Rs 330,000, full documentation of expenditures would be submitted to IDA. For elements of the technical and commercial service component implemented in Nepal, IDA would disburse against 100% of foreign expenditures for equipment, vehicles and raw materials, imported directly and for locally manufactured equipment and raw materials purchased ex-factory; 80% of expenditures for other equipment procured locally; 50% of staffing and related expenditures according to agreed action programs; and 100% for performance contract fees. For those elements of the technical and commercial service component, withdrawal applications would be submitted directly to IDA by the implementing agency, accompanied by certificates of expenditures on staffing, and documentation of other expenditures; the documents for which would be retained by each agency and made available to IDA supervision missions for inspection. In the case of the IMPACT centres in New York and Frankfurt, IDA would disburse 100% of equipment, consulting, staffing costs, and other operating expenditures; HMG would establish contractual arrangements for establishment and operation of the IMPACT centres Repayment Schedules. The maturity for subloans refinanced by the CSI Fund would be 18 months to 7 years, including an appropriate grace period. Repayments of refinance by the participating banks would be on the basis of a composite schedule reflecting the consolidated amortization schedules of individual subloans, fixed at the time refinance is granted and not subject to rescheduling. As the commitment period would be about three years, repayments by the banks would stretch over about 10 years. A fixed amortization schedule of 14 years for repayment of the US$4.5 million subloan component from the CSI Fund would be arranged to enable revolving of the funds and to increase the effective amount available for CSI refinance. Agreement on these terms was reached during negotiations Reporting, Accounts and Auditing. The CSI Fund would submit quarterly reports to IDA and to the CSI Coordinating Committee, through ISC; the main report would aggregate the individual quarterly reports of participating banks, covering activities, portfolio data, collection and arrears performance, and supervision results. In addition, the CSI Fund would submit a quarterly statement of its own refinance activity and details of disbursements. In addition, the CSI Fund would prepare annual reports on performance of the participating banks; these reports would present an analysis of the banks t quarterly reports including a projection of commitments, refinance, and withdrawals from IDA. The CSI Fund also would present its assessment of the standards of the banks and propose improvements. The report also would include the audited annual report of the Fund, prepared in a form acceptable to IDA, and the audited annual reports of the banks. The annual reports should be submitted within six months from the end of the fiscal year. An independent auditor, acceptable to IDA, would audit all project accounts. The auditor also would be responsible for providing audits covering certificates of expenditure (para 5.05). Each bank would maintain separate accounts for subloans and documentation on disbursements 1/ Documentation would include: names and locations of subborrowers; the total of each subloan; disbursements during the period for each subloan; and the branch of the lending institution handling the loan.

50 and procurement procedures. Each technical service agency would be expected to prepare annual implementation programs and quarterly progress reports, which would be submitted to the CSI Coordinating Committee and to IDA. Separate accounts would be maintained by the implementing agencies for the technical and commercial service elements; these project accounts would be audited annually by an independent auditor acceptable to IDA. VI. PROJECT BENEFITS AND RISKS 6.01 Benefits. The proposed project is geared to provide the effective credit, commercial, and technical services needed to tap growth potential in CSI products, particularly for exports. The project is expected to generate significant direct benefits in earnings and exports in the three districts of Kathmandu Valley and the six Districts of Gandaki Zone. Also, major general benefits are expected in the strengthening of private, banking and public institutions, development of replicable product specific schemes, and improvement of industrial policies and procedures Under the subloan component, about 2,600 loans would be made by the three participating banks; the majority of these would be for cottage units, with about 50 to performance contractors and other market agents, 100 to organized small industries, and 3 to 5 subloans to the input supply companies and the Emporium. The average subloan size for cottage units would be Rs 15,000 (US$1,200); average fixed costs per job would be about Rs 3,200 (US$240), or Rs 1,900 (US$145) if agroindustrial units are excluded. About 15,000 workers in CSIs would be recipients of credit, directly and through performance contractors Since the project would cater mainly to cottage industries, commercial and technical services are at least as important as credit. The effective operations of the Cottage industry Export Development Division of the TPC and the IMPACT centres in New York and Frankfurt will be of particular importance in providing needed market contacts, product adaptation inputs, and direct incentives to exporters in expanding their rural base. With the project inputs, it should be possible to maintain the 35% annual growth rate in exports of CI products; without the project, market, raw material and credit constraints are likely to slow export growth to roughly 20%. The project, thus, could generate incremental gross foreign exchange earnings of roughly $27 million (US$15 million net) from ; most of this growth is expected in carpets (44%) metal products (37%), and cotton and woolen garments (15%). This export growth would be equivalent to roughly 10 million additional mandays of employment over the four year period. In addition to increasing mandays worked, a portion of the increased value added of products could be passed on to artisans in the form of increased daily rates; however, these potential benefits have not been included in the calculation. Most of the employment created would be in the form of additional days worked by existing artisans rather than new jobs created; this 10 million mandays is equivalent to 50,000 manyears of incremental employment over the period. Performance contract networks could provide regular employment to about 10,000 rural crafts people by About 30% of the export earnings and employment is expected to come from Gandaki; without the project, the share of rural areas in export production would be likely to remain minimal (Annex 1, Tables 4 and 5).