Mexico - Capital Goods Industries Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 9019 PROJECT COMPLETION REPORT MEXICO CAPITAL GOODS INDUSTRIES DEVELOPMENT PROJECT (LOAN 2142-ME) SEPTEMBER 25, 1990 Trade, Finance and Industry Division Country Department III Latin America and the Caribbean Region This document has a restricted distributCon and may be used by recipients only in the performance of their ofricial duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency unit - Mexican peso (Mex$) 1982 1983 1984 1985 1986 1987 1988 June 30. 1969 USSI MexS 67.2 150.3 186.2 310.3 837.4 1,406.8 2,288.3 2,387.4 G3VERNMENT OF MEXICO FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES Metric System ABBREVIATIONS CFE - Comision Federal de Electricidad (Federal Electricity Comission) CGI - Capital goods Industries COCOFIN - Comit- Coordinador y de Evaluaclon Financiera del Programa de Desarrollo a la Industria de Blenes de Capital (Commttee for the Coordination and Financial Evaluation of the Capital Goods Industrial Development Program) ERR - Economic Rate of Return FISOMEX - Fomento Industrial SOMEX (Industrial Development SOMEX) FOMEX - Fondo para el Fomento do las Exportaciones de Productos Manufacturados (Fund for the Development of Exports of Manufactured Goods) FONEI - Fondo Nacional do Equipamiento Industrial (Industrial Equipment Financing Fund) FRR - Financial Rate of Return GOM - Government of Mexico INFOTEC - Servicio de Informacion Tecnologica (Technological Information Service) NAFIN - Nacional Financiera S.N.C. PCR - Project Completion Report PEMEX - Petroleos Mexicanos PROFIDE - Programa de FinancTamiento on Divines para Ia Exportacion (Foreign Exchange Financing Program for Exports) PROFIRI - Programa de Financiamiento Integral para Ia Modernizacion Industrial (Industrial Reconversion Fund) Program - Capital Goods Industries Development Program SEPAFIN - Secretaria do Patrimonio Nacional y Fomento Industrial (Ministry of National Patrimony and Industrial Development) SOMEX - Sociedad Mexicana de Credito Industrial (Mexican Association of Industrial Credit) UNIDO United Nations Industrial Development Organization  FOR OMCIAL USE ONLY THE WORLD BANK Washington, D C 20433 USA OffOce of Directnt.GeltnI Ope#at:afs IvaluitUm September 25, 1990 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Mexico Capital Goods Industries Development Project (Loall 2142-ME) Attached, for information, is a copy of a report entitled "Project Completion Report on Mexico - Capital Goods Industries Development Project (Loan 2142-ME)" prepared by the Latin America and the Caribbean Regional Office. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This document has a restricted dirstribution and may be usod by fecipients only in the performance Of their offcial duties. Is contents may not otherwisse be discksed without World Banks Authoriztion.  PROJECT COMPLETION REPORT MEXICO CAPITAL GOODS INDUSTRIES DEVELOPMENT PROJECT (LOAN 2142-ME) TABLE OF CONTENTS Page No. PREFACE ........................................ i EVALUATION SUMMARY ..... ........................................... iii Part I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE .... ............ 1 1.1 Project Identity. . ....... .. .... 1 1.2 Background......... 1 1.3 Project Objectives and Description . . 3 1.4 Project Design and Organization.... 4 1.5 Project Implementation ..... .. ..... . .. ... 5 1.6 Project Results.... 7 1.7 Project Sustainability .... 9 1.8 Bank Performance ........... . . .10 1.9 Borrower Performance ......11 1.10 Project Relationships ...... ............................... 12 1.11 Consulting Services ....................................... 13 1.12 Project Documentation and Data ..... ....................... 13 Part II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ... ............ 15 2.1 Comments on Part I ....... ................................. 15 2.2 Bank Performance .. ............................ 16 2.3 Borrower Performance ...................................... 17 2.4 Lessons Learned ........ ................................... 19 Part III: STATISTICAL DATA ........................................ 21 Attachment A: Pro4ect Costs and Financing Plan .................... 23 Attachment B: Amendments/Waivers vs. Disbursement .... ............. 24 Attachment C: Estimated and Actual Cumulative Disbursements ....... 25 Attachment D: Basic Data and Project Timetable .................... 26 Attachment E: Use of Bank Resources ............................... 27 Attachment F: Related Bank Loans .................................. 28 Attachment G: Commitments ......................................... 29 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  TABLE OP CONTENTS (cont'd.) Page No. ANNEXES 1: Distribution of Bank Funds .......... ...................... 30 2. Breakdown of the Costs and Financing of Technical Assistance and Other Complementary Measures ............ . 31 3. Characteristics of Subprojects Financed .. ............... 32 4. Pro4ect Investment by Subsector and Benefits ... ........... 34 5. FONEI Loan Portfolio ....... .... ........................ 35 6. Project Promotion ................ .* ......... . ................ 36 7. Supervision Visits .... . ..*.... . ....................... .......... 37 8. Economic and Financial Review of Loan 2142-ME ............. 38  PROJECT COMPLETION REPORT MEXICO CAPITAL GOODS INDUSTRl.S DEVELOPMENT PROJECT (LOAN 2142-ME) PREFACE This is the Project Completion Report (PCR) for the Capital Goods Industries Project in Mexico, for which Loan 2142-ME in the amount of US$152.3 million was approved on May 13, 1982. The loan was closed on June 30, 1989, three and a half years behind schedule and it was fully disbursed. The PCR was jointly prepared by the Industry, Trade and Finance Division of the Country Department II, Latin America and the Caribbean Region (Preface, Evaluation Summary, Parts I and III), and the Borrower (Part II). Preparation of this PCR was started in June 1989, and it is based, inter alia, on the Staff Appraisal Report, the Loan and Guarantee Agreemen,1 and their successive Amendments, supervision and audit reports, correspondence between the Bank and the Borrower, and internal Bank memoranda.  - iii - PROJECT COMPLETION REPORT MEXICO CAPITAL GOODS INDUSTRIES DEVELOPMENT PROJECT (LOAN 2142-ME) EVALUATION SUMMARY Obiectives 1. The original project objective was to achieve an efficient development of Mexico's capital goods manufacturing subsector and thereby help to improve industrial productivity and employment, the balance of payments and the pace of technology assimilation and development in Mexico (para. 1.3.1). The Loan was designed as a credit line with two components. The first component provided financing for the establishment or expansion of medium- sized capital goods industries (CGI). The second component provided financing for technical assistance, training, technological information services, arnd feasibility studies to assist companies participating in the Capital Goods Industries Development Program (the Program). The project was substantially changed several times during implementation. These changes transformed the project into a credit facility for fixed and working capital finan:ing of all types of industrial projects and implicitly changed the project objectives into more generalized support of industry in an economy starved of foreign exchange. Implementation Experience 2. The project could not achieve the original objectives (para. 1.5.1). The Loan was approved at the outset of Mexico's deep economic crisis of 1982. The magnitude of the macroeconomic instability was underestimated and the consequent changes in purposes for which the loan could be used blurred the performance of the credit line and of the "financial delivery" system of the project. The alternative of cancelling the Loan, which was suggested by early supervision missions, might have been an appropriate project response at the time. However, Mexico's need for external resources during the economic crisis and the expectations of an economic upturn precluded that decision. 3. The initial modification of the project, in April 1983, was made in response to the gravity of the foreign exchange crisis in the early 1980s (para. 1.5.2). The initial amendment to the Loan Agreement reallocated funds to the Export Financing Program (PROFIDE) to assist in the financing of any industrial export project. A second amendment - under the Special Action Program approved by the Bank in early 1984 to accelerate disbursements - made permanent working capital for capital goods producers and enterprises directly controlled by NAFIN, eligible for financing. A third amendment, in April 1986, broadened the eligibility criteria of the Loan to include investment and working capital financing for any enterprise in the industrial sector. The approach of gradually expanding the eligibility criteria of the Loan may have delayed the recognition of the most important financing needs experienced in  - i.v - Mexico at the time, the generalized lack of foreign exchange for working capital and industrial restructuring financing to imorove the utilization of existing capacity. However, successive market-driven changes to expedite disbursements gradually changed the project's focus from CGI development to general financing of any viable industrial project (Attachment F). 4. The petformance of the Bank (paras. 1.8.1 - 1.8.3) through some stages of the project cycle was hardly adequate. Project supervision was infrequent and not intensive enough until 1987. Recommendations made by supervision missions were not transmitted clearly enough to the Mexican authorities, and the Bank was not assertive enough in insisting on improvements in NAFIN's subproject evaluation and supervision procedures. Finally, the project suffered from the overall adverse economic developments that were underestimated during project preparation and early implementation periods. 5. The performance of the Borrower (NAFIN) was similarly inadequate (paras. 1.9.1 - 1.9.3); it shared the responsibility for the actions taken - and those not taken - through the project cycle, which affected the results of the operation. There was a lack of clear lines of responsibilities, and the Committee for the Coordination and Financial Evaluation of the Capital Goods Industrial Development Project (COCOFIN) failed to play its role of project coordinator and facilitator. NAFIN was inexperienced in managing second-tier operations, and failed to improve sufficiently its subproject evaluation, supervision and reporting capabilities during project implementation; its response to Bank suggestions concerning necessary improvements in these areas was weak. Project Results and Sustainability 6. As mentioned above, the project did not achieve its original objectives (para. 1.6.7). The amount and number of subloans granted to CGI represent only about 12 to 15 percent and 25 percent of the total, respectively. Equity financing was only one-third of the planned amount. Finally, the financing of technical assistance and complementary measures fell short of the original plan. The changed economic situation required and justified drastic modifications of the project that had been designed in a relatively stable environment. These modifications led finally to relatively quick disbursements. The Loan financed 191 subprojects in seven industrial subsectors, whereby $100.9 million were used for fixed capital financing and $51.4 million for working capital, export, equity and some technical assistance financing, thus aiding the country's balance of payments and providing funds for improved industrial capacity utilization and restructuring. 7. Since NAFIN did not compile the requisite ex-post data on subprojects, only little information on the composition and performance of the subloan portfolio is available. Based upon portfolio information submitted by FONEI and a very small selection of subprojects examined by NAFIN, a very rough estimate would indicate that at the time the assessment was made, as many as 50 percent of all subprojects may have had achieved lower than expected financial and economic results and that in many cases these results may have been below the opportunity cost of capital. However, considering  that the assessment was made only about 12 - 15 months after the bulk of disbursements/investm,ents was made, and that at this time the Mexican industry was in a process of adjustments, it would be difficult to draw any firm conclusions from this estimate. The available ex-ante information indicates that the project may have helped many enterprises to prepare themselves for and pass through the critical stage of transition from a widely controlled and protected to a more liberal economy. Whether these possible positive effects of the project are sustainable, depends on the impact the recent economic liberalization measures have had on the financed enterprises which were all established under a proLective import substitution regime. Lessons Learned 8. The main lessons that may be learned from the execution of this project (paras. 1.8.4, 1.9.4) are that: (a) the Bank should closely assess and monitor the linkages between a proposed operation and the economic situation in and prospects for the country when timing an operation, and not hesitate to substantially modify or cancel projects, even at a very late stage of project preparation, if basic assumptions underlying the project design have changed or were faulty from the outset; (b) project supervision should be more frequent and intensive; (c) the Bank should act more assertively in its follow-up on weaknesses in project implementation reported by supervision missions; (d) clear operating policies and institutional responsibilities for project implementation should be agreed upon at loan negotiations; and (e) the credit line approach, for financing of larger subprojects and subprojects with long gesta8.ion periods, such as CGI subprojects, could substantially benefit from the pre-identification and pre-appraisal of subprojects, i.e. from the *ireparation of a pipeline of promising subprojects covering a reasonable initial period of project implementation. If this approach had been chosen, as initially intended (para. 1.4.3), many difficulties experienced during later stages in the project cycle, e.g. the lack of demand for CGI financing, could have been detected and dealt with during project preparation. 9. NAFIN's list of lessons learned from Project experience is very similar to the above, with the exception that NAFIN seems to place less weight on the importance of measuring project outcomes against stated objectives and on the value of careful project appraisal and risk analysis (para. 2.3.3). NAFIN, which will in the future act exclusively as a second-tier institution, would be well advised to develop intensive supervision and information services, to upgrade its project evaluation capabilities, and to instill among first-tier financial institutions through which it operates a deep sense of the value of these tools.  PROJEC.T COMPLETION REPOhT MEXICO CAPITAL GOODS INDUSTRIES DEVELOPMENT PROJECT (LOAN 2142-ME) PART It PROJECT REVIEW FROM BANK'S PERSPECTIVE 1.1 Project Identity Name: Capital Goods Irdustries Development Project Loan No: 2142-ME Amount: $152.3 million RVP Unit: LAC Region (LA2TF) Country: Mexico Sector: Industry 1.2 Backgro'md 1.2.1 The 1979 National Industrial Development Plan of Mexico identified the capital goods industries (CGI) sector as critical to the country's development needs. In March 1979, the Government of Mexico (GOM) issued the Capital Goods Promotion Plan and its implementation program (the Program), designed to provide a policy framework supportive of efficient CGI development. 1.2.2 The Program was based on an extensive study conducted by the United Nations Industrial Development Organization (UNIDO) and sponsored by Nacional Financiera (NAFIN)1/ which described the CGI sector as handicapped by low productivity, excessive fragmentation, limited specialization, inadequate technology development, unreliable input supply, and high regional concentration. The study idertified the absence of a coherent policy framework, and the lack of supporting mechanisms for the development of the sector (financir.g, training, research, and information) as the main obstacles to be removed. To rectify these shortfalls, the Program proposea specific measures "to address the technical, structural and information deficiencies of the industry, and to develop an institutional set-up which ensure the adequate mobilization of domestic and external resources". 1.2.3 The Program also proposed changes in the CGI policy framework, particularly in trade policy, fiscal incentives, and public sector procurement. In the area of trade policy, the GOM gradually moved away from '/ "Una Estrategia para Desarrollar la Industria de Bienes de Capital"; United Nations Industrial Development Organization, Mexico, 1977.  an import licensing system to tariffs. In January 1982, in the wake of introducing the peso float, tarifis were reduced to a level where domestic prices in "mature" industries did not exceed those prevailing in foreign markets by more than 15 percent. Infant industries were to be protected through higher tariffs and import licenses for a fixed period of up to five years. Concerning fiscal inzentives, the Program provided for tax rebates to CGI enterprises on the basir. of output, price, and domestic content requirements. New investments were granted an additional tax credit proportional to the numbe; of jobs created. Another tax credit was granted to companies meeting specific export targets and foreign exchange earnings. Companies purchasing domestically produced capital goods were also eligible for tax rebates on the purchase price. 1.2.4 The Program relied on the investment plans of Petroleos Mexicanos (PEMEX) and Confederarion Federal de Electricidad (CFE), as "drivers" of CGI investments. A neu law (January, 1980) established common procurement procedures for all public sector agencies and enterprises. These procedures specified -riteria for the selection' of domestic suppliers and domestic preference margins. 1.2.5 Taking into consideration the results of the above mentioned 'UNIDO study, the Staff Appraisal Report found that the Mexican CGI subsector had been one of the more dynamic groups within the manufacturing sector over the two decades preceding the appraisal. Starting from a relatively low level, output had been increasing during this period by about 9 percent p.a. in real terms, and the rate of growth had been increasing to 11 percent p.a. over 1977 - 1980. Nevertheless, compared with other large middle income countries, the Mexican CGI subsector had remained relatively underdeveloped. It accounted for only 11 percent of total manufacturing output, compared to about 20 percent in Argentina and 15 percent in Brazil. On the other hand, total requirements of capital goods had grown rapidly over the past three years at a rate exceeding 20 percent p.a.. Over the same period, Mexican domestic production supplied about 50 percent of domestic demand, compaved to 80 perce-t in Brazil. Capital goods represented Mexico's largest import category, accounting for almost one-third of merchandise imports, and, according to trade balance coefficient calculations, metal products and machinery were the main contributors to the foreign trade deficit. 1.2.6 In addition to the past growth of capital goods demand, the large increases in investment in both public and private sectors anticipated in the 1979 National Industrial Development Plan indicated an even larger growth potential for the 1980s. For all the above reasons, the Government of Mexico (GOM) considered the capital goods industries to be a strategic subsector whose development was expected to be critical to Mexico's immediate 8s well as long-term needs. This assessment appeared justified to the Bank, and, since the scarcity of credit in foreign exchange to support CGI production and sales placed the Mexican CGI in a disadvantageous competitive position relative to foreign suppliers, NAFIN's request for financial support of the Program was received favorably. This initiated the process which led to the approval of the Loan in May 1982.  - 3 - 1.3 Project Objectives and Description 1.3.1 The project, as it was originally designed, constituted the first phase of the Program, whose objective was to achieve the development of an efficient capital goods subsector, and thus to help improve productivity, employment, balance of payments, and technology assimilation and development in Mexico. The project would help to correct existing technical, structural and other deficiencies in the subsector and would provide financial and technical assistance to capital goods manufacturing industries and support institutions. Total project costs were estimated at $1,147 million, to which the Loan was expected to contribute only $152.3 million or 132. 1.3.2 The Loan had two main components (Attachment A): (a) Financing component: $130 million for medium- and long-term loan financing of fixed investments to support the creation or expansion of medium-sized private capital goods enterprises (through NAFIN, Sociedad Mexicana de Credito Industrial - SOMEX and Fondo de Equipamiento Industrial - FONEI); $17 million for equity investments by NAFIN and Fomento Industrial - SOMEX (FISOMEX) in medium-sized capital goods enterprises; and $2 million to finance technological research and development on a risk sharing basis between the investor and FONEI;2 (l) Complementary activities component: $1 million for training the staff of enterprises and of participating financial institutions in the preparation, promotion, and evaluation of capital goods projects, including specialized training in subcontracting, quality control and export promotion; assessing the existing training programs and facilities (through the Comite Coordinador y de Evaluacion Financiera of the Program - COCOFIN); establishing a mechanism to monitor progress in resolving technical and structural issues affecting the subsector; strengthening the expansion of the information dissemination system for the subsector (through the Secretaria de Patrimonio y Fomento Industrial - SEPAFIN); expanding technological information services (Servicio Informacion Tecnologica - lNFOTEC); and assessing the installed capacity of capital goods manufacturers (COCOFIN). 1.3.3 In addition to the Loan financed components, the project anticipated very substantial contributions from external sources, including bilateral export credits ($147 million), local banks and project sponsors ($622 million), and foreign banks and collaborators ($223 million). In addition to the areas covered by the Loan, these resources were expected to be available also for working capital and sales financing (Attachment A). 1.3.4 The substantial economic recession that curtailed capital goods investment from the very outset and the flaws in original design and organization of the project each affected negatively the acnievement of the project's initial objectives. Therefore, the project was modified to authorize channeiling of the proceeds of the Loan to promote overall indus-rial development. The several modifications in the Loan transformed the sharply focused project into a most general and flexible industrial credit 2/ The Loan amount also included US$2.3 million for capitalized front-end fees.  line whose main o''ective was to provide foreign exchange to finance any viable industrial subproject, thus responding to the changed economic situation resulting from the 1982 crisis (Attachment B). 1.4 Proiect Design and Organization 1.4.i Proiect T'esi n. The CGI study by NAFIN-UNIDO and the Bank's subsequent sectoral reports on Mexico3'/ proved to have two basic shortcomings: (i) the assertion of competitive advantages of CGI in Mexico was not substantiated by adequate quantitative economic or market analysis; and (ii) the lack of a detailed assessment of installed capacity and its utilization contributed to overestimating investment demand. Subsequent economic analysis was insensitive to the overwhelming impact of t)e 1982 economic downturn on CGI investment. 1.4.2 During project preparation, the project concept was changed from financing a portfolio of pre-appraised direct investments in CGI by NAFIN, to a credit line for the financing of mainly private sector sponsored investments in medium-scale CGIs. While both approaches aimed at the domestic production of capital goods for large state-owned enterprises (mainly PEMEX and CFE), the former would have had NAFIN as principal investor, while the latter relied on private entrepreneurs to carry out the investments, with NAFIN acting as wholesaler of funds. 1.4.3 NAFIN's role as main investor and promoter of capital goods industry was not supported by the GOM, as it was expected to encourage vertical integration in a subsector characterized in other countries by a large degree of horizontal integration. On the other hand, a credit line dedicated to finance medium-scale private capital goods investment projects had mixed reception in the Bank - being judged by some staff as a "disservice to the country"-, since it was believed that the long lead-time of the capital goods investment/production cycle would deter private investors. The Bank finally supported the credit line approach, departing from the prevailing Bank experience with CGI projects, which favored loans to finance pre-identified large projects or to provide overall support to the expansion of an existing capital goods subsector. Bank experiences had already demonstrated the complexity and pitfalls of CGI development and its dependence on a favorable macroeconomic framework. 1.4.4 The lack of demand analysis left unresolved whether entrepreneurs would be willing to invest in production facilities for import substitution of capital goods substantially depending on sales to state-owned enterprises. Moreover, CGI financing was not very attractive to financial intermediaries because of the newness of this non-traditional activity. CGI financing required long-term exposure and higher acceptance of risks in relation to other business opportunities, not always compensated for by interest rate spreads. Cyclical issues aside, these considerations proved to be the major deterrent to significant participation of both private irvestors and financial intermediaries in the CGI Program. 3/ "Mexico: Manufacturing Sector - Situation, Prospects and Policies" (Report 11671-ME, May/1977) and "Mexico: Development Strategy, Prospects and Problems" (Report 13605-ME, August/1981).  1.4.5 The series of adjustments to the initial project concept made during the first four years of implementation, lead to a major change in focus, but produced only a minor impact on Loan disbursements (Attachments B and C). Therefore, in April 1986, loan eligibility was extended to include financing of fixed and working capital for any viable industrial enterprise, the free-standing working capital component was substantially increased from $20 to $60 million and the financing of industrial services was permitted. These changes are discussed under project implementation (Section 1.5). 1.4.6 Project OrRanization. In the original design, the organization of the project relied on NAFIN, FONEI and SOMEX as implementing agencies, with NAFIN acting as a second-tier institution for the first time. However, NAFIN's dominant organizational culture was that of a first-tier lender, not of a wholesaler. Therefore, it encountered major organizational resistances. FONEI which had the experience and the administrative and staff capability for the Program, was brought into the operation under Bank auspices as second- tier lender. SOMEX experienced an absolute lack of demand from CGIs and had to be eliminated from the list of eligible intermediaries in February 1984. 1.4.7 COCOFIN, chaired by the Secretary of the Treasury, with a Technical Secretariat established in NAFIN, was responsible for: (a) overseeing the implementation of the Program; (b) establishing the Loan operating policies and guidelines; (c) reviewing the progress achieved by the executing agencies; (d) supervising the use of resources; and (e) coordinating the implementation of the complementary measures (training, information dissemination, and technical assistance). This reliance on a high level coordinating committee to ensure the operational consistency of the Program proved to be a mistake, particularly in a situation of serious economic crisis, when senior officials were occupied with other matters. The committee also failed to act decisively when the Program was halted at its early stages as a result of policy conflicts among the main participating agencies.41 1.5 Project Implementation 1.5.1 Loan Effectiveness and Project Start-up. The Loan was approved in May 1982 but became effective only in March 1983, due to the lack of coordination and agreement on the operational rules of the Program among implementing agencies. In retrospect, the timing of the project was inappropriate, imainly because the original project design was based on the assumption that the high level of economic activity experienced before the 1982 economic crisis would continue. The depth and duration of the 1982 economic downturn and its overwhelming impact on CGI investment demand had not been foreseen at the time of Loan approval. Whether they could have been anticipated is debatable, although "red lights" of an economic downturn were already perceived in the financial community, and in. February 1982, the GOM had announced an economic adjustment program centered on a drastic reduction of the public sector deficit. Changes in the policy framework (QRs reimposed, 4/ Disagreements on interest rate spreads between NAFIN and FONEI deadlocked for almost a year the approval of the guidelines for executing the Program.  unstable exchange rates) and cuts in the investment budgets of parastatal enterprises negatively affected demand and the investment outlook for CGI.5/ 1.5.2 As the economic downturn was expected to be temporary, it was believed initially that the project could be activated through relatively marginal changes in the project design. However, in April 1983, the Loan Agreement was amended to provide up to $100 million to the Export Development Fund (FOMEX) as executor of the PROFIDE Fund, for the financing of inputs of exporting companies. For this purpose, Loan eligibility was extended to all industrial export enterprises. The amendment, though transitory in nature,6/ was a substantial departure from the original project objectives, diluting the already weak institutional drive of NAFIN to implement the Program. The reallocation showed limited results because of the availability o compet_tive credit sources. Besides, the reinstated import control system guaranteed to importers the availability of foreign exchange at the official rate from the Bank of Mexico. At the end of 1985, $73 million were transferred back for capital goods investment financing. 1.5.3 In early 1984, again in response to the gravity of the foreign exchange crisis, the Bank approved a Special Action Program (SAP) to accelerate disbursements under loans to Mexico. In this context, the Loan's eligibility criteria were expanded, allowing the financing of permanent working capital for capital goods producers, increasing the financing percentage of local expenditures, raising the limit for equity financing, and extending eligibility to enterprises directly controlled by NAFIN. Simultaneously, given the reorganization of the Mexican financial sector after the nationalization of banks in 1982, some changes were made in financial intermediation.7/ 1.5.4 The above described changes did not substantially expedite the disbursement of the Loan. Since the pace of commitments remained far below expectations and a revival of credit demand for CGI financing was not in sight, a more drastic change became the inescapable alternative to Loan cancellation. Therefore, in April 1986, another amendment broadened further the eligibility criteria of the Loan to include investment and working capital financing of enterprises in any industrial subsector and the financing of industrial services. Also, from mid-1987 on, the Loan was administered by the 5/ Manufacturing activity which had recorded a 9Z annual growth rate during the 1978-81 period, declined by 3Z in 1982 and a further 7% in 1983. Capacity utilization in CGI manufacturing fell from close to 95Z in the boom years to 30Z in 1983. 6/ The allocation of funds to FOMEX/PROFIDE was thought of as a bridge-loan until the Export Development Loan (EDP) to Mexico became effective. However, the US$22.6 million disbursed under the loan to FOMEX/PROFIDE were not reallocated to the capital goods financing categories when the EDP was approved in June 1983. 7t SOMEX/FISOMEX were terminated as financial agents and BANCOMEXT substituted Banco de Mexico as Trustee of FOMEX.  Department for Special Operations established by NAFIN to conduct industrial restructuring operations (PROFIRI). Following the broad changes in project focus that permitted lending to the whole industrial sector and that allocated a larger share of lending to the financing of free-standing working capital, Loan disbursements surged. The creation of a Project Promotion Department in NAFIN also contributed to expediting Loan disbursements (Attachment B). 1.5.5 Imolementation Schedule. The project was originally scheduled to be completed by December 1985. Confronted with slow disbursement, the closing dates for commitments and disbursements were repeatedly postponed, the last time to June 1988 and June 1989, respectively. The situation was exasperated as NAFIN kept cancelling many approved subloans and using such freed amounts to increase other subloans until May 1989. Disbursements were continued until June 1989 when the project was finally completed with three-and-one-half years delay (Attachment D). 1.5.6 Proiect Costs. At appraisal, the total project costs (aggregate subproject costs plus technical assistance) were estimated at about $1,147.3 million (Attachment A). This estimate was based on highly optimistic assumptions about complementary external financing of the project, including bilateral export credits and suppliers credits, raised through principal implementing agencies, and local resources from private sector sponsors and financial institutions. While these expectations may not have fully materialized, the actual total project cost reported by NAFIN, $209.4 million, appear much too low, which must be attributed to NAFIN's inadequate recording and reporting procedures. According to the Loan Agreement, the proceeds of the Loan were to finance up to 50 percent of total subproject costs, which means that NAFIN would have violated the Loan Agreement, if total project costs were not at least $300 million. However, the Bank checked on the 50 percent requirement each time a subproject was approved, and some subprojects providing for more than 50 percent Bank financing were even rejected or had to be miodified. For these reasons, the total project costs should be somewhere between $300 million and the original estimate. Bank staff is still trying to clarify this question with NAFIN. 1.6 Proiect Results 1.6.1 The Loan of $152.3 million was used to finance 191 subloans totaling $149.7 million, technical assistance amounting to $0.3 million, and the front-end fee of $2.3 million (Annex 1). Of the total Loan, NAFIN used $113.4 million for 146 operations, i.e. more than double the amount planned at project appraisal. FONEI used $14.0 million for 31 operations, instead of the planned $64.5 million, because of competing loans it had received from the Bank. FOMEX/PROFIDE used $22.6 million for 14 export financing operations. Only $30 million were used for CGI financing instead of the entire Loan amount, and only $100.9 million were used for fixed capital financing, instead of the originally planned $130 million. Equity financing amounted to $5.6 million, compared to the planned $17 million and only 10 percent of the technical assistance component was utilized (Annex 2). The funds not used by FONEI and FOMEX were mainly used for export financing ($22.6 million) and permanent working capital financing ($20.6 million). The Loan financed 26 new ventures and 165 existing enterprises (Annex 3). Concerning the terms and conditions of subloans, 77 percent of subloans had repayment periods falling  - 8 - between three and seven years; interest rates were set for individual subloans under the General Interest Rate Agreement between the Bank and the Government (GIRA) and fluctuated in line with the rate of inflation. 1.6.2 Based on the subproject appraisal reports submitted to the Bank by NAFIN, the subprojects financed under the Loan have the following characteristics (Annex 2): (a) almost half of all subprojects financed large enterprises with annual sales above $5 million; (b) subprojects were relatively small, with about 73 percent of all subprojects having a cost of between $0.1 million and $2.0 million; (c) subloan size was concentrated (82Z) on amounts between $0.05 million and $2.0 million. According to the scant information provided by NAFIN (Annex 7), subprojects are relatively evenly spread over seven industrial subsectors, with some concentration on chemicals, auto parts, capital goods and textiles. Information on the regional distribution of subprojects was not provided. 1.6.3 The subproject appraisal reports estimated that 86 percent of all subprojects could be expected to generate internal financial rates of return (FRR) above 15 percent, and 62 percent to have rates above 25 percent. Only 14 percent of subprojects were expected to have rates of less than 15 percent but above the 10 percent cutoff (Annex 3). The ex-ante economic rates of return (ERR) were structured similarly to the FRR. Since the FRRs and ERRs of subprojects, as well as the results listed in para. 1.6.2, were not estimated at the time the project was prepared or amended, a comparison between appraisal and actual project results is not possible. Besides, the above information on FRRs and ERRs is very unreliable, due to weak subproject appraisal procedures, and not an acceptable indicator of actual project results. Therefore, the Bank proposed that NAFIN conduct, for the PCR, a special ex-post review of a representative sample of subloans. Only 10 subprojects were subjected to the calculation of ex-post FRR and ERR by NAFIN. Of this small group of subprojects only four achieved FRR and ERR of above 10 percent. The other six subprojects had either negative or below 10 percent rates of return. In at least two of these six cases, it is too early to draw any conclusions concerning subproject viability, since investments were made recently and are not yet showing their full impact. NAFIN mentions marketing problems as the main reason for low returns in all other cases and states that project sponsors are trying to address this issue by expanding their business into foreign markets. A more detailed discussion of the results of the sample evaluation is given in Annex 8. 1.6.4 The spare data on the condition of NAFIN's loan portfolio (Annex 8) indicate that almost $4.7 million (12Z) of NAFIN's outstanding portfolio of $39.9 million is classified as in arrears. However, this information does not offer an adequate basis for assessing the actual situation of NAFIN's second- tier portfolio, since NAFIN did not systematically collect subproject performance data. 1.6.5 FONEI maintained a reasonably satisfactory portfolio information system and from its files was able to provide partial information on the ex- post status of its second tier portfolio of 31 sublrans (16Z of total - Annex 5). These subloans totaled $13.9 million and supported subprojects that aggregated $51.2 million in total costs. Most FONEI lending was for the importation of machinery and equipment and there is some sectoral concentration on machinery and equipment and metallic products enterprises.  - 9 - FONEI subloans were distributed by the first tier commercial banks among 13 regions, with some concentration in the State of Mexico (25Z) and in Nuevo Leon (24%). 1.6.6 FONEI also provided information on the quality of its subloan portfolio (Annex 5). About 45 percent of its 31 subloans have been fully repaid and 13 percent were in arrears of between 180 and 270 days. Five of the 26 subprojects were reported to have failed outright while 8 other subprojects are classified as operating with difficulties. On the basis of project performance data collected by FONEI it is reasonable to conclude that at least 50 percent of FONEI's second tier operations is suffering from low or negative returns. 1.6.7 Conclusions. As follows from the above, the project did not achieve its original objectives. The amount and number of subloans granted to CGI represented only about 12 - 15 percent and 25 percent of the total, respectively (Annex 4). Equity financing reached only one third of the planned amount, and the project did not have the intended institution-building impact, since only a small part of the technical assistance and complementary measures was implemented and, in addition, was implemented too late to benefit the project (studies on capital goods production capacities and training facilities). The latter undoubtedly influenced the outcome of the project. In particular NAFIN should have benefitted from technical assistance in the preparation, promotion and evaluation of projects. The most important factors explaining the non-achievement of the original project objectives are, as discussed above, the economic downturn which dramatically affected CGI investment demand and, to a much smaller extent, the lack of effective project promotion and coordination by NAFIN/COCOFIN. 1.6.8 The changed economic situation required and justified drastic modifications of the project that had been designed in a relatively stable environment. After the Loan had been turned into a general purpose type credit line and the objective of the project had been changed implicitly from the promotion of CGI, mainly through fixed investment financing, to the promotion of industrial enterprises through fixed investment and working capital financing, the remaining 60 percent of the Loan were committed within two years, thus aiding the country's balance of payments and providing funds for improved capacity utilization and restructuring. Based on the above presented partial evidence provided by NAFIN and FONDEI, a very rough estimate would indicate that at the time the assessment was made, as many as 50 percent of all subprojects may have had achieved lower than expected financial and economic results and that in many cases these results may have been below the opportunity cost of capital. However, considering that the assessment was made only 12 - 15 months after the bulk of disbursements/investments was made and that at this time the Mexican industry was undergoing a process of adjustment, it would be difficult to draw any final conclusions from this estimate. 1.7 Proiect Substainability. 1.7.1 Since NAFIN did not carry out a meaningful ex-post evaluation of subprojects, the actual economic and financial benefits of the subprojects and their substainability could not be quantified by the project completion mission. In addition, information on the composition and quality of the  - 10 - loan/investment portfolio and the subprojects, though repeatedly requested was made available in part only (Annex 5 and 8, page 4). In spite of the lack of data, the provision of foreign exchange for the financing of working capital and modernization investments of mostly existing companies that were operating well below capacity, leads us to infer that the project has helped a large number of enterprises to prepare themselves for and pass through the critical stage of transition from a widely controlled and protected to a more liberal economy. Whether these possible positive effects of the project are sustainable, depends o.a the impact the recent economic liberalization measures have had on the financed enterprises, which were all established under a protective import substitution regime. Only a very thorough economic evaluation of the subprojects, including shadow pricing, could have ensured to some extent that project results are sustainable under a more liberal regime. However, NAFIN's economic project evaluation procedures did not provide this important function (para. 1.9.3). 1.8 Bank Performance 1.8.1 The performance of the Bank through some stages of the project cycle was hardly adequate. During project preparation, the Bank did not thoroughly enough investigate Mexico's competitive advantages for CGI development and relied on reports which omitted to assess existing CGI production capacities. Moreover, it was unrealistic to expect substantial private sector investments in an inward-oriented industry - export possibilities were not considered - with public enterprises, which traditionally had been capital goods importers, representing more than 50 percent of the GCI market, especially when the economy was experiencing the beginning of a trade liberalization drive. These factors might have escaped attention during appraisal, given the lack of substantive involvement of prospective investors in the preparation of the project. Regarding the timing of the operation, the SAR significantly underestimated the downside risks of the negative impacts of an economic slowdown on CGI investment demand. The recommendations of the NAFIN-UNIDO study, though questioned for excessive generality in Bank internal memoranda, did not trigger further economic and sector analysis during project preparation. 1.8.2 During Loan implementation, the Bank hesitated in introducing profound design changes or to cancel the Loan, although this latter possibility was discussed with the Borrower. Also a straight-forwardly transformation of the project into a balance of payments support operation was not 3eriously considered. Instead, the Bank started a "trial and error" process of adjustments to eligibility criteria, and it was not until 1986 that it concentrated on the real problem affecting the industrial sector - the overall shortage of foreign exchange working capital. The Bank's perception of the depth and duration of the economic downturn was continuously overshadowed by positive expectations of an upturn, leading to a process of continuous revisions and extensions of commitment and closing dates (Attachments D and B). 1.8.3 Bank performance in supervising the Loan was not effective until more recent stages of Loan implementation - 1987 onwards (para. 1.9.3). Project monitoring was affected by continuous staff changes and infrequent supervision missions (Attachment E). On the other hand, as early as 1984, supervision missions discussed the possibility of Loan cancellation and  - 11 - informed management that intensive supervision was needed. The signals given by the mission were not clearly enough transmitted to the Mexican authorities. Later, most of the substantial changes were made without any serious checking in the field. No intensive supervision mission was undertaken until June 1987 (the 1985 mission report concentrated exclusively on formal matters, audits, etc.), and even after this date, Bank management was not assertive enough in insisting on improvements in all stages of the project cycle. Given Mexico's foreign exchange problems in the mid-1980s, resource transfer seemed to have become the main objective of the project. As a consequence, preemptive actions were precluded, explaining the 'distress management" approach during project implementation. A good opportunity was foregone to support major institutional improvements in NAFIN, thus helping the institution prepare for future activities. 1.8.4 The lessons that may be learned from the experience gained during preparation and implementation of this project are that: (a) the Bank should closely assess and monitor the linkages between a proposed operation and the economic situation and prospects in the host country when timing operations and not hesitate to reappraise and modify or cancel projects, even in a very late stage of project preparation, if there is a basic change in the assumptions (e.g. on future economic developments) underlying the project design; (b) particularly for projects that experience implementation problems, project supervision should be more frequent and intensive; (c) the Bank should act more assertively on weaknesses in project implementation reported by supervision missions; (d) clear operating policies and institutional responsibilities for project implementation should be agreed upon at loan negotiations; and (e) the credit line approach for the financing of larger subprojects and subprojects with long gestation periods, such as CGI subprojects, could substantially benefit from the pre-identification and pre- appraisal of subprojects, i.e. from the preparation of a pipeline of promising subprojects covering a reasonable initial period of project implementation. If this approach had been chosen, as initially intended (para. 1.4.3), many difficulties actually experienced during later stages in the project cycle, e.g. the lack of demand for CGI financing, could have been detected and dealt with during project preparation. 1.9 Borrower Performance 1.9.1 The performance of the Borrower was also inadequate. NAFIN shared with the Bank the responsibility for the actions taken throughout the project cycle that affected the execution and results of the operation. For example, following a proposal by the Bank in May 1984, the Government decided not to cancel the Loan. The lack of clear lines of responsibility, overlapping institutional jurisdictions, a series of misunderstandings on project guidelines during project preparation and implementation, and NAFIN's insufficient experience with and commitment to a second-tier role, seriously affected the management of the operation. 1.9.2 The institutional project organization did not come up to performing the required level of coordination and decision making on policy, financial and technical issues. COCOFIN delegated its functions to a working level committee without the required decision making authority to expedite solutions. Operational misunderstandings with implementing agencies triggered a lack of interest in the Loan by participating agencies. FONEI, after having  - 12 - successfully committed Loan funds in the amount of US$14 million in less than one year (Attachment G), decided not to participate further in the Program. Promotion of the Program (Annex 6) was not a priority for NAFIN, although several supervision missions had requested improvements in this area. Though promotion activities started in 1986, it was not until 1987 that a more effective promotion unit was created in NAFIN. 1.9.3 NAFIN's lack of experience and the absence of an established network of banking relationships with prospective financial intermediaries to act as a first-tier institution, coupled with inefficiencies in NAFIN's own first-tier activities, rated its performance quite low. NAFIN had full responsibility for the execution of the project, but during the initial years did not make major efforts to implement it. The change in project design during preparation, compounded by the lack of CGI investment demand, slowed down the "institutional momentum" of the Program. Finally, while the speed of Loan utilization improved after the amendments of 1986, the quality of NAFIN's project appraisal and supervision work remained very weak. In too many cases the Bank had to ask for additional information during subproject review and this, as well as the cancellation of many approved subprojects, requiring the appraisal by NAFIN and approval by the Bank of replacement subprojects, continued to delay Loan utilization until May 1989. The impression prevailed that NAFIN's project analysis department was not convinced of the value of project appraisal and supervision methods as management tools needed to avoid resource misallocation in a heavily distorted economic environment, but considered them a requirement of the Bank, necessary to comply with in order to trigger disbu-sements. Consequently, subproject evaluations were carried out in a mechanical fashion without a serious effort to promote economically sound projects. The mid-1987 supervision mission addressed the low quality of NAFIN's subproject appraisal and supervision and the inadequacy of the internal information system. It strongly recommended a training program to upgrade NAFIN's capabilities in project cycle activities. Thereafter, some progress was made, above all, subproject promotion and supervision became more frequent (Annexes 6 and 7). However, serious drawbacks remained in NAFIN's project evaluation and portfolio monitoring capabilities. 1.9.4 The lessons of experience for NAFIN are similar to those indicated for Bank management. Besides, NAFIN, which in the future will act exclusively as a second-tier institution, would be well advised to develop intensive supervision and information services, to upgrade its project evaluation capabilities, and to instill among first-tier financial institutions through which it operates a deep sense of the value of these management tools. 1.10 Proiect Relationships 1.10.1 As discussed in para. 1.8.3, the main weakness of 3ank-Borrower relationships was the failure of the Bank to forcefully impress on the Borrower at an early stage its concerns about the low quality and efficiency of NAFIN's project cycle activities and the failure of NAFIN to respond assertively to the (admittedly weak) signals received from the Bank. Had the Bank been more determined in this respect, and had NAFIN been more receptive of the need ior adjustments, the project could probably have contributed to making significantly more progress in these areas, although at the cost of further delays in project implementation.  - 13 - 1.10.2 The Bank's relationships with F'ONEI were significantly better, since FONEI's project evaluation and supervision capabilities were relatively well developed. However, a weakness in the Bank's lending strategy was that, although the project relied on FONEI to utilize about 40 percent of the Loan, it provided competing Loan facilities directly to FONEI, thus weakening FONEI's interest in the project and reducing its participation in the end to only 9 percent. Consequently, NAFIN's share in loan utilization more than doubled from 36 percent to 75 percent, a change in the relative roles for which NAFIN was not fully prepared. NAFIN's position was further affected, at least during the initial stages of the project, by the lack of consensus within the GOM on how to provide support to CGI (paras. 1.4.2 and 1.4.3). 1.11 Consulting Services 1.11.1 The Complementary Activities component (para. 1.3.2, b) used only US$0.25 million of the US$1 million allocated for two surveys on production and training facilities capacities, plus a study of the Mexican financial sector not initially considered. The rest of the complementary activities, including training for the staff of enterprises and of participating financial institutions, the establishment of a mechanism to resolve technical issues affecting the subsector and improvement of the information dissemination system for the subsector, were not executed. During the last Bank supervision mission, INFOTEC, the autonomous tec'.nological information consulting trust, could not explain why they did not participate in any capital goods related consulting work, as originally planned. 1.12 Proiect Documentation and Data 1.12.1 The many amendments to the legal agreements made it soon difficult to have an overview of all the relevant loan provisions. Although some efforts were made by NAFIN and the Bank to consolidate the amendments in a new legal document for easy reference, the fact that a consolidated document was not prepared, remained a serious weakness. Also, the Staff Appraisal Report lost its value as a framework and guideline for project implementation, due to the total change in project design. 1.12.2 As mentioned before, only a very small part of the usually required data for the preparation of a PCR was readily available at NAFIN. Above all, no comprehensive ex-post documentation on the financial and economic impacts of the financed subprojects and no complete information on the status of the loan/investment portfolio was made available to the PCR mission. A lessorn for similar projects with second-tier mechanisms is that in addition to the supervision of their subprojects, the participating institutions themselves should be subjected to close monitoring by the second- tier institutions, including a regular review of their portfolios. Similarly, the participation of any financial institution in the project should be clearly conditioned upon Bank receipt of appropriate assurances of each participating institution's compliance with banking regulations regarding operating policies and procedures, financial condition and management and the quality and condition of external audits. 1.12.3 The audit reports from NAFIN and FONEI were generally received on time and in an acceptable form. Statements of Expenditures of NAFIN, FONEI and FOMEX/PROFIDE were backed by the required documentation supporting  - 14 - disbursements. Supervision missions had made recommendations for improving NAFIN's internal disbursement control procedures which were adequately adopted. A review of selected subprojects concluded that they complied with the administrative standard procedures in terms of supporting documentation.  - 15 - PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE* 2.1 Comments on Part I Economic Environment 2.1.1 In the 1970s, national and international conditions in which the project was set up were characterized by apparent stability and real growth in the economy. Credit flowed into the country as the supply of funds grew in international financial markets. It should be recalled that the decade was marked by an unprecedented supply of credit to developing countries. The lower cost of external credit and the ease in obtaining it led Mexico to assume that it would have the financing to support the changes the country needed to continue the national economy's development pattern that had been followed up to then. It seemed that funds (which domestic savings had failed to generate) were now available to be dedicated to meeting demands for the people's welfare and to reorient the industrial baee toward an increasingly crowded and competitive international market. However, by the end of the 1970s, financial imbalances in the industrialized countries resulted in pressure to take radical measures to adjust their economies. As iong as they failed to restore their financial balance, the threat of recession remained latent. In that context and after analyzing deficiencies in the capital goods sector, Mexico went to the World Bank to seek financing for a project to develop this sector. 2.1.2 The early 1980s were marked by two events: a drop in i.iternational oil prices and rises in international rates of interest. The consequences of this situation directly, but to differing degrees, affected developing economies. On one hand, countries whose projections of available revenues relied on petroleum sales saw their plans go up in smoke when their income fell drastically. On the other hand, the countries that were receiving foreign credit experienced an unexpected increase in the flight of their currencies, due to rising interest rates. These events affected the Mexican economy during th.L decade to such an extent that the need to apply austere adjustments to public financing had repercussions in every sphere. As resources dried up, adjustments in general and sectoral plans were directed toward maintaining the productive base and employment. The project financed with Loan 2142-ME could not escape these facts. 2.1.3 The decline in foreign exchange revenues from loans and oil exports, together with the growing exit of funds due to debt service, triggered structural adjustments in the economy. Faced with the magnitude of the economic events of the 1980s, public expenditures in investment were practically stopped, in turn constraining private investment, and especially in the capital goods sector, an industry that directs a large portion of its output to public enterprises. Thus, in view of the significant need to reorient other sectors and to keep them operating, the possibility of broadening the utilization of Loan 2142-ME was considered. 2.1.4 Substantial modification of the original objective of the project arose from the broadened utilization of Loan funds. Specifically, beginning in 1983, the objective was altered by orienting funds toward industries *Notes Part II and Annex 8 were prepared by NAFIN and submitted as a single report in Spanish, a copy of which is in the project file.  - 16 - pursuing the following objectives: (a) modernization of productive processes; (b) technology assimilation and/or development; (c) generation and/or maintenance of employment; (d) direct or indirect exports; and (e) training for staff, technicians, or laborers. It should be noted that these changes did not mean that the original objective was eliminated. Given the fact that domestic demand for capital goods was curtailed and international markets implemented protective measures (paragraph 1), it became inadequate to offer financing solely to the capital goods sector. However, we continued to offer support to the capital goods industry with funds from Loan 2142-ME. 2.1.5 In view of this situation, it became practically impossible to attain the objective that had been projected in a stable economy. The loan was modified precisely because the assumptions under which the project was structured underwent changes that were beyond the scope of project implementation. Accordingly, project evaluation should be carried out taking into consideration the achievements attained with regard to both the projected objectives and the unplanned objectives. Even at a point when the original project might have been canceled, circumstances in the national and international economic environment in the 1980s justified applying resources to alternative projects (paragraph 2.1.4). 2.1.6 The solidity and consistency of the capital goods development project are demonstrated by the seriousness of the study carried out by the United Nations Industrial Development Organizat Lon (UNIDO) that was sponsored by NAFIN, identification of the problems in the capital goods sector in Mexico, and the plan to promote and implement the Program to support this sector. However, it is important to point out that the Program, which is a pioneer of this type, revealed defects that are characteristic of a developing economy. For this reason it is indispensable to view the project's development in a context of unpredictable changes and dynamic and interdependent economies. For this reason, the causes and defects of the project, some of which are apparent and others real, were not the sole result of its implementation; rather, changing conditions in the environment played a decisive role. However, both the World Bank and NAFIN should acknowledge the rigidity of administrative structures in scenarios such as those that occurred during implementation of the project. In the final analysis, the changes made in the original project had results that were advantageous for the country and NAFIN (paragraph 2.3.1). 2.2 Bank Performance 2.2.1 Changes in the original project were justified by clearly changed circumstances. Through its consent, the Bank always supported these changes, even though the rapidity of these changes, and the impossibility of foreseeing their results in an economy beset by structural deficiencies and financial difficulties, to some extent hindered quick disbursement of loans. Likewise, the untimely Mexican economic recession perhaps prevented a more thorough analysis of the consequences that canceling both the project and the Loan would have had on the economy. 2.2.2 It is clear that Loan 2142-ME quickly adapted to the new circumst-l.ces; agreements between the Bank and NAFIN were always aimed at this purpose. The project was altered so that it could provide support to  - 17 - subprojects of companies requesting funds in the face of the country's economic situation that made it impossible for them to generate surpluses, not only for fixed capital formation, but also for formation of working capital. The Bank's authorizations for project adjustment brought about favorable results by contributing to the operation of 191 subprojects and by maintaining sources of employment, not considering the catalytic effect that these investments had on the rest of the economy. 2.2.3 In this regard it should be noted that also the underutilization of the Loan categories for risk capital and technical assistance was endorsed by the Bank in view of the events occurring in the economy. 2.2.4 In view of this scenario, the Bank did not fail, because objectives not foreseen in the original project were attained. However, closer supervision of the Loan and communication with NAFIN would have allkwed the Bank to more deeply understand the reasons for modifying the original project, as well as the reorientation given to the project, the immediate objectives for the national economy, and the need for funds which the country was faced with following the situation of the late 1970s, when it was receiving more resources from oil exports and loans in a few months than it had received for decades.8 2.2.5 Perhaps an understanding of these factors would have led the Bank to review its information systems and establish an exchange of information with borrowers, while also maintaining a closer relationship with organi- zations whose funct4on is related with the implementation of projects. 2.3 Borrower Performance 2.3.1 NAFIN's participation in this project respor,ded to two situations: (a) an external situation created by an uncertain economic environment, shortages of liquid funds, volatility in thE foreign exchange markets, and increases in domestic prices and foreign interest rates, and (b) given the functions performed up to that time by NAFIN, it was unable to achieve the optimal level of efficiency in second-tier operations. 2.3.2 NAFIN's organization attempted to structure its functions as a first- and second-tier institution. The experience obtained from this would allow it to restructure NAFIN in 1989 and reorient its activities toward those characteristic of a development financing institution. 2.3.3 The external factor (para. 2.3.1) played a decisive role. NAFIN had to make fast adjustments to entirely unexpected circumstances. Faced with this situation, analysis and evaluation of subprojects also had to be fitted to that reality, and the original project objective was adjusted in accordance with the critical situation thet was besetting the country. The subprojects felt the effects of this situation. Therefore, strict and rigid analysis of the subprojects in a distorted economic environment would merely have led to the funds' not being disbursed. This would have been an obstacle to NAFIN's 8/ Bank staff believes that the reasons for modifying the project were well understood, but that closer supervision would have been needed to ensure improvements in all activities during the project cycle (para. 1.8.3).  - 18 - role as a development baink. On this point, NAFIN's performance should be analyzed from the standpoint of its role as a development bank, since the financing granted under Loan 2142-ME made it possible for the companies that were facing liquidity problems -- due to a distorted economic environment and great uncertainty in the economy -- to be able to withstand the consequences of such an environment. From this perspective, NAFIN's ctions satisfied: (a) the project's objective (paragraph 2.2.1) and (b) its objective as a development financing institution. These are objectives which must be simultaneously fulfilled.9 2.3.4 The efforts of COCOFIN demonstrated, during its nine meetings, this contractual obligation. During the meetings, the Committee dealt with aspects of Loan 2142-ME, such as the definition of the mechanisms for operation of the Program, financial support, analysis of preliminary reports, and taking decisions concerning disposition of funds. 2.3.5 With respect to the supervision of subprojects, a special unit was created for the following purpose: to supervise and monitor appli-ation of the Loan and the development of the project. However, outside conditions (paragraph 2.3.1) obliged public enterprises to curtail their structures and to cut spending. NAFIN could not avoid this situation, which impeded strengthening the supervision function so that it could carry out ex-post evaluation as an implicit function in monitoring the Loan. For this purpose, NAFIN's re-structuring and its reorientation as a second-tier institution within the framework of modernization of the Mexican financial system allows us to have a management system dedicated to evaluating the results of the programs implemented by NAFIN. In this regard, it will be necessary to establish effective channels of communication with the Bank in order to guide efforts toward achieving ex-post evaluation objectives in accordance with the Bank's requirements. 2.3.6 The evaluation of ten subprojects, showed that the investments were made in accordance with the program. Problems were revealed in the operation of five subprojects, primarily due to two causes: (a) contraction of demand and (b) loan collection from customers. With regard to the latter aspect, it should be underscored that the companies we analyzed were primarily faced with this problem in relation to public enterprises. 2.3.7 It should be noted that this sample is not representative and, as a result, it would be inappropriate to draw any generalizations from it. Accordingly, the conclusions should not be extended to the rest of the subprojects. However, based on visits to the subprojects and subsequent analysis of information provided by the firms, it was shown that utilization of investments resulted in generation of 718 jobs, that is, 99 percent of the number projected, and that the average amount of investment per job was $12 million current pesos. Also, the average production was at 64 percent of the targeted capacity . The majority of the technology used in the manufacturing processes is of foreign origin, with emphasis on the process of technology 9/ In the view of Bank staff, ca,eful project evaluation (especially detailed risk analysis) is even more important in a distorted economic environment, both to preserve the assets of the development bank and to protect investors.  - 19 - assimilation and implementation of specific areas of development or maintenance of technology. 2.3.8 Of the 10 loans analyzed, only five loans are still outstanding, one was paid off in accordance with the loan agreement and the remaining four were repaid in advance. 2.3.9 Moreover, our first experience with this activity has been gained, allowing us to try out (new) forms and the methods especially developed for this (loan) purpose. In this regard, we shall request from the Bank its advice concerning establishment of the operating system and its orientation toward the requirements established by the loan agreements, in an attempt to streamline and make more efficient our ex-post evaluation activities, in such manner that satisfactory results can be obtained, in terms of quality and cost, for both NAFIN and the Bank. 2.3.10 Economic and financial information on the subprojects, as well as the current status of the Loan portfolio as of December 31, 1990, are detailed in the document titled "Economic and Financial Review of Loan 2142-ME" (Annex 8). 2.4 Lessons Learned 2.4.1 The lessons learned from this project for both the Bank and NAFIN are: (a) greater involvement of the Bank is needed in projects where delays are detected in implementation and operation, (b) information systems of both parties should be appropriately modified, (c) it should be easier to make necessary adjustments in projects, (d) NAFIN should establish an ex-post evaluation unit (paragraph 2.3.1), (e) excessive amounts of information should not be required, because time periods for compliance are short and we lack legal backing to require information from companies once the contractual obligations are completed,10 (f) NAFIN should establish an integrated information system in order to have appropriate and timely information concerning subprojects and in order to implement appropriate mechanisms 10/ Bank staff believes that NAFIN's relative inexperience in project appraisal techniques caused the impression in NAFIN that the Bank's insistence on solid project appraisals was excessive. However, the stricter review of loan applications suggested by NAFIN itself (point g of this para.) would require a satisfactory analysis of areas such as marketing, financial and economic viability, without which subprojects should not be approved.  - 20 - to require the companies to provide information even though the contractual obligazion has ended, (g) an ex-ante evaluation office should be convinced of the need to carry out stricter review of loan applications in order to avoid improper allocation of funds to companies that have alternative sources of financing, and (h) channels of communication linking the specialized technical areas of both institutions should be strengthened.11 1l/ Bank staff feels that the channels of communication were open and uncomplicated but that project implementation would have benefitted from more frequent supervision missions.  - 21 - PART III STATISTICAL DATA  - 23 - Attachment A PROJECT COV LETIOW REPORT MEXICO - CAPITAL 0ooos INDUSTRIES DEVELOPIENT PROJECT (LOAN 2142-YCE Prolect Costs and Flnancing Plan (millions US$ equivalent) Copmeientary materrol resources Othr e*xternal insluding bilatorol reaourceee expert credito a Local reooureee b for ein asI;*" eredita Local Local orIvat, world rald through private finencial Foreign benak A Sank Principal ipls- secter Inotl- callob- other Total Subprogram fiaessa g *s i log ap " too sponars tutionaC orators sourcee cost Eatimated at AoroloaoI Flied inveateont credits 181.0 180.0 *.9 139.0 *.9 s6.9 4709. Credits for peransent working capital *.6 6.0 6.9 120.0 9.6 I". 6 22060 Equity Investments 17.6 17. O110.1 o.od 42.6 O 06 193.0 Sales financing O.6 O.9 _.o 2so.0 o.6 8.6 250.6 Subloan* 147.? 147.8 116.O SO0.O 43.0 1O 9.6 1,133.e Technicalel *ooitnce A complementary _eauroe 3.6 6.6 4.8 $.6f *.* 0.6 12.6 Front and loan feo 6.6_I.- 6. *O6 2.3 162.3 147.0 114.0 511.1 43.0 166.6 1,147.3 Actual Flied invaatmant credits 1i0. 9 *. O.O 1S.O 6.O O .S 119.7 Credl^ for IV * ' .4nt working Capitol 29.6 0.6 6.9 4.9. *.0 *. 61.S Espert finatning 22.0 6.9 6.6 6.6 *.0 6.6 22.6 Equity lvoweteato 5.9 0.0 6.6 6.6 9.9 0.0 5.6 Sales financing .6- L1 6.6 W6. 9.6O L0 6.6 Subloame 14.7? 6.6 6.6 59.7 6.0 9.9 209.4 Technleal aooletanco a complementary e"8mur 6.3 6.1 6.9 6.7 6.6 9.6 1.6 fee" _* loan ft 2 e e *6 1 0 * i 1. 12.32 6.6 6.8 6.4 9.6 6.6 212.79 Soureest Eoatimated-onk Staff Appraisal Reperti actual--NAFIN, adjusted by Ienk.(ele saee below) / Indicative eatimto booed so prier io periopnr taking accee t of likely ehanges. / Figure. Inelude any lecel currency contributioen by the principal mplemseting agenclee to we ea other inte_dlery beaks. Figura Iinclude any local ecerency eontrIbutena by the princlpal imple_ tIng pr.ei.a so veol ae ether Interediery bonks. g/ Local currency centributiene by NAFINSA and MI3OMix. I/ ncludes USS201.O illion equlvalent financed threh.t..a F OWEX yate.. and WUs36O. millien provided by project epone"re. V Includes USS2.S million of budgptery reeeurces allocated directly to the progrsa fer Implementing prepoe-d cemplementery macurea and USt2.S million of locel resources to be providd through FONES for flinncing technicel asalatance subloans and risk-shring program fer technology development. 3/ The lank ceneld e thse. total cost figure unrellable.  NMt FV AO4 "WSW moodd dS ctanl -^-w-8wsr * _ _.,.1t.@@ _ t_l~~~~~~~~*llst. ONmn - I " um I hI. Tm e~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~s/ls3 U5.1 S (15 1586 34. cm 12131/" .e is-$ - - ' 343M/0 Sew. &W oseessansdfls, fee solhifells of 81COlD for I ew tde aeeps Is *spFOa pof U. Igve a. 1a/ss 11 es 1 0 1 o PIc/3t 38.8 3.1 r . .2 34.6 FM 09D/U/S n.e a3.8 . 82. 34.6 12/PSO Me soest eu csIrIm" *da of 11 f km _ t 4w" s*esF " * lot. r ut I(O hlsew h"j nsessss 3as/Se 3.9 18.2 SIC 4. 36.3 3m.s4 slle * f bsmo o-Aese aleopea. rI-t Dnastde .f3e fe1e. S12m3e wne1l e4pls jt3 24.0 30.3 311 sl moftds _ssg terw e pel.dpDt 0//3 36.9 49.5 0.8 20.0 5 5. EM 11111m0 Ull _1VO.se, rp lseeOs sae. ..l ceases e steemSas... e to _s* re (c6ns) /u.ction1sefforts 09 s/3c. 6eS00/64 1ALS 02.0 128.3 de.? 10.0 12/21/8o la ss bmis for slesh of as fe esv lt Use do llass. 32132104 330.3 12.4 14.3 34.9 U.& 32/38/46 laspe, of "aet of e|_sst asds *J le eoIdae mU * _63/6 _o 2. 63/35/US 3W.8 6"2.0 30.3 34.0 38.0 25/33/U baa mUm of fssde ewe gel.ww as.Inealso. 0e,aota ISO.$ 62.0 308 .9 n2.e -~~~~~~~~~~~~~~~~~~~~~~~~~~-~~~~~~a/19''"9w-1Xv p141s M 09. 84. 1 0f lbs b.s f sometimesls o sees f sp sdeess t te 1Ss31/3 4 93.4 49.. 3*.5 t3/12/U Saos of peeee _ad hfla le Ssuwee. psems to s s -s pietoseraseal f*so1582 *r_ asport15860 Sa .133/U ..0 38.11011 bentwss n sIllbIp" la oeSS asssolspeSaustlaesell 0/30/36 31. 93. * 40.9 441.0 EN Olin/0 , . - .9 UA 32/531M PSeonssst et _ eas *_el. feas, yea w // Ada rt. e12131' . 5.0 - r. 87.4 N/U/VW 8eiw of IeWS t or~f owl le ation fo dSse.Sfs. slscpelein ROD 4on lae. asSlap teeuasseltoe a eAr6ot 3/31/3? - 594.5 810.3 Ar p of assss a saws pemllas. 06/3? - - - 61.1 44.3 _5 _ 00_/2,8/3? _1_. exp r - - - 6r.r 13.2 n2/10/1 Isgbi p*sspaS et eleslant da.t fee pi seset 12//UOmps f sr 31313 - -0.1 "6.3 n50/ Wlvssr of tas he Ole fs o sl as fee _ea se limetod spa D s. a cs h 3 0e31. / . - . 304.3 U. S/OS/ Oslmles f fais- g_ _ _. esmpsaes 06*03 - * 383.3 3.5 DR ( 09/30/U _ - - 42.9 03.0 r ct it no rUba_ I WSO" *_ s o 1. based on *1 ersdie profile for OK projects le eer da, leyo-/3 - - - pe 0.s 5n #. c 23/0/3 Ibie pesglpasss of egaslas* dab" fe *si mile h 0/. 60133/3 - - . 380.5 93.0 Scale of the r dsp an O3s t fecosliglbl l s te esls*la. he 15 wcteejsel.e. 9/30/SO D , 1 - 1 - " 0 93.0 Lt 1.4 Cor solingeoles of _Stpv1*e for DN f1|sim MOUND" 8% set "I 1 SA-4va%1 estO4 to Dc*A.r 31. 199, ane_ D-csb# 1 Rle. Ceelasded Cienteab at ineslsel on- as lo Owetbss 31. t"lsessA pef 31S. fee *-An s_ . J IdU. */ Sfe.fre sssael of 4/7U eleled * st_.  - 25 - Attachment C Ph)JECT COMPLETION REPORT MEXICO - CAPITAL GOODS INDUSTRIES DEVELOPMENT PROJECT (LOAN 2142-ME) Estimated and Actual Cumulative Disbursements IBRD fiscal year Apraisala Actual & quarter ending uSn I m Tion uso million FY83 =ecember 31 1982 2.3 1.5 March 31, 1683 6.8 4.8 June 30, 1983 13.3 8.7 22.2 14.6 FY84 =eptember 30, 1983 23.3 15.3 22.2 14.6 December 31, 1983 35.3 23.2 24.8 16.3 March 31, 1984 50.3 30.0 24.8 16.3 June 30, 1984 66.3 43.5 29.0 19.0 FY85 =eptember 30, 1984 88.3 58.0 29.0 19.0 December 31, 1984 110.3 72.4 34.9 22.9 March 31, 1985 126.3 82.9 34.9 22.9 June 30, 1985 135.3 88.8 40.2 26.4 FY86 =eptember 30, 198