Document of The World Bank FOR OFFMCIAL USE ONLY Repwt No. 8548 PROJFCT COMPLETION REPORT MOROCCO SECOND SMALL-SCALE INDUSTRY PROJECT (LOAN 2038-MOR) APRIL 23, 1990 Industry and Energy Division Country Department II Europe, Middle East and North Africa Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their offical duties. Its contents may not otherwise be disclosed without World Bank authorization. ACRONYMS AMD ABBREVIATIONS. BCP Banque Centrale Populaire BNDE Banque Nationale de Developpement Economique CCG Caisse Centrale de Garantie DFC Development Finance Corporation ITPA Industrial Trade Policy Adjustment NGO Non-Governmental Organization ODI Office de Developpement Industriel OED Operations Evaluation Department of the World Bank PB Participating Bank SAP Special Action Program SAT Service d'Assistance Technique Currenc3 Equivalents At time of Project Appraisal: 1 US$ 5 Dirbam At time of initial PCR preparation (June 1989): 1 US$ - 8.5 Dirham FOR OWICIL UESE ONLY THE WORLD BANK Washmgton. D.C. 20433 US.A. O#1ce CA OiwcUw.CeE Ope#.tum, teaIutuim April 23, 1990 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Morocco Second Small-Scale Industry Project (Loan 2038-MOR) Attached, for information, is a copy of a report entitled "Project Completion Report on Morocco Second Small-Scale Industry Project (Loan 2038-BR)," prepared by the Europe, Middle East and North Africa Regional Office. No audit of this project has been made by the Opierations Evaluation Department at this time. Attachment This document has a restittd distribution and may b used by fecipients only in the perfoannae of their official duties. Its contents may not otherwise be disclsed without World Bank authorition. FOR OMCUIL USE ONLY PROJECT COMPLETION REPORT SECOND SMALL-SCALE INDUSTRY PROJECT (LOAN ?038-MOR) TABLE OF CONTENTS Pare No. PREFACE . .........................................................i EVALUATION SUMMARY . ...............................................i PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE .... .......... 1 I. Background. 1 II. SSI II. 3 III. Project Implemen,:ation and Results. 5 IV. Project Performance Analysis. 6 V. Bank/Borrower Performance. 9 VI. Lessons Learned and Recommendations .10 VII. Summary and Conclusion .11 PART II: PROJECT REVIEW FROM THEBORROWER'S PERSPECTIVE .... ..... 13 A. General ............ ..................................... 13 B. Development of the Mobilization of Loan 2038-NOR ........ 14 C. Conclusions ............................................. 16 PART III: STATISTICAL INFORMATION ............................... 17 Annex 1: Processing of SSI II .................................... 17 Annex 2: Schedule of Cumulative Disbursements ..... ............... 18 Annex 3: Evolution of Sub-Loan Activity: Investments and Related Emploment Creation ...................................... 19 Annex 4: SSI II Sub-Loan Allocation by Sub-Sector .... ............ 20 Annex 5: Profile of Sub-Loan Activity by Participating Bank ...... 21 Annex 6: Profile of Sub-Loans by Type of Operation .... ........... 22 Annex 7: Comparison of Loan Maturities under SSI I and SSI II .... 23 Annex 8: SSI II Supervision Missions ............................. 24 This document has a restricted ditrbution and may be used by recipnts only in te puforma of their offcia duties. Its c-ontents may not otherise be disclosed without World lank authorizaton. - i - PROJECT COMPLETION REPORT SECOND SMAHLL-SCALE INDUSTRY PRQJET (LOAN 2038-MOR) PREFACE Loan 2038-MOR, (SSI II Project; US$70.0 million) was approved by the Bank on July 14, 1981. The Loan closed on June 30, 1988 following two extensions of the closing date. Of the original loan amount, US$26.45 million was disbursed, and US$43.55 million was cancelled. The Industry and Energy Division, Country Department II, of the World Bank's Europe, Middle East and North Africa Regional Office prepared the Preface, Evaluation Summary, and Parts I and III of this Project Completion Report (PCR). The Borrower prepared Part II. The Bank's sections of the PCR are based on a review of the Staff Appraisal Report, the Loan, Guarantee and Project Agreements, supervision reports, data provided by the Borrower, and all other relevant documentation available in the Bank's project files. - iii v PROJECT COMPLETION REPORT MOROCCO SECOND SMALL-SCALE INDUSTRY PROJECT (LOAN 2038-MOR) EVALUATION SUMMARY Introduction 1. During the 1970s, the Government of Morocco perceived a need to reorient some of its investment in industry towards maximizing efficiency in resource allocation and employment creation. The development and employment creation potential of small-scale industries (SSIs) represented an area of particular interest (Part I, paras. 1 and 2). To focus more attention on the SSI sector, the Government, in 1978, created a unit within the Ministry of Commerce and Industry to formulate policies for developing small-scale industry and established a Technical Assistance Unit, Service d'Assistance Technique (SAT), to assist the technical and managerial functioning of SSIs. The Government also sought and obtained a World Bank loan for the Eirst Small-Scale Industries Project (Ln. No. 1687-NOR; US$ 25 million, approved April, 1979), in support of the new strategy for SSI development. Objectives 2. The SSI II project sought to build on the apparent success of SSI I (which initially had a very high rate of commitments) by: (a) extending resources to entrepreneurs who otherwise would have had difficulty securing financing, and (b) providing technical assistance to revise the incentives system for the SSI development and help new and existing SSIs overcome institutional constraints. On July 14, 1981, the Bank approved Loan No. 2038-MOR, to the Kingdom of Morocco, for US$ 70 million in support of this SSI II. The Bank made the loan to the Kingdom of Morocco. Of the total loan proceeds, US$ 68.75 million was for onlending through BEDE and commercial banks to small-scale industries to create new enterprises and expand existing ones. The various categories for onlending are described in Part I, para. 7. The remaining US$ 1.25 million was allocated for technical assistance to new and existing SSIs (up to US$ 1.0 million), and for studies (US$0.25 million) to help overcome institutional constraints and tc revise the incentives system for SSI development. Implementation Experience 3. The implementation experience of the SSI II project confirmed some of the preliminary results of the SSI I PCR of October 1984, (Part I, par s. 14-16). Since the Bank had prepared that PCR dxxring the implementation stage of SSI II, some of the lessons learned under the first project could not be incorporated in the design of the second. The supervision work of SSI II nevertheless, took account of lessons learned under SSI I, and took action on some of them during the implementation of the second operation (Part I, para. 18 ). Effectiveness of SSI II was delayed due to an erroneous legal opinion related to the Project Agreement and to delays in recruiting consultants for technical assistance under the Project. Several extensions of the loan closing date were approved to take into account difficulties in collecting withdrawal documentation for the large number of small sub-projects financed under the loan, and issues related to sub- project eligibility (Part 1, para. 14). Furthermore, the original loan amount was reduced from US 70 million to US$ 30 million due to lower than expected - iv - demand for sub-project financing resulting from the difflcult economic situation prevailing at the outset of Project implementatlon (Part I, para. 18), and from the fact that the PBs motivation to on-lend the funds was diminished due to administrative and procedural factors which rendered SSI lending (Part I, paras. 19-20) less attractive to them. Results 4. Sub-project financing under SSI II totaled DH 129.7 million, supporting DE 638 million In SSI investment. 17,266 jobs were created under this operation --- 15X more than the appraisal estimate and significantly more than under SSI I (8,500). The relatively high levels of employment generated by SSI II is explained by the fact that a high proportion of the investment sub-projects financed were in labor-intensive enterprises. It is notevorthy that the investment cost-per-job under this operation (DH 36,946, or US$ 5,474) was considerably lover than the average for manufacturing industries in general (about DR 190,000, or US$ 28,358). Sustaiuabili 5. Despite the reticence to fully utilize the loan proceeds of SSI II, commercial banks have become more active in SSI lending as a result of this operation. In order for such lending to continue to expand in the future however, the authorities, the Bank, and the PB's should take account of the lessons learned from the SSI II experience. Lessons Learned 6. This PCR discusses the combined lessons learned under the SSI I and SSI II projects. Both projects have been successful in terms of creating jet) at a lover cost than the Industrial average (para 22). However, neither SSI I or SSI II was able to decentralize SSI investment for reasons discusseu in para. 25 (b). The report concludes that the policy-making unit of the Ministry of Commerce and Industry and the financial institutions should review industrial investment decentralization policies with a view to formulating a more realistic and effective strategy. In addition, a new strategy is required to improve the system of loan guarantees and collateral since the comrercial banks continue to overburden clients with personal and physical guarantees (Part I, para. 20). Concerning the SAT, neither SSI project was able to bring about substantial improvements in its operation; the problems hindering its effectiveness are discussed in paras. 16 and 25 (d). ,The Government should review the assistance needs of the SSI sector to determine the most effective way of meeting them (Part I, para. 25 d). For very small projects, the Bank's demonstrated flexibility during SSI II implementation should be (and has been) continued when lending to SSIs is possible under its other lines of credit. Given some of the difficulties experienced in SSI project implementation and supervision, which seem to stem from the nature of such credit operations, OED should consider undertaking a cross-country comparative analysis of Bank-financed lines of credit to determine the broader lessons that emerge, and the improvements which can be brought to this type of Bank lending activities. PROJECT COMPLETION REPORT MOROCCO SECOND SMALL-SCALE INDUSTRY PROJECT (LOAN 2038-MOR) PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE Project Identity Proiect Name: Second Small-Scale Industry Project (SSI II) Loan No.: 2038-NOR RVP Unit: EMENA Country: Morocco Sector: Industry Subsector: Small-Scale Industry I. Backeround 1. Introduction. During the 1970s, Morocco had embarked on an ambitious public investment program spurred by large increases in the price of phosphates, the country's main export commodity. Public investment programs during this period had emphasized capital intensive projects with limited returns and costly infrastructure projects in excess of the country's needs. Productivity of capital was low and employment creation was limited. The Government began to recognize the need to reorient some investment towards maximizing efficiency in resource allocation and employment creation. The development and employment creation potential of SSI's represented one area where such investment was appropriate. 2. Small-Scale Industries Subsector. In the late 1970's, estimates were that there were some 3,000 small-scale enterprises, employing about 42,000 workers (between 10 and 49 workers per enterprise) and representing some 20X of manufacturing employment and value added. To focus more attention on the SSI sector, in 1978 the Government created a policy-making unit within the Ministry of Commerce and Industry to formulate policies for developing small-scale industry and established a Technical Assistance Unit. Service d'Assistance Technique (SAT), to assist the technical and managerial functioning of SSIs. The Government also sought and obtained the First Small-Scale Industries loan (US$ 25 million; Ln. No. 1687-NOR; approved April, 1979) from the Bank, in support of these new directions. 3. SSI I. Preparation of the SSI I project began during the final year of Morocco's Third Development Plan (1973-77) which had emphasized the development of large public enterprises, and medium to large privately owned industries. In response to the country's changing economic environment, the Fourth Development Plan (1978-80) called for enhanced Government rupport to 2 small and medium-scale industries to create employment at a cost per job lower than that of large industries, and to help improve integ;ation of the existing industrial structure. The preparation of SSI I drew on the successful experience of an SSI pilot component in the eighth Bank loan to the Banque Nationale de D6veloppement Economique (BNDE VIII, Ln. 1428-MOR) which had allocated US$5 million for SSI financing as well as funds for the carrying out of a sector study to identify the potential for, and the constraints to, SSI development. 4. SSI I consisted of $24 million in credit to small businesses, and $1 million in expert assistance to the technical assistance unit, Service d'Assistance Technique (SAT) of the Office du D6veloppement Industriel(ODI). The line of credit was committed a year ahead of schedule. It financed 214 subprojects compared to the 200 sub-projects estimated at the time of appraisal. Of this total, 65X were for new enterprises. Under 5SI I, 8,500 jobs were created amounting to 70 X more than the appraisal estimates. The average cost per job created vas US $5,000 or about five to six times less than the average cost in the manufacturing industry as a whole. About US$ 7.9 million of the original loan amount of US$ 25 million was cancelled due to the appreciation of the dollar in relation to the Dirham (DH), and to disuse of the available funds due to the unavailability of complete documentation to participating banks (PBs) on sub-project expenditures (paras 18(b), 25(e)). 5. Despite impressive results in terms of employment creation, SSI I fell short of its objectives in the following areas: (a) decentralization of industrial investments from Casablanca to economically depressed regions; (b) support to very small industries (the actual amount lent was only half the original target); and (c) improvements in the terms and conditions of SSI lending (i.e. maturities, grace periods, and sub-loan guarantees). rFurthermore, the SSI I Project Completion Report (PCR) indicated that more selective monitoring of sub-project implementation could have improved the terms and conditions of SSI lending. The same PCR stated that a thorough assessment of achievements would await the results of the SSI II project. However, at the time that that PCR was prepared (October 1984), the combined experience under both projects indicated that: (a) SSI development in Morocco was capable of generating employment at a very low cost and with limited Government intervention; (b) the institutional framework established under the first project and carried over to the second project was sound and could be maintained for future operations; (c) the institutions concerned with SSI development needed to enhance their capability to respond to changes in the investment climate, particularly in the areas of strengthening and focusing of investment activities; and (d) there was a need to strengthen conditions leading to more active use of the Bank's line of credit, such as a greater interest spread to participsting Banks and improved procedures for guarantees on defaulting subloans. 3 II. SSI XII 6. Pro1ect Description. The SSI II loan (Ln. No. 2038-NOR, approved July 14, 1981, originally for US$ 70 million, was made to the Kingdom of Morocco for onlending through BNDE and commercial banks to small-scale industries to c-.eate new enterprises and expand existing ones. The rationale was to provide resources to entrepreneurs (US$ 68.75 million) who otherwise would have had difficulty securing financing. It also provided for technical assistance to new and existing SSIs (up to US$ 1.0 million) and financed studies (US$0.25 million) to help overcome institutional constraints and to revise the incentives system for the development of SSIs. 7. The terms of the sub-loans were to conform to the economic life of the sub-projects, with a max4Im" maturity of 12 years and a maximum grace period of three years. Interest rates were to be those prevailing for loans to industry. As with SSI I, BNDE was both a direct lender and the apex institution (agent of the Government) to supervise the commercial banks' management of the funds onlent to them. Sub-loans were approved and project implementation was coordinated by a Technical Committee with representation from the Ministeries of Finance, Commerce and Industry, as well as the Central Bank (Banque du Maroc), Office de Developpement Industrial (ODI), and BNDE. Eighty percent of the sub-loans were to be guaranteed by the Government's Caisse Centrale de Garantie (CCG), and guarantees were to be approved at the same time as the sub-loan approval. 8.Sub-loans were to finance up to 50 percent of the total cost of SSI II sub-projects, representing their estimated foreign exchange content. Proceeds of the Bank loan were to be allocated among three categories of sub-projects: (a) very small firms with assets of up to DH 1 million (US$ 200,000) --25X of the loan amount; (b) small firms, with assets between DH one million and DH 3 million :between US$ 200,000 and US$ 600,000) -- at least 60 X of the loan amount; (c) medium firms, with assets between DH 3 million and DH 5 million (US$ 600,000 and US$ 1,000,000) -- at most 151 of the loan amount. 9. To promote maximum employment creation, ceilings were placed on the investment per job requirement. The investment cost per job ceiling was set slightly higher for enterprises outside Casablanca (DH 50,000, equivalent to about US$ 10,000) than for enterprises around Casablanca (DH 45,000) in order to encourage decentralization of investment and employment creation. For the electromechanical industry, the ceiling was set at DH 55,000 (US$11,000) in the Casablanca area and DH 60,000 (US$ 12,000) elsewhere. SSI II was expected to create 15,000 jobs at an average investment cost of US$ 9,500 per job. 10. The technical assistance component of the Project was to be implemented by the policy making unit at the Ministry of Commerce and Industry and by the SAT. At project appraisal, it had been agreed that the SAT would: (a) 4 identify and prepare projects in coordination with the participating banks and the Ministry of Industry; (b) provide technical advice to entrepreneurs in procuring equipment and in preparing financing requests; (c) facilitate project implementation through assistance in identification and resolution of infrastructural and administrative problems; and (d) assist in identifying and solving management, production, accounting and financial problems of SSIs. In particular, the assistance under SSI II was to focus on identifying investment opportunities by creating a data bank and helping promoters with project processing, holding seminars on important management topics and assisting SSIs in the their cbjice of equipment and technology. The t.a. component was to have been comprised of 96 man-months of technical assistance including 18 man-months of consultant time to review existing policy and incentives related to SSIs. 11. During negotiations, the Bank and the Government also agreed to strengthen the SSI Policy Making tnit within the Ministry of Commerce and Industry. The Unit was to focus on: (a) the involvement of SSIs in executing public contracts; (b) quality control of products and marketing, particularl' for exports; and (c) improvements in the structure of SSI incentives. The Bank loan was to meet the foreign exchange cost of studies necessary to address these issues (estimated at US$ 250,000). 12. Prolect Design and Orzanization. The Project's design and organization derived from those of SSI I, which had proven successful in encouraging the development of SSIs. However, projected loan demand under SSI II was significantly below the planned target, as evidenced in the Project's rer ts (para. 13). Loan demand mainly was affected by changes in the economy during the implementation of the SSI II Project, as well as by certain administrative bottlenecks due to the design of the financing procedures under the loan. These bottlenecks, identified in a 1985 study undertaken by BNDE and the Ministry of Finance, were as follows: (a) unfamiliarity of entrepreneurs with financing procedures; (b) lengthy waiting periods for processing loan applications; (c) cumbersome procedures for justifying expenses under sub-projects; and (d) slowness in the authorization of sub-projects due to the absence of a free-limit procedure. 13. To alleviate these bottlenecks and improve the implementation of SSI II, the Bank agreed in 1985 to increase project eligibility limits and to create a free-limit procedure for sub-project authorization. It also agreed to simplified disbursement procedures for special cases where complete documentation was unavailable. These actions are described in greater detail in the following section. 5 1III Project Implementation and Results 14. General. The SSI II loan (US$70 million) was declared effective in February, 1983 following a one year delay due to submission to the Bank of an erroneous legal opinion of the Project Agreement, and to delays in recruiting consultants to undertake studies The original loan amount was reduced twice, to US$40 million in Ju. - 34, and to US$30 million in December 1986, because of lower than expected comnitments relative to loan size. The original and actual disbursement schedule of the loan are presented in Annex 2. The limit date for commitments under the loan was extended from December 31, 1983, to December 31, 1987. The closing date, originally June 30, 1986, was extended several times (to June 30, 1988) to take account of difficulties in collecting documentation for a large number of small sub-projects, and of ongoing special efforts of PBs to expedite outstending withdrawal applications to maximize disbursements. The loan account finally closed on March 31, 1989. By account closing, disbursements had reached US@26.4 million (88% of the revised loan amount) and the undisbursed portion fUS$3.6 million) of the loan was cancelled. 15. Profile of Sub-Loan Activity. During the period 1983-1987, 514 small enterprises received sub-loans under SSI II, equivalent to DH 129.7 million, to support investments on the order of DH 638 million (about US $75 million). Two-thirds of the loan proceeds were for the creation or expansion of enterprises in the food, textiles and leather subsectors. About 701 of the proceeds were for the creation of new enterprises end 30% fox the extension/modernization of existing ones. As outlined in the project description (para. 6), the loan proceeds were on-lent through BNDE and ten commercial banks (Annex 5). BCP accounted for 58% of thv total number of sub-loans while the relative share of the ten other PBs averaged about 4.2X. About half of the sub-loans were made to enterprises in the Casablanca area. 16. Profile of Technical Assistance and Studies. The Service d'Assistance Technique (SAT), a unit located within the Office de Developpement Industriel (ODI), was established in 1978 to provide technical assistance under both the SSI I and SSI II projects. Aside from providing some consultant assistance to a small number of enterprises, SAT was not able to provide the support enterprises needed to prepare new projects and resolve problems faced by borrowers in securing loans due to budgetary constraints. Futhermore, demand from enterprises for SAT services declined when in 1985/86 the SAT was obliged to begin charging for its services. Of the US$1.25 million allocated to technical assistance and studies under the SSI II, only about US$32,000 was disbursed. 17. The components for technical assistance and studies under the Project are described in paragraph 10. The technical assistance component was implemented in 1984-85. The study on incentives to SSIs was completed in 1986, and it confirmed the fihdings of the July, 1983 and October, 1984 Bank missions which produced the Bank Study of Small-Scale and Artisanal Industry (IBRD Report No. 5056 MOR), issued in March, 1986. The latter study proposed a definition of 6 SSIs based on the size of the assets of the enterprises rather than the branch of activity. It revealed that SSIs with assets of up to DH 7 million employed more than 110,000 persons and generated about one fifth of the value added by the manufacturing sector, accolrnting for less than 41 of Morocco's GDP. According to SSI surveys, the major problems SSIs had faced since 1980 were: growing competition in traditional export markets; limited access to medium- and long-term financing for investment and working capital; lack of skilled labor and technicians; cumbersome administrative procedures for investment and trade; and lack of institutional assistance to identify new product lines and market opportunities. Although the report found that some improvements had been made in these areas, it also projected difficulties in expanding SSI investment considering Morocco's existing economic and financial problems, and the weak institutional support and policy framework for SSIs. The report also analyzed the labor intensity of SSI investment and recommended that the job creation objectives of the Bank loan should be maintained, if significant changes were to be made to the project eligibility criterion for SSI projects (para. 18b) financed by the Bank. A long series of detailed recommendations are provided in Volume II of that Report. IV. Project Performance Analysis 18. Demand for Sub-Loans. SSI II Project data on the level of demand for sub-loans, in terms of numbers of loans, reflect changes in the overall economic situation as well as administrative changes in financing procedures for sub-project approval and disbursement. During the 1983-84 economic crisis (see a below), the number of small-scale enterprises granted sub-loans under the Project increased by only 161 (from 64 to 74). Between 1984 and 1985 however, the annual number of new sub-loans increased by 27X (from 74 to 94) and then jumped by nearly 501 in 1986 (from 94 to 140). The latter level was roughly maintained in 1987 (Annex 3). The sharp increase in demand for sub-loans demonstrated after 1985 reflects the impact of: (a) the economic reform grogrsms. begun in 1984 in response to the economic crisis. In 1983, following the decline in international phosphate prices and increases in world oil prices of the early 1980s, Morocco experienced a major economic crisis. Outstanding external debt reached 10O1 of GDP, foreign exchange reserves were nearly depleted, and GNP growth declined to about 21 per year, barely keeping pace with population growth. The country was forced into debt rescheduling, emergency import restrictions were imposed, and public expenditures were massively curtailed. In 1984, the Government embarked upon a multi-phased stabilization cum adjustment program, with Bank and IMF assistance. While reduced, the public investment programs included in the Government's new strategy sought to ensure that only priority proje.cts, which contributed to the twin objectives of stabilization and growth, were undertaken. Consistent with this, the Government maintained its move towards the development of SSIs (para. 1). The reform programs were designed to reduce aggregate 7 demand, increase efficient production through reforms in incentives for agriculture and industry, enhance private sector savings and improve investment allocation through financial sector reforms. The Bank supported these structural reforms through two Industrial Trade Policy Adjustment (ITPA) loans. The ITPA programs emphasized export promotion and import liberalization, and included reforms in fiscal policies, as well as in the financial and public enterprise sectors. Of particular relevance to small-scale industry were export incentives introduced through these programs, and the reduction of import protection (thus enhancing competition). Special tax advantages for the development of small-scale enterprises were also introduced. (b) the changes in procedures for sub-project approvals. disbursement against the nroceeds of the Bank loan, and sub-lending bank margins for loan transactions. In the course of project implementation the Bank had agreed to: (i) increase the margin of the commercial banks from 21 to 31; (ii) raise the eligibility limits for sub-projects by size of firm'/, (iii) introduce a free limit for Project commitments set at DH 500,000 (or US$50,000) in 1985, and (iv) permit full disbursement of a sub-loan based on documentation of 851 of expenditures with certification by the PB that a visit to a given sub-project site had revealed satisfactory implementation2/. 19. As noted earlier (para. 14), the demand, in dollar terms, for SSI II funds was lower than expected as a result of which about 601 of the original loan was cancelled. While the difficult economic situation prevailing at the outset of Project implementation had an undeniable impact on the use of the Bank loan, demand for Bank funds was also hindered by a lack of strong motivation on the part of the PBs to utilize these funds because: (a) their project pipelines were dominated by new creation activities which are 1/ During project implementation, sub-project eligibility ceilings were raised from those originally in the Project design (para. 8): for small scale industries the ceiling was raised to DH 4 million; for medium-scale, labor- intensive enterprises, the ceiling was raised to DH 8 million. This was done in order to broaden the base of eligible subprojects in light of inflation affecting the Dirham. I/ This procedure was established in 1988 along with special procedures in exceptional cases, i.e. for construction, civil works or equipment. In such cases, if the sub-borrower was unable to provide evidence of payment, the Bank could allow withdrawal from the Bank loan upon certification and evaluation by a qualified engineer, of completion of works based on sub- project inspection. 8 generally more risky than extension projects; (b) arrears were still owed to them against losses under SSI I sub-projects; (c) they lacked ronfidence in the Government guarantee agency, Caisse Centrale de Garantie (CCG) to provide sufficient default protection to them for problematic subloans; and (d) their transactions margin remained low in relation to costs associated with processing and monitoring the sub-loans, even after the 1985 increase to 3% (compared to 4-5X on transaction involving their own funds). In addition, the positive results of the SSI I project had, by the start of SSI II, rendered aid financing for SSIs readily available, on terms and conditions more advantageous than those of the Bank loan. Although, in dollar terms, demand for SSI II funds was lower than expected, it is noteworthy that the number of sub-loans made with only 38X of the original loan amount was close to that originally expected (514 vs. 600). Accordingly, the Bank loan was used principally for smaller enterprises, and the cost per job created was lower than anticipated at appraisal (para.22). 20. A 1984 BNDE survey of projects financed under SSI I had revealed that: (a) only 571 of the sub-loans had normal disbursement profiles; (b) arrears of the PBs were about 161 of the combined outstanding balance of sub-loans; and (c) 71 of the sub-loans were under legal proceedings. BCP, the PB most affected by arrears under SSI I, was also the PB which financed the bulk of projects under SSI II. Although the arrears situation showed improvement, PBs remained understandably concerned about the need for effective loan guarantees and collateral during SSI II, given the SSI I portfolio performance. In case of sub-loan default, the CCG guarantee system was to cover 801 of losses incurred by a client PB. In order to qualify for the CCG guarantee however, the PBs were required to pay a one-time fee ranging from 1.25X to 3.25% of the sub-loan principal, depending on the duration of the sub-loan. In addition, the guarantee procedures only entitled PBs' reimbursement by CCG after PB's had exhausted all possible means to collect the funds on their own (including legal proceedings), which generally takes at least two years after the initial declaration of default. Moreover, the bureaucratic slowness of the CCG in reimbursement for claims, and its occasional failure to pay claims at all, had led many PBs to adopt an even more conservative approach to lending, further limiting the use of the Bank's funds. 21. Enterprise Creation vs. Exnansion. A breakdown of SSI II sub-loans by those granted for the creation of new firms (701) and those for the expansion/modernization of existing ones (301) is given in Annex 6. The main reason for the large share of new enterprises in the loan portfolio derives from the fact that the preferred source of financing for enterprise expansion had been through leasing arrangements. The evolution of loan demand, however, indicates that an increasing number of extensions and modernizations vs. new firms were financed under the Project after 1985. According to BNDE, the evolution may be explained by: (a) increases in the assets of these firms due to new purchases of equipment; and (b) establishment of a free-limit procedure for obtaining funds under the Bank loan, obviating the need for detailed documentation for authorizations up to a certain limit, and enhancing the use of Bank funds to purchase, rather than lease, equipment. 9 22. Employment Creation. The sub-loans under the Bank project were associated with investments totalling DH 637 million (US$ 94 million), and a related investment cost per job of DH 36,946 (US$ 5,474). This compares to an average cost per job of DH 190,000 (US$28,358) for employment in the manufacturing industry in general. As previously indicated, the large cancellations from the original Bank loan amount did not adversely affect actual performance in terms of job creation and average cost per job. In fact, performance in these areas was better than originally expected in the SAR. To wit, 15X more jobs were created than originally expected, at an average cost per job of only 581 of that estimated at appraisal. In addition, though ultimately similar in loan amounts, the number of jobs created under SSI II (17,266) was significantly higher than under SSI I (8,500). This could also be largely explained by the fact that sub-loans under SSI I were distributed more evenly across several sub-sectors, while those under SSI II were concentrated in more labor-intensive and export-oriented sub-sectors (i.e., agro-industries, textiles and leather). 23. Terms and Conditions of Sub-Loans. To ease pressure on the cash flows of enterprises with difficult debt situations, and to reduce the risk of default to PB's, the Bank had recommended (under SSI I) that loan maturities be lengthened. It was therefore agreed during SSI II that the optimal maturity for loans to SSI's should be about eight years. Annex 7 compares the loan maturities under SSI I and SSI II. That annex indicates that under SSI I, only 16X of the sub-loans had maturities of more than five years, and less than 1X had loan maturities of eight years or more. Under SSI II, 46X of all sub-loans had maturities of greater than five years, and 321 of the total had maturities of eight years or more. In turn, the higher maturities under SSI II resulted in a reduced incidence of arrears to PB's and of default by SSI's. V. Bank/Borrower Performance 24. Despite Bank/Borrower efforts to improve the procedures under the Bank loan, the anticipated dollar level of loan demand did not materialize and the motivation of the PBs to use the Bank loan was diminished for the reasons outlined in paragraph 19. There is an inherent difficulty in the supervision of SSI lines of credit due to the large number and variety of sub-projects under such loans. In addition, follow-up with respect to technical assistance activities seems to have been of lesser priority under this project than efforts made to ensure the maximum use of Bank funds, given the large discrepancy between demand and loan amount. These factors notwithstanding, supervision efforts by the Bank and the PB's overall were satisfactory. Eight 10 supervision missions were launched by Bank staff, and an estimated 40 staffweeks, were devoted to this effortl/. BNDE's supervision activities improved considerably during project implementation. Regular status reports were submitted to the Bank which were based on a large and varied enough sample of sub-projects to be meaningful. When bottlenecks appeared in commitments and disbursements due to elements related to, or which could be addressed through, project design, the Bank and the PBs responded and agreed on revisions which improved performance in these areas. In addition, the Bank's push to convince the PBs to increase sub-loan maturities ameliorated their portfolio performance, Thus, although the size of the loan was reduced and changes were made to the scope of the project, the PBs' use of Bank funds and supervision activities were satisfactory. Tardiness however was occasionally experienced in the submission of audit reports. Finally, the need to finalize the Ministry of Commerce and Industry's report on incentives to SSIs was pre-empted by the more generalized industrial and trade discussions and reforms which developed following the 1983 economic crisis (para. 18 (a)). VI. Lessons Learned and Recommendations 25. SSI II was prepared before the conc3,isive results of SSI I were available. The success of SSI I at the time carried excessive weight in the formulation of the demand projections contained in SSI II. Many of the lessons which could have been learned from the first project, therefore, were unfortunately not incorporated in the original design of the second project. The combined experience under both projects nevertheless indicates that: (a) the Bank's two SSI projects have been successful in terms of job creation, at a lower cost than the industrial average (see para 22). Demand for SSI investment financing was in line with projections under SSI I. It was however far below expectations under SSI II. This shortfall was due mainly to the cotintry's economic situation, to the design of authorization and disbursement procedures for subloans, and to the availability of other competitive forms of credit. At the time of appraisal, these risks were not explicitly stated among the risks anticipated under the project. (b) despite the fact that on-lending to SSIs continues to be heavily dominated by one bank (BCP the main bank which caters to small-scale enterprises), commercial banks have generally become more active in SSI lending since the implementation of these Bank lines of credit. The continued lower participation of other PB's in SSI financing may be due to the relatively more limited access of SSI's to those banks, '/ This estimate is derived from the number of missions and of mission members, and assumes an additional 3-4 SW's per year for headquarters supervision work. The actual reported time indicated in the Time Recording System (28.5 SW's) appears erroneous, as mission work alone would exceed that amount. A supervision summary is provided in Annex 8. 11 and to the nature of the clientele preferred to be served by other banks. These factors alone limit the degree to which the objectives of decentralization and diversification in the development of SSI's can be attained, as the SSI projects were unable to significantly improve the regional distribution of such investments. This issue should therefore be reviewed with the policy-making unit of the Ministry of Commerce and Industry and with the financial institutions with a view to formulate a more realistic and effective strategy for decentralization. (c) a new strategy needs to be developed to improve the system of loan guarantees and collateral (para. 20). In spite of the Government's determination to reform and rehabilitate the CCG as expressed during the appraisal of SSI II, little was achieved and the commercial banks continue to overburden clients with personal and physical guarantees. (d) neither SSI project was able to bring about substantial improvements in the operation of the SAT. The SAT's effectiveness has continued to be hindered by the: (a) lack of budget and staff to meet the needs of a large caseload of enterprises with technical and managerial difficulties; and (b) lack of interest on the part of enterprises in these services, particularly since most do not seem to be willing to pay for them. The Government should review the assistance needs of the SSI sector to determine the most effective way of meeting them, either through an improved SAT or through selected consultant services such as those offered by non-governmental organizations (NGO's) and bi-lateral agencies. (e) for very small projects, the Bank's procedural and documentation requirements for commitments and disbursements have consistently proven cumbersome and unrealistic in the case of SSI lending to Morocco. This has been particularly true for small construction projects, where documents such as invoices and other evidence of payment cannot be provided by the sub-borrower. The Bank's demonstrated flexibility in this regard during the implementation of this project should be (and has been) continued when lending to SSIs is possible under its other lines of credit. The need for such simplified procedures should be considered however, as early as possible in project preparation/implementation. VII. Summarv and Conclusion 26. Overall, the implementation experience of the SSI II project confirmed some of the preliminary results of the SSI I PCR of October 1984, and provided certain additional lessons which could have been incorporated in the original project's design, had project preparation not overlapped with the implementation of the first project. Nervertheless, during implementation, many of the lessons learned under SSI I were incorporated in SSI II project supervision activities, and changes were made to the second projects' design. In addition, the magnitude of the 1983-84 economic crisis and its impact on the implementation of the second project provided for a very different 12 environment in which implementation took place than that of SSI I. In response to the economic crisis, between 1984 and 1986, 15 of the Bank's projects in Morocco were formally restructu-ed and loan amounts reduced (under the Bank's Special Action Program--SAP), to adjust the projects and their financing arrangements to the constrained economic conditions. While SSI II was not formally included in the group of Moroccan projects approved for the SAP's provisions, the reduction in its loan amount and changes made to its design during this period were consistent with the Bank's overall approach to projects in the country, and were appropriate. They were also consistent with the recommendations of the Bank's Small Scale and Artisanal Industry Study (para. 17) in that they allowed for project implementation to continue toward the attainment of the Project's objectives, though at a more limited level and moderate pace. 27. Finally, the combined implementation experience of the Morocco SSI projects reaffirm the view that certain of their characteristics and lessons learned could be generally relevant to other SSI Bank credit operationsl/. An OED cross-country comparative analysis of DFC implementation experiences could provide a useful basis on which to review Bank practices such that the attainment of project objectives could be facilitated. / i.e. The targeting of loan proceeds by size of enterprise rather than branch of activity, and the simplification of disbursement procedures. 13 PART II: PROJECT &?VIEW FRON THE BORROWER'S PERSPECTyVE A. GENE The agreement relating to Loan 2038-40R was signed in 1981 but did not become effective until 1983. The original amount of US$70 million was twice downscaled, first to US$40 milliC.a ar4 then to US$30 million. The closing dates were also extended several times, being ultimately set at December 12, 1987 for commitments and March 31, 1989 for withdrawals. Finally, a number of amendments were made: - Introduction of the Free Limit; - Uniform cost-per-job ceiling; - Raising (on two occasions) of the net assets figure; - Increasing of the banks' margin; - Modification of the release procedure; - Adaptation of the procedure for justifying investment expenditures to the actual circumstances of the SSI1, etc. Thanks to these measures and notwithstanding the unfavorable economic situation that prevailed from 1983 to 1985, Loan 2038-4OR was 100% utilized as regards commitmenzs and nearly 90S as regards withdrawals. The considerable number of measures taken and their impact on SSI promotion in Morocco are evidence of the attention and interest of the parties participating in the system and particularly the State, IBRD and 8NDE. IBRD was consistently concerned to ensure effective execution of project 2038-MOR and followed its implementation with great attention and diligence, sending about seven experts to Morocco in the course of the project. These experts always worked in close collaboration with the BNDE management and staff and the other members of the Technical Committee. Visits to the participating banks and companies were arranged by BENDE whenever necessary. The aides-memoire and the interesting reports drawn up by these experts were most helpful for better monitoring of the management of the loan proceeds, and also 'erved as the basis for certain of the amendments made to the procedure. BNDE, as Secretary of the Technical Committee and manager of the loan resources, played a central role in the project. Its experience in the financing of industrial projects and the mobilization of international lines of credit was very beneficial for the system set up for the financing, promotion and accurate monitoring of SSts in Morocco, for which purpose it provided itself with the human and material resources needed to perform this function. 14 KNDE *Cted as intermediary both between the State and IBRO and between thbs And the Moroccan comeercial banks. It helped to resolve certain problems, of wbich the main ones were detailed in the study it made with the Ministry of Finance in February 1985. In the course of the project BNDE provided all sorts of financial, economic and technical assistance to all the partners in the system established. To sum up, the State-tURD-BNDE collaboration made it possible to set up a very original financing procedure, by means of which the banks were induced to start financing SSIs. The SSI subsector is currently receiving the benefits of a set of specific financing procedures. 8. DEVELOPMELNT OF THE MCBILIZATION OF LOAN 2038-NOR As of December 31, 1987, commitment requtests addressed to IBRD related to 514 investment projects which received ON 262 million in direct credits in conjunction with rediscountable credits totaling DS 130 million. This assistance generated investments aggregating 0R 638 million, enabling creation of 17,266 jobs once fully operational and the maintaining of a further 198. B.1 INCREASE IN COMIITMENT REQUESTS We give below the breakdown by year of commitment requests: No. of Investments Direct Rediscountable Jobs Year proiects (DH) credits (Of) credits (DE) created maintained 1983 64 61,140,663 26,484,062 12,453,088 1837 38 1984 74 75,097,000 32,716,000 16,353,000 2525 87 1985 94 113,034,000 48,075,500 24,030,000 2481 i4 1986 142 178,303,000 71,189,000 35,298,000 4391 31 1987 140 210,343,000 83,674,000 41,572,000 6032 23 TOTAL 514 637,917,663 262,138,562 129,706,088 17266 198 The above figures show that the number of projects and direct credits grew by 22% and 342 on average per year, respectively, during the period 1983-87. The pace of this Increasa quickened as of 1985, the year in which the above-mentioned measures were put into effect. 15 It should be noted that 732 of the projects and 772 of ihe credits were pogted between 196S and 1987, which is evidence of a special effort by both the CO =0 cial banks and BNDt in those three years. Rowever, the considerable increase in commitments in dichams was offset somewhat by the apprecidtion of the U.S. dollar, the average rate for which rose from DH 7.46 in 1983 to DOH8.64 in L984 then soared to OH 9.94 in 1985. On the other hand, the rate for the dollar averaged DR 9.06 in 1986, then dropped to Di 8.22 on average in 1987. Expressed in U.S. dollars, commitments increased by 32% per year over the period 1983-87, as shown in the following tab:e: 1983 1984 1985 1986 1987 Commitment requests (x ON 1000) 26,484 32,716 48,076 71,189 83,674 Comitment requests (x US$1000) 3,549 3,788 4,836 7,858 10,183 Average exchange rate for USS 7.46 8.64 9.94 9.06 8.22 Moreover, as a result of the favorable trend in approvals and the relative fall in the rate for the dollar, 601 of the commitments were made in the last two years (1986-87). 82 WITHDRAWALS (situatlon as of M4ay 29, 1989) Withdr'wals increased appreciably over the years, as is apparent from the followitcg table: Year Amounts Withdrawn (US$1) 1983 246,639 1984 1,123,873 1985 1,382,328 1986 3,973,067 198t7 3,192, 315 1988 l1,728,023 1989 (3 months) 4.775,750 Subtotal 26,422,015 Assistance component 32 374 26,454,589 16 Tbe $urge between 1983 and 1984 was exceptf,nal and was due to the buildup of applications owing to the delay in setti.. ap the line. 1110 poSf@rmance achieved since 1986, an the other hand, is the result of tho effort made by BNDE to make the commercial banks aware of the need to speed withdrawals together with the easing of the investment requirements as of 1985. The requirements in question related to the two items for which it is most difficult to produce justification, namely construction and development costs and working capital costs; the relaxations introduced consisted of establishment of quantity surveying for the former and r!lease of the proportion relating to the latter after completion of the physical program. It must nevertheless be noted that the question of justification of expenditures remains a problem for SSIs in Morocco. The fact is that despite all the relaxations introduced and notwithstanding the various extensions of the closing date for withdrawal requests, ultimately set at March 31, 1989, the unused balaine as of that date had to be cancelled. This balanee was of the order of US$3,545,411 (IBRD telex of April 11, 1989) and rep.esented 11.82 of the loan. By way of comparison, the cancelled balance of the f .rst SSI loan, No. 1687-MOR, was 32S. C. CONCLUSIONS The Second Small-Scale Industries Project, Loan 2038-MOR, made it possible to finance some 500 subprojects and to create about 17,000 jobs. Eleven banks participated in the project. This project encountered a number of problems, most of which have been overcome. Its implementation took longer than originally expected, but as a result many lessons were learned. Thanks to the project, the different agencies and organizations concerned are now able to operate the SSI financing system more efficiently. In addition, the terms and conditions governing execution of SSI projects have been improved compared with those applicable to projects financed under the first loan, which demonstrates the attention that is being paid to SSIs. A large number of administrative requirements have been eased and several financing procedures specific to SSIs have been created. Through this second project, IBRD has made the banks aware of the value of ssr financing. BNDE, being experienced in this form of financing, was also able to mobilize other lines of credit at the same time. 17 PART III: STATISTICAL INFORMATION Processing of SSI II Original Plan Project Brief 9/80 same Departmental Approval 10/22/80 same Yellow Cover 03/27/81 same Documents to Loan Committee 04/27/81 same Starting Date of Negotiations 05/18/81 same Board 07/14/81 07/14/81 Signing 11/03/81 same Effectiveness 02/03/82 02/24/83 Loan Closing Date 06/30/86 06/30/88 Loan Account Closing 12/31/86 03/31/89 Sourge: World Bank 18 &Qo2 Small-Scale IndustrY II SChedule of Cumulative Disbursements (in USS million) x m ustimatad S Of L gta1 ctUal SOf Total 1982 5.74 8 0.00 1983 26.95 38 0.00 1984 53.62 76 1.17 4 1985 67.34 96 2.26 8 1986 70.00 100 3.87 14 1987 - 8.55 32 1988 14.61 55 1989 26.45 88 Original Loan Amount: US$70.00 million Cancellation: US$43.55 million Actual Loan Amount: US$26.45 million 19 hAnnox3 Evolution of Subloan Activity: Investments and Related Emolgymnt Creto No. of Subloans 64 74 94 142 140 Imestment Total (Million DH) 61 75 113 178 210 Bank Credit (Million DH) 26 33 48 71 84 Other Credit (lillion DH) 12 16 24 35 42 Jobs Created 1837 2525 2481 4391 6032 Source: BNDE 20 Annex 4 all 11 Subloan Allocations bySu-Sector Total X Food 184 36 Textiles and Leather l8 30 Electromechanical Industries vs 13 Wood and Wood Products 32 6 Cnemicals 27 5 Non-metallic Minerals 31 6 Paper and Paper Products 20 4 Total 514 100 Source: BNDE 21 Profile of Subloan ActiXity bv ParticiEating Bam' No. of Subloans X BCP 300 58.0 BMCI 51 10.0 SGMB 42 8.0 B(CE 38 7.0 BCM 25 5.0 CDM 8 2.0 WAFA Bank 34 7.0 Arab Bank 2 0.5 SMDC 5 1.0 BIAO 2 0.5 BNDE 7 514 100.0 ZSource: BNDE 22 Annex_-6 P pfile of Sub-loans by_ yIn of Ooeratio' Greatle Extensio ggdernization 1983 46 14 4 0 1984 60 10 3 1 1985 66 25 3 0 1986 98 39 3 2 1987 89 50 0 1 Total 359 138 13 4 Source: BNDE 23 Com Arinon of Loan Maturities Undr SSI I and SSI II Sublgan Katur=iS in Years Is&L Thum 1I Zis lAU Lyg ZLgb Eight Plus SSI I 23 63 93 18 15 2 - SSI II 9 151 68 91 163 31 Soureg: BNDE 8SI II: SU IuII Nl88ISKm Perfosmanc* mnth Da" in no. ot ati / J 1ield Persona ~~~~naoia4tatiqpe lSawu~rn) oe In§ { 1 a" 1DpuLty Div. Cuif part-Utn) 2/2 W ned dm or investment SO Of cited Santo" Operations Officor 2/83 14 1 Senior Operations Officer 2J2 Loa made, ffective 2/24/83. Probable tegst tot revision of eligibility criteria cosistet with 58I definition undr new Investment code. treauy and D1 s! m in agrement o*trse lwiVt for approving emall projacts without revie by the technical comittee. Preblms Xind abeorptiv capacity of 881 tund ad recruitamat ot tfore4 experts for TA. 6/83 14 1 Senior Operations Officer 2f2 Moderate problemswo as those tound In 2f83 mission. Loan surplus of USS 35-40 million expected. Th fis ding ba bew ¢ontfid by 85I study. 6/84 10 1 Santor Operations Officer 2/2 Haf problems are (a) declining Interest in projects by participating bank and (b) inadeqacy of aubprojesct appraisal and supoevision by Me. Two covenants in defoult in recruitment of conaultant to assist HCIT's policy-makting unit and a foreign oxpert to support 001 in provision of TA to SSts. Ameduent of project atresnnt betwee PBs and BNDE to improve sub-project appraisal and supervision -- redefinition of loan categories and coat per job ceilings. MCIT has sel¢cted satisfactory xpert. ODI reluctant to use foreign xpert. Recomendation that Dank and ODI look for alternative m*es of meting technical assistance needs. 12/84 10 1 Senior Operations Officeor 2/3 Loan *unt reduced from US$ 70 ailion to US$ 40 million. Proe dural difficulties have delayed employment of FCIT experts for policy uit. OD0 baa identified experts for tecbhneal Assistance unit, plan, to award contract in early 1985. wa/85 12 1 Senior Operations Office 2/3 not available OD Katb/ De"y in go. of 8pecilisatioms Reting ^/ YTr Field Fors o _ tStatu/Sfrend} Cannts 5t86 2 1 Op.rctim Officer 2/2 RecanosdUtien tbat Bank co_ideS in09esimUs frte limit and miwnZ inVest.unt 01s.0 Lo d nd ftor loon funds become of demanding *ppreiel. sad proces*sig requirements. Pf conider it financing of let reort. 12/87 6 II 2 Operations Officer. 2/2 Woak demcnd for credit except from textiles Consultnt sub-scctor. Dibursemmt_ low du* to documntation difficulties. CG suarantee syst not vekicng well due to procedurel prcblm *nd lack of funding. DR end STreaury wiLl re lw_ disbursemects problem in detail. 1 09 ytm betng rostructurd. A/ Ratings: 1 - Problem free/mdnor problems 2 - fbderate problems 3 - Major problems _t Estirt ftrom total of 18 days i&Icb includes supervision of otbor projects. u