FiLE COPY. Document of The World Bank FOR OFFICIAL USE ONLY SECOND SMALL-SCALE INDUSTRY PROJECT BANGLADESH. Public Disclosure Authorized

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 ON THE SECOND SMALL-SCALE INDUSTRY PROJECT BANGLADESH May 26, 1978 Industrial Development and Finance Division South Asia Projects FiLE COPY Report No. 1802a-BD 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 The Bangladesh Taka is officially valued at to the Pound Sterling. The Pound now floats relative to the US Dollar and consequently the Taka/US Dollar rate is subject to change (as of January 30, 1978, US$l = Tk 14.5). The rate below has been used throughout this report. US$1 Tk 15.5 Tk 1 = US$ Tk 1 million US$64,500 PRINCIPAL ABBREVIATIONS USED CI - Cottage Industry BHMC - Bangladesh Handicrafts Marketing Corporation BSIC - Bangladesh Small Industries Corporation BCIC - Bangladesh Cottage Industries Corporation BSCIC - Bangladesh Small and Cottage Industries Corporation BSB - Bangladesh Shilpa Bank EPSIC - East Pakistan Small Industries Corporation GOB - Government of Bangladesh SCI - Small-Scale and Cottage Industry SSI - Small-Scale Industry excluding cottage FISCAL YEAR GOB July 1 - June 30 BSCIC July 1 - June 30 Commercial Banks January 1 - December 31

3 FOR OFFICIAL USE ONLY BANGLADESH APPRAISAL OF A SECOND SMALL-SCALE INDUSTRY PROJECT TABLE OF CONTENTS Page No. I. BACKGROUND... 1 II. THE INDUSTRIAL SECTOR AND ROLE OF SMALL AND COTTAGE INDUSTRIES... 2 Economic Setting... 2 Characteristics and Role of SCIs Policy Framework and Industrial Finance... 5 III. THE PROJECT... 7 The Project Components... 7 Credit Component - Size... 9 Credit Component - Subproject Selection Methods and Criteria Special Capital Fund Cottage Industry Component Technical Assistance Component IV. INSTITUTIONAL ASPECTS Institutional Framework Commercial Banks BSCIC V. PROMOTION AND EXTENSION General SSI Promotion SSI Extension CI Extension VI. ADMINISTRATION OF THE CREDIT VII. ECONOMIC JUSTIFICATION AND RISKS Benefits Risks VIII. SUMMARY OF AGREEMENTS REACHED ANNEX 1 - LIST OF RELATED DOCUMENTS AND DATA AVAILABLE IN THE PROJECT FILE This report was prepared by E. Elejalde, J. Balkind, and N. Barry (ASPID) and B. Hansen (Consultant) following their visit to Bangladesh in August 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.

4

5 BANGLADESH APPRAISAL OF A SECOND SMALL-SCALE INDUSTRY PROJECT I. BACKGROUND 1.01 The Government of Bangladesh (GOB) has requested an IDA Credit to finance small-scale and cottage industries (SCIs). This report recommends a credit of $7 million. The purpose of the Credit will be to support the development of SCIs by: (i) introducing reforms in the institutional framework for assisting SCIs by strengthening institutions, redefining their roles, and initiating programs to tackle problems constraining SCI development; and (ii) providing finance for fixed investment and permanent working capital requirements of new, expansion, balancing and modernization SCI subprojects, and for technical assistance In June 1970, IDA made a $3 million Credit (Credit 192-PAK) to Pakistan to provide foreign exchange for imported equipment and technical assistance for small-scale industry (SSI) in the country's east wing. The East Pakistan Small Industries Corporation (EPSIC) was responsible for the administration of the Credit and a consortium of commercial banks also participated in the project. In January 1973, IDA reactivated the Credit to GOB (Credit 353-BD) with the Bangladesh Small Industries Corporation (BSIC), EPSIC's successor, responsible for administering the Credit and again with the participation of a consortium of commercial banks. The Credit was disbursed by March The first Credit financed 43 subprojects. The average fixed cost per enterprise was about Tk.1.2 million, with an average requirement of imported machinery and equipment (average IDA subloan) of about $70,000, and in a number of cases, fairly large recurring foreign exchange needs for spares and raw materials. The sub-projects were heavily concentrated in two subsectors: food and allied industries (40%), primarily financing oil and flour mills; and textiles (40%), primarily financing warp-knitting and synthetic textiles. The remainder were in engineering (9%), printing and packaging (7%), and ice plant (4%) industries. The enterprises were concentrated in the Dacca and Chittagong areas and in terms of size, belonged to the upper end of SSI. Employment generation was modest at an average fixed cost per direct job created of about $5, The first Credit achieved modest results partly because of the situation prevailing in Bangladesh immediately following independence, which made effective project implementation difficult. The other reason was a lack of emphasis on project identification, promotion and extension work. The roles of the implementing institutions were not clearly defined and the overlap of BSIC/commercial bank functions deterred BSIC from carrying out promotional work and made procedures for appraisal and follow-up cumbersome.

6 - 2 - While the subprojects financed by the first Credit are only now reaching the repayment stage, other loans financed under these arrangements are heavily in arrears, mainly due to lack of proper supervision. The proposed project, therefore, seeks to strengthen promotion and extension work and to redefine the roles and responsibilities of the implementing institutions. It will be directed to smaller type of enterprises including cottage industries (CIs) and will generate a significantly higher number of jobs at a lower capital cost. Commercial banks, which have comparative advantages in the financing of SCIs, will participate directly in SCI development and will gradually assume full responsibility for appraisal and supervision. Simultaneously, BSCIC's 1/ activities will be concentrated mainly on promotion and extension, which it is better equipped to undertake. II. THE INDUSTRIAL SECTOR AND ROLE OF SMALL AND COTTAGE INDUSTRIES Economic Setting 2/ 2.01 The industrial sector, although accounting for only 8% of GDP, is an important element in the Bangladesh economy; it produces basic consumer goods, provides agriculture with key inputs, accounts for about 60% of export earnings and employs some two million workers. Industrial performance has at the same time a critical bearing on resource mobilization and is potentially an important source of rapid economic growth. The basic industrial structure has changed little since independence. Large and medium scale enterprises are mainly in the public sector and concentrated in jute, cotton textiles, paper, steel, sugar refining and engineering. They represent about 65% of the total value added in industry. After independence, the ownership and management of "abandoned" units were transferred to the Government. In addition, all jute, cotton textile and sugar mills were nationalized. At present, the public sector comprises about 400 units, including those under several public sector corporations; GOB has recently disinvested about 80 of these units. Small scale and cottage units, which account for about 35% industrial value added, are composed of roughly 50,000 SSIs and 500,000 CIs. 3/ 2.02 A large number of industries in Bangladesh originally had been established to supply the two wings which comprised Pakistan. The separation of these markets after independence substantially reduced capacity utilization. This was compounded by a shortage of foreign exchange to purchase raw materials and spares, inexperienced management, and labor unrest, especially in 1/ BSIC was merged with the Bangladesh Cottage Industries Corporation (BCIC) in May 1976, becoming the Bangladesh Small and Cottage Industries Corporation (BSCIC), which is now the agency primarily responsible for promoting the development of SCIs in Bangladesh. 2/ The general economic background is covered in the most recent economic report on Bangladesh, No BD, dated April 13, / Paragraph 2.03 gives the definition for SSI and CI units.

7 -3- the years immediately following independence. The overall index of industrial production only recently reached the FY70 level, although capacity utilization remains at less than 50% in many industries. After achieving about 90% of the FY70 level in FY74, total industrial production showed no sizable increase in the following year, largely because of a slackening of consumer cemand. As consumer demand has picked up gradually, industrial production during the last three years is estimated to have increased by about 10%. Characteristics and Role of SCIs 2.03 Definition. SSIs are officially defined in Bangladesh as enterprises with fixed assets, including land, of up to Tk 2.5 million. CIs are defined as activities that are carried on wholly or mainly by members of a family either as a full- or part-time occupation. As in many other countries, the data base on Bangladesh's SCI sector is weak. However, the following table is indicative of the characteristics of Bangladesh's SCIs compared with large-scale industrial units: Total Total Avg. No. of Total No. of Fixed Production Value Fixed Asset Value /a Workers/Unit Establishments Assets Workers Added (US$ ('000) -(US-$ million) million) Below US$6,500 (unregistered; CI generally) - sample 4 60, US$6,500-US$160,000 (SSI registered) - census 9 1, Over US$160,000 (Medium and Large) - census 295 1, /a Source: Manufacturing Census for registered firms, and Sample Survey of Unregistered Industries, (Project File 1). The data for units with fixed asset value below US$6,500 is on the basis of a sample and is not adjusted to cover the total population Employment Impact. SCIs provide work for about 80% of the total industrial labor force. From the above table, the average fixed assets per job is Tk 2,460 (US$160) for cottage/informal industries, and Tk 10,370 (US$570) for registered SSIs, compared with Tk 18,300 (US$1,200) for larger firms. However, it should be noted that the fixed assets/job figures are understated, perhaps by a factor of three to four, as they represent depreciated capital stock at book value and are based on weak data. Nevertheless, the figures are indicative and provide a basis for comparing SSI, CI and large firms.

8 The ratio of value added to fixed assets (book value) is 0.74 for large-scale industries, 0.97 for registered SSIs and 1.76 for cottage and informal industries (which are characterized by extremely low fixed investment). A review of 20 major subsector/product lines in which SSIs represent over 10% of value added indicates that value added/fixed assets in registered SSIs are about equal to that of large scale industries in the same subsector in 16 of the 20 subsectors (e.g., bakery products, knitted goods, plastic products, leather tanning, various metal products). Value added/fixed assets were higher in eight product lines (e.g., fruit and vegetable processing, wooden and metal furniture, leather products, glass products, and various repair operations). Subsectors in which value added per fixed assets was markedly higher for large industries were: edible oil processing, silk textiles, and pharmaceutical products. On the other hand, labor productivity is somewhat lower with value added per production worker at 16.5, 11.0 and 9.5 for large, registered SSIs, and CIs, respectively Economic Efficiency. Data on capacity utilization and cost structure which would help determine SCIs' efficiency more rigorously are not available. Data on profitability may also be misleading given relatively poor accounting practices and because profits are not always calculated on the same basis for small and large firms, particularly when computing labor costs. Further, the "mortality" rate of SCIs is high and data is available only for the more successful ones. Nevertheless, the figures give an indication since SCIs do not receive any special protection or subsidy. Excluding interest, overall profits/sales were 17% for large industries, 20% for registered SSIs, and 29% for cottage and other unregistered industries. Profits/fixed assets averaged 39%, 62% and 156%, respectively. Cottage and other unregistered small industries had highest profits to sales and to fixed assets in almost all sub-sectors in which they have major participation, with rural industries more profitable than urban unregistered industries. In the 20 major SSI subsectors, (accounting for about 80% of total assets of all registered SSIs), registered small industries showed higher profits/sales than large industries in 9 sub-sectors: fruit processing, oil milling, sawmilling, wooden furniture, tanning, leather products, perfumes, glass products, and umbrellas. SSIs averaged lower profits/sales in 9 subsectors: bakeries, flour milling, silk spinning, housing, metal furniture, pharmaceuticals, paper board printing, and soaps. SSIs and large industries had about the same profits to sales in 2 subsectors: grain milling and footwear. The most profitable subsectors (over 20%) were: fruit processing, bakeries, oil processing, tanning, pharmaceuticals and perfumes Present and Potential Exports. At present, SCIs contribute about 5% of Bangladesh's exports of some Tk 6 billion (about $400 million). SCIs account for at least 20% by value in several major export product lines, e.g. frozen and dried fish and seafood; handicrafts; hides and skins. Export growth potential for SCI products exists in several lines, e.g. fruit and fish processing, ready-made garments, sericulture, cane and bamboo, jute and other handicrafts. Each of these subsectors is labor intensive and the minimum economic size falls within the SCI definition.

9 Linkages with Agriculture. Industries processing agricultural and livestock products which have immediate expansion potential include fish drying, bakery and confectionery, fruit processing, poultry and rice milling in selected areas. Also, agricultural implement manufacture and repair have productive li-s to agriculture. In all agro-based industries, potential exists only in certain regions; resource availability and existing regional processing/manufacturing capabilities would have to be assessed carefully on a project-by-project basis Local Raw Material, Local Demand. Some 70% of SCIs are located in rural areas and 30% in urban areas. Given the small and dispersed nature of agricultural and natural resources, the appropriate size for processing industries often falls within the SCI definition (e.g. saw mills, fish drying, fruit bottling, rice and oil mills). Also, given the small size of the domestic market for many products, SCIs have the flexibility to respond to limited needs for goods, services and specialized products of the rural population (e.g. tailors, carpenters, repair workshops, blacksmiths, shoemakers) Need for Promotion. Project File 2 contains reviews of the problems and prospects for developing export oriented SSI projects in fruit processing, fish processing, and garments. It also contains descriptions of some subsectors with potential for CI development (wood, footwear, agricultural implements) for sales in the local market. These reviews illustrate the need for strengthening promotion and extension work to improve project design and implementation, particularly when sales are geared to export markets. Policy Framework and Industrial Finance 2.11 Policy Framework. GOB's industrial policies do not differentiate between SCIs and larger units although GOB recognizes the importance of SCIs in creating employment and furthering urban and regional development. GOB's industrial strategy has recently emphasized two main aspects. One consists of measures to improve the climate for private investment such as raising the ceilings for such investments, the payment of compensation for nationalized property, the encouragement of foreign private investment collaboration in selected fields, the expansion of opportunities for private participation in domestic and foreign trade, and steps to activate the capital market. The second element in the strategy consists of efforts to improve the efficiency of public enterprises although more is needed concerning measures such as the granting of more autonomy in matters of staff management and pricing policies. These are appropriate priorities. They should have salutary indirect effects on SCIs to the extent that they will stimulate demand for outputs of ancillary and service industries The main problems constraining the development of SCIs have been poor sub-project selection, lack of access to working capital and investment credit, a cumbersome institutional framework, and weaknesses in support institutions responsible for promotion and extension in areas such as marketing and technical assistance (para 2.14). Another major constraint has been the unavailability of imported raw materials due to a restrictive import policy resulting from the shortage of foreign exchange. The import policy has been

10 - 6 - liberalized substantially in the past year and firms are now able to apply for permits for up to 2 shifts (they apply for one shift and if they use up their permit amount within the first six months they are granted a second allocation). This new measure is encouraging although it is too early to determine whether the policy is being implemented effectively. Regarding procedures for sanctioning investment, the Investment Board has to approve projects requiring raw material imports over 20% and for SCIs, those of over Tk 1 million investment cost. IDA has recommended that SCI projects above Tk 1 million (or up to Tk 2.5 million) not requiring raw material imports over 20% be exempted from this approval process and this is under consideration by GOB Financial and Service Institutions. The Bangladesh Shilpa Bank (BSB) and the Bangladesh Shilpa Rin Sangstha (BSRS), each with loans outstanding of about Tk 1.2 billion as of March 1977, are the main sources of long-term loans for industry. About half by number of BSB's loans are to SSIs, but given the objectives and nature of the institution, loans are to relatively large SSIs with project costs ranging between Tk 1 million to Tk 2.5 million. Nine commercial banks (six of which are local and nationalized) with deposits totalling Tk 13 billion as of June 1977, concentrate mainly on the provision of working capital although they also make available long-term loans through roll-over of funds. Loans to SCIs (including small businesses) account for roughly 2%-4% of commercial banks' advances. The Investment Corporation of Bangladesh was recently reconstituted to provide mainly underwriting facilities and help reactivate the capital market, but its operations are not yet significant. Outside the banking system, GOB, through the Department of Industries, also provides long-term financing in the form of equity participation in public sector enterprises Apart from BSB, the main source of long-term finance for SSIs 1/ has been BSCIC in consortium with the commercial banks. This consortium arrangement has not worked well. BSCIC was responsible for appraisal of technical, organizational, economic, financial and market viability of the projects and for providing extension services and follow-up as well as project promotion. The commercial banks' role was limited to appraisal of creditworthiness of clients and handling disbursements and collections. This split of responsibility in areas that are interrelated (such as follow-up and collections), and the fact that different institutions made decisions on different aspects of a project during appraisal, resulted in the institutions blaming each other for mistakes without taking responsibility for failures. Thus, most of the loans made under this arrangement are in arrears. 2/ Further, given its involvement in lending operations, BSCIC did not give proper attention to promotion and extension work and most of the projects financed were in the upper range of the SSI definition in which, given BSB's involvement, the need was not as acute. A major objective of the proposed project therefore will be to have 1/ CIs generally have not had access to institutional finance and borrow from money lenders (para 2.15). 2/ Except for projects financed under Credit 353-BD where most of the amounts are not yet due (para 4.20 discusses actions being taken on BSCIC arrears).

11 BSCIC focus on promotion and extension work and to build up gradually the capability of the commercial banks to undertake appraisal and follow-up work independently (para 4.01) starting with the Sonali and Janata Banks. The commercial banks, with a developed branch network, more knowledge of local conditions better access to local entrepreneurs, and which in any event finance their working capital requirements, should be more effective in the appraisal and supervision of loans to this type of small enterprise. Further, small entrepreneurs will not have to approach different agencies for investment and working capital and commercial banks will have more leverage on collections given their day to day contact with clients for working capital financing Interest Rates. The Central Bank rate has been at 8% since July Working capital loans are generally given by commercial banks at 11-12% and BSB and BSRS generally charge 11.5% on their local and foreign long-term loans for industry. CIs, which generally have little access to institutional finance, borrow from moneylenders, whose rates vary although the effective rate is generally above 25% p.a. Deposit rates range up to 10.25% depending on the term although BSB and the Krishi Bank 1/ are allowed to pay one percentage point more than commercial banks on fixed deposits. The proposed Credit will be on-lent at not less than 11.5%, the current long-term lending rate, with GOB taking the exchange risk (para 6.01). The rate of inflation was about 35% in FY74 and FY75 although prices declined by 11% in FY76 and increased by about 11% in FY77. Inflation is projected at about 10%-14% for FY78 and if it remains at about 10% in the longer-run, which is a reasonable assumption given GOB's policies, onlending under the proposed Credit would be at a positive real rate of interest of 1.4%. III. THE PROJECT The Project Components 3.01 While the above discussion indicates considerable potential for SCI development, a modest credit of only $7.0 million is proposed. While the demand for credit probably exceeds these amounts, there are institutional constraints, particularly in promotion and extension work. The proposed credit seeks to remove these constraints. However, until experience with the revised institutional arrangements is seen, a cautious approach is justified. The proposed credit would finance the following components: (a) 100% of subloan disbursements to SCIs by two participating commercial banks (up to 70% of subproject costs) for parts of the local plus all foreign costs of fixed investments including permanent working capital 2/ (US$6.1 million; paras ); 1/ A specialized institution for agricultural finance. 2/ Defined as capital requirements for the initial stock of raw materials and supplies needed for the commencement of operations or the increase in such stock needed for the expansion of operations.

12 - 8 - (b) 100% of disbursements from a special capital fund (up to 15% of subproject costs) to provide part of the equity resources in cases in which the entrepreneur is unable to make the normal 30% contribution to the project cost (US$0.2 million; paras ); (c) An average 75% of the costs of a cottage industry component for jute and cane/bamboo handicrafts including credit; new product development; and specified extension from the Bangladesh Handicraft Marketing Corporation 1/ (US$0.4 million; paras ); and (d) 100% of the foreign and local costs of technical assistance to BSCIC, mainly for promotion and extension (US$0.3 million; paras ) Total project costs and IDA's contribution can be summarized as follows (Tk million): GOB/CB/ IDA IDA BSCIC Sponsors Foreign Local Local Local Total Subloans /a - SSI /b CI /b Special Capital Fund /c /c 3.1 Technical Assistance - BSCIC /d 7.7 Cottage Industry Component /e /a SSIs in this table are defined as projects costing between Tk 100,000- Tk 2.5 million. CIs are projects costing less than Tk 100,000. About 200 SSI subprojects and about 1,000 CI would be financed. /b Reflects incremental staff and overhead costs for disbursement period of the credit. /c Actually, the special capital fund would finance up to 50% of the contribution to the subproject costs by selected sponsors (or up to 15% of sub-project costs) but the figures reflecting sponsors contribution, as well as incremental staff, are shown in this table under the subloans component. /d This relates to incremental activities by BSCIC for promotion and extension (para 5.12). /e Storage and incremental bank staff. 1/ BHMC - see para 3.19 footnote.

13 Taking all components together, foreign costs would account for about 40% of total project costs. The proportion of foreign costs is relatively high because the domestic capital goods industry is not well developed. The proposed Credit would finance 70% of the total project costs including local currency financing which will facilitate reaching small industries and allow the use of local raw materials and equipment when available. To ensure that smaller industries are assisted, and since subprojects costing between Tk 1.0 million and Tk 2.5 million can obtain funds from BSB (para 3.03 b footnote), at least 40% of the subloans component of the proposed credit is allocated to sub-projects/enterprises with fixed assets below Tk 1 million. Credit Component - Size 3.03 The size of the credit component was derived by: (a) Screening BSCIC's pipeline of current SSI subloan applications. Proposals for subprojects with fixed assets greater than Tk 2.5 million were eliminated, as were applications in subsectors in which significant underutilized capacity exists; (b) Assessing potential for smaller SSI and cottage industry subprojects; BSB 1/ makes funds available for SSIs with fixed assets between Tk 1.0 million and Tk 2.5 million. Hence, the proposed project should pay more attention to smaller, more labor-intensive projects which often lack access to credit yet hold untapped opportunities; (c) Examining investment opportunities in high priority subsectors (export-oriented, labor-intensive, using local raw materials, supplying basic needs); opportunities and requirements of three subsectors (fish- and fruitprocessing and readymade garments) were investigated, and potential in 10 additional subsectors was reviewed. Subsector work by BSB, BSCIC, and other government bodies was used, as well as information collected from private industry; (d) Estimating BMRE 2/ and related permanent working capital needs of existing SSIs, to improve capacity utilization and output; r (e) Appraising the project administration capabilities of BSCIC and the two commercial banks. 1/ In FY77, Bangladesh Shilpa Bank made 45 loans to firms with fixed assets of less than Tk 2.5 million, up from 39 loans in FY76. Total project costs averaged Tk 1.5 million in almost all subsectors, with about 80% of the loans and 90% of the lending going for SSIs with total fixed assets above Tk 1 million. 2/ Balancing, modernization, replacement and expansion.

14 BSCIC's Pipeline. BSCIC's invitation for loan applications in early 1977 yielded about 114 SSI subprojects with total credit requirements of Tk 94.6 million (US$6.1 million). The majority of applications were for subprojects under Tk 1 million each; 73 of these small loans had total credit requirements of Tk 34.2 million (US$2.2 million) with subloans averaging Tk 330,000. The other 41 applications, for subprojects of Tk 1 to 2.5 million, were for credit of Tk 61.5 million (US$4.0 million). The mission did a further subsector screening of BSCIC's 60 accepted applications (see para 3.12 for criteria); at least 45 subproject proposals, with credit requirements of Tk. 43 million (US$2.8 million) appear viable: New BMRE No. Amount No. Amount (Tk million) (Tk million) Food/Allied Garments/Textiles Engineering Printing Chemical Miscellaneous Total While the credit would not necessarily finance these specific subprojects, this response (at least 45 reasonable proposals, totalling US$2.8 million) from a single invitation made by BSCIC, indicates effective demand for credit by SSIs The mission's analysis of SSI subsectors 1/ indicated investment potential in nearly 40 product areas, listed below. The 13 starred product lines (see table below) are those in which short term expansion potential has been identified; BSCIC's promotion and Commercial Bank prescreening would focus on these priority product areas. It is reasonable to expect that at least 5 subprojects would be developed for each priority subsector during the 2 year commitment period of the project; at this rate, 65 subprojects, averaging Tk 300,000 in credit requirements, would be financed, for a total of Tk 19.5 million (US$1.3 million). In addition, 2 subprojects could be expected from the remaining potential SSI subsectors mentioned below; in that case, 46 1/ S2T subsectors with high potential were identified by (a) analyzing market studies by the Export Promotion Bureau, Trading Corporation of Bangladesh, EEC Missions to Bangladesh, product-market export studies of key subsectors from various sources; (b) interviewing SSI entrepreneurs, BSB and BSCIC officers, large industries and private importexports; (c) assessing growth patterns of SSIs, regional demand, and raw material availability (using Census data, district and subsector studies by BSB and BSCIC; and data from public agencies and private enterprises); and (d) performing detailed analysis of 3 subsectors (garments, fish and fruit processing) and a shorter review of ten additional subsectors.

15 more subprojects would be financed, for a total of Tk 14.7 million (US$0.9 million). Using these estimates, about 110 projects with credit requirements totalling US$2.2 million would emerge, in addition to the projects in BSCIC's pipeline. While most of these 110 subloans would be for new projects, some engineering, textiles, and food processing industry loans would be for BMRE. Average Average Subsectors F v Assets (TOn) Subsectors Fixed Assets (r-k.,co Food/Allied Leather-based *Fruit bulk birrelling Footwear 120 for export 600 *Leather goods 90 *Fruit bottling, drying 150 Sporting goods 320 *Fish drying, smoking 100 Feed plants 625 Chemical *lice milling 350 *Bakeries 550 Cosmetics 180 Oil milling 325 Soap 120 Salt crushing 200 Plastic products 140 Confectionary 420 butter, ghee, cheese 300 Glass/Ceramics Poultry 225 Ice plants 525 Building materials 210 Glass products 120 Specialized Textiles Engineering Sericulture seeding/hunting 425 *Knitted goods - manual *Agricultural implements 220 machines 100 Cast iron foundries 240 *Xeadymade garments - Non-ferrous metal foundries 300 export, local 480 *Aluminum utensils 150 Hardware 330 Wood-based *Bicycle parts 110 Sawmills 225 *Light engineering workshop Structured metal products *Yurniture, other wood Pumps, compressors 120 products 120 Pqinting The amount of the proposed credit also takes into account the staff capability of the Sonali and Janata commercial banks, which have established Loan Cells to implement the project (para 4.05). In year 1, it is assumed that 1.5 Loan Cell man months would be required per SSI appraisal, moving to 1 man month in year 2. At this rate, with 18 members (out of 30) in the two Loan Cells working full time on prescreening and/or appraisal it is feasible to expect that 140 prescreened projects would be appraised in year 1, resulting in 100 approvals; in year 2, expected appraisal would be 200, with 150 approvals. This would be sufficient capacity for about 200 SSI subprojects expected to be approved during the two year commitment period of the project Cottage and Informal Industry Loans. In addition to the SSI proposals appraised at Headquarters, regional and branch managers would appraise smaller projects falling within their lending limits (Tk 5, ,000). Sonali has 23 Regional and 450 Branch Offices, while Janata has 14 Regional Offices and 420 Branches. Some regional offices have already begun making term and working capital loans to CIs (e.g., tailors, carpenters, small mills, rickshaw pullers). With local industry profiles prepared by BSCIC and short regional training

16 courses, each of the 37 Regional Offices (averaging 25 Branches each) could be expected to lend for at least 10 small projects during year 1, and at least 20 during year 2, for about 1,110 subloans during the two-year period. 1/ Assuming 1,000 subloans at an average of Tk 17,000 each, there would be some Tk 17 million or US$1.1 million in subloans to very small borrowers Conclusion. Combining these elements, the demand for IDA finance would be about US$6.0 million for fixed investment and permanent working capital credit. IDA financing would cover credit requirements for both organized SSI subprojects and cottage and informal industries. Project costs for modern SSI (Tk 100, million) subprojects requiring credit over a twoyear project life are estimated as follows: (Tk million) New Subprojects BM&R for Existing SSIs Project Project No. Costs No. Costs Food Allied Engineering Garments/Textiles Wood-based Leather-based Chemicals/Plastics Glass/Ceramics Total It is expected that 70% of fixed investment and permanent working capital finance will be covered by credit, and hence, about Tk 77.5 million (US$5.0 million) would be required. IDA finance would cover 100% of the credit provided by the commercial banks up to 70% of subproject costs. In addition, Tk 17 million (US$1.1 million) would be required for cottage and local industry subloans for investment and working capital, to be appraised at the branch level (para 3.08), totalling US$6.1 million to be allocated to the Credit Component of the Project Industrial Investment Schedule. The priority SCI subsectors and the distribution of projected credit fall within GOB's Industrial Investment Schedule. 1/ In addition to the these funds, GOB has allocated Tk 60 million to provide term and working capital finance to cottage and informal industries under a scheme which generally follows the same procedures as the proposed project (para 4.Old footnote) although all banks (with branches totalling over 2,000) would participate. The allocation of these funds would provide a relatively small proportion of potential demand for credit (including working capital) by cottage industries but, given institutional contraints, represents a realistic effort.

17 Credit Component - Subproject Selection Methods and Criteria 3.12 Several measures would be taken under the project to redirect lending efforts away from subsectors which are import-dependent, capital intensive and burdened -;th over-capacity. SSI applications would be screened and considered according to the following criteria, which were agreed at negotiations. While sub-project appraisals will also evaluate factors such as the entrepreneurs' capability and the projects' profitability, (a) to (c) above will be minimum criteria for selection, while (d) to (g) will be used as additional factors to determine priorities for investment. (a) Exports. Net foreign exchange earner with strong assurances of export sales (at least letters of intent from major importers) or local sales at competitive prices. (b) Local Raw Materials. At least 60% of the value of raw materials (except for net foreign exchange earners) and other inputs are local, with sufficient regional availability established during appraisal. (c) (d) (e) (f) (g) Labor-intensive Methods. Fixed investment per job created not greater than US$3,000. Relative Efficiency. Subproject with profit and value added to fixed investment at least as high as averages for large scale firms in the same product line. Capacity Utilization. Expected two shift production schedule, taking into account markets, raw material availability, and operations of other units in same subsector. Decentralization. Production sites would be located outside Dacca, Chittagong or Khulna. Agricultural Links. Related to agricultural productivity/ earnings either in processing crops or providing inputs or services to agriculture Subsector Work. As indicated by the above discussion, the proposed project should have a considerable impact in shifting past patterns in subproject selection. To encourage investment in priority subsectors and to forge links between BSCIC promotion and extension and commercial bank lending, several steps would be taken as discussed in Chapter V, particularly paras 5.05 and Special Capital Fund 3.14 Purpose. The purpose of the special capital fund would be to provide support to qualified small entrepreneurs who do not have sufficient

18 resources of their own to provide the normal equity contribution required by banks, i.e 30% of subproject costs. The fund would provide, in the form of "bridge finance", up to 50% of these equity requirements, or up to 15% of the sub-project costs. The "bridge" loans would be made at 7.5% interest with repayment within a period of 12 years or less, depending on the projection of liquidity and capital structure done during appraisal. The entrepreneur will be required to provide not less than 15% of the sub-project cost from personal resources, so that the entrepreneur has something at stake Size. The size of the fund will be Tk 3.1 million (US$200,000) which corresponds to about 2%-3% of the total costs of the SCI projects estimated to be financed. Since it is difficult to determine the demand for this supplemental financing, the Tk 3.1 million is a notional amount covering about 7% of the total contribution expected from sub-project sponsors. In case the amount allocated is too high, unused resources of the fund will be transferred to the loan funds of the commercial banks. If the amount is too small, GOB could replenish the fund from its own resources Eligibility Criteria. Projects eligible for finance from the fund will be SCI projects in which all relevant aspects, including the entrepreneur's capability and ability, have been appraised by one of the participating commercial banks and the project has been found eligible for finance except for the equity gap. In addition, the entrepreneur will be subject to a more detailed scrutiny than is normally the case, to ensure need and above-average capabilities/skills Administration. The special capital fund will be administered by the participating commercial banks under the Bangladesh Bank's (Central Bank) supervision. The commercial banks are the most appropriate institutions to administer the resources, since this will relieve the entrepreneurs of the burden of applying for funds from and being appraised by another institution. Also, the commercial banks would be appraising the project in any case, and would normally be in the best position to judge the character and integrity of the entrepreneur Since the commercial banks would be providing loan funds of up to 70%, special arrangements would be needed to induce the banks to implement the scheme; banks might be hesitant in exposing themselves to such an extent (85%) in any one project. Therefore, in addition to the normal risk coverage of 30% (para 4.04), banks will receive an interest spread of 5.5% for the special capital fund compared with 5% for normal SCI lending operations (para 6.01). The interest spread/quarantee coverage package will be kept under close review by IDA so that appropriate revisions can be suggested to GOB as necessary. Cottage Industry Component 3.19 General. The project would address the needs of cottage industries in three ways: (a) cottage industries would be eligible for investment and working capital finance under the general credit component; (b) BSCIC would create a department responsible for integrated support to specific cottage industry subsectors to develop designs, help organize village-level extension

19 services, and work closely with BHMC 1/ and other marketing organizations (paras 3.25, 4.15); and (c) a pilot cottage industry component, described below, would consist of about US$400,000 for marketing, technical, financial and some raw material support to jute, cane and bamboo handicraft units in selected areas A separate cottage industry component should be included in the project since: (a) considerable progress in exports of jute, cane, and bamboo handicrafts has been made, and export expansion potential exists for these products; (b) there is need for credit accompanied by subsector specific forward and backward linkages (marketing, design, raw material supply, organization), to increase sales, employment and earnings; (c) the component would benefit the poorest portion of the population, primarily living in rural areas, in the form of higher income and employment; (d) it would be based mainly on local raw materials; and (e) these subsectors are very laborintensive Elements and Costs. The cottage industry component would consist of: (a) Credit to jute, cane, and bamboo artisans and cooperatives, with IDA financing 100% of disbursement for subloans made to provide up to 70% of funds for working capital and equipment needs; 2/ average credit would be US$100 per artisan in jute handicrafts and US$250 in cane and bamboo totalling Tk 4.2 million (US$27,000) over 2 years; 1/ Bangladesh Handicrafts Marketing Corporation (BHMC), founded in 1972 as the Bangladesh Handloom Export Corporation (BHEC), is a subsidiary of BSCIC with a separate Board of Directors headed by BSCIC's Chairman and with 16 officers. BHMC (a) undertakes marketing, organization and export sales of handicraft and handloom products (purchasing, quality control, reselling, warehousing, contracting agents); (b) establishes, organizes and promotes production units catering to these markets. During the last 4 years BHMC has exported about Tk 2.3 million in handicrafts, holding about fourth position in Bangladesh handicraft exports. BHMC plans to export Tk 2.5 million in and already has Tk 1.9 million in firm orders. In both jute and raw cane/bamboo products BHMC has penetrated new markets and organized local production units, acting as a catalyst for private exports. GOB has been considering whether BHMC should limit its role to more passive export promotion, e.g., catalogues, fairs, display centers. However, there are advantages in BHMC continuing its role in selling and organizing producers, working closely with the Trading Corporation of Bangladesh, the Export Promotion Bureau and private firms in areas (such as for the proposed component) where there is need to extend geographical and/or market coverage and to act as catalyst for private sector involvement. 2/ Mainly raw materials and tools.

20 (b) Market Development. BHMC's (para 3.19 footnote) sales trips and participation in fairs in present and potential markets would be funded with IDA financing 100% of foreign exchange costs to-make arrangements with sales agents and export outlets; this direct sales approach has been the most effective for BHMC. BHMC is expected to perform a catalytic role for private marketing agencies, establishing broader market channels, developing new export product lines, and extending the production network to more remote areas of Bangladesh (Tk 400,000; US$26,000); (c) Design Consultancy. While importers may specify designs, BHMC and BSCIC need to establish a set of popular standard items to allow producing to stock without major risk of obsolescence; designers, hired for short terms and specific tasks, can develop these lines and prescribe production techniques; IDA would finance 100% of foreign and local costs (Tk 500,000; US$32,000) 1/; (d) Training and Extension. IDA would finance one project coordinator, 3 central trainers and 12 thana-level extension workers to work with production units (1,500 participants) in the jute, cane and bamboo schemes. The project also would finance 5 additional quality control inspectors for BHMC. This incremental staff would be recruited as a condition of disbursement for the cottage industry component of the IDA Credit. Subsector specific extension workers and inspectors would enable BHMC/BSCIC to expand the local production network by building on existing and new production groups in selected thanas and linking them to marketing organizations, credit sources, and assistance on improved methods and handtools (Tk 900,000; US$59,000); and (e) Storage. GOB has agreed to become responsible for providing storage facilities which would be built or rented. Construction costs for finished product storage facilities in the three fairly remote districts that could allow bulk handling are estimated at Tk 150,000 (US$10,000) To summarize, elements, responsibilities, and costs of the cottage industry component would be: 1/ UNDP is now preparing a technical assistance project for cottage industries which may include provision of designs specialists for jute and cane/bamboo handicrafts. If this is included, the allocation of $32,000 would be transferred to the credit component of the project.

21 Agency IDA IDA Responsible Foreign Local Total (Tk million)---- (i) Credit to artisans Commercial banks /a (ii) Market development BHMC (iii) Design consultancy BSCIC/BHMC (iv) Extension/Inspection BHMC/BSCIC (v) Storage (counterpart funding) BHMC Total /a Excluding Tk 0.1 million for staff and overhead costs and Tk 1.8 million for sponsors contribution Benefits would include: (a) increased employment and earnings for 1,500 direct participants 1/; (b) annual increased exports of at least Tk 3.5 million by the fourth year of the project; (c) expanded channels in high volume, utility product markets; and (d)improved packages of sub-sectorspecific institutional support Coverage. The cottage industry component would focus on districts with concentrations of jute, cane and bamboo handicrafts. In jute, the major handicraft centers presently are Comilla, Barisal, Tangail and Dacca, although many districts have a tradition of jute handicrafts. While certain villages in the Dacca District are well covered by existing market organizations, most areas are sparsely covered. The major cane handicraft areas are in Sylhet, Bogra and Pabna. Bamboo handicraft production centers exist in Comilla, Dinajpur, Sylhet and Mymensingh. GOB/BSCIC has provisionally selected three districts for each product, with about 35 village units and 1,500 participants in total. Each village unit would be a cooperative or producer association consisting of members each. In addition to the direct benefits to 1,500 craftsmen, other artisans would benefit indirectly through the strengthening of BHMC, BSCIC, and other institutions and through the increased availability of credit from commercial banks Implementation. The cottage industry component would be implemented by BHMC, BSCIC, and participating commercial banks. In addition, BSCIC would work closely with the Integrated Rural Development Program, voluntary organizations operating at the field level, and commercial marketing agencies. BSCIC would be responsible for adapting designs, developing new products and for providing general support to BHMC. BHMC, with BSCIC's assistance, 1/ Cost per job created of about US$180 for credit, and US$86 for marketing and technical support.

22 would work on the: (a) development and diffusion of new techniques to improve quality, productivity and marketability, and (b) organization of production groups, training and raw materials. BHMC would also be responsible for (a) arranging export sales, using BHMC, private, voluntary or government channels (Trading Corporation of Bangladesh or Export Promotion Bureau); (b) direct selling and market intelligence or arrangements with agents or importers; (c) collection or reception of finished goods, inspection and quality control; and (d) packing and shipment. Janata and Sonali Banks would be responsible for making small loans to individual producers or associations which are covered in the marketing scheme. Although at the present time this is not a major constraint, the banks could also, once a number of specific product lines become well established in foreign markets, provide credit from their own funds to BHMC to finance advances and stocks in addition to that covered by normal letters of credit Background on Jute Handicrafts. Jute handicrafts represent a strong handicraft subsector with immediate expansion potential: 3.27 As evidence of its market potential, jute handicrafts accounted for about 60% of Bangladesh's Tk 14.1 million handicraft exports during FY77 and was the leader in the rapid growth of Bangladesh's handicraft exports from Tk 306,000 in FY74 to present levels. BHMC was the first to penetrate jute handicraft export markets (FY74) and this successful experience led private market agencies into the area. However, expansion to date has been based largely on a few product lines and narrow, specialized handicraft outlets; market organizations have begun diversification into new utility lines e.g. plant hangers, handbags, placemats, wall hangings and accessories. This diversification should be expanded by BHMC and other marketing organizations and market channels should be increased to include direct sales to department stores and chains Over 100,000 women, located in villages throughout Bangladesh, are involved in handicraft production using jute as the major raw material; according to BHMC and private exporters, women can earn Tk 330 (US$21) a month if marketing and production organization is present. In addition to improving earnings for these women, expansion of jute handicrafts could build on a larger production base if training were introduced. 1/ 3.29 Equipment involved in jute handicraft production is minimal, and value added per fixed asset is extremely high. While the Philippines and other handicraft exporters have introduced selective mechanization for similar items. Bangladesh with its low labor costs and abundance of low cost, local raw materials can compete successfully relying largely on existing manual methods. Unlike other handicrafts areas, where technical problems are a barrier, jute handicrafts can build upon the existing methods and skill base. 1/ Under the project BSCIC would coordinate with the Bangladesh Women's Rehabilitation and Welfare Foundation, the Integrated Rural Development Project, and voluntary agencies which train women's groups, as the extension workers for jute handicrafts are expected to be women.

23 Geographical concentrations of jute handicrafts do exist which enable development of extension, marketing, organization and storage. Most programs today are concentrated in parts of the Narsindi Zone of the Dacca District, which accounts for about 80% of BHMCs and over 60% of Jute Works' (para 3.31) coverage. Other centers for jute handicrafts exist in Comilla, Tangail, Khulna, Rangpur, Noakhali and other areas of the Dacca District. Jute raw materials are available in abundance throughout Bangladesh. 1/ It would be possible to expand jute handicraft production and marketing to appropriate areas by developing decentralized collection and storage of goods, quality control agents, extensionists, and other services including advancement of working capital to the producers Organization of village cooperatives linked to a marketing organization has already proven successful. One example is the case of Jute Works, a private voluntary organization which markets these products. Each of the 100 women's cooperatives (6,000 members) in the Jute Works' network elects a leader who is responsible for collection of the handicraft products, quality control at the village level, and movement of the products to the Dacca head office. Jute Works' central organization provide sales organization, designs, quality control, export shipment, and training of cooperative leaders. The extension worker for the jute cottage industry component of this project are expected to be women Background on Cane and Bamboo Handicrafts. According to the "Report of the Committee by the Development of Cottage Industry, July 1977", there are about 20,000 units involved in bamboo handicrafts, employing about 53,700 workers producing about Tk 13.1 million in bamboo products. Most of the products are sold locally; this year has been marked by rapid growth in exports of bamboo products. In cane, 2,500 units employ about 6,000 workers, producing goods valued at about Tk 11.5 million. Cane products account for over Tk 2 million in export sales during 1976/77. Cane and bamboo handicrafts now account for about 30% of all Bangladesh's handicraft exports On marketing and production, the assessment by private exporters and BHMC is that cane and bamboo products represent a major area for immediate future growth in handicraft exports although these products face competition from other areas of East Asia, especially China and the Philippines. As with jute, cane and bamboo handicrafts represent utility products (baskets, knockdown furniture, trays) which have proven marketability in Europe and the United States. Both cane and bamboo products require simple hand tools utilized by skilled craftsmen, although improvements in more specialized hand tools and selective introduction of equipment could have a marked effect on productivity and costs in the subsector. Production of cane products is highly concentrated in Sylhet and Pabna, which would facilitate marketing and production organization. Bamboo handicraft units are more scattered with large concentrations in Khulna, Sylhet, Pabna, Comilla, Dinajpur, Noakhali, Jessore and Mymensingh. 1/ With the exception of North Bengal.

24 Regarding raw materials, unlike jute products, problems of quality and quantity of raw materials exist in bamboo and cane. In cane, most of the raw materials locally available are of the smaller variety which are appropriate to making of trays, smaller pieces of furniture and baskets. In the high potential area of knocked down furniture, however, about 40% of the raw material is imported from Singapore and India. As a result of large demand for bamboo products in the local market, bamboo producers have begun to cut the bamboo before its three-year gestation period, thus requiring additional treatment in varnishing and chemical disinfectant in order to produce bamboo handicrafts of export quality. In the long term, it would be necessary to develop a project in which a component for systematic production of cane and bamboo is included. However, supply is sufficient to develop a limited project focusing on smaller cane and bamboo items, using largely local raw materials, which would have only marginal impact on local price and availability of the major raw materials. Imported cane from India and Singapore, available at nearly the same price as local cane, could be used for furniture. Technical Assistance Component 3.35 The Project's technical assistance component will focus on BSCIC's promotion and extension activities (Chapter V), as this is a difficult endeavor and a key to the future development of SCIs. Other aspects where technical assistance is required, such as training of commercial bank staff, have been arranged separately (para 4.07) and are partly financed from IDA's Technical Assistance Credit 622-BD. Advisors in the following fields and for the periods indicated would be stationed with BSCIC and financed from the proposed Credit as a GOB grant: (i) Promotion of SSI; 12 months Assist and guide BSCIC in undertaking sector studies and preparing project profiles as outlined in Chapter V; (ii) Industrial Engineering for SSI; 12 months Assist and guide BSCIC in building up a unit which will collect, analyze and publish information on local and foreign machinery and production processes which are suitable for SSIs in Bangladesh. The data collected will be made available to the commercial banks to improve the procurement process. (iii) Short-term consultants, total of 24 months To meet various requirements for specific tasks (particularly, specific subsector studies including at least 6 man-months for extension work through diagnosis of problems of existing industries in certain subsectors) which will strengthen BSCIC and/or its services. Guidelines on the types of studies to be undertaken are given in paras 5.05 and 5.08.

25 The proposed technical assistance concentrates more on SSIs than in CIs as there is a separate cottage industry component and because UNDP is preparing with BSCIC a technical assistance project for cottage industries (para 5.11). Paragraph 5.12 shows detailed costs estimates for the above consultants, and for other BSCIC promotion and extension activities. IV. INSTITUTIONAL ASPECTS Institutional Framework 4.01 As discussed in para 2.14, the institutional framework for assisting SCI's so far has been weak. The proposed project introduces institutional reforms involving assignment of more specific responsibilities among the agencies involved in the sector and the establishment of programs to strengthen SCI support and the institutions concerned. The following summarizes the institutional arrangements. (a) Project implementation responsibilities would be allocated as follows: Activity (i) Promotion (ii) Appraisal Agency Responsible BSCIC Commercial banks/bscic (first eighteen months) and commercial banks (second year) (iii) Loan Sanction/ Disbursements Commercial banks (iv) Follow-up (v) Extension Services (vi) Credit Guarantee Scheme (vii) Special Capital Fund (viii) Jute Handicrafts and Cane/Bamboo (Cottage Component) Commercial banks BSCIC Bangladesh Bank Commercial Banks BHMC, BSCIC, commercial banks (b) BSCIC will be mainly responsible for project identification and promotion and for providing extension services. Promotion would entail undertaking area/subsector surveys to identify economically viable investment opportunities, determining linkages, preparing feasibility studies and

26 project profiles, conducting market surveys, and coordinating the work of the agencies involved in assisting SCIs. Organized promotion will be a key aspect of the proposed project and advisory services to assist BSCIC in this work will be made available from the Credit. Improvement of BSCIC's extension service functions (providing marketing advice and other technical assistance) will be another key aspect of the project. BSCIC's promotion and extension service work for the proposed project are discussed in Chapter V. (c) Appraisal work will be undertaken jointly by BSCIC and the commercial banks as a transitional arrangement of 18 months (see below). 1/ As SSI projects generally are not complex and as BSCIC would have already prepared, in many cases, a feasibility study, the appraisal should not take much time (maximum 1 month). It would concentrate mainly on key aspects such as management, market, and liquidity prospects. In contrast to past practice (para 2.14), the commercial banks will be accountable for all aspects of the investment decision, would perform all the follow-up work and will take full responsibility for loan defaults and other problems faced at the project supervision stage. Project appraisals will be reviewed by a committee with representatives of BSCIC, the commercial bank involved, and possibly the Bangladesh Bank,the Central Bank (refer to f). If there is disagreement between BSCIC and a commercial bank on financing a project, the commercial bank's management will make the final decision. BSCIC will not be responsible for follow-up although it will keep contact with the client through its extension service responsibility. However, wherever required, BSCIC will provide technical support to the banks in follow-up. The concept of joint appraisal has disadvantages; having two institutions decide whether a loan should be made or not will be somewhat cumbersome and may create friction if there is disagreement between the commercial banks and BSCIC. However, it will be prudent to maintain it for the time being, given the lack of experience by commercial banks in project appraisal (para 4.04). The fact that, in contrast to the past, one institution will now take responsibility for the final decision and the same will be responsible for project supervision, will allow pinpointing responsibilities for success and failure, easier management and control, and taking corrective action on a timely basis. Joint appraisal will be a temporary arrangement until the commercial banks acquire enough experience to undertake this activity on their own, which is estimated to take about one 1/ Loan applications could be received either at a bank's branch or at BSCIC which would be copied to the other institution. An appraisal team would be staffed with representatives of both BSCIC and the commercial bank which will produce a joint report.

27 year to eighteen months. An understanding has been reached with GOB that the commercial banks will take over full responsibility for appraisal eighteen months after Credit effectiveness unless a review by IDA and GOB determines that the commercial banks have not acquired the capability to carry out this activity on their own. Further, this division of responsibility between commercial banks and BSCIC will also apply to other SCI financing proposals. (d) As a start, only 2 commercial banks would participate in the project: Sonali and Janata. Focussing on two banks will facilitate institution building work and will still allow for competition. Both banks have established Small Industry Loan Departments at their head office each to be staffed initially with some 15 officers transferred from other departments in the bank and/or recruited from outside. The banks' branch network would keep contact with the entrepreneurs, assist them in filling application forms and ensure follow-up. As most bank branches are not sufficiently strong at the moment, the Small Loan Departments at the Head Offices will be responsible for appraising projects costing over Tk 100,000. Smaller projects usually would be appraised by the banks' regional offices within the current sanctioning limits of the banks. 1/ (e) The bank's security requirement would be relaxed somewhat to a debt/equity ratio of 70:30. The special capital fund (paras ) would provide part of the equity resources if a project is a worthwhile one and the client, although suitable, does not have enough funds to make its 30% contribution. (f) A credit guarantee scheme managed by the Bangladesh Bank (the Central Bank) will be established to cover 30% of the commercial bank's risk in SCI lending for a fee of 1% per annum on amounts outstanding paid by the commercial banks. The purpose of the scheme will be to provide an incentive to commercial banks for increased lending to SCI given that this is a more risky operation than the normal commercial bank activities. The Bangladesh Bank has prepared a draft Credit Guarantee Scheme (Project File 4) which is satisfactory. The establishment of the Credit Guarantee Scheme by the Bangladesh Bank is a condition of effectiveness. 1/ GOB recently allocated Tk 60 million for financing small projects (loans below Tk 100,000) at the district level. Such projects will be jointly appraised by BSCIC and the commercial bank branches and submitted to a committee comprising representatives of both (with the final decision in case of disagreement with the commercial banks). The procedures followed for these loans would be the same as discussed in para 4.01 (c), although all commercial banks are involved, decisions are made at the district level and the committee also includes the District Commissioner. These arrangements have the same 18-month transition period mentioned in para 4.01(c).

28 (g) Concerning the cottage industry component for jute handicrafts and cane and bamboo, as discussed in more detail in para 3.25, BHMC will be responsible for marketing,quality control and technical assistance, BSCIC in coordination with BHMC for product development, and the commercial banks for providing credit. (h) Arrangements have been made for representatives from BSCIC, the Janata and Sonali banks, and the Bangladesh Bank to meet at least quarterly starting when the credit is approved by IDA, to discuss and solve problems in project implementation and to ensure coordination in the work of the different institutions involved Therefore, the main implementing agencies for the proposed project will be the Sonali and Janata Banks and BSCIC. Project File 5 contains a report on the two banks and Project File 6 on BSCIC. The issues concerning these institutions, as they relate to the proposed project, are discussed below. Commercial Banks 4.03 Background. The Sonali and Janata Banks are the two largest commercial banks in Bangladesh. Sonali was formed in 1971 from the National Bank of Pakistan and as of December 31, 1976 had 8,566 staff (of whom 1,162 were officers) and 450 branches; its total assets were then Tk 6.4 billion. Janata came into being in 1972 when it took over the East Pakistan assets and liabilities of the United Bank Limited of Pakistan. As of December 31, 1976 Janata had 6,140 staff (of whom 1,405 were officers), 420 branches, and total assets of Tk 8.4 billion. Both Sonali and Janata have been already associated with IDA as implementing agencies for a number of agricultural projects Commercial banks in Bangladesh have little experience in making longterm loans for SCIs although they have general experience in loan supervision matters. While there are some organizational and financial weaknesses (para 4.09), the Sonali and Janata Banks have adequate management, staff and procedures. Both are headed by individuals with strong experience in development banking. The staff is well qualified with the business background necessary to appraise SCI credits where the key issues are an evaluation of the management, market prospects, liquidity, and technical assistance needs and, after completion of the proposed training program (para 4.07), should be able to carry out effectively the tasks under the institutional arrangements for the project. However, it would be prudent for BSCIC to provide assistance in appraisals initially. BSCIC's appraisal standards are weak in market analysis although the technical assessment is usually adequate. As the commercial banks already have some expertise in market and financial analysis, this will help ensure complementary expertise and that the key aspects of SCI projects will be satisfactorily evaluated Organization. Sonali and Janata Banks each have established a new Small Industries Loan Department with responsibilities for preparing appraisals and undertaking follow-up work for term loans and working capital to SCIs.

29 Staffing. The following staff and skills for the Small Industries Loan Department at the Head Office of each bank are being recruited and/or transferred from other sections of the bank: I Department Manager 3 Financial analysts for appraisal 4 Financial officers with business background for project supervision (follow-up) 2 Procurement/disbursement officers 3 Engineers/technical officers 2 Project economists 15 Officers Given that appraisals of SCIs are generally simple, this staff should be sufficient for the project. 1/ Both the Sonali Bank and the Janata Bank have agreed to recruit or transfer to its Small Industries Loan Department the officers mentioned above in time to complete their training (para 4.07) Training. The training program for these officers would comprise: (a) External. 3-4 officers from the Small Industries Loan Department in each bank would be seconded for two to three months to commercial banks in India with experience in this field, for on the job training. Two Indian banks have indicated their interest in providing the training starting in early An application from GOB for funding the costs of the external training out of IDA's Technical Assistance Credit to Bangladesh has been received and it is expected that the training will start in June (b) Internal. All officers in the Small Industries Loan Department in each bank would attend a course of about 2-3 weeks duration on project evaluation and project supervision. BSB and BSCIC as well as BIDS, which collaborated with EDI on a recent Industrial Projects Evaluation Course in Dacca, would assist in the training. Regional Managers would receive 1-2 weeks training on appraisal of CIs and the use of sub-sector profiles. 1/ Para 3.07 gives the basis for estimating staffing requirements. Appraisal of CI projects, which do not require much technical input, will be done at the branches. For SSIs, the bank appraisals would concentrate on the marketing strategies, distribution channels, management and entrepreneurial capacity, projected technical assistance needs, and liquidity position since these are the usual critical areas of small-scale projects. Regarding follow-up, each project will be visited at least quarterly and thereafter performance reports would be prepared by the commercial banks' staff although much of this can be done at the branches. Further, clients will be required to submit quarterly reports to the banks.

30 The Sonali and Janata Banks have agreed to complete the above training programs by September 30, Strategy Statements. On the basis of discussions with IDA, the Sonali and Janata Banks have prepared Strategy Statements (Project File 7) which (i) outline the banks' credit policies regarding SCIs; (ii) quantify its expected involvement in this subsector over the next 3 years; (iii) outline its priorities in project selection; and (iv) describe expected developments in its organization concerning SCIs, including appraisal and follow-up procedures, role of branches, and staffing/training plans, as discussed above. These Strategy Statements were approved in May 1978 by the commercial banks' Board of Directors Other Issues Relating to the Commercial Banks. Although not directly related to the project, the following matters have also been discussed with the banks and GOB: (i) in view of GOB's policy to develop the outlying regions, it has asked banks to expand their branches substantially 1/ which will affect their profitability and staff situation; (ii) financial planning, particularly long-term, is weak; (iii) due to lack of data, the impact on the banks' financial position of their assets and liabilities relating to Pakistan is not clear; 2/ and (iv) while the banks have a weak equity base and relatively low profitability, much of their profits are not being allowed to be ploughed back as reserves. 3/ These issues are complex and as they affect all the commercial banks, it is difficult to treat them in isolation for the two commercial banks concerned with the proposed project. Further, they do not affect the proposed project directly, a situation similar to the IDA agricultural credit projects which have used the Sonali and Janata Banks. However, IDA will keep these matters under review and will continue discussions on them with GOB in the course of this project's supervision. 1/ While the principle of branch expansion is appropriate, this should be planned on the basis of the banks' implementing capacity. For example, the Sonali and Janata Banks have been asked to expand their branches by and respectively, by end However, the view of the banking sector is that actual implementation will be in accordance with the capacity of the banks. 2/ Although GOB has stated that those banks that were negatively affected are being supported and will continue to be supported by GOB. 3/ About 55% of the commercial banks profits are set aside as provision for taxation (after deducting other provisions) and thereafter approximately 30%-50% of the remaining amount is transferred to Government in terms of Section 25 of Bangladesh Bank Nationalization Order of In effect, this represents a dividend pay out ratio of 30%-50% and a dividend rate of about 50% which is high relative to the banks' profitability and may not allow an adequate build up of their reserves.

31 BSCIC 4.10 General. While BSCIC has been given the responsibility of promoting and assisting the development of SCIs in Bangladesh, and BSIC's Charter provided for promotion and technical assistance to SSI, activities in this field have been limited. BSIC's main activities over the last few years were: (i) distribution of raw materials; (ii) evaluation of requests by SSIs for foreign exchange for the purchase of imported machinery; and (iii) management of industrial estates. Further, BSIC's staff and work have been centralized in Dacca with little contact with realities in the field. The proposed project, therefore, as one of its principal objectives, aims at assisting BSCIC to focus more on its promotional and technical assistance role for the development of SCIs. These are key aspects given the low level of entrepreneurship and management skills in the country. The responsibility for (i) is no longer with BSCIC, and for the proposed project, the responsibility for (ii) will be transferred to the commercial banks. While (iii) will continue to be a BSCIC function, it should not impose an undue strain in the organization as it is a separate activity and not expected to expand in terms of staff requirements. At the same time, GOB supports regional decentralization of BSCIC's activities. Thus, BSCIC should be able to devote much more of its attention to promotion and extension. The proposed approach for improved promotion and extension is outlined in Chapter V and the discussion below deals with the main institutional issues facing BSCIC that should be dealt with to achieve this objective Board of Directors. BSCIC's Board consists of six members, the Chairman and five full-time Directors. The Chairman is the Chief Executive and the Directors are all functional, each in charge of a Division and responsible for the day-to-day management of BSCIC. The Chairman and the Directors are all appointed by GOB and in the past, the quality of Directors has been mixed and there has been little management continuity due to frequent transfers. GOB has indicated that in future BSCIC Directors will normally be retained in their post for at least 3 years to ensure management continuity Overseeing Committee. Except for its reporting responsibility to the Ministry of Industries, BSCIC has not had an overseeing body with the specific task of guiding its policies and activities, coordinating them with those of other organizations and reviewing and evaluating performance. As its Board is functional, and the Directors are busy with the day-to-day matters, such a body could play a useful role. GOB has recently assigned this task to the 10 member "National Committee for Development of Small and Cottage Industries", which in general has adequate representation and meets quarterly Quality of Staff. To enable BSCIC to attract and retain better quality staff, BSCIC's salary scales were recently increased and set at the same level as other public sector organizations dealing with industry. In addition, there is need to improve staff policies and training. On the basis of discussions with IDA, BSCIC is taking steps to improve the motivation of its staff. These include improvements in the staff evaluation system, and the establishment of policies to reward good performances through extra increments and rapid promotion, and where possible no increments for or layoffs of continually poor performers. Also, BSCIC has prepared a training

32 program. The training program covers a range of topics and would be useful particularly for new staff although there is need to complement this with specific training in promotion and extension. As part of this effort, BSCIC has prepared a manual for extension work and is using the training facilities offered by Technonet Asia. Further, the advisors discussed in para 3.35 will prepare training programs, particularly for promotion work. These training activities will be kept under review so that thay can be suitably expanded in due course Quantity of Staff. BSCIC's total personnel stands at about 1,300, of whom 292 are officers. Of the officers, some 140 are stationed at head office while 91 are at the regional offices or industrial estates, and 61 are employed in special projects. Starting in 1977, BSCIC has been making efforts to decentralize its organization and to increase the proportion of operational staff by stationing more personnel at the district level, particularly extension workers (para 4.16). BSCIC's personnel is somewhat excessive, particularly non-officers, and BSCIC expects to reduce this through attrition Organization. Apart from decentralization, discussed below, BSCIC has agreed to some changes to its organizational structure which would entail mainly having BSCIC activities for small and cottage industries placed under different directors. This will allow a more coordinated approach for promotion and extension work and will help BSCIC in giving appropriate attention to both small and cottage industries as the approach to their development is different. 1/ The proposed adjustments would not result in any major changes in BSCIC's organization in number of directors and staffing requirements, although there would be need for some transfers of staff between departments to strengthen promotion and extension Decentralization. Most BSCIC functions, staff, authority, and responsibility have been overly concentrated in the Dacca Head Office. In an effort to decentralize, BSCIC took some steps during 1977, particularly to strengthen the Rajshahi Regional Office 2/. More recently, it established procedures for delegating more powers to district officers and has started a program to station industrial extension officers at the district level. Some 17 extension officers have been posted at BSCIC's districts offices and BSCIC has indicated its plans to station another 30 extension offices at the districts during FY79 (para 5.07). 1/ Chapter V outlines the recommended organization and staffing for promotion and extension and para 4.16, the decentralization from the head office. 2/ SIDA (Swedish International Development Agency) is considering providing finance for projects and technical assistance in Rajshahi although this is not expected until after FY78. Nevertheless, IDA and SIDA should coordinate closely to ensure consistency and avoid duplication of efforts.

33 Policy Statement. On the basis of discussions with IDA, BSCIC has revised its Policy Statement on SSIs to reflect its new focus on promotion and extension work, (refer to Chapter V) and its withdrawal from appraisal and financing of loans. Also, the new Policy Statement (Project File 7) describes BSCIC's role in cottage industries reflecting the merging of BSIC and BCIC Accounts. The audit reports on BSCIC's accounts and activities have stated a number of shortcomings mainly concerning the availability of supporting documents related to specific transactions and the adequacy of provisions for doubtful accounts. BSCIC has discussed with its auditors their observations on BSCIC's accounts and is taking the necessary steps to eliminate these reservations for future audits. Concerning provisions for doubtful accounts, adequate allocations are expected once the project by project review discussed below is completed Loan Portfolio. BSCIC's portfolio comprises projects financed by BSCIC itself, through joint financing with BSB, and through joint financing with a consortium of commercial banks. These projects have not been adequately supervised mainly because the responsibilities for this activity were previously not clearly defined. Thus, the bulk of these loans are in arrears, many of them for over 8 years. Further, the auditors of BSCIC's accounts were unable to establish whether provisions for doubtful debts are adequate. This problem has to be tackled. However, an involvement by BSCIC in banking activities would distract it from promotion and extension work. As in any event other agencies also have had responsibility for supervising a substantial portion of this portfolio, it should be possible to find an appropriate solution to this problem. IDA has recommended that (i) a joint team of BSCIC and relevant commercial banks staff review all overdue loans under BSCIC/ Consortium Banks; (ii) a team of BSCIC and BSB review the BSCIC/ BSB portfolio; and (iii) BSCIC itself review all overdue loans financed from its own funds. The review of BSCIC/Consortium banks overdue loans (about 80) has not yet started but the reviews of BSCIC/BSB overdue loans (about 350) as well as BSCIC's own overdue loans (about 600) are underway and should be completed by March 31, Concerning the BSCIC consortium banks portfolio, the review did not start yet because the commercial banks are already fully responsible for loan supervision of this portfolio and there is no dispute on the extent of work to be done, which is manageable given the number of loans. Concerning the BSCIC/BSB as well BSCIC's own portfolio, however, there are a large number of loans outstanding, which have been in arrears for a longer period; here there is need for a more organized approach. Thus, following the project by project review, amounts not recoverable will be written off and adequate provisions for bad and doubtful loans will be established on that basis. After this has been done, the full responsibility for supervision and collection of BSCIC/BSB loans will be transferred to BSB. The risk on these loans would, however, have to be shared as per original agreement. Concerning loans from BSCIC's own funds, it is difficult for legal reasons and because of the lack of interest by other agencies, to transfer responsibility for supervision of these loans to such other agencies. However, a large number of these loans will undoubtedly have to be written off (80% of the amounts have been in arrears for over 8 years). Thus, the number of loans to be supervised

34 by BSCIC is likely to be greatly reduced. BSCIC has allocated about 10 officers to carry out this supervision work and this recovery workload should not affect the quality of BSCIC's other activities as the recovery cell would be kept separate Funding. BSCIC's income would be supplemented by a 1.0% fee paid by the commercial banks on outstanding loans under the proposed IDA-Credit, which should cover BSCIC's contribution to appraisal, promotion and extension under the proposed project (paras 5.12, 6.01) Monitoring and Evaluation. The effective implementation, the achievements, and the development impact of BSCIC's various activities and of overall assistance to SCIs need to be better monitored and evaluated. BSCIC has agreed to establish a unit (of about three officers), which is neither responsible for nor in any way involved in the implementation or running of these activities, to undertake such monitoring and evaluation. The unit would report directly to the Chairman and initially would concentrate on reviewing BSCIC's own activities and the impact of the proposed project. General V. PROMOTION AND EXTENSION 5.01 An important objective of this project is to reorient and strengthen BSCIC's role in promotion and extension. BSCIC's promotion and extension services would be strengthened by: (a) developing separate organization, staffing and schemes for SSI and cottage industries; (b) assigning local staff and the advisors specific responsibilities and timetables for completion of subsector and area studies to be used in promotion and extension; (c) separating promotion and extension work from credit with BSCIC providing the commercial banks with studies on SSI subsector/regional profiles for very small industries, and diagnoses of problems and BM&R needs of existing enterprises; and (d) implementing integrated schemes for specific cottage industries (e.g., jute, cane, and bamboo handicrafts). SSI Promotion 5.02 The Promotion Unit of the SSI Division would be responsible for (a) preparing studies and schemes of subsector/area potential, (b) completing product and market feasibility studies, (c) developing simple profiles/checklists of standard cottage industries with fixed assets of less than Tk 100,000 1/, (d) conducting promotion among small industrialists and other entrepreneurs, and (e) providing assistance in project preparation and initial organization of industries. 1/ While nearly all promotion and extension for CIs would be performed by the Cottage Industry Division (para 5.10), the SSI Promotion Unit would prepare profiles/checklists for selected cottage industries (e.g., carpentry, repair workshops, tailoring) since this Unit would have the best equipped technical staff to provide information on regional markets, appropriate technology, raw material requirements, etc.

35 The SSI Promotion Unit at Head Office would consist of one unit chief, seven other senior officers and three junior officers. These officers have already been recruited, as have an additional ten for the District Offices. In addition, an advisor would be assigned to this unit for the first year of the project. SSI Division SSI Promotion Unit Chief (1) Area/Subsector.Feasi bility Very Small Studies and Promotion Studies Industry Projects -(4) (4) (2) 5.04 Each officer would be assigned a region as well as one or two subsectors for specialization. Each member of the Promotion Unit at Head Office would be expected to spend at least two weeks out of every two months in the districts. Regional staff would be actively involved in project identification exercises to assure continuity BSCIC has agreed to undertake the following promotional activities during year one of the project: (a) Five to ten SSI subsector studies; these would include: (i) fruit processing (bulk processing for exports, dried, jams); (ii) fish processing (drying, smoking, freezing); (iii) readymade garments especially for export; (iv) agricultural implement manufacture and repair; (v) poultry; (vi) footwear and leather goods; (vii) subcontracting in readymade garments, agricultural equipment, or other subsectors suggested by BSCIC. The first stage of these studies would concentrate on market/resource possibilities, size and technology, requirements and potential, and recommended subproject organization and design. Specialized consultants would be funded for a period of about three months (para 3.35), to provide expert advice, especially on markets and technology; (b) About 5 SSI area studies would be conducted in conjunction with representatives of BSB, the Ministry of Industry, District Development Boards, Export Promotion Bureau; in all cases BSCIC's regional staff would be actively involved. The studies would identify market and resource processing potential as well as locate promising entrepreneurs. Areas selected would be those which, despite resource and other potential, have limited industrial development.

36 (c) At least 10 profiles/checklists of very small and cottage industry subprojects (with less than Tk 100,000 fixed assets and catering to local demand) to be used by commercial bank branch managers in appraising very small projects; these profiles could include, inter alia, tailoring, carpentry, blacksmith, repair workshop, earthen ware and building materials, small grain mills, bakeries, aluminum utensils, footwear and leather goods, fish and fruit drying. SSI Extension 5.06 Most extension staff should be located in the districts; a goal would be set that by the end of the disbursement for the project (about 4 years) at least 2 to 3 extension officers would be placed in each district, while about 5 to 7 specialists will be kept at headquarters in an SSI Extension Unit (see Chart below) BSCIC has recruited 36 officers in the last year primarily for extension work, 9 of whom are already at Dacca headquarters, 17 are in the districts and ten are yet to join. The staffing plans provide for recruitment of another 30 workers for the districts in FY79. SSI Division.I SSI Extension Unit Chief Specialists in: - Management training - Subcontracting and SSI marketing District Level - Textiles/garments Extension Officers - Food storage and processing - Engineering - Chemical/Other 5.08 Diagnosis. Since many SSI problems are subsector-specific, the first step by headquarters would be to do a diagnosis of a sample of existing SSI units. The subsector diagnoses should focus on problems and potential solutions of most industries in the subsector, rather than collecting census data available elsewhere. As complete surveys are too time consuming, samples of about 20 firms per industry in the major subsectors would be taken in areas of concentration to provide a reasonable basis for diagnosing subsector needs. At least 5 surveys would be completed in year 1. The subsector diagnoses would take place in stages, with extension staff concentrating on specific subsectors; for example, food storage and processing (about 4 months); textiles/ garments (3 months); engineering industries (3 months); and another important subsector (2 months). Once the diagnoses are completed, regional level services and/or BMRE loans would be arranged with interested SSIs, with possible credit for diversification, improved capacity utilization and building of economies of scale.

37 Training. On the basis of these diagnoses, BSCIC's extension work would include provision of short, 2 to 4 day courses for entrepreneurs, to be held monthly on such topics as equipment needs, accounting, bookkeeping etc. These sessions would help demonstrate the benefits of BSCIC advice, provide points of contact to entrepreneurs and opportunities for mutual discussion of SSI problems. Cottage Industry Extension 5.10 While the SSI Promotion Unit would develop standard profiles/ checklists for use by commercial banks in lending to cottage and very small industries which cater to local markets, all other support to cottage industries would be the responsibility of the Cottage Industry Division of BSCIC. Cottage industries require active, subsector-specific institutional support in marketing organization, joint raw material purchasing arrangements, extension services, financing, design and technology development. Therefore, the Cottage Industry Division would be organized along subsector lines (e.g., jute handicrafts, fish drying, sericulture) and special schemes would be developed over time for each of the major cottage subsectors A UNDP Advisor to BSCIC recently presented a project proposal to strengthen BSCIC's support to cottage industries. Principal components of the UNDP project would be (a) design and technology advisors at Headquarters in about 7 product areas developing schemes for subsector specific support, and (b) general Cottage Industry Development Officers at district level, to review potential for subsector specific training and support in the longer term. The IDA Cottage Industry Component would complement UNDP efforts, by providing a pilot scheme for support (in marketing, design, credit, extension) to jute, cane and bamboo units in selected thanas, where concentrations exist (see paras for description of Cottage Industry Component) The costs of the promotion and extension work, discussed above, including the technical assistance provided by IDA (para 3.35) are estimated below. 1/ 1/ These costs exclude services for the cottage industry component, which have been discussed in para 3.21.

38 A. Consultants Quantity Cost Cost (Tk) ($'OOO) (Tk'000) 1. Salaries 48 man-months 200 /a 3, Per Diems 48 man-months 85 /a 1, Transport (coach) to and from Dacca 12 round trips 20 /a Vehicles 2 15 /a Other travel, Bangladesh 100 trips 6 7a Equipment and contingencies /a 7% of /a 372 Subtotal 350 5,424 B. Staff and Other Costs for Promotion and Extension, BSCIC 1. Salaries, Promotion staff 11 members, 4 yrs Salaries, Extension staff 10 members, 4 yrs members, 2 yrs. 3. Landrovers Field Trips 464 trips/4 yrs Overhead, office, etc. 10% of Subtotal 148 2,299 TOTAL 498 7,723 /a Financed by IDA Credit /b These include imported equipment (typewriters, etc.) for consulting services. IDA would finance about 70% of these costs with responsibility for repayment by GOB. The balance of about Tk 2 million would be recovered over the project's disbursement period by the income of 1.0% on subloans outstanding, generated by BSCIC from the IDA Credit (para 3.35, 4.21, 6.01). VI. ADMINISTRATION OF THE CREDIT 6.01 Lending Arrangements. The Credit would be made to GOB which would relend the subloans component to commercial banks for on-lending to the SCIs. GOB would relend to the commercial banks at 4.5% and the commercial banks would on-lend to the ultimate borrowers at a minimum of 11.5%, the current lending

39 rate for SCIs. If the inflation rate remains at about 10% in the long-run, which is a reasonable assumption given GOB policies, the ultimate borrowers will pay a positive real rate of interest of about 1.4%. The commercial banks would pay the Bangladesh Bank a guarantee fee commission of 1% on amounts outstanding and a service charge to BSCIC of 1.5% that would cover the costs of its promotional and extension activities for the subprojects financed; this would allow the commercial banks a spread of 5% which is justified in view of the administrative costs and inherent risks associated with SCIs and will serve as an incentive for the banks to finance this sector. As many small entrepreneurs could not afford nor understand the implications of incurring heavy losses from foreign exchange fluctuations, the foreign exchange risk would be borne by GOB. With regards the special capital fund the margins will be: GOB, 1%; Bangladesh Bank, 1%; and commercial banks, 5.5%; thus, the interest rate to the sub-borrowers for the special capital fund will be 7.5%. The funds for technical assistance would be provided by GOB to BSCIC/ BHMC, as a grant Free Limits and Prior Approval. The maximum subproject size under the proposed Credit would be Tk 2.5 million, and the maximum subloan, about Tk 1.75 million (US$110,000). The average subloan size for projects costing between Tk 100,000-Tk 2.5 million would be about Tk 440,000 (US$30,000) and about 200 subprojects in this size range would be financed from the proposed Credit. About 1,000 subprojects costing below Tk 100,000 would also be financed with an average subloan size of about Tk 20,000 (US$1,300). Because the number of subprojects will be large and the size small, and as SCIs should not be burdened with an unduly complex review system, the normal DFC subproject approval procedures should be simplified. For subprojects costing less than Tk 100,000, no prior review by IDA would be required. 1/ Instead, the commercial banks would send to IDA for a post-review a quarterly statement of approvals with key information on the subprojects financed. For larger subprojects the banks would send to IDA, on a monthly basis, for a no-objection, the applications for authorization to make withdrawals from the Credit, with a summary of each subproject and the terms and conditions of the proposed financing, and a discussion of the basis for the procurement decision. 2/ There would be no prior review of appraisal reports (all subloans would be "free limit" subloans) but appraisal standards would be discussed during supervision missions on the basis of reviews of representative samples by size and subsector Procurement. In view of the size of the individual transactions, IDA's normal requirements for international competitive bidding would be inappropriate for the proposed project. Instead, procurement of foreign equipment for projects estimated to cost Tk 100,000 or more would be based on at least 3 quotations from different suppliers. To ensure adequate and 1/ However, IDA would be reviewing, in the course of the dialogue with BSCIC on promotion, the project profiles prepared by it on which the appraisal of most of these subprojects would be based. 2/ The form in which these sets of information would be presented to IDA has been agreed during negotiations.

40 consistent procurement, the basis for the procurement decision will be reviewed prior to subproject authorization by IDA for projects costing above Tk 100,000 (para 6.02). For smaller sub-projects and the procurement of locally produced components of subprojects, IDA would rely on the local banks' appraisal of reasonableness of costs together with the local entrepreneur's incentive to keep costs low. Further, supervision missions will review procurement procedures and standards including the relevant documentation for a sample of subprojects financed Disbursements. Disbursements from the proposed credit would be made as follows: (a) Concerning credit, IDA would finance 100% of loan disbursements by the participating commercial banks for the general subloans component and for the credit element of the cottage industry component, in both cases, up to 70% of subproject costs. 1/ IDA also would finance 100% of disbursements, up to 157% of subproject costs for the special capital fund. Withdrawal applications would be submitted directly to IDA by the two participating commercial banks. In view of the large number of sub-projects involved and the small sub-loan sizes, the normal DFC procedure of disbursing against expenditures on individual sub-projects accounts, would be inappropriate. Thus, all disbursements for foreign or local expenditures made by the commercial banks will be financed equally up to 70% of subproject costs. Disbursements for sub-loans above US$45,000 (with sub-project costs from Tk 1.0 million to Tk 2.5 million), accounting for about 40% by amount of the project's credit component, will be fully documented, except for permanent working capital. Disbursements for permanent working capital and for the balance of sub-projects as well as the special capital fund would be reimbursed against a quarterly statement of expenditure, certified by the commercial banks, showing their own subloan disbursements for the specific sub-projects. (b) For the balance of the cottage industry component, IDA will finance 100% of foreign expenditures for sales trips and design consultancy and 100% of local costs to pay for BSCIC/BHMC incremental staff. Disbursements would be submitted to IDA by BSCIC and would also be based on a quarterly statement of expenditures, certified by BSCIC, except for the foreign costs, where withdrawal applications will be fully documented. 1/ The subproject sponsors will normally be contributing 30% to subproject costs, so the commercial banks and IDA would be financing up to 70%.

41 (c) Concerning the technical assistance to BSCIC, IDA would finance 100% of the direct foreign and local expenditures for the consultants hired, including equipment needs. Withdrawal applications would be submitted to IDA by BSCIC and would be fully documented The documentation on (a) and on local costs for (b) above would be retained by the commercial banks and by BSCIC respectively, and would be available for review by supervision missions. All statements of expenditure would be audited at least once every 12 months by auditors acceptable to IDA and a report would be submitted to IDA promptly thereafter. The disbursement schedule is estimated as follows (US$ thousand): FY79 FY80 FY81 FY82 First Quarter Second Quarter Third Quarter Fourth Quarter ,650 2, Cumulative total 750 3,400 6,100 7,000 of which (cumulative) Credit Component 660 3,010 5,260 6,100 Special Capital Fund Technical Assistance BSCIC Cottage Industry Component ,400 6,100 7, Amortization of Subloans. It is unlikely that any subproject will require an amortization schedule of over 12 years including a grace period of say, 3 years. As the commitment period for the subloans will be about 2 years, an amortization schedule between the banks and GOB of 15 years, including 3 years of grace would be appropriate. The amortization schedule between the banks and GOB will be fixed because back-to-back repayments for such a large number of subloans would be administratively too complex. Benefits VII. ECONOMIC JUSTIFICATION AND RISKS 7.01 General. The proposed project addresses the main problems constraining the development of SCIs such as pdor subproject selection, lack of access

42 to working capital and investment credit, cumbersome institutional framework, weaknesses of support institutions responsible for promotion and extension, and failure to provide forward and backward linkages. SCIs can play a key developmental role given the characteristics of Bangladesh's industrial sector. SCIs have relatively higher labor intensity and can provide for better utilization of the available entrepreneurial skills and limited capital resources of the private sector. Further, given the low per capita income in the country, SCIs have the ability to cater to the very limited needs for specialized items in the small domestic market and can provide for dispersal of industrial activity throughout the country. Finally, SCIs give possibilities for better labor relations as a result of closer contacts between management and labor. Thus, the proposed Credit, although small in size, has considerable significance for Bangladesh Institutional Improvements. Apart from redefining the roles of support institutions to make assistance for SCIs more efficient, the project will provide needed technical assistance to BSCIC which is the primary agency responsible for promoting development of SCIs. There is little doubt that promotion and extension are of key concern for SCI development and the technical assistance program for BSCIC will thus focus on these areas. In addition, the involvement of commercial banks in SCI financing will make access to credit much easier and will relieve small entrepreneurs from approaching different institutions for investment and working capital. Commercial banks, with their extensive branch network, more knowledge of local conditions and better access to local entrepreneurs can play a considerable role in SCI financing. The use of appropriate expertise, as planned, should not only generate significant improvements in SCI promotion and development work, and in the provision of technical assistance to entrepreneurs, but also in subproject selection, appraisal and supervision of enterprises Further, the proposed project introduces new schemes such as the special capital fund and the credit guarantee scheme which should provide opportunities for further growth of SCIs. The pilot cottage industry component for jute, cane and bamboo handicrafts provides a model integrating promotional, financial, marketing and technical assistance services, which could be replicated by GOB to develop other subsector specific, export oriented schemes for CIs Employment Effects. It is not possible to quantify directly the benefits that would flow from the above institutional improvements and from the technical assistance components of the project. Clearly, they should provide for more rapid growth of SCIs, and associated improvements in employment as well as productivity and SCI managerial development. Para 3.23 quantified some of the expected benefits of the jute, cane and bamboo cottage industry component. With regards the credit component, it is expected that a fairly large number of small and labor intensive enterprises (some 200 SSIs and about 1,000 CIs), using chiefly local raw materials and some of which are export oriented, or aimed at providing basic needs at lower prices, would be financed. Financial and economic benefits cannot be easily quantified in a component such as this, involving a large number of enterprises in disparate activities. However, it is roughly estimated that the project