Mexico - Fourth Small and Medium Scale Industry Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 16819 IMPLEMENTATION COMPLETION REPORT MEXICO FOURTH SMALL AND MEDIUM SCALE INDUSTRY PROJECT Loan 2858-ME June 27, 1997 Sector Leadership Group Mexico Country Department Latin America and the Caribbean Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY Currency Unit New mexican Peso (N$) S$1.00 = Mex$367 December 1985 S$1.00 = N$7.58 December 1996 $1.00 = N$7.8 June 1997 FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES One Hectare (ha) = 10,000 square meters One Metric Ton (mt) = 2,200 pounds One Metric Ton (mt) = 1,000kg One Kilometer (km) = 0,6215 miles One Kilometer (km) = 1,000 meters One Cubic Meter (m3) = 1,000 liters One Liter = 1,000 mililiters ABREVIATIONS AND ACRONYSM USED ACF = Costo Promedio de los Fondos = Average Cost Funds ERR = Tasa Economica de Retorno = Economic Rate of Return FRR = Tasa Financiera de Retorno = Financial Rate of Return FIDEIN = Fideicomiso de Conjuntos, Parques y Ciudades Industriales = Trust Fund for Industrial Park Dvelopment FOGAIN = Fondo de Garantia y Fomento a la Industria Mediana y Pequefia =Guarantee Trust Fund for Small and Medium Industry Development FOMIN = Fondo Nacional de Fomento Industrial = National Fund for Industrial Development GIRA = Acuerdo General de Tasas de Interes = General Interest Rate Agreement ICR = Informe de Terminacion del Proyecto = Implementation Completion Report LCB = Licitacion Publica Nacional = Local Compettitive Bidding NAFIN = Nacional Financiera S.A = Government National Industrial Development Bank PAI = Programa de Apoyo a la Industria Mediana y Pequefia = National Program for SMI Development PCR = Informe de Terminacion del Proyecto = Project Completion Report PFI = Intermediarios Financieros del Proyecto = Participating Financial Intermediary SA = Cuentas Especiales = Special Account SECOFI = Secretaria de Comercio y Fomento Industrial = Ministry of Trade and Industrial Development SINCAS = Sociedades de Inversion = Society of Investmenst SOE = Estados de Gastos = Statement of Expenses SMI = Industria Pequefia y Mediana = Small and Medium Industry TA = Asistencia Tecnica = Technical Asistance Vice President Shavid Javed Burki Director Olivier Lafourcade Sector Leader Carl Dahhnan Task Manager Enrique Vanegas  Table of Contents FOR OFFICIAL USE ONLY P reface .... ....... ................. ........ .... ............................................... ......... .... . ................ . . .. ........ .. ..........i Evaluation Summary ......... ......................... .........i............ .......... ................................................... i PART I: PROJECT IMPLEMENTATION ASSESSMENT 1. In tro d u ctio n .......................... ...................................... ...................... ....... ......... ........... ............. 1 2. Statement of Objectives and Design ...............................................2. 2 3. Achievement of Objectives ..... . ..... .... ...................................................................................4.....4 4. Implementation Record and Major Factors Affecting the Project .7 5. Project Sustainability .......... ...... ... .......................................... ..................9... ............ ...........9 6. Bank Performance ..................1...................... ....................................... 10 7. Borrower Performance ....................................... 11 8. Assessment of Outcome ..... . .. ... . .....1....................................................................................... I 1 9. Future Operations .12 10. Key Lessons Learned. ............................12 ... ............. .... . .... .... ...... ............... 12 PART II: STATISTICAL TABLES Table 1: Summary of Assessments ........................ .... ..... ..... ..... .. ..... . 14 Table 2: Related Bank Loan/Credits ................................ 16 Table 3: Project Timetable .. .. .. .... ..................1..............7............................... ...... 17 Table 4: Loan Disbursements: Cumulative and Actual ............................... ...... .. .......... 18 Table 5: Key Indicators for Project Implementation ................................................. .. 19 Table 6: Key Indicators for Technical Assistance ........... .... ... ........... ........... .. .... ...........0. ..... 20 Tablable 7: Studies Programmed and Actual ... . . ................ .. ...... 21 Table 8: Project Cost ... ... ... ..... ..... .... ..................................................................... ............ 22 Table 9: Project Financing .......................................... ......................... .... ............. ... ..... 23 Table 10: Use of Funds by Category .................4....... .... ........ .. . ... .... .. . ..... . . ... 24 Table 11: Number of Subloans by Size of Firms ..25 Table 12. Total of Subprojects for Technical Assistance ..26 Table 13: Total of Subprojects by Amount ..... ... ........... 27 Table 14: Number of Subloans by Category ...................... .... .......... ........ .. ..... ... ...... 28 Table 15: Number of Subloans by States ......................... .... .......... .... .... ..29 Table 16: Number of Subloans by Matunty ............................. ........ .. .... ...... .. .30 Table 17: Number of Subloans by Banks and Financial Intermediaries . . 31 Table 18: Status Of Legal Covenants ........................................I.... ... ......... ........32 Table 19: Compliance with Operational Manual Statement ...... ........................... ................... .33 Table 20: Bank Resources: Staff Inputs ... ... .. ... ......................... ......... ... .....3............... 34 Table 21: Bank Resources: Missions ........... ..... ........ ... .... ... .... ..... .... 35 APPENDIXES A. - Aide Memoir of Mission B. Borrower's Contribution to the ICR 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.  I I  IMPLEMENTATION COMPLETION REPORT MEXICO FOURTH SMALL AND MEDIUM SCALE INDUSTRY PROJECT (Loan 2858-ME) Preface This is the Implementation Completion Report (ICR) for the Fourth Small and Medium Scale Industry Project in Mexico, for which Loan 2858-ME in the amount of US$185 million equivalent -w-s approved by the Bank on June 25, 1987 and made effective on November 28, '989. The Loan was reduced to US$100.0 million in September 1988 per request of the GOM and was closed on February 28, 1995. Final disbursement took place in July 1995. The ICR was prepared by Mr. Enrique Vanegas (Task Manager) supported by Mr. Juan L. Moreno, (Consultant), and reviewed by Carl Dahlman (Sector Leader) of the Finance Private & Public Sector Leadership Group of Mexico Country Department, Latin America and Caribbean Region. Preparation of the ICR was begun during the Bank's final supervision/completion mission in July 1996. It is based, inter alia, on material in the project file and discussions with Bank staff and representatives of the Borrower. The Borrower contributed to the preparation of the ICR by submitting a Final Implementation Report on the project with an evaluation of its execution which is included as Appendix B  I  ii IMPLEMENTATION COMPLETION REPORT MEXICO FOURTH SMALL AND MEDIUM SCALE INDUSTRY PROJECT (Loan 2858-ME) Evaluation Summary Introduction 1. In 1978, NAFIN established a program of assistance to small and medium industries (SMIs), known as Programa de Apoyo Integral a la Industria Mediana y Pequefia (PAI). The principal participants were: i) Fondo de Garantia y Fomento a la Industria Media y Pequenia (FOGAIN), a credit rediscount fund involving commercial banks and operated by a second tier institution; ii) Fondo Nacional de Fomento Industrial (FOMIN), a risk capital financing fund; iii) Fideicomiso de Conjuntos, Parques y Ciudades Industriales (FIDEIN), a credit and technical assistance fund for industrial estate development; iv) an industrial extension services for SMI, provided by a newly created corps of agents; and v) a technical secretariat to oversee the working of the PAI program as a whole. The PAI program was supported under three previous Bank loans (SMI I - III), which were the first made in Mexico for SMI development (see paras. 3-4) According to the Performance Audit Report prepared by OED, two loans (SMI I & II) were rated unsatisfactory and SMI III marginally satisfactory (see para. 5). Project Objective and Design 2. This project (SMI IV) was to continue to support GOM in its efforts to improve the quality and range of financial/technical assistance to SMI. Its explicit objectives were to: i) foster economic growth and employment by supporting SMI - particularly those that were operationally viable, but in financial distress - by establishing an institutional and financial framework conducive to financial restructuring; ii) promote regional decentralization; iii) assist in the financing of microenterprises, particularly those managed by women entrepreneurs; and iv) improve the policy framework for the development of SMI. These objectives were to be met through six components, of which five provided credit to: (i) FOGAIN to finance SMI's fixed assets and permanent working capital; (ii) FOMIN to support its equity and quasi-equity investment programs, (iii) a pilot program operated by FOMIN to fund financial packages including debt/equity swaps and financial restructuring; (iv) FIDEIN to finance the development of industrial parks and buildings; and (v) NAFIN to provide financing, on a pilot basis, to microenterprises that were unable to borrow through the normal banking channels. The sixth component consisted of technical assistance and extension services to be provided through PAI This component  iii also included funds for institutional strengthening and to enable SECOFI in undertaking key subsector studies to identify development constraints and to formulate policies to promote SMI development (see para. 7) Implementation Experience and Results 3. Overall, the credit component was successful in reaching SMIs. The project provided financial support when it was needed for their modernization and the opening of the economy. Because ERR data were not available for the subprojects, we do not know if the reported achievements of enhancement of SMI productive potential was realized. The loan financed 3,512 subprojects by disbursing US$100 million. Among the factors affecting project implementation were: (i) GOM's economic policies that resulted in a period of low inflation and trade liberalization, free of market distortions and subsidies; (ii) the long delay in loan effectiveness (26 months), which made the project design complexity evident (see para. 26-27); (iii) NAFIN's organizational restructuring to become a second tier bank (see para. 35); and (iv) the merger and later dissolution of the three trust funds (FOGAIN, FOMIN, and FIDEIN) 4. When assessing the project from its institution building perspective and improving the policy framework, the objectives were not met. The development risk of the project was not explicitly defined up front and well articulated protocols to limit, monitor and manage exposure to development risk were absent. This was shown by the failure of the technical assistance component to improve institutional development (see para. 24). Lack of information and data on PFIs portfolio quality and Bank subprojects did not make possible an analysis of SMI structure and its characteristics 5. The performance of the Bank (see paras. 33-34) through some stages of the project cycle was less than satisfactory. There was excessive emphasis on disbursement performance, while the emphasis on institutional improvements was insufficient. With the dissolution of the three main executing agencies, the Bank did not define an indicator (s) to ensure that the project original objectives were understood and supported by NAFIN (see para. 38e). This resulted in key activities not being carried out in areas such as portfolio management and sector studies 6. The performance of the Borrower/executing agency (see paras. 35-37) was similarly inadequate. NAFIN's capabilities in project cycle activities were weak and did not adequately maintain operational information on subprojects performance This limited the ex-post review of the project However, NAFIN's performance in terms of solving previous administrative problems was satisfactory, i.e., streamlining approval procedures and decentralizing its operations. Project Sustainability 7 With regard to the project subborrowers, field interviews conducted by the ICR mission with beneficiaries and PFIs provided positive feedback on the loan results.  IV However, sustainability can not be ascertained because of the lack of ERR data for the subprojects, as previously mentioned. 8. With regard to the organizational framework, the abolition of PAI and trust funds reduced the sustainability of the institutional development impact. The provision of technical and financial services to SMIs, as now established under NAFIN, is weakened by the absence of subproject evaluation and NAFIN's limited supervision and reporting capabilities. Key Lessons Learned 9. The lessons that may be learned from the experience gained during preparation and implementation of SMI IV are as follows: (a) Project Design. The design of SMI IV failed to identify and address the project's high risks and low rewards, as experienced under the previous three operations. SMI IV's novel features further complicated the institutional situation, which existed at the time of appraisal. Adding new components and executing agencies did not simplify the institutional structure. On the contrary, these features made it more complex Moreover, efficiency in terms of staff resources and implementation times was lost. The long delay in loan effectiveness should have been a "red-flag" for the flaws in project concept. (b) Ownership. The lesson here is that the government needs to take ownership of projects. When NAFIN became the sole executing agency - which was not anticipated during or after appraisal - the Bank did not reassess the development objectives of the loan. There was no evidence of NAFIN's responsiveness to the benefits that could be derived from the activities supported by the institutional building component and the Bank failed to recognize the limited leverage it had with a large development bank Implicit was the lack of NAFIN's ownership and commitment to the project as initially envisaged which was aimed at capacity building of the three trust funds, not NAFIN, and a policy framework to develop an SMI culture. In retrospect, the Bank should have canceled the loan or transformed it into a balance of payment support operation (c) Feedback. The Bank should strengthen its feedback mechanism and its ability for learning from its mistakes, so that the same problems do not persist through a string of follow up projects SMI IV did not incorporate many of the lessons learned under SMI I, II, and III.  I I  IMPLEMENTATION COMPLETION REPORT MEXICO FOURTH SMALL AND MEDIUM SCALE INDUSTRY PROJECT (Loan 2858-ME) PART 1: PROJECT IMPLEMENTATION ASSESSMENT Introduction 1. In 1978, NAFIN established a program of assistance to SMIs, known as Programa de Apoyo Integral a la Industria Mediana y Pequefia (PAI)'. This program was to be implemented through three NAFIN-administered trust funds, which were already supporting SMI activities. The program's major objectives were to: (i) encourage more rapid growth in output and employment in SMIs; (ii) facilitate the achievement of national plans for industrial decentralization and regional development; (iii) foster more comprehensive and effective financial and technical assistance to SMIs; and (iv) develop a mechanism to adapt SMI assistance programs to national industrial development goals. 2. The principal participants in the PAI program were: (i) the Fondo de Garantia y Fomento a la Industria Mediana y Pequefia (FOGAIN)2, a credit rediscount fund involving commercial banks; (ii) the Fondo Nacional de Fomento Industrial (FOMIN)3, a risk capital financing fund; (iii) the Fideicomicio de Conjuntos, Parques y Ciudades Industriales (FIDEIN)4, a credit and technical assistance fund for industrial estate development; (iv) an extension service group to help SMIs improve their operations and to obtain assistance from the above funds or from other agencies in the country or both; and (v) the Coordinating Committee with the Technical Secretariat, to oversee the working of the PAI program as a whole. / PAI was the coordinating committee for the project agencies, which was part of the integrated program for SMIs. It was governed by a high level Program Coordinating Committee, headed by the Director General of NAFIN that included the heads of all participating institutions, as well as representatives from the Ministnes of Finance, Commerce and Industiy, and Banco de Mexico. 2 / FOGAIN was established in 1954 as NAFIN's trust fund to act as a second-tier credit rediscounting loans to micro, small, and medium industnes for equipment & machinery. 3 / FOMIVN was established in 1972 as a government's trust fund to assist small and medium size companies in carrying out investment plans fostenng industrial diversification and regional development. 4/ FIDEIN was established in 1970 as a govemment's trust fund to promote and implement a national system of industrial states to foster regional development and industrial decentralization FIDEIN's role was modified under SMI III. It withdrew from the direct construction of factory buildings and the leasing of machinery and equipment and transferred to FOGAIN responsibility for the provision of credit to firms wishing to finance their own buildings and facilities.  2 3. The PAI program was supported under three previous Bank loans, which were the first made in Mexico for SMI development: SMI I (Loan 1552 -ME of 1978 for US$47 million); SMI II (Loan 1881-ME of 1980 for US$100 million) and SMI III (Loan 2325-ME of 1983 for US$175 million). The main objectives of the three loans were to provide a range of financial services, together with technical and managerial support, to SMIs, and to promote regional decentralization. 4. The loans were prepared and disbursed during an extremely turbulent period in Mexico's economic history that culminated in a fundamental reorientation of government economic policies. The macroeconomic situation featured growing economic stagnation in the early to mid-seventies, an oil boom in the late seventies, and a major financial and balance of payment crisis in 1982. This was followed by partial recovery in 1984-85 and a further financial crisis, accompanied by high inflation in 1986 A sustainable recovery, based on massive cuts in government spending, began in 1987. The public sector deficit was eliminated and inflation was reduced to single digit levels by 1993. Since 1985, the GOM has carried out sweeping reforms designed to transform the Mexican economy into one that is efficient, market-responsive and outward-oriented. 5. According to the PAR (see Performance Audit Report No. 13257, June 30, 1994) for the three loans, the resource transfer objective was met, but the projects' outcomes were limited in terms of improving technical and managerial assistance to SMIs. The first two projects were rated marginally unsatisfactory, and the third marginally satisfactory. This was mainly because under the third project: (i) Mexico's need for resource transfers was greater; (ii) interest rates on new and outstanding subloans were more appropriate; (iii) the implicit subsidization of SMIs was lower; (iv) the institutional performance of one intermediary better; and (v) the operations of the industrial extension service more cost-effective. Sustainability of the three projects was rated as uncertain. Statement of Objectives and Design 6. The SMI IV's broad objective was to continue to support GOM in its efforts to improve the quality and range of financial/technical assistance to SMIs. SMI IV had two novel components: (i) a pilot component in financial restructuring and (ii) a micro-enterprise credit. The project was targeted to SMIs whose financial viability was being undermined by high inflation, lack of credit and the increased competitive pressure following GOM's recently introduced trade liberalization measures. Accordingly, the project's specific objectives were to : (i) enable private small and medium scale investors to make their existing units more competitive by investing in the modernization, and expansion of their units; and helping them establish new enterprises; (ii) revive production of existing SMIs, which were operationally viable but overleveraged, by establishing an institutional and financial framework to support the restructuring of selected firms; (iii) help disadvantaged segments of the industrial sector (informal micro-enterprises mainly managed by women entrepreneurs) through a pilot credit and training scheme; and (iv) improve the policy framework for SMI development by conducting a series of studies on the potential constraints of SMIs. 7. The project objectives were to be met through six components:  3 (a) Credit Component. US$100 0 million, representing about half the loan, to finance fixed assets investments, covering plant and machinery purchases, associated working capital needs and related civil works. Up to 50% of the credit funds (US$50.0 million) could be applied to financing free-standing permanent working capital requirements. This type of financing was considered particularly critical since it would stimulate employment by raising capacity utilization. (b) Risk Capital Component. US$20.0 million to support and broaden FOMIN's equity and quasi-equity investment program, which was financing expansions by existing firms and venture capital type investments by new entrepreneurs. For the first time, this component would offer joint financing opportunities with the private sector: investment companies, brokerage houses and state-owned venture capital companies. As in the past, FOMIN would finance investments in fixed assets and related permanent working capital (no more than US$10 0 million). (c) Financial Restructuring Component. US$25.0 million to be operated by FOMIN on a pilot basis. It would provide funds for financial packages including debt/equity swaps, lengthening of accumulative commercial banks debt and fresh working capital and fixed assets investment resources. This was designed to meet the long-term needs of companies that were operationally healthy and had good prospects. An integral part was the technical assistance (TA) to develop institutional capacity and associated skills for restructuring firms by FOMIN and commercial banks. Bank funds were to finance fresh working capital and fixed assets needs of restructuring companies, while FOMIN's own funds were to be used to refinance existing assets in exceptional cases. (d) Factory Building and Industrial Park Development Component. US$10.0 million to finance the construction of factory buildings by lease operators within industrial parks in the high priority industrial zones. The funds were to be available on a selective basis for the expansion and/or modernization of industrial states in the private or public sector, and establishing of new industrial parks primarily in the export processing zones including those locations of in-bond maquila industries (e) Pilot Micro-Enterprise Component. US$10.0 million to finance fixed asset investments and long-term working capital needs of micro-enterprises, which were unable to borrow through the normal banking channels NAFIN would on-lend to public and private sector promotional agencies which had active micro-enterprise lending programs.  4 (f) Technical Assistance Component US$5.5 million to finance technical assistance and extension services through FOMIN, FIDEIN and PAI, and for computer equipment and related materials. Also, to develop key subsector studies to promote SMI development. This component included the strengthening of FOGAIN's management information system on portfolio management and FIDEIN's staff training in credit evaluation and subproject preparation. 8. Total project cost was estimated at US$350.3 million, of which about US$157.3 million represented foreign exchange. The project financing plan anticipated substantial contributions from other sources: (i) US$104.9 million from subproject sponsors and PFIs (ii) US$25.0 million from FOGAIN; (iii) US$30.0 million from FOMIN/Government; (iv) US$2.0 million from FIDEIN; (v) US$3.0 million from NAFIN; (vi) US$0.5 million from Secretaria de Comercio y Fomento Industrial (SECOFI); and (vii) US$0.3 million from UNDP. The Bank loan of US$185 0 million would cover 100% of the project's estimated foreign exchange component and US$28.0 million of local costs. The total cost of the technical assistance (US$8.3 million) would be financed by the Bank loan (US$5.5 million), and UNDP grant of US$0.3 million, and the remaining US$2.5 million by NAFIN and the three trust funds. 9. Following a long delay in effectiveness (see para. 26), the loan was reduced to US$100 million in September 1988 per request of GOM, because of the long gestation period and FOGAIN's high liquidity. 10. Loan disbursements were projected to take up to five years from mid-1988 to 1993. The projected disbursement period was conservative compared to the average LAC industrial projects and to the experience under the three previous projects. However, disbursements took longer than anticipated because of the 26-month delay in loan effectiveness. Consequently, the original closing date of December 31, 1993 was extended first to December 31, 1994 and then to February 28, 1995, to allow for full loan disbursement (see Tables 3 and 4). The Bank agreed to the establishment of a Special Account (SA) and made an initial deposit of US$15 million. It also allowed expenditures for subprojects below the free limit to be eligible for disbursement from the SA and the use of Statement of Expenditures (SOEs). 11. All the procurement was to be done in accordance with prevailing Bank guidelines. NAFIN, together with FOGAIN, FIDEIN, and FOMIN, would ensure directly or through PFIs, that goods and services financed from the loan would be competitive in quality and price and appropriate for their client's needs. Procurement packages above $500,000 requiring Local Competitive Bidding (LCB) would be reviewed by the Bank. Consultant services would be open to recruitment of international consultants, selected in accordance with Bank Guidelines for the Use of Consultants. As with the three previous projects, the accounts of NAFIN and the executing agencies, (FOGAIN, FIDEIN, FOMIN and the PAI technical assistance unit), including the SA and SOEs, would be audited annually by independent auditors acceptable to the Bank. Achievement of Objectives  5 12. From the resource transfer point of view, the project achieved its objective. It provided financial support to SMIs when it was needed for their modernization and the opening of the economy. While ERR data was not available for the subprojects, the reported achievement of its target group implies that substantial enhancement of SMI productive potential was realized by the project (see para. 30). In addition, the Pilot Micro-Enterprise Component was successful in providing credit to micro-enterprises and low income producers (see para. 22). 13. From the point of view of improving the policy framework and achieving its institution building objective, however, the results were limited. The development risk of the project was not explicitly defined up front and well articulated protocols to limit, monitor and manage exposure to development risk were absent. The sector development functions of PAI were replaced with NAFIN's programs, but with different emphasis and limited success. Moreover, PAI's extension program that provided technical support at the enterprise level was discontinued and its staff disbanded. The sector studies - a joint SECOFI/UNDP undertaking to identify SMI constraints as the basis to formulate policies to promote SMI development - were not carried out. The development of restructuring programs for financially weak SMIs did not take place and a system to gather data on the quality of portfolio from PFIs never materialized. 14 The project components' outcomes, once revised resource allocations were made after the loan reduction from US$185 million to US$100 million, (see Allocation of Loan Table) were as follows: 15. The Credit Component (Revised allocation: US$35.1 million; Utilization: US$62.6 million). This component (63% of loan) proved to have a higher demand than initially anticipated under both SAR (50% of loan) and the revised allocation (35%). The loan provided financial support at a time when commercial banks were reluctant to lend to SMIs due to their higher risk and funding cost. It even encouraged them to rediscount with NAFIN some medium term subloans to SMI that otherwise would have been unavailable, because commercial bank operations tend to be short term in nature. Field observation indicated that Bank resources were instrumental in expanding the firms involved, increasing employment, and/or improving the performance and competitiveness of beneficiaries. 16. The allocated amount was modified several times to accommodate demand. First it was increased to US$20.0 million in 1989, then to US$35.0 million in January 1992, ending at US$62.6 million by 1994. A total of 1,416 subloans were made with an average of US$44,000 and a median of about US$25,000, only 13.5 % of the subloans were in excess of US$100,000. The majority of the subloans carried out a relatively short maturity (less than three years), but there were some (13%) over 5-year's. About 48% of the subloans were for financing working capital and about 98% were for financing micro and small enterprises (see Table 11). The larger number of subprojects were located in Mexico City, Jalisco, Nuevo Leon, Guanajuato and Puebla (see Table 15). Eighteen commercial banks (see Table-17) participated, in addition to credit unions with 269 subloans (19%). No data was provided by NAFIN on SMI industry subsectors. In the period 1988 -1989, the interest rates to PFIs were at Certificados de la Tesoreria (CETES), and the spread varied depending on the perceived credit risk and the type of PFI. Afterwards, NAFIN' s interest rate policy mandated a 6% spread for all subloan categories (see para. 25).  6 17. The Risk Capital Component (Revised allocation: US$24.9 million; Utilization: US$1.9 million). This component was to support NAFIN's direct equity participation in SMI for planned expansion by existing firms and investments in new firms. Under the three previous operations, equity financing had had disappointing results, especially in the SMI III when it failed to encourage the emergence of new minority investors and to establish a secondary market. Nonetheless, FOMIN had developed into a fairly efficient organization with reasonable operating costs and a good cadre of professionals and the Bank continued to support FOMIN's operations. 18. Under SMI IV, the barriers to find small firms willing to organize themselves with equity or to accept external partners, were not overcome. Only US$1.9 million for a total of 12 subprojects was disbursed, out of a revised allocation of US$24.9 million. Seven subprojects were for quasi-equity to micro-enterprises, with an average amount of US$9,000 and five relatively large subprojects had an average of US$350,000 (see Table 13). 19. The Financial Restructuring Component (Revised allocation: US$0 million; Utilization: US$0 million). There was no demand for this component and it was eventually canceled. 20. The Factory Building and Industrial Park Development Component (Revised allocation: US$20 million; Utilization: US$15.1 million). At the beginning of project execution, demand for financing was higher than estimated, but declined later because expansion of industrial parks and factory buildings reached a degree of saturation by 1991. In addition, FIDEIN, the specialized institution, was dissolved, and NAFIN's priorities were more oriented toward regular credit to SMIs. 21. Twenty eight infrastructure subprojects were financed with a total investment cost of US$31.6 million during the period 1990-1991. The subprojects, averaging US$1/2 million, were normally in US dollars and received large cofinancing from commercial banks. This ran counter to the loan objectives of fostering new SMIs - average subloan amount under the other components were much lower. Most of the subprojects were for export activities (maquiladora sector) and located in Baja California, Puebla, and Chihuahua (see Table 15). 22. The Pilot Micro-Enterprise Component (Revised allocation: US$15 million; Utilization: US$15 million). This component was successful in achieving its objective of reaching micro-enterprises and it was novel in the Bank's operation with SMI in Mexico. The component was considered important because of its orientation to low income producers, with an implication for poverty alleviation and credit to woman producers. Collateral requirement was a constraint for micro-enterprises to get credit from commercial banks. During implementation a higher free-limit threshold was allowed from US$20,000 at appraisal to US$26,000. 23. Disbursements were slow at the beginning of implementation up until mid-1991, due to Bank requirement to assess some PFIs' financial soundness, as a condition for disbursement. However, Bank supervision later was lax or non-existent. To expedite disbursements, the Bank  7 agreed to NAFIN's request to include credit unions5 as PFIs. NAFIN's procedures for credit approval were cumbersome, but this situation improved later when they were simplified. A total of 2,056 subloans with an average amount of US$7,000 were granted to micro-enterprises. Of the total, 71% of the subloans were for less than US$10,000. Most subloans had maturity of less than three years (only 18 subloans had maturities of over five years) due to the concentration of working capital lending. Credit unions handled 39% of the number of subprojects approved, and the balance was handled through State Supported Development Agencies6 (Entidades de Fomento). NGOs were not PFIs, as anticipated at appraisal, because at the time they were not registered to issue lending contracts. No special effort was given for promoting credit to women entrepreneurs because given the nature of the subloans, they were significantly represented and employed in this subsector. The geographical distribution of the subloans was widespread in the country, with a high number in the industrial regions of Nuevo Leon, San Luis Potosi, Sinaloa, Guanajuato, Jalisco, Coahuila, Veracruz, Zacatecas and Oaxaca (see Tables. 13-15). 24. The Technical Assistance Component (Revised allocation: US$5 million; Utilization: US$5.4 million). This component did not achieve its objective of improving institutional development. Key activities that were not carried out included: a) the establishment of a Control and Information System for portfolio management to keep track of the quality and status of PFIs subloan portfolio; b) the Evaluation of Subprojects to broaden the scope of NAFIN's monitoring and supervision to include the development impact of subprojects and their repayment potential; c) studies related to incentives, and the simplification of procedures required to enhance or set up new enterprises; and d) PAI's industrial extension programs to continue throughout the life of the project. The only objective achieved was the decentralization of NAFIN's operations as a means to speed up subloan approvals. The TA component changed during implementation and funds were reallocated (see para. 38d). Some goals, such as stronger credit appraisal capabilities of the trust funds that were absorbed by NAFIN (see para. 26), previously contemplated, became irrelevant. This was because after the reorganization of NAFIN as a second-tier institution, PFIs were in charge of credit evaluation and bore the credit risk, except non-bank institutions such as credit unions and Entidades de Fomento. The main TA financed (about US$2.3 million), not anticipated at appraisal, was the establishment of an Asset/Liability Management System to monitor NAFIN's financial and operating risks (see Tables 7 and 12). Implementation Record and Major Factors Affecting the Project 25. The first key factor that affected implementation performance was complex project design. Although coordination arrangements' were similar to the previous three projects, the Bank should have devoted more time in addressing the added riskiness, complexity and demandingness when 5 / Credit unions are pnvate business organizations of firms in the same activity or related industries. 6 / State Supported Development Agencies are private and publicly supported promotional agencies with especial programs to assist micro-enterpnses. 7 / GOM decided the Project Coordinating Committee to be presided by the Sub-Secretary of Mimstry of Finance (SHCP) and members of NAFIN, FOGAIN, FIDEIN, FOMIN, and SHCP. The Comnumttee's responsibilities included setting priorities and policies for the Program operations and the use and allocation of the resources, review the results of the operation, and closely monitoring the project's progress.  8 two new components (financial restructuring and pilot microenterprise) were introduced to the four historical ones and three additional executing agencies for a total of six (FOGAIN, FOMIN, FIDEIN, SECOFI, NAFIN and PAI) were managing the project Despite the fact that the three trust finds were eventually merged with NAFIN, this policy decision by GOM was not anticipated at the time of appraisal nor later consulted with the Bank. The Bank should have taken a simpler approach in project design by reducing the number and scope of components. The Bank should have focused on less ambitious, but sustainable objectives such as improving and expanding industrial extension services and broadening understanding of the potential and constraints of SMIs. Instead, in addition to achieving those goals, the loan was expected to facilitate industrial deconcentration, to develop the institutional capability and associated skills for restructuring firms by PFIs, and to support micro-enterprises in the informal sector, primarily woman entrepreneurs, through a credit program. However, on the positive side, the Bank did address some of the institutional and policy problems previously identified under the three previous operations. Interest rates were set in real terms". NAFIN's decentralization of its operations was accomplished. Subproject processing was simplified and became more efficient, partly by reducing the process needed for subloans approval (see para. 37). Also, as previously mentioned, the unanticipated absorption of the trust funds into NAFIN resolved the coordination difficulties between the various institutions. 26 The second factor was the long gestation period. The project was identified during the final stages of SMI III execution, approved in June 1987, signed in September 1987, and became effective in November 1989. This long delay in effectiveness was caused by: a) NAFIN's organizational restructuring to become a second tier bank; b) the liquidation and merger with NAFIN of the three executing agencies (FOGAIN, FOMIN and FIDEIN), which involved a long legal/financial process, especially in the allocation of the foreign exchange risk between the Federal Government and NAFIN; and c) the reallocation of funds among the different components from the reduction of loan amount of US$185 million approved by the Board to US$100 million as requested by GOM. Allocation of Loan (US$million) Credit 100.0 351 62 6 Risk Capital 20 0 24.9 1 9 Pilot Financial Restructuring 25.0 0 0 Factory Building & Industrial Parks 10.0 20 0 15.1 Pilot Micro Enterprise 10 0 15.0 15.0 Technical Assistance 5 5 5 0 5.4 Unallocated 14.5 Total 185 0 100.0 100.0 8/ The onlending terms had long been an issue m Mexico with all the previous projects, especially regarding FOGAIN which was using fixed rates and were substantially negative in real terms It was agreed that onlending to sub-borrowers should use interest rates which conform to General Interest Rate Agreement (GIRA) between the Bank and GOM, and to be vanable over the life of the subloans.  9 27. The new institutional arrangements did not affect the objective of providing financial assistance to SMI, but did affect the project outcome.. The project's strong institution building program was predicated on the trust funds (FOGAIN, FOMIN, FIDEIN) being the main providers of technical and financial services to SMIs. Bank resources constituted their main source of funds. NAFIN's role was supposed to be marginal (managing the Pilot Micro-Enterprise Component). The overall implementation risk and the risk related to financial and economic efficiency of resource allocation was reduced by FOGAIN and FOMIN, the two leading executing agencies. They had developed into reasonably efficient and mature financial institutions. When they were dissolved, NAFIN became the Borrower and sole executing agency. Although the Bank supported the rationing of trust funds and streamlining of DFI's activities, the financial sector reforms that took place did not differentiate the various level of professionalism embedded in some of these institutions. With the merger, trust funds' senior staff disappeared and technical capabilities were scattered and rendered impotent. Bank resources which represented the main source of funds under the original scheme were now a fraction of the executing agency's and consequently of reduced importance and influence (NAFIN's annual lending increased from US$800 million prior to 1990 to US$2 billion later on). At the time, NAFIN was a large state bank serving the industrial sector. It later became a second tier financial agent targeting SMIs and functioning as a "discount window" rather than a traditional development bank involved in analysis and evaluation of investment projects. The loan also did not meet its resource mobilization target. Complementary external financing considered significant at appraisal, including resources from private sector sponsors, financial institutions, GOM, and multilateral agencies, was lower than 2:1 28. Once loan became effective, disbursements moved faster up until 1991 when 73% of loan funds had been disbursed. It then slowed in 1991-1992 due to NAFIN's reorganization and resumed in 1994, by which time disbursements had reached 98% of the loan. To facilitate implementation, the Bank agreed to frequent funds reallocation. The Bank also agreed to expedite subloan approvals by exempting calculation of ERR and FRR to those below US$ 100,000 and to delegate approval authority to NAFIN regional offices. Consequently, the economic and financial analysis carried out by PFIs of the subprojects was minimal. Credit Unions' appraisal consisted basically on evaluating the capacity of subborrowers to take on additional debt rather than their credit worthiness. All these factors revealed that the Bank had a greater preoccupation with disbursement than the economics of the project. 29. Two changes in GOM's economic policies also helped to expedite loan disbursements. First, the economic environment of the project was substantially different during implementation as compared to appraisal. Project implementation was carried out during a period of low inflation and economic liberalization. Second, commercial banks were reprivatized in 1991 contributing to a more efficient financial system. In addition, banks' reserve requirements were lowered, monetary policy relaxed, and massive private capital inflows led to a large credit expansion by private banks. Project Sustainability  10 30. Since NAFIN did not carry out ex-post evaluation9 of subprojects, the actual economic and financial benefits of the subprojects and their sustainability could not be quantified by the ICR mission. As mentioned above, credit risk was absorbed by PFIs so NAFIN was repaid independent of subloan status. However, in the case of non-bank intermediaries such as credit unions and Entidades de Fomento, NAFIN de facto took the credit risk causing subsequent substantial losses to NAFIN's portfolio. NAFIN's supervision consisted in ensuring that Bank funds were used for the purposes they were intended and there was no attempt to measure their development impact. 31. Nonetheless, field interviews conducted by the ICR mission with beneficiaries of the various loan components and with PFIs provided overall positive feedback, in aspects such as employment creation, product sales increases and improvement in operating efficiency. In addition, lending was provided under a liberal trade regime, free of market distortions and subsidies. 32. The abolition of PAI and the trust funds, however, reduced the sustainability of the institutional development impact. NAFIN did not seem interested in the institutional features proposed by the loan, as demonstrated by the TA failure (see para. 24). Regarding Bank SMI lending sustainability, it is no longer relevant. With the reprivatization of the commercial banks and the passage of NAFTA, there is increasing recognition by the banks of the profit potential of SMI financing. SMI IV was the last Bank SMI project in Mexico. Bank Performance 33. The performance of the Bank through some stages of the project cycle was hardly adequate. With the metamorphosis of the trust funds into departments of NAFIN, the Bank did not define an indicator(s) to ensure that project's original objectives were understood and supported by NAFIN. Bank supervision in terms of number of missions was satisfactory (about 2 missions per year) with matching number of staff resources (about 3.2 staffweeks). However, Bank missions were not effective in bringing about the necessary basic changes for implementation progress, such as: (a) the introduction of a management information system to determine the status of subloans portfolio; (b) the adoption of procedures to measure the development impact of subprojects; and (c) the undertaking of sector studies for SMI development. Supervision missions focused excessively on disbursement performance and inadequately on institutional improvements, proper data collection and reporting. Neither did subproject visits carry out any monitoring of the financial performance of the PFIs, particularly credit unions and Entidades de Fomento. 34. During loan implementation, the Bank hesitated in taking drastic action or to cancel the loan given the availability of Mexico's large capital inflows, although this latter possibility was 9 / In the period September-October, 1996, NAFIN carried out a survey of 104 beneficianes that had indicated the creation of 2,600 jobs at an average cost of US$9,800. However, the survey did not include calculation of the internal rate of return or their financial situation.  11 discussed with the Borrower. Also, a straight-forwardly transformation of the project into a balance of payment support operation was not seriously considered. Borrower Performance 35. There were also significant shortcomings in the performance of the Borrower/executing agency. NAFIN's capabilities in project cycle activities were weak and did not adequately maintain operational information on subprojects performance following the merger of the trust funds. This limited the ex-post review of the project. NAFIN failed initially to adequately supervise non-bank lending practices but later contracted out an audit reviewl