Mexico - First and Second Small and Medium Scale Industry Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 7552 PROJECT COMPLETION REPORT MEXICO FIRST AND SECOND SMALL AND MEDIUM SCALE INDUSTRY (SMI) DEVELOP PROJECTS (LOANS 1552-ME AND 1881-ME) DECEMBER 27, 1988 Projects Department Latin America and the Caribbean Regional Office This document has a resicted disibution and may be used by redpients only In the perfonumace of their offcial duties. Its contents may not othedwse be dbclosed withot Wodd Bank authoration.  CURRENCY EQUIVALENTS Currency Unit - Mexican Pesos (Mex$) - 100 cents 1978: 1US$ - Mex$29.6 pesos 1979: 1US$ - Mex$30.4 pesos 1980: 1US$ - Mex$29.6 pesos 1981: 1US$ - Mex$30.5 pesos 1982: 1US$ - Mex$106.4 pesos 1983: 1US$ - Mex$150.7 pesos 1984: 1US$ - Mex$188.8 pesos 1985: 1US$ - Mex$291.5 pesos PRINCIPAL ABBREVIATIONS AND ACRONYMS USED ACF Index of Average Cost of Funds BANIXCO Banco. de Mexico, S.A. CANACINTRA National Industrial Chamber CENAPRO National Productivity Center CONACYT National Science and Technology Institute FIDEIN Trust Fund for Industrial Parks Development FOGAIN Trust Fund for Industrial Guarantees and Credit FOMIN Trust Fund for Industrial Development ,| FONEP Trust Fund for Pre-Investment Studies GIRA General Interest Rate Agreement IDB Inter-American Development Bank IMIT Institute for Technological Investigation INFOTEC Institute for Information and Technological Development IPADE Pan American Institute for Industrial Management NAFIN National Industrial Development Bank PAI Program for Integrated SMI Development SECOFI Secretariat of Industry and Commerce SFB Standard Factory Building SMI Small and Medium Scale Industry FISCAL YEAR January 1 - December 31  FOR OMCIAL USE ONLY THE WORLD BANK Wash;ngton. DC 204 33 USA Ogle. Ml Diectev.C"Wf4I Opevatmm EwabAtuin December 27, 1938 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Pass-Through Audit Report on Mexico First and Second Small and Medium Scale Industry Development Projects (Loans 1552-ME and 1881-ME) Attached, for information, is a copy of a report entitled "Pass-Through Audit Report on Mexico-First and Second Small Scale Industry Development Projects (Loans 1552-ME and 1881-ME)" prepared by the Latin America and the Caribbean Regional Office. Full evaluation of this project has not been made by the Onerations Evaluation Department. Attachment This document has a retricted distribution and may be use by recipbents only in the perfomnance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICiLAL USE ONLY MEXICO PROJECT COMPLETION REPORT First and Second Small and Medium Scale Industry (SMI) Development Proiects Loans 1552-ME and 1881-ME TABLE OF CONTENTS Page No. Basic Data Sheet ..... *..............*** i aighlDghtshee... ....... .... ..... . iv lo INTRODUCTION .......... oo.o ..................... ............... 1 Bank Lending for Small and Medium Scale Industry (SMI).. 1 Macro-economic and Sectoral Performance-oooo....... ..... 1 Interest Rate Policy and Supply of Financial Resources.. 2 II. PREPARATION AND APPRAISAL OBJECTIVES OF THE LOANS-oooo 3 Loan Preparation.oo .. . ..-.. 3 Appraisal Objectives..j.....e 3 III. INSTITUTIONAL DEVELOPMENT. ...o ....o.o.o.o. 4 Program Coordinating Committee (PAI)o................. . 4 Fondo de Garantia y Fomento a la Industria Mediana y Pequena (FOGAIN)o.o..oo0.0 ........ ................... . . . . . 5 Operating Procedureso. ....... ......... ......... 6 Fondo Nacional de Fomento Industrial (FOMIN)...... ..... 8 Operating Policies and Procedures ........ .......... 9 Fideicomiso de Parques Industriales (FIDEIN) ............ 11 IV. UTILIZATION OF LOAN PROCEEDS UNDER LOANS 1552-ME AND 1881-ME .... 12 Resource Transfer ......... .. ........ ... . 12 Lending Characteristics Under Loans 1552-ME and 1881-MEo ................. ....................................... 13 FOGAIN Cjwwonent .............. ........... ............. ............ 14 FOMIN Component ................................. o ....... 16 FIDEIN Component ........ o .. .. . .............. . 19 This document has a restricted distribution and may be usd by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  TABLE OF CONTENTS (cont'd.) Page No. V. FINANCIAL PERFORMANCE OF INSTITUTIONS ................... 20 PAI ..*...................... o............. ........ 20 FOGAIN .............. . ..**** 21 FOMIN ......................................... ..... ......... 22 FIDEIN..*.* ....*.............*............ .. .o. ......... . 23 VI. TECHNICAL ASSISTANCE ACTIVITIES .......... ............... 24 PAI Extension Service .................................. 24 Technical Assistance Institutions ............... ....... 26 VII. CONCLUSIONS AND LESSONSLEARNED ..................... .... 27 PkI*# *.* # ... ...... #.... ....................... . ............. 28 Technical Assistance.. ................... 28 FOGAIN* ... 6*.* ..o.... . 28 FOMIN ....................................... ......... 29 FIDEIN ...**.... .................... ....................... . ....... 30 ANNEXES 1. FOGAIN: Characteristics of Subprojects Financed Under Loan 1552-ME ........................ .... . 32 2. Table 1 - FOGAIN: Economic Characteristics of Subprojects Financed Under Loan 1552-ME ....... 34 Table 2 - FOGAIN: Financial Characteristics of Subprojects Financed Under Loan 1552-ME ....... 36 3. FOGAIN: Characteristics of Subprojects Financed Under Loan 1881-ME* ............. ........................ ..##* ...... 39 4. Table 1 - FOGAIN: Economic Characteristics of Subprojects Financed Under Loan 1881-ME ....... 41 Table 2 - FOGAIN: Financial Characteristics of Subprojects Financed Under Loan 1881-ME ....... 44 5. Table 1 - FOMIN: List of Subprojects Financed Under Both Loans (Loan 1552-ME) ..................... 47 Table 2 - FOMIN: List of Subprojects Financed Under Both Loans (Loan 1881-ME) ....... 49 6. FOMINs Characteristics of Subprojects Financed Under Both Loans ............. .. 52 7. FOMIN: Financial Characteristics of Subprojects ........ 54 8. FOMIN: Status of Financed Projects (as of December 31, 1985) ................. .. 58 9. FOMIN: Analysis of Equity Investments ........ 59 10. FIDEINs Characteristics of Subprojects Financed******** 62  TABLE OF CONTENTS (cont'd.) PaRe No. 11. FIDEINs Financial Characteristics of Subprojects ....... 64 12. FIDEINs Analysis of Arrears as of December 31, 1985 .... 66 13. Analysis of FOGAIN Loan Approvals as of December 1985 ... 67 14. Analysis of FOGAIN Loans Approved in 1985 .......O-*..... 69 15. FOGAIN's Key Financial Indicators 1982-85 ....... ........ 71 16. Loan Portfolio of FOGAIN ...........00 ...... ...... to ...# 72 17. FOGAIN's Income Statement ................. ......... 73 18. FOGAIN's Balance Sheets . .... ... .. ..... . .. . 74 19. FOMIN: Summary of Operations (1980-85) ......... .... .... 75 20. FOMIN: Analysis of Arrears as of December 31, 1985 ..... 76 21. FOMIN: Projected and Real Income Statements, 1978-84 ... 77 22. FOMIN: Projected and Real Balance Sheets, 1978-84 ...... 78 23. FIDEIN: Summary of Operations .. ........ .. 80 24. FIDEIN: List of Operations (Standard Factory Buildings) .............. * ....................................... 81 25. FIDEIN: Income StatemertJ,, Projected and Actual, 1979-84 ................. 83 26. PAI Income Statement ............ . ............... 84 27. PAI Balance Sheets ..... ............. 85 28. PAI Summary of Financial Statements ..... ................. 86 29. FOGAIN's Interest Rates to Final Bo,-rowers and Margins of Intermediaries in Effect Since September 1979 ...... 87 30. Table 1 - Index of Manufacturing Production 1975-86 ..... 88 Table 2 - Industrial Physical Production ................ 88 FIGURES 1. Interest Rates and Inflation .............. .......... .... 89 2. Domestic $ Interest Rate and Financial Savings .......... 89  MEXICO PROJECT COMPLETION REPORT First and Second Small and Medium Scale Industry (SMI) Development Projects Loans 1552-ME and 1881-ME PREFACE This report reviews the performance of the First and Second Small and Medium Scale Industry (SMI) Development Loans channeled through NAFIN (Industrial Development Bank) for onlending by the following three NAFIN trust funds: Fondo de Garantia y Fomento a la Industria Mediana y Pequena (FOGAIN), Fondo Nacional de Eomento Industrial (FOMIN) and Fideicomiso de Conjuntos, Parques y Ciudades Industriales (FIDEIN). The coordinating committee for the project agencies which was created as part of the Integrated Program for SMIs, Is known as PAI. The report covers the implementation period from the inception of the PAI program in 1978 until 1985 which is the termination date for the Second SMI Loan (1881-ME). The PAI scheme of integrated support to SMI's is a unique experiment, in that it is the only SMI loan of its type supported by the Bank. Under the project, different types of financial services were brought together, though the beneficiaries were allowed the flexibility to choose from amongst the line of credit they needed and in this sense the acronym "Integrated" could be considered a misnomer.. The PAl extension service however lent some credence to an integrated approach in that it served the clients of FOGAIN, FOMIN and FIDEIN, in promoting their programs and providing badly needed technical advice. Loan 1552-HE was approved on April 27, 1978, declared effective on January 12, 1979, and disbursed until September 30, 1982. The delays in loan effectiveness were mainly due to the lack of agreement to implement interest rates agreed upon during negotiations. Loan 1881 was approved on June 24, 1980, and became effective on December 22, 1980. The loan was completed on June 30, 1985. This report is based upon the data obtained during two Bank missions to Mexico in May and June 1986. The preparatory work for this PCR was carried out under the overall direction of PAI's Technical Secretariat. The final report was prepared by the Industrial Development and Finance Division 2, of the Projects Department, Latin America and the Caribbean Region. In accordance with the revised procedures for project performance audit reporting, this Project Completion Report xms read by the Operations Evaluation Department (OED), but the project was not audited by OED staff. The draft report was sent to the Borrower for comments; however, none were received.  - ii - MEXICO PROJECT COMPLETION REPORT First and Second Small and Medium Scale Industrr (SMI) Development Projects Loans 1552-ME and 1881-ME BASIC DATA (Amount in US$ million) I. PROJECT DATA Loan 1552--M Loan 1881-RE Original Loan Actual/Reiised Original Loan Actual/Revised Date Date Date Date Board Approval 4/27/78 4/27/78 6/24/80 6/24/80 Loan Signing 5/04/78 5/04/78 9/29/80 9/29/80 Effectiveness 10/01/78 1/12/79 12/22/80 12/22/80 Completion of Comitments 6/30/81 9/30/81 12/31/83 12/31/84 Loan Closing 6/30/82 9/30/82 12/31/84 6/30/85 II. CUMKULATIVE DISBURSEMNTS Original Disbursed Cancelled As of Dec. 31, 1985 Loan No. 1552-ME 47.0 47.0 - Loan No. 1881-ME 100.0 100.0 - Loan 1152-ME Fiscal Year FY78 FY79 FY80 FY81 FY82 (i) Estimated 10.0 17.5 13.5 6.0 - (ii) Actual - 27.2 12.7 5.4 1.7 (ill) (ii) as x of (i) - 1.55 0.94 0.90 - Loan 1881-ME Fiscal Year FY80 FY81 FY82 FY83 FY84 FY85 (i) Estimated 8.0 33.0 30.0 14.0 3.0 - (iL) Actual 24.0 22.0 26.0 20.5 7.5 - tili) (ii) as 2 of (i) 3.0 0.66 0.52 1.46 2.5 -  - tM - III. MISSION DATA No. of No. ov- Dates of Item Month/-r Staff Weeks Persons Ra-Orts Pro-appraissl 1/77 62.2 a 9/77 Appraisal I 8/77 84.8 4 A/78 Nsotlations 1978 10.8 Supervision 1 11/78 .1 a 12/78 Supervision 1I 2/79 3.6 1 3/79 Supervision III 11/79 4.0 4 12/79 Total 136.2 Pro--pp*isal 23.3 Appraisl I 1/ 2/80 47.3 7 i/80 Negotiations 8.7 Supervision IV 6/80 3.7 2 7/80 Supervision V 9/80 4.0 l 10/80 Supervision VI 1/81 6.0 1 2/81 Supervision VII 10/81 7.0 4 12/01 Supervision VIII 8/82 10.7 8 0/82 Supervision IX 2/ 9/83 3.3 1 11/83 Supervision X 2/ 3/84 4.1 1 7/84 Completion 6/86 5.0 2 6/87 STAFF W (FY 77-87) lana 1552-if FY77 FY78 FY79 FY80 FY81 FY82 FM83 FY84 FY85 FY86 FY87 MTAL Preappraisal 51.3 0.8 - - - - - - - - - 52.2 Appraisal - 63.2 - - - - 1.4 - - - - 64.6 Negotiatiom - 10.8 - - - - - - - - - 10.8 Supervisiam - 0.1 7.5 7.7 5.9 2.7 - - 0.1 10.7 6.4 41.2 Other - - 1.4 2.2 0.1 1.0 0.1 - - - - 4.9 Total 51.3 74.9 9.0 9.9 6.0 3.7 1.5 - 0.1 10.7 6.4 173.6 Lon 1881-4 FY77 FY78 FY79 FY80 Flb1 FY82 FY83 FY84 FY85 FY86 FY87 TOML Preappraisal - - - 23.3 - - - - - - - 23.3 Appraisal - - - 47.3 - - - - - - - 47.3 Negotiatisn - - - 6.7 - - - - - - - 6.7 Supervisiun - - - - 7.1 8.0 3.3 4.1 4.0 11.8 6.8 45.1 Other _ - _ 2.1 0.1 - - - - - - 2.2 Total - - - 79.3 7.2 8.0 3.3 4.1 4.0 11.8 6.8 124.6 IV. FOLLOW-ON LOANS Third and Fourth Small and Medium-Scale Industry Development Project LOAN 2325-MIE LOAN 2858-ME (US1175.0 million) (US218.0 million) Approved 6/23/83 6/26/87 Loan Agrement 86/1583 8/26/87 Effectiveness 1/16/84 6/2 /87 Completion of Comitments 12/31/86 6/26/87 Loan Closing 8/30/87 6/25/87 1/ Appraisal mission for Loan 1881-li 2/ Includes supervision for Loan 2326-ME  - iv - MEXICO PROJECT COMPLETION REPORT First and Second Small and Medium Scale Industry (SMI) Development Projects Loans 1552-ME and 1881-ME HIGHLIGHTS 1. The PAI program was the first Bank involvement in an integrated development program for SMI's in the industrial sector and as such the implementation experience is unique, in so far as it relates to the provision of various financing packages to a common clientele (SMI's) and also the experience with the coordinating mechanism set up to manage a complex project with several componentsc The two loans under review in this report were the first two industrial credit loans for SMI development made to Mexico in support of the Government's Integrated Program for SMI development known as PAI. The loans were in the amounts of US$47 million and US$100 million, respectively. The project was implemented through the following three key trust funds managed by NAFIN: FOGAIN, (the trust fund for credit to SMI's); FOMIN, (the trust fund for equity and quasi-equity financing) and FIDEIN, (the trust fund responsible for industrial park development). In addition, for the first time an industrial extension service was established which worked in close collaboration with the implementing agencies and other existing specialized technical assistance agencies in Mexico. 2. The Bank's principal objective under the two loans was to foster more rapid growth in output and employment and to promote regional development by strengthening the existing system of providing term financing for financially sound and economically efficient investment projects of small and medium-sized firms (para 2.02). Secondary objectives focused .on the institutional development of the executing agencies, the operation of an industrial extension service, and support to other technical assistance and research inatitutions involved in the industrial sector. Project implementation has been reasonably smooth and loan fund utilization timely (para. 4.01-4.03), particularly in the early implementation period which coincided with the boom conditions in the Mexican economy (para. 1.02-1.03). 3. The subprojects financed by FOGAIN have had a positive economic impact with significant employment generation. The bulk of subloans have gone to the small end of the industrial size spectrum (paras. 4.07-4.14). The FOMIN equity investment portfolio had mixed results in terms of profitability, but most loans were in support of high priority investment activities. Nevertheless, FOMIN was not instrumental in financing new enterprises and most of its equity investments were confined to existing enterprises or holding companies. Under both loans, FIDEIN operations went through an experimental stage, and eventually it withdrew from its direct construction role (para. 3.24). While FIDEIN operations were helpful to the eytent that they had a positive demonstration effect, the investments did not make any significant contribution to the decentralization objective rarely did  existing enterprises relocate in industrial zones, only new enterprises sought space in industrial parks which offered clear market and other advantages. Similarly, the industrial estates, developed by FIDEIN had varied success, with a high occupancy rate for only half of those developed, while the rest failed to expand their available capacities fully. 4. While the two SMI loans were conceived and implemented without any broad financial sector policy objectives, the Bank must be credited with influencing interest rate policy for lending to the SMI sector (para. 3.08 - 3.10). At the time the first project was appraised FOGAIN operated a fixed interest rate structure which was not subjected to periodic reviews to bring it in line with either market interaes rates or the cost of FOGAIN funds. The Bank assisted projects gradually initiated the rationalization of interest rates by FOGAIN moving it to link its subloans to the average cost of funds (ACr) index and under the 3econd SMI project FOGAIN narrowed the interest rate differentials which existed for different sizes of borrowers and for different locations. Through this gradual process of reform the Bank helped to introduce variable interest rates for FOGAIN lending under the third SMI project (Loan 2325-ME). The introduction of variable interest rates clearly demonstrated a positive impact on the financial operations of FOGAIN which had previously experienced operating losses due to the fixed interest rate (para. 5.06). 5. The loans have been considerably more successful with respect to their primary than to their secondary obiective which was to improve the functioning of the three participating institutions. The Bank loans have had a limited impact on the functioning of the institutions which were already well established at the time the PAI program was initiatedi and in fact in the case of FOGAIN, Bank funds only represented a minor portion of its total operational resources. While both these factors reduced the Bank's leverage in inducing any radical changes in the operating practices of the executing agencies, some irportant organizational and procedural improvements were initiated at the Bank's advice. The degree of receptivity to changes however varied for each institution. While FOGAIN undertook some positive organizational and staffing changes and made improvements in its operating procedures (para. 3.11-3.13), FIDEIN continued to suffer from institutional weaknesses throughout the implementation of the two loans (para. 3.20-3.26) and made little effort to adopt changes as suggested by Bank missions. FOMIN took some good initiatives in procedural matters (para. 3.18-3.22), but also suffered from staff weaknesses. 6. The participating financial intermediaries (PFI) made limited efforts under the loans with respect to improving their subproject preparation, appraisal and supervision practices. The PFI's seem to have concentrated their subproject evaluations on the assessment of credit worthiness and the availability of adequate collateral. There was very little emphasis on determining the financial and economic viability of the investments. As an apex rediscounting agency dispensing subsidized credit resources FOGAIN could have exercised influence in improving the operating procedures of the PFI's. Regrettably FOGAIN chose not to disturb the delicate balance between a very strong commercial banking sector (which was fully responsible for the credit risk) and a government trust fund whose role according to the banks was to channel funds to a target group of beneficiaries namely the SMI. FOGAIN on its own part, confined itself to  - vi - reviewing the end use of the funds and the eligibility of the borrowers and more or less accepted the credit evaluation of the PFIs. Under the influence of the Bank, however, FOGAIN began to carry out selected subproject evaluations, but it was not until the third SMI loan that formal subloan evaluation criteria were agreed upon. On the other hand, the impact of the Bank advice was much more apparent in the case of FOMIN which adopted improved evaluation practices and streamlined its supervision system. The performance of FIDEIN in the area of project evaluation .remained weak not only during the implementation of the first two projects but also in the early periods of the ongoing loan. A clearer focus of technical assistance activities in the above mentioned areas, could have had a positive impact on the institutional building. 7. The variation in institutional improvement, particularly the lack of progress by FIDEIN, can be attributed to the undefined character of the agency and the lack of strong commitment to strengthening the institution by NAFIN. Experience has shown that unless there is strong commitment to change at the government and institutional levels, interventions in institutional strengthening are bound to fail. Either at the NAFIN or FIDEIN level there was always a reason for inaction. The very future institutional status of FIDEIN has been in question, and several proposals to either merge it or expand it were considered by NAFIN during the implementation of the two projects. Under these circumstances, the projects should have included concrete changes prior to implementation rather than proposed studies and agreements on changes at some future date. 8. The overall coordinating mechanism set up under the project including the financing arrangements proved to be the least successful aspect of the project and frequently led to institutional friction (paras. 3.01-3.03). This situation was further complicated by the fact that for' most of the implementation period the same person was director of PAI and FOGAIN, All three executing agencies were answerable to their own executive committees and while they depended on the financing resources of PAI they resisted any direct management role by the PAI Technical Secretariat. The resistance was particularly strong in the case of FOGAIN which represented almost 70% of the PAI program, while the financing from the PAI fund onlj provided 6% of FOGAIN funding when the second loan was appraised. In retrospect the implementation of the SMI projects has demonstrated that in setting up an overall coordinating mechanism adequate consideration should be given not only to the relationships between institutions but also to the fiiancing arrangements which in this case proved to be an overriding factor in determining the effectiveness of integration under the Program. Another important lesson from the operation of an integrated program with a central coordinating body and a common financing fund is to confine the coordinating role simply to providing policy guidance and to monitoring the program components, and not to attempt to manage them directly. Besides it is preferable to ca^italize the executing agencies directly and to concentrate project technical assistance resources in strengthening the executing agencies. 9. 'An impressive and instructive part of the project wee the initiation and operation of an industrial extension service which has developed into an important support function for SMI operations in Mexico and can be replicated in other countries (paras. 6.01-6.08). The PAI extension service has now been fully institutionalized and is part of  - vii - NAFIN's regular industrial promotion activities. These activities are largely supported through Government budget, but consideration is being given to imposing a nominal fee for users of this important and useful service. 10. The two loans aimed at supporting the technical assistance activities of other institutions involved in industrial development (para. 6.09). The target group of beneficiaries under this component were in practice quite different from that originally envisaged, and above all the institutions were not exclusively focused in serving the SMI. A major drawback of a loosely defined technical assistance program was that the three main executing agencies did not benefit adequately from the available resources which were managed by the PAI technical secretariat. Under the Fourth SMI loan a detailed technical program has been prepared for each participating institution which will ht.-zceforth be responsible for Implementing its own technical assistance program.  MEXICO PROJECT COMPLETION REPORT First and Second Small and Medium Scale Industry (SMI) Development Projects Loans 1552-ME and 1881-ME I. DNTRODUCTION 1.01 Bank Lending for Small and Medium Scale Industry (SMI). In 1978, lAPIN (the National Industrial Development Bank) initiated an Integrated Program for SMI development, known as PAI.1/ This program was to be liplemented through three NAFIN-administered trust funds which were already2/ supporting individually SMI activities. The main components in the PAf program were: (1) a credit rediscount scheme involving commercial banks and operated by a second tier Institution the "Fondo de Garantia y Fomento a la Industria Mediana y Pequena' (FOGAIN); (2) risk capital investments by the "Fondo Nacional al Fomento Industrial" (FOMIN); (3) credit and a technical assistance for industrial estate development by *1ideicomlso de Conjuntos, Parques y Ciudades Industriales' (FIDEIN); (4) technical assistance to be delivered by an extension service of PAI to help SKI improve its operation and to obtain assistance from the above Funds and/or from other agencies In the country. To date, the Bank has financed three projects under the PAI program for a total amount of US$322 million. A fourth SMI project (US$185 million) has been recently negotiated with overall objectives similar to the earlier projects, but with a deeign that incorporates the major lessons of this PCR, and the implementation experience gained from the ongoing third SMI project (Loan 2325-HE). 1.02 Macro-Economic and Sectoral Performance. The implementation perlod for the first and second SMI projects (1978-84) spans two distinct phases of economic activity -- the year 1978 witnessed a resurgence of productive activity and the economy experienced four years of exceptionally high growth rates, averaging 82 p.a. for GDP and about 9% p.a. for manufacturing output. The second and latter project implementation period is marked by the severe economic and financial crisis which commenced in 1982 and in some respects persists to date. The earlier boom conditions were associated with the initiation of intensive exploitation of the country's hydrocarbon resources, a substantial growth in public expendlture, and a similar expansion in private sector investment. During the high growth period, industrial imports grew rapidly, equivalent to over 172 p.a. in real terms. The 1982 crisis, however, brought industrial lt Programa Apoyo Integral - Integrated Support Program 2/ FOGAIN was established in 1954, FOMIN started functioning in 1972 and FIDEIN in 1970.  - 2 - growth to an abrupt halt and in fact reversed the period of sustained g'rowth. Many medium to large-scale Mexican companies came close to bankruptcy, and even though small scale industries showed more resilience because of their ability to adapt their output to the low demand situation, they reduced their capacity utilization significantly. 1.03 Since disaggregated data on manufacturing sector performance is unavailable, the performance of the SMI sector is assessed by comparing the overall manufacturing sector growth to the performance of subsectors dominated by SMIs. The highest concentration of Mexican SMIs is in consumer goods (mainly non-durables) and the intermediate goods subsector. While the annual average rate of growth of physical production in the manufacturing sector was 3.6% for the period between December 1975 and June 1986, consumer goods grew at an annual rate of 3.7%, and intermediate goods at an annual rate of 4.2%. Within the consumer goods industries, however, the physical production of non-durable consumer goods (preponderantly SMIs) grew by 3.6% annually whereas durables (mainly produced by large scale industries) grew by 1.9%. Another important feature is that when compared to the overall growth of the manufacturing sector, the growth of both the non-durable consumer goods and intermediate goods subsectors has been much faster during the upswings, and the reduction in growth during the recessionary periods has been more moderate (Annex 30). In sum, SMIs were concentrated in sectors with a smaller downward risk when compared with the manufacturing sector as a whole, a factor which explains their resilience during the-1982/83 crisis. 1.04 Interest Rate Policy and Supply of Financial Resources. Interest rate arrangements on Bank loans in Mexico have traditionally been designed,. negotiated and supervised on a project by project basis and the rates under the two loans reviewed were no exception to this rule. The principal instrument of Mexican interest rate policy is the Central Bank's authority to set rates on deposits of various maturities. A weighted average of the actual deposit rates then determines the average cost of funds (ACF). The ACF In turn serves as a reference rate for many types of loans. The interest rates under the two SMI projects were equal to the ACF during the earlier implementation years (1978-79) but overtime a gap developed between the ACF and the preferential interest rates charged by FOGAIN (para. 3.08- 3.10). Yore importantly FOGAIN's interest rates were fixed over the life of the subloans and annual adjustments to the level of interest rates were applicable only in the case of new subloans. 1.05 Mexico's interest rate policy has undergone significant changes during the implementation period of the two projects. Prior to 1979, interest rates remained stable during long periods; then, between August 1979 and the beginning of 1983, the policy was to adjust interest rates vis-a-vis inflation with a lag. While from 1978 to late 1981 the interest rate levels as determined by the ACF index remained more or less in line with the inflation rate the level of interest rates had fallen far below the inflation rate and the rate of exchange rate devaluation during the years 1982-83 (Figure 1). The severe inflation and movements in the real exchange rate during 1982-83 greatly aggravated and extended the severity of distortions in the level and dispersion of interest rates and the Bank encountered increasing resistance to covenants requiring adjustments in the interest rates levels. In the context of the SMI program, while agreement was reached in late 1982 with NAFIN to adjust FOGAIN interest rates (on a  quarterly basis), this agreement was not implemented until mid-1984, in the face of rapidly accelerating inflation, which pushed up the ACF to unprecedented levels. FOGAIN also considered it counterproductive to raise interest rates, at a time of severe economic recession and low level investment demand. 1.06 Deposit interest rates also fluctuated widely over the implementation period of the two projects, showing a consistently upward movement since 1982. During 1979 and 1980, the effective real cost ot the average term deposit to the commerciel banks was negative; after dropping to about -5% in 1980, it recovered to 1% in 1981, before plummeting again to a low of -22% in 1982. Since 1983, however, the average real cost of funds (compounded monthly) to the banking system (ACF) has become positive and reached a real annual rate of 5% in 1984 and 6% in 1985. An important feature of the interest rate policy in Mexico has been the decision to set effective longer term rates below short term rates. The objective was to influence inflationary expectations by putting a ceiling on long term rates below that on short term rates. Its immediate effect, however, has been to dampen the growth of badly needed long-term resources to the financial sybtem. For example, in 1980, term deposits with maturities of one year and over accounted for 47% of total term deposits from the pablic in commercial banks, while in 1985, they accounted for 3X. Consequently, subsidized credit lines through trust funds like FOGAIN were an important source of financing, particularly for the SMI's which had limited access to scarce commercial bank resources. II. PREPARATION AND APPRAISAL OBJECTIVES OF THE LOANS 2.01 Loan Preparation. Preparatory work for the first Small and Medium Scale Industry (SMI) Development Project was carried out within the framework of the InduL,rial Sector mission during October 1976 followed by pre-appraisal mission in January 1977. The appraisal took place in August 1977 and a loan (1552-ME) for US$47 million, was approved by the Board on April 27, 1978. The loan was signed on May 4, 1978 but only became effective December 1, 1979 due to a prolonged discussion on interest rates (the Mexican Government was not ready to accept an upward adjustment in rates as agreed during negotiations), and a subsequent delay in the ratification of the Guarantee Agreement by the Mexican Congress. Since the first SKI project disbursed rapidly (para. 4.01) and the demand for credit was quite high the Mexican Government requested in November 1980 that a follow-up project be prepared. The second SMI project (Loan 1881-ME) for US$100.0 million was approved by the Board on June 24, 1980 and became effective on December 12, 1980. The loan comprised the same components as in Loan 1552-ME except that in addition to the PAI extension service other specialized technical assistance institutions such as CENAPRO (Centro Nacional de Productividad) INFOTEC (technical information and advice) and FONEP (feasibility studies) could also benefit from the T.A. funds. 2.02 Appraisal Objectives. The objectives of the two loans, were essentially similar namely to: (i) foster more rapid growth in industrial output and employment by providing finance for SMI's wishing to increase their productive capacity; (ii) assist in implementing the national plans for industrial decentralization and regional development by making financial and technical assistance available to firms outside the main  -4- metropolitan areas; (iii) strengthen the support system for SMI by coordinating more closely the activities of existing financing and technical assistance institutions, including the operation of an extension service, which was to be expanded under the follow up project to lesser developed regions. 2.03 The extent to which the appraisal objectives, including the specific institution building objectives, were achieved, is described and reviewed in the following chapters. III. INSTITUTIONAL DEVELOPMENT Program Coordinating Committee (PAI) 3.01 Organization and Coordination. The PAI program was to be governed by a high level Program Coordinating Committee, headed by the Director General of NAFIN and including the heads of all participating institutions, as well as representatives from the MIinistries of Finance, Commerce and Industry, and Banco de Mexico. A Technical Coordinating Committee with a Technical Secretariat was set up to oversee the operational aspects of the program. While initially the Secretariat was based in NAFIN, it was later moved to FOGAIN, with the Director General of FOGAIN also acting as the technical secretary. This organizational change was aimed at increasing the linkage between the Secretariat and the largest operating unit (FOGAIN). Because of the experimental nature of the coordination role of PAI, the Bank appraisal missions considered the institutional arrangements as "transitional" and necessary to establish the program. Bank supervision missions expressed the need to separate the functions of PAI staff from those of FOGAIN, because PAI extensionists were becoming too closely involved with the credit promotion activities, at the cost of their technical advice functions. While some improvements were introduced in January 1979 by appointing separate deputy directors for PAI and FOGAIN operations, a complete separation between them did not take place until 1984. PAI was then put under the direction of a NAFIN director who doubled up as its Technical Secretary. The new management expanded PAI staff and began to take a direct role in overseeing the operations of the executing agencies anid acting as a filter for subprojects submitted to the Bank for approval and/or reimbursement. The institutional arrangement with various layers (Executing Agency, NAFIN, PAI) to control 'subproject quality", caused unwanted delays in subproject approval. 3.02 The underlying cause of the institutional friction was the PAI financing arrangement whereby the trust funds were expected to pass on to a common PAI Financing Fund the loan recoveries from the Bank loan as well as a major portion of their interest rate spread. This requirement seemed reasonable when account is taken of the fact that PAI not only provided the trust funds with the counterpart funds but assumed the foreign, exchange risk (for which it received an interest rate spread). Nevertheless, FOGAIN, the largest participant, refused to fully honour its financial commitment; it witheld transfers of loan recoveries to PAI and paid only a  small part of its interest spread to PAI.3/ The institutional problem worsened when PAI's Coordinating Committee did not meet officially after 1982 and in its absence the technical secretariat started to assume direct responsibility for coordination. 3.03 The Bank's supervision missions were fully aware of the PAI coordination problems, and several attempts were made at top levels in NAFIN to resolve these problems. However, since the coordination difficulties did not materially affect the actual implementation pace of the program (which was essentially in the hands of the executing agencies), the institutional relationships between PAI and the executing agencies were left unchanged even under the Third Loan 2325-ME. The financial arrangements between PAI and the executing agencies did adversely affect the financial soundness of the whole program (para. 5.03) and obliged the Government to review the entire institutional and financial relationships under the PAI program. Consequently, agreement has been reached, within the context of the Fourth SMI loan, with Government to abolish the PAI fund and to capitalize the implementing agencies directly. In addition, measures have been taken to not only ensure a more effective coordination between the executing agencies, but to oversee the implementation of the entire Program. Fondo de Garantia y Foviento a la Industria Mediana y Peguena (FOGAIN) 3.04 The Institution. FOGAIN, the most important financial institution for small and medium industry (SMI) in Mexico, was established in 1954 as a Government trust fund (Fideicomiso) administered by NAFIN. It is a second-tier institution, rediscounting credits made to SMI for equipm.ent, machinery, factory construction and working capital. Under Loan 1552-ME the FOGAIN component amounted to US$30.0 million, increasing to US$70.0 million under the follow-up loan. 3.05 Management and Organization. FOGAIN's main decision making body is its National Technical Committee while its day-to-day operations are the responsibility of a management team headed by a Director General. From 1978 to 1984 FOGAIN had three Director Generals: from 1978 to 1982, it was headed by Mr. Tomas Gonzalez Hinojasa, formerly Credit Manager of a leading Mexican Financiera (SOMEX), who introduced a comprehensive reorganization as recommended by the Bank under the first SMI project. The reorganization consolidated related functions and responsibilities, upgraded key middle management and increased the decentralization of staff at the regional offices with the objective of transferring loan processing authority to field staff. 3.06 With the change of Director in 1982, there was a significant outflow of professional staff from both PAI and FOGAIN, causing a serious slowdown in operations. In February 1984, separate directors were appointed for PAI and FOGAIN which had hitherto operated jointly. Mr. Sergio Luis Cano, who was previously head of FOMIN, was appointed the new Director of FOGAIN. He made substantial organizational changes including the setting up of an Internal Audit Office, a Legal Department, 3/ It is important to note that 80% of FOGAIN's funds came from resources that did not administratively pass through PAI.  - 6 - and a Credit Monitoring Department which began operating in 1985. Through the successive three Bank loans, FOGAIN has developed into a stronger and more mature financial institution. Nevertheless, further decentralization of FOGAIN subloan approval and processing functions at the regional level is needed (para. 3.11-3.12). 3.07 Staffing and Staff Training. In 1977 at the time Loan 1552-ME was appraised, FOGAIN staff was about 100 employees, most of whom were concentrated in Mexico City with a few representatives stationed at NAFIN's regional offices. By 1979, when Loan 1881-ME was appraised, FOGAIN staff had increased to 130, but by 1982 this number went up to 150. When FOGAIN separated from PAI in February 1984, the staff numbers had jumped to 454 professionals of which only 21 were at the regional offices. As of end 1986, FOGAIN staff totaled 449, consisting of 39 economists, 70 public accounts and administrators, 10 lawyers, 20 engineers, 115 administrative officers and- 195 other professionals. The expansive growth in staff numbers in-late 1985 and early 1986 was in keeping with the rapid increase in FOGAIN'S lending activities (para. 4.06). Overall FOGAIN staff is of high quality with the majority having good qualifications. Furthermore during 1980-82 special training programs were developed to help the staff acquire the necessary skills to adapt to an expanded credit program. In fact recently an internal training division has been set up to identify, on a continuing basis, the staff training requirements and to arrange such training. Operating Procedures 3.08 FOGAIN Interest Rates. At the commencement of Loan 1552-ME, FOGAIN was using fixed interest rates (at levels set in 1977) which varied by geographic zones and averaged 13 percent. During negotiations for Loan 1552-ME, the Bank proposed that interest rates be no less than 16% until December 1978, and three points above the ACF index thereafter. The pro- posed rates were accepted reluctantly at negotiations but were not formal- ized by Government after loan signing. After delaying project start up for almost one year, it was agreed to adopt a more flexible interest rate structure which permitted FOGAIN to fix final interest rates between one to three points above the ACF. This system was a marked improvement over the earlier fixed interest rate regime, in that it allowed for periodic revisions in the interest rates to reflect both the prevailing market rates and allowed rates to respond to any significant changes in the ACF index. 3.09 Further progress on FOGAIN interest rate policy was achieved during loan negotiations for 1881-ME. The Bank proposed another increase in interest rates from July 1980, so that the average would be one percentage point above the average ACF for the preceding twelve month period (May 1979-April 1980). For the first time agreement was reached to adjust interest rates more frequently than the annual adjustment called for in the Loan Agreement. An interest rate adjustment was to be made if the ACF index deviated by more than 3 points in the last quarter. In addition discussions were initiated to reduce the interest differentials within categories of borrowers. The Mexican Government opposed the increase in rates on grounds that the larger and more established enterprises could borrow competitively and freely in international markets (where rates were relatively lower). They also wanted to maintain the differentials in interest rate by location, since it was an important instrument to promote  -7- industrial decentralization. The interest rate mechanism finally agreed (Annex 29) included clauses for periodic review of rates,, and one which showed an overall increase in rates for the medium-scale enterprises; the differential by zones was maintained. 3.10 During implementation of Loan 1881-ME, the Mexican authorities failed to adjust interest rates (in February 1981) and requested that the Bank grant a moratorium until the end of 1981 on all adjustments. The main reason for the non-compliance with interest rate adjustment clause was the intense political presures generated at a time when the presidential candidate was being nominated. In May 1981 the Bank suspended disbursements against the Loan, and finally succeeded in obtaining an upward adjustment in FOGAIN onlending rates. In summary, the Bank worked closely with Government and NAFIN authorities under the two loans to bring FOGAIN interest rates closer to market rates. However, FOGAIN continued with a fixed interest rate for its outstanding loans, which as inflation accelerated after 1982 caused serious portfolio losses. It was not until the negotiation of Loan 2325-ME In 1983 that agreement was reached with Mexican authorities to adjust interest rates of outstanding loans to correspond with the interest rates of new subloans. Despite the limited leverage the Bank had in influencing FOGAIN's interest rate policy, positive changes were introduced under the three SMI loans in Mexico. And the Bank may rightly clalm that it helped rationalize interest rates for the SMI sector to reflect more closely the cost of funds. However, since the issue affected Bank-assisted projects across sectors, it was concluded that this issue was best addressed within the context of Bank's macro-economic policy dialogue with the Mexican Government (para. 1.05). The policy dialogue on interest rates and subsidies between the Bank and Mexican authorities finally culminated in the General Interest Rate Agreement (GIRA) concluded in 1984. 3.11 Appraisals. Under the FOGAIN rediscounting system, the participating financial intermediaries (PFIs) are primarily responsible for evaluating the creditworthiness of loan applicants and determining their financial viability before submitting a financing request to FOGAIN. The analysis of subloans at FOGAIN level was thus limited to ascertaining: credit eligibility, expected economic impact as regards employment generation and sectoral factors; marketing prospects; and the amount of credit needed and the financing terms. While FOGAIN's subproject appraisal procedures have been evolving slowly, it has made little effort to promote standard evaluation practices at the PFI level. Even itself FOGAIN did not calculate ERR's, and FRR's were estimated only on a ex-post basis and that too for the larger subloans. Recently FOGAIN has initiated, as a result of the Bank's insistence, to calculate rates of return for subprojects of more than US$100,000. Similarly, the revised criteria for appraising permanent working capital subloans was agreed upon with the Bank to take more account of inflation. Lack of adherence to subproject evaluation criteria has been a disappointing aspect of FOGAIN's operations. 3.12 Project Supervision. As a second-tier institution, FOGAIN considers its supervision role to be one of "controlling' the end use of its funds by PFIs. Thus, supervision of subloans has been performed primarily by the PFI's who have generally concentrated on loans in arrears. FOGAIN has performed some analyses of samples of FOGAIN-financed enterprises to ascertain if its funds were being used for the purposes  stated in the subloan submissions. It was not until 1984-85 that FOGAIN strengthened its supervision efforts by creating the Credit Monitoring Department and by hiring outside consultants to assist in this function. Bank supervision missions recommend that subloan follow-up not only verify the ultimate use of loan proceeds, but also identify problems confronting the beneficiaries. The most serious gap in FOGAIN's monitoring role is that it does not maintain any detailed information on the status of its loan portfolio iith the PFI's. Bank supervision missions have consistently commented on this weakness, but no concrete measure has been initiated until the Fourth SMI loan which requires a time bound plan to collect information on arrears. The recent computerization of FOGAIN's activities has improved subproject quality control, but there is stiU a strong need to strengthen the capability of commercial banks in this regard, particularly with respect to helping them focus on the development impact of subprojects, instead of their overriding concerns with simply ensuring collateral. Under the recently approved Fourth SMI project technical assistance resources have been provided to strengthen FOGAIN's information system including its review of subprojects performance. 3.13 Reporting Requirements. FOGAIN's internal auditing has been carried out by a special unit of NAFIN. Additionally, FOGAIN has been independently audited by outside firms. The audit report for 1981 was qualified but arrangements made for the following-audit years were satisfactory. Independent auditors are designated every year by the Comptroller General of the Republic under a procedure for selecting and monitoring auditors who are required to follow the principles of Government audits. In general the Bank found FOGAIN audit procedures adequate and reports were submitted in time. Fondo Nacional de Fomento Industrial (FOMIN) 3.14 The institution. FOMIN was established in 1972 as a Government trust fund administered by NAFIN. Its role is to stimulate the creation of new SMI enterprises or the expansion or strengthening of existing ones by contributing equity capital as a minority shareholder (initially less than 33% and, from 1980, less than 49% of the company's equity). FOMIN was allocated US$10.0 million of loan resources under the first loan and US$12.0 million under the follow-up loan. 3.15 Management, Organization and Staffing. During most of the implementation period of Loans 1552 and 1881, FOMIN had the same Director, Sr. Jorge Luis Cano, an experienced manager who stayed with FOMIN until 1984. FOMIN has also had, throughout, a relatively capable and well-motivated staff, but turnover has been high and vacancies difficult to fill. The main reason for the turnover is the special skills FOMIN staff acquire in equity financing, which has been in high demand in the banking sector. FOMIN's total staff increased from 49 employees in 1978, of whom 24 were professionals, to 82 in 1985, of which 43 are professionals; in mid-1986 only three of them had been with the institution for more than two years. Although the lack of continuity in staffing showed no apparent adverse efforts on the level and quality of FOMIN operations, it put an excessive burden on FOMIN's top level staff to train and motivate the newly recru.ted staff. The Bank missions have consistently noted the need to enhance the renumerations of FOMIN staff, but NAFIN maintains that it has to have a uniform salary policy for staff all the trust funds it administers.  - 9 - 3.16 At the time of appraisal of Loan 1552-ME, FOMIN was organized in two major departments: one in charge of appraisals and legal procedures, the other portfolio control and sales of equity participations. By 1981, during implementation of Loan 1881-ME due to growth and staff shortages, appraisals and supervisions were being done by the same analysts. Bank supervision missions in 1981 and 1982 reported an imbalance between the evaluation of prospective new clients and effective portfolio control. FOMIN introduced changes on the Banks suggestion and under the present organizational structure, although appraisal and control are under the same Deputy Director (Sub-Direction General Tecnica) these activities are in different departments. Similarly supervision has been strungly reinforced over the last two years, with frequent visits to client companies and active participation in their boards of directors. 3.17 Staff Training. Part of the technical assistance zomponent of both loans was designated for the additional training of personnel but the training courses provided by the training institutions in Mexico were not ideally suited to the needs of a risk-capital investment company such as FOMIN. Besides, since government regulations restricted FOMIN's employees from travelling abroad to visit other more experienced risk-capital institutions, training has been confined largely to on-the-job training in the institution. Only one foreign expert in venture capital investment was brought in under the technical assistance components of both loans. Operating Policies and Procedures 3.18 Lending Policies. Initially FOMIN's irnvestments were mostly in the form of new issues of common stock. Based on advice from Bank and IFC, FOMIN moved in early 1981 towards investments in preferred shares and in convertible subordinated credits. This diversification broadened the range of FOMIN's potential customers and facilitated investment packages more adapted to the cash flows of the enterprises. Above all the use of convertible credit has also generated a more stable income for FOMIN. In practice, convertible credits have been favored over preferred stock by firms because interest on convertible loans is tax deductible. Until 1981-82 FOMIN's established practice was not to invest in financial restructuring of firms or to do so only in very exceptional cases. However, the crisis of 1982-83 created an increasing demand on FOMIN in this area which resulted in some incremental equity investments in its earlier clients and also new investments in firms undergoing financial difficulties. 3.19 Regarding-clientele focus, at the time of the approval of Loan 1552-ME, entrepreneurs were celuctant to accept a public institution, like FOPIN, as a partner even as a minority shareholder and FOMIN was forced, in practice, to work mainly with already established firms, many of which belonged to medium-size holding groups. These saw FOMIN as a means to expand capacity, to decentralize existing plants or to start new ventures. The time consuming and costly assessment of risk factors, and some provisions of its charter (such as the requirement for external auditing), generally precluded FOMIN from playing a significant role in financing small firms. Therefore, FOMIN's assistance focussed primarily on medium size enterprises (equity contributions of Mex$12.3 million on average during Loan 1552-ME and during Loan 1881-ME Mex$32.9 million on average (in  - 10 - the range of $700,000 to $1,000,000 in 1980-82 and around US$500,000 in 1980). The economic difficulties of 1982-83 also pressed FOMIN into providing more finance to cover working capital needs (including debt restructuring). This increased working capital financing caused cash flow problems for FOMIN, as some of these projects were not eligible for Bank financing and FOMIN was forced to use its own resources. 3.20 Project Appraisal and Supervision. While FOMIN's appraisal capabilities were well established prior to 1978, the following improvements were introduced during Loans 1552-ME and 1881-ME: calculation of ERR for subprojects using more than US$250,000 in Bank funds; increased emphasis on evaluating the technical aspects of projects, introduction of more imaginative financial packages through the combination of equity and convertible loans and attempts at co-financing with commercial banks. These changes notwithstanding FOMIN's appraisal procedures still use traditional approaches as in credit project appraisals with too little consideration for the abilities of the entrepreneurs and the market prospect for products. FOMIN's supervision was initially lax, but following Bank suggestions, supervision staff were organized according to sectoral activity rather than by geographical location of firms. And by the end of the implementation of Loan 1552-ME, FOMIN started issuing a monthly report with financial indicators on the status of enterprises in its portfolio. As part of the appraisal of the ongoing Loan 2325-ME a plan of action was prepared to rehabilitate the recoverable enterprises and to liquidate the failed investments. Since 1983, FOMIN started using outside consultants to carry out technical supervisions on a continuous basis. 3.21 Procurement and-Disbursement. A check of documentation for a sample of subprojects showed that FOMIN's disbursements procedures have been adequate. FOMIN's disbursement policies provide for the first 20% of FOMIN's investment to be disbursed when the firm has met all the agreed conditions. The remainder is disbursed against a time table agreed with the firm on a pari-pasu basis with other shareholders of the firm. Regarding procurement practices, as a general policy, FOMIN obtains price quotations from several suppliers and from time to time reviews the quality and technical adequacy of goods purchased. However, during the initial stages of execution of Loans 1552-ME and 1881-ME (1979-81) it was not normal practice for FOMIN to insist on an independent valuation of the equipment purchased. Prices were agreed by the foreign seller and the local entrepreneurs based mainly on reference prices of new equipment. More recently, FOMIN has insisted on the independent valuation of used equipment, either by registered foreign appraisers or by NAFIN's capital goods specialists. Under the proposed Bank Fourth SMI operation, FOMIN would establish explicit procedures for procurement of used equipment. 3.22 Auditing and Accounting Procedures. FOMIN's financial statements have been independently audited by reputable accounting firms. At the Bank's request, FOMIN has required annual audited statements and the use of acceptable accounting systems from all firms in its portfolio. Every year's financial statements require reformulation one year later to make the adjustments resulting from revaluation of equity participation in the client firms after their own audits have been completed. There is urgent need for streamlining FOMIN's management information base and reporting system but improvements have been hampered by bureaucratic delays in the acquisition of computer equipment. Technical assistance under the proposed Fourth SMI Loan would address this issue.  - 11 - Fideicomiso de Pargqus Industriales (FIDEIN) 3.23 The Institution. FIDEIN was established in 1970 as a Government trust fund administered by NAFIN to plan, promote and implement a national system of industrial estates as part of a national program of industrial decentralization. Under Loan 1551-ME the FIDEIN component was US$5.0 million increasing to US$3.0 million in the Second SMI loan. 3.24 Development Strategy and Operating Procedures. Between 1978 and 1985 FIDEIN underwent several changes in its lending policies and procedures; from a central coordinating/advisory institution to the "Fideicomisos Especiales" (the legal local entities owning and operating the individual regional publicly financed estates) assisting the Secretaria de. Asentamientos Humanos y Obras Publicas (SAHOP) in selecting the location of industrial estates and in feasibility studies of the projects. After approval of the locations SAHOP, in consultation with FIDEIN, prepared overall layout plans of the 500 ha site acquired in each case and proceeded to develop the initial 20 ha. The local "Fideicomiso Especial" was then responsible for implementation and management of the project. FIDEIN assistance was supposed to include: promotion and marketing of the itdustrial plots, accounting and financial control (including annual budgets and-financial statements), administrative and legal procedures related to development of the estates, etc. In some cases FIDEIN provided the Fideicomisos with small amounts of short-term credits. 3.25 The appraisal of Loan 1552-ME identified the lack of long-term credit for SMIs to finance building construction, as a key constraint in industrial estate development. As a result, FIDEIN's activities under Loans 1552-ME and 1881-HE were expanded to cover three areas of direct financial assistance: (i) construction of standard factory buildings (SFB) for rental on industrial estates with option to purchase, or for direct sales to SMIs; (ii) construction and leasing of common facilities to provide services to groups of SMIs on the estates, and (iii) financing of equipment and machinery leasing to SMI operating on the estates. FIDEIN was expected to act as a second-tier agency to provide financing and only exceptionally, was it prepared to finance and administer directly the construction of factory buildings and common service centers. The three new tasks given to FIDEIN produced only disappointing results. In the SFB eub-program, FIDEIN opted for direct contracts with client firms, and financing directly sales, rent and leases of the installations to SMIs. Only the development of the land sites remained the responsibility of the local "Fideicomisos". 3.26 A comprehensive program of staff expansion and training was discussed at negotiations of Loan 1881-ME and a program of advanced construction by FIDEIN of SFB to be subsequently sold or leased was agreed upon. Under this program FIDEIN was to undertake the advanced construction of at least one or two factory units per estate, of 1,000 to 2,000 m2 each which was expected to have a demonstration effect. The operational results were disappointing (para.4.23-4.25), and further policy changes had t,c be introduced. The three investment activities that were included in the two Bank loans under review were excluded from the third SMI Bank loan, and FIDEIN's financing role was restricted to supporting investments by public and private sector entities.  - 12 - 3.27 Management, Organization, Staffing. FIDEIN suffered from management problems during most of the implementation period of both loans. Immediately after the appraisal, a new Director General, Mr. A. Navarrete, was appointed, but since he was sick during the major part of his tenure, his subdirector, Mr. Felix Lopez acted as the Director until the end of 1981 when Mr. Gustavo Varela took over. Serious financial irregularities during the tenure of the previous management were identified by an independent financial and technical audit. Consequently NAFIN empowered Mr. Varela to reorganize FIDEIN, remove several key staff, and initiate judicial proceedings as appropriate. The new Director General implemented significant organizational changes in FIDEIN in early 1983 at a time when the Government redefined FIDEIN's role (para. 3.25). 3.28 From the time Loan 1552-ME was appraised to 1980, FIDEIN's professional staff increased from 15 to 27 organized under two managers, one in charge of legal, administrative and financial matters, and the other responsible for technical and operational matters. Despite the increase, the quality of FIDEIN staff was considered weak at the time of appraisal of Loan 1881-ME. To meet the needs of expanding activities, staff was increased from 43 (including 27 professionals) in 1980 to 84 employees (38 professionals) by December 1982. Unfortunately many of the new staff were inexperienced and coupled with a lack of appropriate management direction, FIDEIN's performance remained poor in the quality and management of its programs (para. 4.25). In December 1984, an Acting Director General was appointed under whose guidance FIDEIN gradually evolved into its new role as technical assistance agency. FIDEIN staff was reduced to 62 by December 1985, of which 30 were Lechnicians and professionals. While FIDEIN has developed a good technical capacitj in the design of industrial estates and in the execution and supervision of civil works, its financial analysis and project appraisal work still needs further strengthening. This aspect will be addressed under the recently approved Fourth SMI project. IV. UTILIZATION OF LOAN PROCEEDS UNDER LOAN 1552-ME and 1881 ME Resource Transfer 4.01 There were no disbursements during the first implementation year of Loan 1552-ME, because of a lack of agreement on the FOGAIN interest rate clause (para. 3.08) but PAI held up disbursement requests from the other implementing agencies as well. After this matter was resolved the rate of loan utilization increased rapidly. By the end of 1979 disbursements already amounted to US$27.2 million, 155% more than the levels anticipated at appraisal. The fast disbursement rate reflected the increase in industrial sector investments which continued throughout 1980 (para. 1.02) and by start of 1981, about 97% of the total amount (US$45.3 million) had been disbursed. 4.02 The experience with the utilization of Bank Loan 1881-ME was somewhat different in that during the first year of effectiveness, the loan was committed almost 50% more than anticipated at appraisal and total disbursements were US$46.0 million. After one year the disbursement of loan 1881-ME was actually 300% above the disbursement level projected at appraisal. The strong demand for loan resources was in line with the  - 13 - buoyant economy at the time in Mexico, helped by the active promotion efforts of the PAI staff. After the crisis of late 1982, however, loan disbursements slowed but in 1983 these still amounted to US$20.5 million, i.e., 146% more than the level projected at appraisal. Accumulated disbursement reached US$92.5 million by the end of 1983, 8% above disbursement levels anticipated at appraisal. However, in November 1983 the Loan was amended under the Bank's Special Action Program (SAP) to assist Mexico confront the severe economic problems it was experiencing. since late 1982. Two important measures which speeded the disbursement rate under the loan were an increase in the disbursement percentage from 55% to 73% for subloans, and the agreement to permit financing for permanent working capital through the projects equity component. By mid-1984, the loan was fully committed and the closing date of the loan was extended for six months to June 1985, to allow time for the disbursement of some subprojects. 4.03 The loan disbursement experience under both projects varied for each of the subcomponents. While the FOGAIN credit resources moved rapidly, under both loans, the FOMIN equity component and FIDEIN industrial park investments lagged. The movement of funds under the FIDEIN component were particularly slow so that when Loan 1881-ME was appraised in 1980 only US$1.1 million out of the US$5.0 million had been committed. The relatively tardy utilization of loan resources by FIDEIN was partly due to the newness of the operation, but largely due to the inherent weakness in FIDEIN's implementation capacity (para.3.27). On the other hand, FOMIN investments moved slowly due to the cautious investment policy adopted by its management. The disbursement experience with the two SMI loans and the subsequent operation (Loan 2325-ME) was taken into consideration in estimating the Implementation speed of the the proposed Fourth SMI project. Lending CharacterisL ^s under Loans 1552-ME and 1881-ME 4.04 The proceeds of Loan 1552-ME financed 915 subprojects at a cost of Mex$1,829.9 million (US$78.7 million)4/ distributed as follows: 864 subprojects through FOGAIN with a cost of Mex$1,248.4 million (US$53.7 million); 29 subprojects through FOKIN with a cost of Mex$356.4 million (US$15.3 million); and 22 subprojects through FIDEIN with a cost of Mex$223.7 million (US$9.6 million). Also the loan financed Mex$56.9 million (US$2.4 million) for a program of Technical Assistance. The total cost of the subprojects reached Mex$1,885.8 million (US$81.1 million) for the entire project. The results were close to the loan targets since the appraisal estimated financing 1,160 subprojects. 4.05 Loan 1881-ME helped finance a total of 2,120 subprojects for a cost of Mex$10,902 million (US$87.5 million)5/ distributed as follows: 2,019subprojects through FOGAIN for a cost oT Mex$8,574.5 million (US$68 million); 79 subprojects through FOMIN for a cost of Mex$1,910.1 million(US$15.3 million) and 27 subprojects through FIDEIN for a cost of 4/ Exchange rate of US$lmMex$23.25, representing the average rate during the loan implementation period 1979-81. 5/ Exchange rate of US$1-124.54 Mexican Pesos, representing the average during the loan implementation period 1982-84.  - 14 - Mex$X$417.4 million (US$3.9 million). Also the loan financed technical assistance for Mex$123.0 million (US$0.9 million). The 2,120 subprojects financed under the loan compares very favorably with the appraisal estimate of 870 projects. As in Loan 1552-ME the foreign exchange risk was borne by the Government at no cost to the sub-borrowers and as this loan (1881-ME) was disbursed during a period of a rapidly appreciating exchange rate, it was clearly an important factor in the high demand for subloans. FOGAIN Component 4.06 Lending Operations. Since FOGAIN's information system on subloans does not permit a disaggregation of credits by source of financing (IDB, Bank or its own resources), it was not possible to analyze the characteristics of all subprojects financed with the loan resources. Nevertheless, a broad picture of FOGAIN's lending characteristics is obtained from the analysis of its overall operations analyzed in Annex 13-14 which summarizes FOGAIN's loan approvals during its 31 years of operation from 1954 to the end of 1985. During this period FOGAIN subloans totalled 66,925 loans for Mex$298.7 billion. The majority of loans were between Mex$1.0 million and Mex$5.0 million and these accounted for 34.0% of the total amount approved. Loans of over Mex$30.0 million accounted for only 1.02 of the approved loan amount. Twenty percent of the credits approved during the whole period were in 1985 indicating the major expansion in FOGAIN's clientele was during that year. FOGAIN loan approvals during the period 1954-85 cover all sectors, notably food products (21%), metal processing and machine fabrication (17%), and manufacture of foo%.wear and other leather products (3%). Fifty percent of the loans approved were for production of consumer goods, 38% for intermediate goods, and 12% for capital goods. FOGAIN subloans were mainly to small scale industries - 83% of the credits (64% by amount) were channelled to enterprises with less than 50 employees. Small scale industry as defined by annual sales (below Mex$400 million), accounted for 81% of the credit (56% by amount). During 1985, a total of 91% of the loans (75% by amount) were directed to small scale industries. 4.07 Lending Characteristics - Loan 1552-ME. To compare FOGAIN's overall lending characteristics with subloans financed with Bank resources, a sample survey of 84 subprojects was carried out by FOGAIN under Loan 1552-ME, including 13 new enterprises. The results of the survey are detailed in Annex I. It shows that 27.4% of the subloans had a maturity range of one to two years, with only 2.6% of subprojects benefitting from payment periods of over five years. Most subprojects (70Z) had maturity terms between three to five years. This range of maturities were less than expected at appraisal and reflected the conservatism of the commercial banks rather than any FOGAIN policy to keep maturities short. Nevertheless, FOGAIN benefitted from the shorter maturities to the extent this permited FOGAIN to roll-over the loan recoveries several times. 4.08 Subproject Types. As regards the project objective to finance mainly small enterprises with 25 or less employees, of the 84 enterprises sampled it was found that 70 enterprises (or 83% of the total) had 10 or less employees. Only one firm had more than 100 employees. The total cost of the 84 subprojects included in the survey amounted to Mex$230.8 million (US$9.9 million), with an average subloan size of about Mex$500,000 (US$21,500). Only four subloans of the 84 were over the free limit of  - 15 - US$150,000. Any single industrial subsector did not represent more than 25% of the subloans: food, beverages and tobacco industries represented 23% of the total subloans with the remaining 77% distributed to eight sub- sectors, primarily non-electrical machinery, 10%; textiles, apparel and footwear, 12%; chemical products, 10%; non-metallic minerals, 10%; and others with smaller participation. The geogriphical dispersion of the sub- projects shows that only 17.8% of the total subprojects were located in the high priority Zone I with the remaining 82.2%, representing 83.8% of total lending, were located in Zones II and III. Under the follow up pro- ject greater attention as given to financing investments in Zones I and II. 4.09 Subproject Impact. Economic and financial characteristics of the subprojects financed under Bank loan 1552-ME are summarized in Annex 2. The economic impact of the subprojects can be considered satisfactory in terms of employment creation and the viability of the investments financed. The 84 subprojects surveyed generated a total of 2,852 new jobs with an average investment cost per job of Mex$80,900 (US$3,479). In the survey, the investment cost per job ranged from Mex$9,700 (US$417) for a labor intensive furniture firm to Mex$2,269,000 (USS95,951) for a medium-sized capital intensive engineering plant. In some of the enterprises the cost of creating additional employment was high because investments were mainly for expansion (71 firms), raising productivity rather than an increase in the labor force, and less for new enterprises (13 firms). 4.10 FOGAIN carried out an 'ex-post' analysis of a sample of the SMI projects financed based on the subloan appraisal data, updated for actual performance data obtained as part of the PCR work. In addition PAI calculated IRR and ERR for a sample of subprojects based on data available in the files. Based on the data for these two samples, the ERR for the sampled subloans ranged from 13.8% to 293.8%, and more than 80% had an average ERR's of 47.0%. The IRR ranged from 1.4% to 207.2%. On a priori basis, therefore, FOGAIN financed economically sound and efficient subprojects under Loan 1552-ME with the average rates of return well within the overall range of 15%. 4.11 Lending Characteristics - Loan 1881-ME. Details from a representative sample of 240 Bank subprojects (about 10% of subprojects) financed under Loan 1881-ME, including 56 new enterprises, are shown in Annex 4. The total cost of the 240 subprojects of the survey amounted to Mex$739.37 million (US$5.9 million) with an average subloan size of MexS3.5 million (US$28,103). None of these were over the free limit of US$250,000. The objective of financing smaller enterprises was well achieved; more than 90 percent of the financed subprojects analyzed in the survey had 25 or less employees. 4.12 As regards the geographic distribution of subprojects, about 45.8% (100 subprojects), of a total 240 subprojects surveyed, were in Zone I 6/ and the remaining 54.2% (140 subprojects) were located in Zones 6/ Geographic Zones are defined in the decree published in Diario Oficial, February 2, 1986. Zone I (Priority Preferential); Zone II (State Priority); Zone III (Controlled and Consolidated Growth); Rest of the country.  - 16 II and III which represented 56.3% of the total financing. There was thus a marked improvement compared to the first loan in locating investments in high priority zones. The pattern of financing by subsectors was similar under the two projects, with no industrial subsectors representing more than 20% of the total disbursed. The food industries subsector benefitted most with 38 subloans (15.8%) of the total subloans. The remaining 84.2% is divided between 16 subsectors including non-metallic minerals (10.8%), metallic products (10.8%), non-electric machinery (8.7%), chemical products (8.3%) and others with smaller participation. 4.13 Subproject Impact. The economic and financial characteristics of the subprojects financed under this Bank Loan are summarized in Annex 4. The 240 subprojects surveyed generated an estimated 1,471 new jobs at an investment cost per job of Mex$0.6 million (US$5,369); investment cost per job ranged from Mex$0.1 million (US$939) to Nex$10.6 million (US$99,624). Again, the cost of creating additional employment in some of the subprojects was relatively high because the investments were mainly for plant expansions (184 firms) and less for new enterprises (56 firms). In this loan FOGAIN calculated ex-ante ERR which ranged from 4% to 385%. The subloans had an ex-ante IRR ranging from 7.6% to 387.0X. 4.14 Current Status of Subloans. To assess the latest position of subprojects, the PCR mission visited a randomly-selected sample of 12 sub-borrowers. All the enterprises were experiencing working capital shortages due to severe restrictions on commercial bank credits. None of the enterprises had received technical assistance in project preparation nor in any aspects of management. While the original calculations of the financial and economic rates of return of the subprojects being financed appeared justified originally, the mission observed that these groved to be optimistic on an ex-post basis. This is partly due to the deterioration in the financial condition of the enterprises, which suffered due to the accelerating rate of inflation and associated high financing costs and the depressed level of demand in domestic and international markets; cost overruns and unsatisfactory sales performance were common among subprojects surveyed. FOMIN Component 4.15 Subprojects Financed. Under Loan 1552-ME FOMIN financed 29 subprojects at a cost of Mex$356.6 million (US$15.3 million) using US$8.4 million from the Bank loan. Of these, 22 were direct equity investments; six took the form of convertible loans; and one received part as straight equity and part as a convertible loan. Seven additional subprojects were approved by FOMIN and the Bank, but were eventually cancelled before any disbursements were made (See Annex 5). Seventeen of the subprojects required equity investments over the free limit of US$150,000 in Bank financing. 4.16 Under Loan 1881-ME, 61 FOMIN subprojects received financing at a cost of Mex$2,005.8 million (US$16.1 million) and a disbursement of US$10.0 million from the Bank loan. Of the 61 subprojects, 41 were straight equity investments, 12 received both equity and convertible subloans and eight were straight convertible type loans. Six additional projects were cancelled. Fifteen firms received equity investments and/or convertible subloans over the free limit of US$250,000 in Bank financing,  - 17 - with the highest FOMIN investment being Mex$150 million and the highest Bank participation Mex$109.5 million (US$580,000 at 1984 exchange rate), or 73% of FOMIN's participation. Bank participation in terms of total project costs under the FOMIN component was 55% for Loan 1552-ME and 63% for Loan 1881-ME. The increase during Loan 1881-ME is due to a Government request in mid-1983 to raise Bank Rarticipation in PAI disbursement to trust funds from 502 to 73% in line with the policy approved for the follow-up Third SMI project (Loan 2325-ME). 4.17 SubproJect characteristics: Seventy percent of the subprojects under Loan 1552-ME and 79% of those under Loan 1881-ME went to medium scale enterprises (each receiving investments of Mex$5 million or over) which used 95% and 97% of the funds respectively. This confirmed expectations at appraisal (para. 3.19) based on the recognition that small scale firtms depend on their own equity and in fact are reluctant to have a Government agency as a partner. However, only four subprojects reached the maximum of Mex$150 million. The size of FOMIN investments being Mex$12.2 million (US$530,000) during Loan 1552-ME and Mex$32.8 million (US$264,000) during Loan 1881-ME. Only three participants in Loan 1552-ME and four under Loan 1881-HE were first time investors and started new enterprises. This number increases to eight and 27 respectively if new projects of existing enterprises are considered 'new". The majority of FOMIN financing went to expanding or restructuring existing enterprises. In some cases, the assisted firms belonged to a larger conglomerate operating as a holding company and in some of those cases (e.g., Grupo DEMSA-Notyformas) FOMIN assisted several firms belonging to the same group. There are no specific covenants in the loan agreements against investing in holding groups, these procedures run counter to FOMIN's explicit role of stimulating the creation of new SMI enterprises, and also conflicts with FOMIN goal of diversifying its portfolio. FOMIN has begun to gradually diversify its clientele by investing in newer enterpries, but mwre concrete efforts are needed in this direction under the Fourth SMI project. 4.18 Geographically, although a high percentage (39%) of subprojects financed were located in Zone III, a substantial number of enterprises operated in Zones I (26%) and II (35%). Considering the high concentration of professional and entrepreneurial talent around Mexico City, Monterrey and Guadalajara, this degree of FOMIN project dispersion can be considered satisfactory. An additional positive feature was the concentration of FOMIN investments in specific geographic areas which were hitherto regarded relatively underdeveloped (e.g. San Luis Potosi where FOMIN has, at present, 12 investments, and Sinaloa where it has seven investments). More than 90% of the FOMIN projects were designed to supply the domestic market, with only 20% specifically intended to substitute for imports. Only 9% of subprojects, using 22% of FOMIN's resources under both loans, were planned for export. 4.19 Current Status of FOMIN Subprojects. Of the 89 projects financed by FOMIN under both loans, only six subprojects are still under implementation (one under Loan 1552-ME and five under Loan 1881-ME) while the rest are all in operation. Some have already been sold or liquidated (15 from Loan 1552-ME and eight from 1881-ME) (See Annex 9). To assess the impact and present status of the different components under both Bank loans, PAI undertook a special survey of FOMIN subprojects. However, the Information provided by FOMIN-assisted enterprises was, in most cases,  - 18 - patchy and inconsistent and in the cases of some firms where FOMIN had already sold its participation, the enterprises refused to provide information. FOMIN needs to strengthen its information base and collect on a regular basis data on the operational aspects of the firms it supports. This should be relatively easy since FOMIN maintains a relatively long term association with its business partners through its equity investments. More recently FOMIN has developed a good data base on its enterprises and under the recently approved Fourth SMI Loan technical assistance resources are being specifically provided to further develop FOMIN's information systems. 4.20 As of December 1985, of all the investments or credits approved under both loans (see Annex 9), only 17% are operating with profits, 46% with losses, and 30% have been sold or liquidated, a situation which though quite disappointing is reflective of the effects of the industrial recession, which began in mid-1982. By December 1982, 54 companies (47% of FOMIN's portfolio) were experiencing financial difficulties, many as a result of exchange losses (following the large September 1982 devaluation), high financial costs and associated working capital shortages. To some extent the poor performance of FOMIN assisted enterprises is also due to deficient financial management and an inability to handle crisis situations. 4.21 The rates of. return calculations prepared for the PCR are however quite optimistic ranging between 15.0 and 80.0% with an average of 33.2% for Loan 1552-ME and 37.8% for Loan 1881-ME. Although the appraisals did not project expected IRR, the actual rates of return can be considered very acceptable by international venture capital standards. A random review of recent supervision reports prepared by FOMIN staff and some visits to enterprises indicate that (i) ex-post results are, for the most part, in line with "ex ante" estimates, and (ii) at least 50% of the enterprises currently facing financial difficulties can probably be rehabilitated. 4.22 Results of equity sales show that all 16 firms sold by FOMIN, since inception of PAI-I resulted in capital gains, if investment is valued at historical costs (Annex 9). However, a sample of cases calculated in real terms shows most of these sales merely broke even or represent losses in real terms. Even while accounting for the adverse effects of the economic situation, FOMIN's meager sales performance on its equity portfolio generally indicates poor investments and a weak divesture policy characterized by sales of shares to existing owners with limited bargaining position in the hands of FOMIN. With respect to the FOMIN convertible loans under both loans 1552-ME and 1881-ME, Mex$493.5 (out of a total of Mex$686.7 million) are currently outstanding in 14 subprojects.. As of December 1985, six of these projects were in arrears of three months or more, with total arrears of principal and interest amounting to MexS77.6 million, which represents 11% of the total amount provided as convertible subloans. However, 62% of the arrears is concentrated in two firms which are presently receiving technical assistance to help towards recovery. Thus, while excluding the above two firms, the average of "arrears of three months and over amounts to Mex$29.5 million, or only 4% of total financing provided, and 6% of the outstanding balance, a reasonable average considering the economic conditions prevailing.  - 19 - FIDEIN Component 4.23 Overall operations. Under Loan 1552-ME, FIDEIN provided financing for 31 factory buildings, with a total PAI participation of Mex$198.4 million (US$8.5 million) and a Bank participation of Mex$109.1 million (US$4.7 milion), or 55% of the total PAI financing. Also, three equipment leasing subprojects were financed with a PAI participation of Mex$25.3 million and Mex$13.9 (US$320,000) of loan proceeds. Under Loan 1881-HE, financial assistance was provided for the construction of 24 factory buildings, with a PAI participation of Mex$390 million and using US$1.7 million of loan proceeds, three leasing operations were financed under this loan requiring about Mex$27.4 million (US$220,000) of PAI assistance and about US$120,000 of Loan proceeds. 4.24 Annex 10 summarizes some characteristics of the industrial projects established within FIDEIN financed factory buildings. A detailed analysis of subproject characteristics is not relevant in this case since the performance of these beneficiaries can only be marginally attributed to the buildings, or to the SFB (Standard Factory Buildings) subloans. Nevertheless, the following features are of interest: (i) over 80% of factory buildings were sold to small companies with less than Mex$25 million (in 1985 pesos) of equity; (ii) nearly 60X of enterprises assisted under the program were new enterprises and nearly 70% of the enterprises assisted were established on industrial estates located in Zone I of the highest decentralization priority; (iii) loans to final beneficiaries had maturities in the range of two to five years, with the majority being of the four to five year range, which represents maturities much shorter than expected at appraisal. The cash flow from investments in factory buildings proved to be quite positive, and about 702 of the amounts disbursed under both loans have already been repaid to FIDEIN; (iv) to mid-1986 subprojects were well balanced between different branches of industry, with a slight predominance of machinery manufacturing firms which represented 13% of the total allocations in number of enterprises, and 20% in total allocations of funds under both loans. 4.25 FIDEIN subprojects were, in reality, a termrsale of a fixed asset to industrial clients. Appraisals of the enterprises were carried out by FIDEIN before sale of the buildings, but assessment of repayment capabilities of the enterprises was limited to the building which, in most Investment cases, represented only a small part of the investing companies total operations and asset base. Furthermore, a few SFB were even paid for in cash, in which case subproject appraisal became a purely academic exercise. The only meaningful information that can be derived from the results of the FIDEIN survey of subprojects is that out of 50 enterprises surveyed (which represent about 80% of the total number of FIDEIN subprojects), only two have gone out of business and their factory buildings are closed and unused. The remaining 48 are still operating and are using the SFBs for their intended purpose.  - 20 - V. FINANCIAL PERFORMANCE OF INSTITUTIONS PAI 5.01 Financing Operations. The original agreement between SHCP and NAFIN setting up PAI and its financing Fund provided a seed capital of Mex$510 million (US$22.5 million equivalent) and stipulated that the principal of the IBRD loan (US$47 million) was to be repaid by the Mexican Government so that PAI could capitalize the loan resources. PAI was only responsible for servicing the interest payments for the FOGAIN credit component (US$30 million). In addition, NAFIN obtained from Banco de Mexico a line of credit of MexS420 million (US$18.5 million) for additional rediscounting operations of FOGAIN. It is not clear from the documents whether PAI was obligated also to pay only the interest on the FOGAIN portion of Loan 1881-ME. 5.02 PAI also received annual Government subventions (approximately US$60 million equivalent since 1978) mainly to meet its administrative expenses, over and above the amount needed for repayments to IBRD since 1981. To meet its financial needs, PAI has also been borrowing funds from NAFIN and Banco de Mexico. The funds borrowed from NAFIN were high cost while the funds from Banco de Mexico have been slightly below the Average Cost of Funds (ACF). As of May 31, 1986, FOGAIN represented 69% of PAI's portfolio, FOMIN accounted for 13% and 19% was held by FIDEIN. In theory, PAI should have received from the sub-programs the recoveries of the credits or investments made by these institutions, but FOGAIN (its largest client) did not transfer any funds. As already stated, (para. 3.05) during 1978-1982, the head of FOGAIN and PAI were the same person, and the arrangements tended to favor FOGAIN allowing it to retain a high spread from the interest rate charged to the final beneficiary. FOGAIN paid PAI 35% p.a. interest rate on the outstanding balance of the funds provided by PAI, and kept the subloan repayments for its future lending. 5.03 Financial Performance. Since its inception PAI has been losing money and its equity base has been eroded despite the total infusion of about US$350 million in quasi-equity, from the Mexican Government in the form of seed capital or as subsidies in the form of repayments of principal and as a portion of the interest on Bank's loans (PAI only pays interest on a portion of the IBRD's loans). The following reasons have attributed to PAI's losses: (i) PAI was responsible for the payment of the foreign exchange risk on a substantial part of IBRD's loans; (ii) the low interest rates paid by FOGAIN to PAI, was inadequate for PAI to cover interest and foreign exchange losses, requiring it to borrow funds from NAFIN and Banco de Mexico to cover operating expenses or losses; (iii) the subsidized interest rates granted to FIDEIN's clients meant lower earnings for PAI on these operations; and (iv) the relatively poor performance of FOMIN's portfolio, and the retention of PAI resources by FOMIN for further investment. 5.04 As seen from Annexes 26-27, PAI's interest rate income represents a maximum of about 20% of the value of the portfolio, about a quarter of the CPP level in 1986. This is insufficient for PAI to cover interest payments and the foreign exchange losses. The portfolio in mid-1986 was worth only about US$65 million equivalent, while the funds provided to PAI have been worth at the time about US$350 million. The difference between  - 21 - the full value of the IBRD loans (US$325 million, of which more than US$297 million have been disbursed to date) and that of the existing portfolio has been due to the massive devaluations of the peso against the dollar, and the relatively low interest rate spread received by PAI from FOGAIN, and the non interest earnings equity investments financed through FOMIN. Above all, while the credits or investments made by the institutions using PAI resources are denominated in pesos, PAI has assumed the foreign exchange risk which is not adequately covered by the interest rate spread it receives. If PAI's interest rates had been closer to the CPP rate, the income from interest income would have been gone much further towards covering the foreign exchange losses and there would have been less erosion of its capital base. FOGAIN 5.05 Financial Performance. The FOGAIN income statement, balance sheet, and financial indicators for 1979-85 are presented in Annexes 15, 17 and 18. The rapid growth of FOGAIN lending operations during 1979-85 (para.4.06) had a positive impact upon its financial performance. FOGAIN's loan portfolio grew at an explosive rate of 25% in dollar terms in the 1979-82 period. On the income side, with an exception of 1980 in which FOGAIN had a moderate profit Mex $3.9 million (US$130-125 million) due the beginning of the government annual contributions, from 1981 to 1983 it made net losses. In 1981 FOGAIN incurred net losses of Mex$ 6.8 million (US$2.2 million) and in 1983 Mex $419.5 million (US$2.7 million). This was mainly due to the foreign exchanti losses incurred on its IDB borrowings, which were not covered by the fixed interest rates maintained over the life of the subloans. In addition, FOGAIN's cost of borrowing increased along with inflation -- in the period under review its operating portfolio carried an average peso interest rate of about 18% compared to its average cost funds, excluding its equity base, of about 30%. The losses were only partially covered by Government subsidies and in 1984, FOGAIN was absolved from carrying the foreign exchange risk. 5.06 The appraisal reports for Loans 1552-ME and 1881-ME had predicted operating losses in case domestic inflation rates accelerated and if FOGAIN interest rates were left unadjusted. In the earlier years (1979-83) FOGAIN interest rates did fall behind inflation rates with consequent income losses. However, starting in 1984, FOGAIN's made a profit of Mex$20.4 billion (US$12.8 million) increasing to Mex $39.8 billion in 1985 (US$18.0 million) and to Mex $15.2 billion in 1986 (US$52.1 million). The positive operating results were essentially a direct consequence of the variable interest rate policy (adopted by FOGAIN in late 1984) linked to a inflation base which was rapidly on the rise, since 1983. 5.07 As far as expenditures are concerned, interest charges (i.e., interest paid on funds received) is the major cost item. This financial charge decreased in relation to FOGAIN's total income from 87% in 1981 to 54% in 1985 as a result of a progressive reduction in interest rates linked to increased borrowing from IDB/Bank whoes resources were obtained at fixed rates denominated in pesos. FOGAIN's administrative costs were less than 2% of average total assets an acceptable level compared with similar institutions elsewhere.  - 22 - 5.08 FOGAIN's total assets show an increase in current terms in 1980 from Mex $9.7 billion (US$689.2 million) to Mex $130.8 billion (US$448.7 million) in 1985, however, a decline in dollar terms. An analysis of the balance sheets presented in Annex 18 shows that most significant asset, the loan portfolio, grew rapidly between 1982 and 1985 in current terms from Mex $33.7 billion (US$317.8 million) to Mex $123.7 billion (US $24.3 million), but fell in real terms, as a consequence of domestic inflation and a reduction of FOGAIN's borrowing levels. The loan portfolio constitutes approximately 95.0% of the asset base. Liabilities as an element in the overall position of FOGAIN shows an increase in current Mex$4.9 billion to Mex$61.6 billion, but a decline in US$ terms from US$165.6 million to US$153.4 million. From a time perspective the structure of FOGAIN's liabilities has improved: whereas in 1984, 36% of all liabilities were short term, only 24% were in this category in 1985. This is the result of a deliberate policy to replace short-term high cost borrowing instruments with foreign financed fixed interest resources. 5.09 FOGAIN's equity capital consists mainly of Federal Government contributions in the form of budget appropriations and the recoveries of loans negotiated with international institutions. Prior to 1984 its equity capital was eroded through exchange losses and through inflation, but for the entire period 1980-85 it grew in current nominal terms, with the equity capital rising from Mex $1.9 billion to Mex $69.2 billion. Capital as a factor in liabilities plus equity has increased steadily from 20% in 1980 to 53% in 1985. Also, the debt/equity ratio dropped from 2.58 in 1980 to 0.89 ln 1985, a level that put FOGAIN in a sound a financial position. FOMIN 5.10 Investment Portfolio. FOMIN's operations during the implementation of both loans grew rapidly; in nominal terms its investment portfolio of Mex$181.9 million as of December 1977, grew to an aggregate investment of Mex$6,777.4 million, as of December 1985. In real terms FOMIN's cperational growth is less impressive, however, and shows a portfolio increase from US$6.2 million in 1977 to US$22.3 million in 1985. Thus, FOMIN's portfolio grew at an annual rate of 57% in nominal terms and 18.0% in real terms between 1977 and 1985. 5.11 FOMIN's increase in investment was made possible through a major capitalization program which is reflected in an equity increase from Mex$377 million in 1978 to Mex$5,710.6 million in 1985, in nominal terms. Although an important part of this increase is due to the revaluation of fixed assets in line with inflation, Government capital contributions also increased from Mex$435.7 million in 1978 to Mex$2,630.4 in 1984. In dollar terms, FOMIN's equity rose from US$12.7 million in 1978, to US$19.6 million in 1985, or a growth of 6.4% per year. The Government contributions and, to a lesser extent, retained earnings from FOMIN's own operations (MexS496.0 million as of December 1985) contributed to the growth in equity. The second most important element contributing to the increase in assets was PAI's financial support to FOMIN, which went up from zero in 1978-79, when the NAFINSA-World Bank program started, to Mex$2,732 million (US$9.4 million) in 1985. In 1985 FOMIN's own resources, consisting of retained earnings, investment sales, and interest from subordinated loans, represented 43Z of its total funds.  - 23 - 5.12 Financial Performance. After incurring losses until 1979, FOMIN's revenues increased significantlv, due to the rapid growth of its portfolio, due to the introduction of convertible credits and to a more profit-oriented asset sales policy. From 1980, net revenues covered operating administrative expenses. By 1981 operating expenses (Mex$63 million) represented 62% of current income (Mex$102 million) and at year-end 1985, operating expenses (Mex$328 million) represented only 52% of current income (Mex$630 million) and only 4.8% of FOMIN's total portfolio (See Annex 18). 5.13 Based on the above results, FOMIN's financial situation appears to be sound, even while discounting for inflationary distortions, and the impossible task of adequately assessing the current market value of its portfolio. On the liabilities side, FOMIN does not bear the full cost of losses on its investments (it only repays PAI to the extent of its investment recoveries) and in the case of profitable investments it repays PAI only in nominal terms plus half-of the capital gains. Therefore, FOMIN's growth in equity does not necessarily mean-a real improvement in the effectiveness of its investment operations. The revaluation of the portfolio does not always reflect fully the increase in value or in actual prices and therefore cannot be said to show the real increase in equity. On the assets side, while FOMIN has increased the number and value of its operations, the present financial status of the companies in FOMIN's portfolio makes it highly improbable that the values given as to the different items in the balance sheets represent their real market value. Any assessment of FOMIN's current financial position must take account of the above and in the future, as agreed during appraisal of Loan 2325-ME, procedures should be followed to require FOMIN to update regularly the market value of its portfolio and to produce audited financial statements for management purposes. FIDEIN 5.14 Financial Performance. FIDEIN's past financial statements is in Annex 25. An assessment of FIDEIN's net worth is difficult since audited statements for most years during both loans were qualified mainly due to the problems related to the transfer of ownership of industrial estates to and from State Governments, without clear agreements on the financial conditions. Evaluation of FIDEIN's assets was extremely dificult. Also the audited statements, since 1984, have been prepared in nominal terms. 5.15 An analysis of the limited financial data available shows an improvement in FIDEIN's financial situation during 1979-84, but again the manner of resource transfer from PAI to FIDEIN obscures the real situation. In the loans under review NAFIN agreed with the Bank that FIDEIN would not pay interest on the Bank funds used for its operations until it was generating sufficient income to make this possible. FIDEIN was expected to repay to NAFIN the principal of the accounts received under the loan, as and when it recovered such accounts from individual invest- ments. However, FIDEIN's inventory of buildings created liabilities from FIDEIN to PAL at the time of construction. Several years later, when the buildings were sold they generated income at prices prevailing in the year of sales. The difference produced a significant nominal increase in FIDEIN's assets so that in nominal terms assets grew by more than six times between 1980 and 1985, but its equity remained almost constant in nominal  - 24 - terms and decreased substantially in real terms. While FIDEIN's Income Statements show profits from 1981 onwards, they include as income substantial Government subsidies which hide the real operational losses. 5.16 As of December 1985, only ten of the 54 FIDEIN clients under both loans were in arrears of three months or over amounting to MexS13.8 million in principal and Mex$52.9 million in interest, out Of a total outstanding balance of Mex$491.0 million and a total amount of credit operations of Mex$1,235.7 million. However, nearly 50% of the principal in arrears (Mex$6.4 million) and over 70% (Mex$37,300) of interest in arrears, were concentr4ted in two companies (Becal, S.A. and Ceramicas Filtrantes, S.A.) both already under litigation. The other arrears are mainly short-term (less than six months) and have not increased during the last three years. These short-term arrears amount to Mex$32.9 million, including both principal and interest, and represent 7% of the total outstanding balance and a 3% of total lending by FIDEIN, an acceptable figure considering the crisis in the Mexican industry. VI. TECHNICAL ASSISTANCE ACTIVITIES PAI Extension Service 6.01 A central feature of the PAT program was the creation of an industrial extension service which was designed by NAFIN with Bank assistance. The role of the extension service was to make small scale industrialists aware of the sources of financial assistance, and to provide them directly with some basic technical/managerial assistance. In addition, for resolution of specific problems the extensionists were to refer entrepreneurs to specialized technical assistance institutions cooperating in the program. The results of the extension services can be considered as quite significant both in terms of the positive institution building and of services provided to SMI's. The Bank made substantial contributions to the development of the extension service through comments and suggestions following some in-depth supervisions. However, the key element in the strengthening of the service was the full commitment of NAFIN to the concept of an "integrated program' of assistance to SMIs which included a common extension service for the three executing agencies. 6.02 Staff and Staff Training. The extension service started operations in mid-1979 with the training of 20 extension officers in a six-week program designed and implemented with the assistance of specialized training institutions (e.g., IPADE, CENAPRO). A further group of extension officers were trained in August/September 1979; and by year-end 1979, 40 qualified officers had been deployed to 14 offices throughout Mexico. By the end of 1981, PAI's extension service had 93 industrial extension agents and eight auxiliary technical staff which, in addition to eight regional subdirectors and 8 regional coordinators, amounted to a corps of 117 extension professionals deployed in 26 branches. The system was operating under the gu'tance of eight regional subdirectors and a small central coordinating staff at headquarters. All of the 117 staff had been recruited and trained through a rigorous program during 1979/1980. The conditions offered to the extension officers were attractive at the time and the number of applicants exceeded by a wide margin the number of places available. PAI was thus able to choose from a  - 25 - select group of capable young professionals with varying backgrounds in engineering, economics and business administration. 6.03 Each PAI extension branch is staffed with two to three extension officers, depending on the level of priority assigned to each region. Regional offices visited (Guadalajara, Monterrey, Puebla, San Luis Potosi, Leon and Mazatlan) revealed a high degree of motivation, professional competence and knowledge of the respective industrial groups on the part of all extension officers. The quality and quantity of work performed by the regional offices appears satisfactory. It should be noted that such work has been performed since 1982 with limited means and logistic support. The extension service has become fully institutionalized and has recently been put under the management of NAFIN's Diiectorate of Industrial Promotion, which plans to further strengthen and expand this services to cover hitherto unattended regions. Under the recently approved Fourth SMI project, the largest porticn of the technical assistance funds would be used to support the expanded activities of the extension service. 6.04 The initial training provided to the staff -- using loan resources -- was comprehensive and of high quality. The training given together with the high motivation of the staff resulted in low turnover in the first years and an excellent rapport between extensionists and entrepreneurs during the period 1979-82. After 1983 staff turnover in the extension service was high following management changes in PAI and FOGAIN and a deterioration in the conditions of employment due to Government's austerity measures. In terms of geographical coverage the original plan was to locate extension officers in NAFIN's regional offices and thereby reduce administrative costs. However, this policy had to change ini order to reach designated priority areas and by year-end 1985 the number of regional branches had increased to 33. Furthermore, the most recent policy review (January 1986) set the number of offices at 38 classified in four different categories. The total number of extensionists in December 1985 was 94 plus the regional subdirectors and central coordinators together with 20 or so headquarters staff. 6.05 Operational Policies and Procedures. The basic policy and the philosophy of the PAI extension service is to make contact with as many SMI's in the area as possible and to make them aware of the range of institutions and programs being offered to help them. The types of activities to be undertaken by extension officers were designed to fall under five different categories: (i) promotion (making first contact with enterprises); (ii) technical advice (making a quick diagnosis and providing direct assistance); (iii) referrals (to specialized technical institutions); (iv) training (for entrepreneurs 4t;d bank officers); and (v) group actions (assistance to groups of entrepteneurs). Two different periods in the extension service development can be identified. One, when PAI and FOGAIN were together under the same Director General and then after 1983 when the PAI-FOGAIN separation started and the extension services began to redefine its role. Although the separation of PAI and FOGAIN actually took place in early 1984, the second stage of review and redefinition of PAI's role lasted until mid-1985 and, in some respects, is still going on.  - 26 - 6.06 During the earlier phase of its operations, a major part of PAl's extension activities was concentrated in promoting FOGAIN's credit operations. The close relationship between extension work and credit was further stimulated by the existence of subsidized interest rates on FOGAIN credits which encouraged many SMI clients to cooperate in various technical assistance programs in the hope of gaining access to a FOGAIN credit. In retrospect, it is clear that in spite of some shortcomings in this mode of operation, the close relationship with credit served as an effective opening to "sell' the idea of technical/managerial assistance to the entrepreneurs. As a result, when PAI started some innovative approaches (e.g. collective assistance through "clusters"), entrepreneurs were ready to cooperate. Later, when PAI was separated from FOGAIN, in early 1984, the extension service was already mature, experienced and above all known to try new approaches and to broaden its clientele. Under the current policy, PAI's technical/managerial assistance is less concerned with promoting FOGAIN credits which is seen as only one of the many tools of assistance available to the enterprise. 6.07 In terms of delivery procedures, the most significant development, in the period under review, was the change from assistance to individuals to groups of enterpreneurs. While group assistance procedures are not new in extensionism, the methods used in Mexico were developed, with the assistance of a foreign expert, to fit the Mexican environment. The 'clusters" approach of organizing groups of firms in the same geographical area and with similar product lines (shoes, textiles, furniture, glass, etc.), and imparting technical assistance to the group, has achieved good success over the years. As an outcome of the "clusters" work, entrepreneurs have not only improved their technical and management skills, but have also organized purchasing cooperatives, credit unions, joint exhibits, quality control groups, subcontractors' exchanges and other forms of cooperation. Groups of entrepreneurs have expressed their appreciation for the assistance received in this way from the extension service and from each other. They recognized the value of that assistance and have expressed their willingness to pay a moderate price for further assistance. 6.08 During 1985 PAI conducted 66 general and 70 specialized courses, most of which were of about 20 hours duration. During 1984, 2,524 SMI enterpreneurs participated in these 136 courses which were offered free or at very modest cost. These training courses have suffered until now for lack of local training material. PAI headquarters, recognizing this problem, has recently prepared a collection of cases based on SMI in Mexico which should help to improve the relevance of the training given. Technical Assistance Institutions 6.09 Each of the Loans included a US$2.0 million technical assistance component to finance the foreign exchange costs of training the extension agents and the equipment needed for specialized technical assistance institutions. While the First SMI loan did not specify a breakdown of the technical assistance component, the second SMI loan estimated the resource allocation as follows: 20% for INFOTEC, 22% for CENAPRO, 24% for PAI, 17% to be divided be:ween FOGAIN, FIDEIN and FOMIN, and 16% for other institution (e.g. IMIT, FONEP, training schools).  - 27 - 6.10 In practice the use of T.A. funds under the First and Second Loans was substantially different than anticipated. INFOTEC used 50% of the funds, 40% of which was used to buy a computer system and 60% for foreign expenditure costs of foreign publications, patents and consultant services. Although INFOTEC definitely increased its work for SMI during this period there is no evidence that the funds used from this component were used for help only to SMI; FOGAIN used 29% of the funds, of which half was spent in calculators, typewriters and leasing of computing equipment, ard the other half in office furniture, office remodeling and publications for promotional purposes; PAI used 15% of the funds, of which nearly one-half was spent in paying for rental of conference rooms and oifice remodeling and the other half in paying for a foreign consultant to help develop the 'clusters" program; CENAPRO used 5% of the funds in audiovisual equipment, which was subsequently transferred, when CENAPRO was dissolved, to the Labor Ministry. FOMIN and FIDEIN used less than 1% of the funds altogether and operated without even some essential elements for their every day operations (e.g., computers and calculators). 6.11 In sum, the technical assistance component which was managed exclusively by PAI was not successful in coordinating the resource allocations and to evaluating the expenditures as related to the goals of SMI support programs. It should be noted also that Bank supervision missions did not raise the issue of the use of the T.A. funds with the PAI Secretariat although questions were raised by the Bank when the expenditures were submitted for approval. In retrospect, eventhough the technical assistance program was loosely defined the Bank supervision missions should have insisted that PAI develop a forward-looking monitorable technical assistance program, with adequate attention and resources applied to strengthening the operations of the main implementing agencies (FOGAIN, FOMIN and FIDEIN). In fact, a major drawback of the centralized control of technical assistance funds was that the implementing agencies did not benefit adequately from the resources for their institution building activities. To correct these anomalies, the recently appraised SMI project includes two features different from the earlier projects: first each participating institution will manage its own technical assistance program and secondly the technical assistance package has been prepared In detail with elaborate implementation schedules which can be jointly monitored by NAFIN and the Bank. VII. CONCLUSIONS AND LESSONS LEARNED 7.01 General. Overall the Bank played a positive role in the institutional development of the three executing agencies FOGAIN, FOMIN, and FIDEIN and at the same time it provided much needed long-term investment resources for SMIs. During implementation of the two projects, the executing agencies consolidated their policies, improved their operational procedures and built up a well-trained staff. Nevertheless, they still have to overcome weaknesses in project promotion, evaluation and supervision, and other important lessons have been learned from the implementation experience which are discussed below.  - 28 - PAI 7.02 Throughout the implementation of the loans, program coordination was an area of concern to the Bank. The projects were complex involving three participating institutions, an overall controlling authority (NAFIN) and PAI, an integrating organization. The Bank was fully conscious of the problems in getting all these elements to work together and efforts were made to improve coordination by raising this issue with the appropriate authorities during supervisions and by maintaining a direct contact with all the trust funds and with PAI. Despite these efforts, it must be acknowledged that the coordination did not prove effective in getting all the groups involved in both projects to work together. A major implementation lesson from the operation of an integrated program with a central coordinating body and a common financing fund for the several programs is to confine the coordinating role simply to providing policy guidance and to monitor projects, and not to directly manage the programs of independent executing agencies. Above all it is preferable to capitalize the executing agencies directly and to concentrate project resources in their institutional strengthening. Accordingly, under the Fourth SMI project, the Government has decided to: abolish the PAI financing fund and restrict the role of the technical secretariat to collecting information. Overall project policy guidance will be provided by a small high level coordinating committee. Technical Assistance 7.03 The most positive aspect of the two loans was the setting up and operation of an extension service for SMI enterprises. Besides its direct support, the extension service mobilized the assistance of other technical institutions involved with industrial issues. Under the two projects, innovative extension approaches were tested, and successfully replicated in different geographical areas. The support for other technical assistance institutions proved to be diffuse, and it is difficult to ascertain the extent to which SMIs benefitted directly from the activities of these. institutions. An Important lesson is that the technical assistance program in projects should be developed in detail with specific targets which can be monitored (para. 6.11). FOGAIN 7.04 Since the two loans under review were first appraised, FOGAIN has been successful, both financially (para. 5.05) and operationally (para. 4.06). It has made profits, and expanded its operations very rapidly. The acceleration in its loan portfolio, however, had some costs in that FOGAIN's operating procedures were in some cases applied loosely, particularly in areas of subproject evaluation and control (para. 3.11). Despite certain shortcomings, the subprojects have had positive economic impact in employment .creation and in industrial decentralization (para. 4.09). FOGAIN was also successful in mobilizing financial resources, both domestically and from international agencies. Its management over the period (1978-85) has generally been dynamic and its staff capable and dedicated. Evidence of FOGAIN maturity and capacity was shown by its ability to adapt and in fact expand its lending operations in the face of the severe economic constraints faced by the country in the 1982-83 period.  - 29 - 7.05 One of FOGAIN's weaker points has been its subproject supervision and the lack of data on actual subproject performance and the quality of the loan portfolio at the commercial bank level (para 3.12). Part of this problem is the extreme sensitivity in relations between FOGAIN and the nationalized commercial banks, which consider FOGAIN's information needs, as an "unnecessary irritant' particularly when the entire credit risk is borne by the banks themselves. On the other hand, the commercial banks appear to be more concerned about the collateral aspects of lending and have not adequately monitored the performance of subprojects, especially after the construction or equipment installation phase ends. Under the Fourth SMI project assistance has been provided to help FOGAIN and the financial intermediaries improve their subprojected information system. In addition, a time-bound plan was agreed to initiate a system for collection of key information on the status of subloan rcoveries at the intermediary level. 7.06 Thus a major implementation lesson from the FOGAIN component is that the Bank should concentrate from initiation on requiring improvements in the operating procedures (subproject evaluation, supervision and monitoring) of the institutions it finances and for this purpose the technical assistance resources should be clearly targetted. This is particularly important in the case where the Bank is supporting an already well established institution like FOGAIN, with in-built resistance to changes in procedures which have been operating for 25 years. FOMIN 7.07 FOMIN started its equity financing operations under the 1552-ME components in 1978 at a t*ime when the concept of providing risk capital for new SMI investments was virtually unknown in Mexico, and moreover the idea of a financing partnership with a Government-financed institution was considered inappropriate by private SMI entrepreneurs. Consequently, FOMIN's initial investment strategy had to concentrate on image building working with established entrepreneurs. It performed this task exceedingly well and helped in the expansion of existing plants and in creating new projects. The Mexican economic crisis of 1982-83 forced FOMIN into some unplanned involvement in financial restructuring investments, which has in restrospect been a good learning experience which is expected to prove useful under the FOMIN financial restructuring component under the Fourth SMI project. 7.08 FOMIN has built a significant institutional knowledge and experience in the difficult field of promotion, appraisal and supervision of equity investments in SMI. It has filled a critical gap in equity finance, in the absence of an equity market and any other institutional source of equity for medium size firms. FOMIN has also promoted the development of venture capital companies, of which the first five were authorized in May 1986. In retrospect, while FOMIN has successfully progressed to an innovative investment institution offering imaginative financial packages for SMI development, it needs to introduce more rigour in the selection of its investments which have not enjoyed complete success; in fact a substantial portion of FOMIN's portfolio has been adversely affected by the recent deterioration in economic conditions.  - 30 - 7.09 On balance, FOMIN's financial situation appears to be sound although again the discounting procedures between PAI and FOMIN, and inflationary distortions may obscure the real financial situation (para. 5.10, 5.12). Since FOMIN does not itself bear the losses on bad investments, FOMIN shows a continual increase in its equity which includes capitalization of nominal rather than real capital gains. In practice, divesture procedures have been the weakest point in FOMIN's operations. In the absence of a well-established capital market, transactions of minority shareholdings in medium-size companies remain restricted mainly to "buy-backs" by the majority shareholders. In determining the sales price of its participations FOMIN has usually considered the revaluation of its initial contribution or the book value of shares plus a modest (5%) premium on risk. Only recently, under Loan 2523-ME has FOMIN begun to explore other methods of valuation (e.g. projections of firm's future earnings). Income from interest on convertible loans and interest on sale of shares, has, in nominal teris, more than covered FOMIN's operating expenses. 7.10 FOMIN's appraisal procedures are still limited to traditional credit project appraisals and the venture capital aspects of appraisal (i.e. entrepreneurial abilities of majority shareholders, market life of new technologies) are not given proper assessment (para. 3.20). Also, the venture capital approach to investment selection is not yet fully recognized and there should be a more rapid rejection of nonviable investment proposals.. Greater promotional effort is required to reach new potential entrepreneurs and .to use the resources of INFOTEC, IMIT and other technical institutions, to generate innovative projects. FIDEIN 7.11 The FIDEIN program has produced .mixed results; While its park development schemes have contributed to industrial decentralization the results of its standard factory building (SFB)construction program, in common services facilities, and leasing of buildings and equipment were far less than anticipated at appraisal. While the development of industrial parks played an important role in industrial location, there were very few cases of relocation of existing units since the majority of the enterprises financed by FIDEIN were new and in some cases subsidiaries of existing firms located in the highly concentrated industrial zones (Mexico City, Monterrey, and Guadalajara). FIDEIN's SFB program had a good demonstration effect but as a business proposition public interest in SFBs was limited since investors preferred custom-made industrial buildings. As a result, FIDEIN was unable to market its SFBs and its inventory of buildings was finally liquidated in early 1986 (paras 4.23 - 4.25). 7.12 Management problems plagued FIDEIN operations throughout, due mainly to a lack of direction which was aggravated by the end of 1981 through (a) an apparently conflicting perspective towards FIDEIN from its two guiding institutions (SAHOP, the technical supervisor, and NAFIN, the financial trustee), and (b) irregularities in FIDEIN's financial management, during 1981, which prompted NAFIN to change FIDEIN's Director General and its main operational officers. After this there were various changes in policies and in FIDEIN's role (para 3.24 - 3.26). In its new role after 1982 FIDEIN reinforced its Technical Department and financing was limited to infrastructure development on industrial estates. The successive changes in policy and management since 1979 have hampered FIDEIN's development and damaged its institutional effectiveness.  - 31 - 7.13 The implementation experience of the FIDEIN component clearly demonstrates the operating difficulties of an institution which was seeking to change its role from a technical assistance (engineering type) institution to a financial intermediary without the necessary wherewithal to carry out its new role. The Bank appraisal missions recognized the need for institutionpl strengthening if FIDEIN was to carry out its new role, and hence appropriate plans for staff strengthening were drawn up under both loans. During implementation these plans were given lower priority. Given the complexity of the two projects (with several components) the small FIDEIN investment component received relatively less attention both from the Mexican authorities and the Bank supervision missions. Because of the lack of action in strengthening FIDEIN, this component was excluded during the appraisal of the Fourth SMI project. The Government and NAFIN pressed hard for the inclusion of FIDEIN as an excuting agency for the industrial park component. Recently NAFIN/FIDEIN have taken several positive measures to improve FIDEIN's performance. During the negotiations of the Fourth SMI project several additional time-bound measures were agreed upon for FIDEIN's institutional strengthening.  - 32 - MEXICO ANNEX I PROJECT COMPLETION REPORT Page I of 2 First and Second S.ll and Medium Scale Industry (SMI) Dsvelovent Prolects Loans 1662-ME and 1881-ME FOGAIN: Characteristics of Subprojects Financed Under Loan 1552-mri I/ (in Thousands of MexS) A. By Size of Approval Number % Amount 7 Upto50 5 6.0 811 0.6 51 to 100 7 8.3 1,690 1.2 101 to 200 20 23.8 10,421 7.5 201 to 500 21 25.0 19,423 14.0 501 to 1000 15 17.8 25,488 18.4 1001 to 2000 12 14.3 54,832 39.5 2001 to 5000 4 4.8 26,000 18.8 5001 to 10000 - - - - Above 10001 TOTAL 84 T10.0 1,50 B. By Size of Enterprise Number 2 Amount 2 Up to 25 3 3.6 893 0.f6 26 to 50 3 3.6 2,976 2.2 51 to 100 3 3.6 3,088 2.2 101 to 300 6 7.1 2,695 1.9 301 to 500 7 8.3 4,363 1.I 501 to 1000 14 16.7 8,850 6.4 1001 to 2000 16 19.0 i,566 A.3 2001 to 3500 13 15.5 24,926 18.n 3501 to 5000 10 11.9 43,73R 31.5 5001 to 10000 7 8.3 23,853 17.2 10001 to 15000 2 2.4 11,717 R.6i 15001 to 20000 - - - 20001 to 25000 25001 to 30000 30001 to 35000 - Above 35001 TOTAL 84 100.0 138,665 10n.n C. By Nature of Enterprise Number % Amount ! New Enterprise 13 15.5 28,911 20.8 Existing Enterprise 71 84.5 109,754 74.2 TOTAL Tr. 138,665 Inn. D. By End Use Number 2 Amount Fixed Assets 7.7Y 114,130 A2.1 Working Capital 29 25.7 24,535 17.7 TOTAL TmT T?