Mexico - Chiapas Rural Roads Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 9406 PROJECT COMPLETION REPORT MEXICO CHIAPAS RURAL ROADS PROJECT (-OAN 2525-ME) MARCH 1, 1991 Infrastructure and Energy Operations Division Country Department II Latin America and the Caribbean Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY Appraisal US$1 Mex$ 256.9 (March 1985) Interveninz Years Average 1985 US$1 = Mex$ 256.9 1986 US$1 = Mex$ 611.8 1987 US$1 = Mex$ 1378.2 1988 US$1 - Mex$ 2273.1 1989 US$1 = Mex$ 2461.3 FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES (Metric system in use throughout report) ABBREVIATIONS AND ACRONYMS NAFIN - National Financiera SCT - Secretariat of Communications and Transport SEDUE - Secretariat of Urban Development and Ecology SARH - Secretariat of Agriculture and Water Resourtes SPP - Secretariat of Programming and Budgeting  FOR OFFICIAL USE ONLY THE WORLD BANK Washington. DC 20433 Ohice of 01rect(W-C~.1a Opetaimn. 9akiatonn March 1, 1991 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report - Mexico Chiapas Rural Roads Project (Loan 2525-ME) Attached, for information, is a copy of a report entitled "Project Completion Report - Mexico Chiapas Rural Roads Project (Loan 2525-ME", prepared by the Latin America and Caribbean Regional Office. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment g 2 . 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.  FOR OMCIAL USE ONLY CHIAPAS RURAL ROADS PROJECT (LOAN 2525-ME) PROJECT COMELCION RPT TABLE OF CONTENTS Page No. Preface ............................... Evaluation Summary ................ Li PART I PROJECT REVIEW FRON BANK'S PERSPECTIVE ................. 1 Project Identity ...................................... 1 Background .......... 1 Project Objectives and Description .................... I Project Design and Organization ....................... 2 Project Implementation .....................; 3 Project Impact and Sustainability ..................... 4 Performance of the Borrower ........................... 5 Performance of the Bank ............................... 5 Project Relationships ................................. 6 PART I PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ............ 7 I. Introduction ..................................... 7 II. Project Identification, Preparation and Appraisal 8 III. Project Description .............................. 9 IV. Project Implementation ........................... 11 V. Economic Reevaluzation ............................ 13 VI. Conclusions and Recommendations .................. 14 PART III STATISTICAL INFORMATION ............................... 15 Related Bank Loans .................................... 15 Project Timetable ..................................... 15 Loan Disbursements .................................... 16 Project Implementation ................................ 16 Project Costs and Financing ........................... 17 Project Results ....................................... 17 Status of Covenants ................................... 18 Use of Bank Resources .................................. 18 Mission Data .......................................... 19 Map IBRD 18421 rThis document has a restricted distribution and may be used by recipients only in the perforrnance |of their offcial duties. Its contents may not otherwise be disclosed without World Bank authorization.|  MEXICO CHIAPAS RURAL ROADS PROJECT (LOAN 2525-ME) PROJECT COMPLETION REPORT PREFACE The following is a Project Completion Report on the Chiapas Rural Roads Project for which a loan in the amount of US$22.0 million was approved by the Board of Directors on April 30, '985. Only US$1.06 million of Lhe loan funds were disbursed. The remainder was cancelled in two stages: US$10.0 million on July 27, 1987 and US$10.94 million on December 17, 1988 retroactive to May 30, 1988. The Project Completion Report was jointly prepared by the Infrastructure and Energy Operations Division, Country Department II, of the Latin America and the Caribbean Regional Office. (Preface, Evaluation Summary, Parts I and III) and the Executing Agency, Mexico's Secretariat of Communications and Transport (Part II). Preparation of this Project Completion Report is based, inter alia, on the Staff Appraisal Report, the Loan and Guarantee Agreements; supervision reports; correspondence between the Bank and the Borrower; and internal Bank memoranda.  - iii - CHIAPAS RURAL ROADS PROJECT (LOAN 2525-ME) PROJECT COMPLETION REPORT Evaluation Summary Objectives 1. The Chiapas Rural Roads Project was to support the development of the state of Chiapas through a program of construction, rehabilitation, improvement and maintenance of rural roads coordinated with simultaneous agricultural and other social and economic development efforts.1 The project was to provide for: (a) the construction of about 1,000 km of rural roads; (b) the rehabilitation and improvement of approximately 2,000 km of existing roads; and (c) consulting services for detailed engineering and technical assistance. The selected roads were expected to provide improved access to rural areas so as to reduce transportation costs and increase agricultural production, and also to improve access for the efficient provision of social services and for employment opportunities. At the time of project appraisal it was envisaged that the project, where appropriate, would promote the use of labor-intensive construction techniques to increase the employment rtunities for unemployed -nd underemployed local residents (para. 3 - ImDlementation Experience 2. Between 1984 and 1988 only 443 km of rural roads were constructed in Chiapas by SCT, and about 1,000 km were reconstructed or rehabilitated. Although this amounts to about halt of the appraisal physical targets, only 5% of the loan was disbursed. This is because many of the works carried out did not have the documentation necessary for Bank disbursement (i.e., economic analysis and/or environmental impact analyses by the Secretariat of Urban Development and Ecology (SEDUE)) or were carried out according to procurement practices not acceptable to the Bank. In addition, delay in submitting documentation for disbursements eroded the dollar equivalent of the expenditures made since at the time the Mexican peso underwent a continuing process of devaluation (para. 5.04). 1/ The Chiapas Agricultural Development Project (Loar; 2526-ME), a complementary project, was approved by the Bank's Board of Directors on the same day as the Chiapas Rural Roads Project. Loan 2526-ME included also a rural roads component, to be carried out by the Secretariat of Agriculture and Water Resources (SARH) in a different area of influence than those financed under Loan 2525-ME.  - iv - 3. Despite thorough project preparation which included a detailed Project Implementation Document and a list of tasks to be carried out in order to complete the assessment of different groups of roads to be financed under the loa:-, SCT's rural road investments in Chiapas were mostly carried out without Bank participation. The US$2 million disbursed for the Special Account inmediately after the loan became effective were partially drawn-down only after the Government decided to completely cancel the loan. The Government, by DecemEer 1988. could justify eligible expenditures for US$1.06 million, and US$0.94 million were reimbursed to the Bank (paras. 5.01 and 5.03). Results 4. The project financed the construction of about 160 km of rural roads with an estimated ex-post economic rate of return of 13%, as compared to the 19% envisaged at appraisal. The less-than-expected returns are mostly explained by the lack of complementary agricultural services that were to have an important impact on agricultural output. Because the institutional component of the project never took off, its impact on that area did not go beyond the limited on-the-job training that the exposure to Bank evaluation techniques provided to the staff in SCT-Chiapas (para. 6.01). Sustainabilitv 5. Sustainability of the limited accomplishments under the project will mainly depend on Chiapas's future efforts in road maintenance. At present, road maintenance is inadequate. But the Bank is considering a road project with heavy emphasis on strengthening future road ma4ntenance capability of the States. As a result, future road maintenance may improve and positively affect the sustainability of the Chiapas Rural Roads Project (para. 6.01). Findingj and Lessons Learned 6. Project implementation and withdrawal from the Bank loan account required SCT-Chiapas, SCT's regional office, to have an important role in data gathering and in conducting the analysis of project justification, tasks for which they were ill-prepared. Lack of experience in these matters could have been a surmountable problem and the implementation of the technical assistance component of the project, which never took off, could have helped in this regard. However, the executing agencies had very low motivation, andt many disincentives to use loan funds since the external financing did not result in additional funding to the agency but was meant to substitute Federal funding, which was available with fewer strings attached (i.e.: as regards project justification and procurement requirements) (para. 7.01). 7. The Bank's requirements with respect to the economic justification of subprojects eligible for financing under the project were perceived as obscure by those officials in Chiapas unfamiliar with cost-benefit analyses. However, as already stated, this problem could have been surmounted if the Government would have given priority to the project (para. 9.01).  v 8. The main lesson to be learned from the project is not a new one. It relates to the need to ascertain, early in the project cycle, the strong commitment of those agencies and/or levels of technical staff that are in the critical path of successful project implementation. In addition, the Bank should try to balance the need to ensure the economic viability of the projects it funds with the limited capability of those in charge of conducting the assessment. Finally, from the moment the Borrower and the Bank decided that the best course of action was to cancel the loan, two years had elapsed until it was actually canceled. This was the result of the time required by the different Government agencies involved, (NAFIN, SCT, the Secretariat of Finance and Public Credit and the Secretariat of Programming and Budgeting) to reach consensus on the future of the loan. These delays probably reflected the Borrower's expectation that the project could be turned around, but with hindsight there were no objective reasons to expect this (para. 8.02).  MEICIQ CHIAPAS RURAL ROA.DS PROJECT (LOAN 2525-ME) PROJECT CO'1PLETION REPORT PART I: PROJECT RFEiVIEW FROM BANK'S PERSPECTIVE 1. Proiect-Identity Name: Chiapas Rural Roads Project Loan Number: 2525-ME RVP Unit: LAC Region Country: Mexico Sector: Transport 2. Background 2.01 The State of Chiapas is, economically and socially, one of the least developed of Mexico. Per-capita income is estimated at US$ 350, as compared to the national average of US$2250. From its population of about 2.3 million, near one-third live :'.n urban areas, and the remaining two- thirds in rural areas, an opposite, relationship to the national population distribution. Statewide, poverty is prevalent and malnourisiament among children is widespread. The child mortality rate (between the ages of 1 and 4) is of 85.2 deaths per thousand children, quite above the national average of 42.9. The population has the highest illiteracy rate (34%) compared to the national rate of 20%. One of the main obstacles for improving the socio-economic environment has been the difficult terrain which hinders access to the widely disperse population. This, coupled with adverse climatic conditions, lack of proper drainage, limited extension and other agricultural services, lack of storage and inadequate transport access, makes increases in agricultural production (the main source of income) very difficult. 2.02 To help define a comprehensive development strategy for Chiapas, a multisectoral Bank mission visited Mexico in late 1981. The recommendations of the joint Bank/Government Working Group formed the basis for the Chiapas Development Plan (Plan Chiapas), officially adopted by the Government in May 1983. Two priority elements were selected as crucial to the initial stages of the program and provided an appropriate instrument for Bank financing: agricultural development and rural roads. Two projects: the Chiapas Agricultural Development Project and the Chiapas Rural Roads Project, were thus prepared in parallel. 3. Proiect Objectives and DescriRtion 3.01 The Chiapas Rural Roads Project was to support the development of the State of the Chiapas through the implementation of a rural roads construction, rehabilitation, improvement and maintenance program coordinated with simultaneous agricultural and other social and economic  -2- development efforts. 1 The project was to provide for: (a) the construction of about 1000 km of rural roads: (b) the rehabilitation and improvement of about 2000 km of existing rural roads; and (c) related consulting services for detailed engineering and technical assistance. The selected roeds were expected to provide improved access to rural areas to reduce costs and increase production and/or improve access for the efficient provision of social services and for employment opportunities. At the time of project appraisal it was envisaged that the project, when appropriate, would promote the use of labor-intensive construction techniques to increase the employment opportunities for unemployed and underemployed local residents. 3.02 To ensure the full realization of the expected benefits, the Rural Roads Project was also going to focus upon the Governments efforts to: (a) strengthen the state's planning and decision making capabilities in Chiapas; (b) improve planning, programming, construction, rehabilitation, improvement and maintenance programs of rural roads in Chiapas; (c) assess rural transport services for freight and passengers in Chiapas; and (d) extend lessons learned in Chiapas to other states. 3.03 Project preparation was delayed by the change in Government and the need to confirm the support of the new administration to the objectives of the project. The project was appraised in February, 1984 and netotiations took place in Washington from February 25 to March 20, 1985. The loan became effective in April 4, 1986, about six months later than originally scheduled. In retrospect, these delays were an early indication that the new Government, beleaguered by Mexico's financial crisis and severe cuts in public investments, did not assign high priority to this particular project. 4. Project Design and Organization 4.01 The project was designed as a component of a multisectoral effort to address widespread poverty in Chiapas. Thoroughly prepared, it included a detailed Project Implementation Document detailing methodologies and procedures for subproject selection and execution. This document is still used as a valuable example for rural roads projects in the Region. __i main accomplishments were its innovative approach to balance the demand for resources to address social needs in those cases in which economic benefits are not readily quantifiable and its attempt to simplify, to the extent possible, the measurement of the project's benefits. The approach. described in para. 3.03 of Part II of this report, consisted of a staged evaluaSion of the generated agricultural producer 1/ The Chiapas Agricultural Development Project (Loan 2526-ME), a complementary project, was approved by the Bank's Board of Directors on the same day than the Chiapas Rural Roads Project. Loan 2526-ME included also a rural roads component, to be carried out by the Secretariat of Agriculture and Water Resources (SARH) in a different area of influence than those financed under Loan 2525-ME.  - 3 - surplus. However, those roads in areas with limited agricultural potential and high population concentration were considered "social" roads. The number of -oads with a predominantly social justification could not exceed 15% of the construction program and each such road could not exceed a construction cost of US$450 per person served. 4.02 Project identification and design involved several Government Amencies at the State and Federal levels. While the project was clearly understood by those participating in this process, in retrospect it appears that it did not elicit the same degree of support among them. In particular, the Planning and Budgeting Secretariat (SPP) was the driving force behind the Plan Chiapas. On the other hand, the executing agency, the Communications and Transportation Secretariat (SCT) might have seen the implementation of this project as cumbersome, involving evaluationi exercises in a state with limited data and even more limited capability to process it. 5. Project Imnlemeatatio.. 5.01 Despite the substantial effort during project preparation which, in addition to the already mentioned Project Implementation Document, included a detailed list of tasks to be carried out in order to complete the assessment of different groups of roads identified for possible Bank financing, SCT's rural road investments in Chiapas were mostly carried out without Bank participation. The US$2.0 million disbursed for the Special Account immediately after the loan became effective were only partially drawn-down once the Government decided to cancel the loan. The following paragraphs discuss the efforts to implement the project and the steps leading to the cancellation of most of the loan funds. 5.02 From the onset the project suffered from the severe reduction in public expenditures that Mexico undertook as part of its macroeconomic adjustment program. In 1986, SCT's construction of rural roads in Chiapas amounted to about 35km, as compared to the 150-200km envisaged at appraisal. Not only was the lack of counterpart funds resulting in a slow rate of project execution, but no documentation was submitted to the Bank for the financing of the small amount of eligible expenditures that were incurred. This is also true of the construction works carried out between July 1984 and July 1986, for which the loan provided retroactive financing at the exchange rate of the date in which the expenditure took place. 5.03 In November of 1986, only seven months after the loan became effective, the Bank was recommending its cancellation. By then, two Bank missions had visited Chiapas to familiarize local SCT professionals with the procedures related to project implementation and the use of loan funds. However, there was no progress in these areas. At the request of the Government, US$10.0 million of loan funds were canceled on July 27, 1988. Subsequently, the Government presented to the Bank possible alternatives to revise the scope of the project including, inter-alia, the financing of rural roads in other states and the financing within Chiapas of roads with higher specifications than those originally agreed. The Bank did not find these alternatives acceptable since they were not within the project objectives. On May 30, 1988 the Government requested cancellation of another US$10.0 million which was 'ostponed until the presentation to the Bank of documentation to justify the use of funds in the Special Account. On December 17, 1988, US$ 10.94 were canceled retroactive to May 30, 1988.  4 - 5.04 Loan funds amounting to US$1.06 million were disbursed under the project. In fact, between 1984 and 1988 only 443km of rural roads were constructed in Chiapas,2 and about 10OOkm were reconstructed or rehabilitated. Although this amounts to about half of the appraisal physical targets, only about 5% of the loan was disbursed. This is because many of the works carried out did not have the documentation necessary for Bank approval (i.e. economic evaluations and/or SEDUE's clearance regarding the environmental impact of the road works (para. 1.13)) or were carried out according to procurement practices not acceptable to the Bank. In addition delay in sAbmitting documentation for disbursements eroded the dollar equivalent of the expenditure:s made since, at the time, the Mexican peso underwent a continuing process of devaluation. 5.05 The project adequately addressed the problem of protecting Chiapas' unique ecological reserves. To that end, it was agreed to use Mexico's established review process applicable to all construction works, with the Secretariat for Urban Development and Ecology (SEDUE) acting as the responsible agency. During project preparation, data on the initial list of subprojects to be financed under the project was submitted to SEDUE, and all but seven of those roads received clearance. The seven roads which required further study were located in the Lagos de Montebello National Park area. Although SCT-Chiapas continued requesting SEDUE's opinion on these roads, SEDUE never responded to these requests. As already discussed, lack of act4on on SEDUE's part also contributed to poor project implementation. 6. Project Impact and SustaliWhAWtv 6.01 The project financed the construction of about 443 km of rural roads an6 road rehabilitation and improvements. The estimated ex-post economic rate of return (ERR) for the road construction component is 13X as compared to the 191 envisaged at appraisal. Ex-ante ERR's ranged from 5.7S to 262%. The ex-post rates ranged from 8.31 to 37.21 (Part III, Table 6). The less than expected returns are mostly ex-lained by the lack of complementary agricultural services that were to have an important impact on agricultural output. Because the institutional component of the project never took off, its impact did not go beyond the limited on-the-job training that the exposure to Bank evaluation techniques provided to the staff in SCT-Chiapas. Sustainability of the project will depend on Chiapas's efforts in rural roads maintenance. At present, road maintenance is inadequate. But the Bank is considering a road project with heavy emphasis on strengthening future road maintenance capability of the States. As a result, future road maintenance may improve and positively affect the sustainability of the Chiapas Rural Roads Project. i Of these 443 km the Bain; '_nally provided financing for 241 km, or 23% of the appraisal physical target. Delays in submitting documentation for disbursements coupled with the process of continuing devaluation of the Mexican peso eroded the dollar equivalent of the expenditures made in those roads. As a result, Bank financing represented, in real terms. approximately 26% of the estimated total cost of the 241 km of roads.  5- 7. Performance of the Borrower 7.01 Satisfactory project implementation would have required SCT- Chiapas, the regional SCT office, to have an important role in data gathering and project justification, tasks for which they were ill- prepared. In this regard, SCT's Directorate General of Planning, could have played a more active training and dissemination role. It should be noticed.-however, that lack of experience in these matters could have been a surmountable problem but the executing agency had very low incentives, and many'disincentives to use loan funds. This was because the external financing did not result in additional funding to the agency but was meant to substitute Federal funding which was available with fewer strings attached (i.e.: as regard project justification and procurement requirements). 8. Performance of the Bank 8.01 The Bank acted with relative innovativeness by helping to develop the Plan Chiapas and supporting it through two independent but complementary operatious that were processed in parallel. With hindsight, it is clear that the project did not have across-the-board support from the Government, mainly because of changed priorities and the perception that the project's benefits were outweighed by the cumbersome procedures required to implement it. It is difficult to ascertain, however, to what extent it was possible for the Bank to realize that this was the case during the project preparation stage. Once it became evident that the project was not going to be implemented as originally envisaged, the Bank pressed the Mexican Government to act promptly to cancel the loan and provided the necessary assistance to streamlir.e the processing of the information to maximize the disbursement of loan funds. On the other hand, SARH continues to successfully implement, albeit with delays, the rural roads component of the Chiapas Agricultural Development Project. To date about 50% of the physical target has been reached. 8.02 The main lesson to be learned from the project is not a new one. It relates to the need to ascertain, early in the project cycle, the strong commitment of those dgencies and/or levels of technical staff that are in the critical path of successful project implementation. In addition, the Bank should try to balance the need to ensure the economic viability of the project with the limited capability or those in charge of conducting the studies to quantify it. Finally, from the moment the Borrower and the Bank decided that the best course of action was to cancel the loan, two years had elapsed until it was actually canceled. This was the result of the time required ',y the different Government agencies involved (NAFIN, SCT, the Secretariat of Finance and Public Credit and the Secretariat of Programming and Budgeting) to reach consensus Ofn the future of the project. These delays probably reflected the Borrower's expectation that the project could be turned around, but with hindsight there were no' objective reasons to expect this.  - 6- 9. Project Relationships 9.01 Throughout project preparation and implementation, relations between the Bank and the Borrower were always excellent. The Borrower and SCT, however, always made clear that they felt the Bank had to relax its conditions for subproject approval, citing it as the main reason for lack of project progress. The Bank, on the other hand, was of the opinion that the problem of project evaluation, even when it existed, was a surmountable one. The fact that SCT was able to justify the inclusion under the project of works in several road links, seem to support the Bank's position.  -7- PART II PROJECT REVIEW FROM BORROWER'S PERSPECTIVE I. Introduction 1.01 In order to identify strategies to adequately develop the State of Chiapas, the Mexican Government, supported by the World Bank, in 1981 started to review the social and economic problems affecting the State. The studies resulted in the design of a development plan called 'Plan Chiapas", that was adopted in May 1983. From such Plan, two elements were selected as priorities that translated into World Bank projects, with financing made possible by foreign loans: an agricultural development project and a rural roads development and rehabilitation project. 1.02 The proposed rural roads project would provide a basis for direct growth in the State of Chiapas. Its implementation was to be consistent with agricultural development, and with the objective of strengthening the social and economic aspects of the State. 1.03 Nevertheless, carrying out the project required substantial budgetary support, which, because of the economic crisis affecting the country, mainly due to the drop in oil prices, was not possible. This resulted in the project's goals not being met, and remaining very much below the planned levels. Compounding budgetary limitations were an untold number of operational-type problems that inhibited the optimal use of scarce available resources. 1.04 The above-cited resulted in only Mex$2,703.2 million being considered for financing from Loan 2525-ME, of a total of Mex$4,859.5 million spent on the program, i.e. 56% of the total amount susceptible of financing under the project. However, it is important to note that the entities involved in project implementation always attempted to provide their best effort. There was no way to anticipate the problems, which will be described later, that came up during the execution of the project. Perhaps the answer to future problems of this kind would be to provide adequate training to all staff involved in handling foreign credits, but especially to those who will be implementing externally-financed projects. 1.05 This report is based on Bank Aide-Memoires, various documents submitted to the World Bank on the "Plan Chiapas", a Project Evaluation Report, the General Agreements on IBRD loans, the Loan and Guarantee Agreements, and the Letter signed by the World Bank, dated May 24, 1984.  - 8 - II. Proiect Identification. Preparation and Appraisal 2.01 Road construction, besides achieving the objective of linking communities, implies that populations will come out of isolation and start an integrated development through education, health, trade, industry, agriculture, tourism, and public services. Roads provide agriculture a larger geographic access, i.e., agricultural products are no longer exclusively for local consumption but can be consumed at regional and national levels, and even be exported. Roads also serve to exploit the riches of forestry and the safe and timely transport of production inputs, reflecting an accelerated rate of increase in production. 2.02 The Chiapas Plan was specifically designed to achieve adequate development for the State and regarded as a priority the construction and rehabilitation of road infrastructure. Thus, with the purpose of supporting the implementation of the "Chiapas Plan", and therefore of co- operating with the State Government for the State's economic development -- mainly in the agricultural area -- in 1982 the Federal Government started negotiations with the World Bank to obtain loans to financially support the Program. The result were two complementary loans: one to develop new agricultural areas, with the Secretariat of Agriculture and Hydraulic Resources as the responsible agency, and the other one to construct and rehabilitate rural roads in the State, with the Secretariat of Communications and Transport (SCT) as the responsible agency. These projects were executed independently. 2.03 The economic feasibility studies and the reports required by the World Bank for approval of the loan were carried out by the Commission or State Planning and Development (COPLADE), the General Planning Directorate (formerly the General Directorate on Investment Analysis), and the project's implementing agencies. The studies, carried out for the Bank, amounted to seven volumes with technical and economic feasibility information on the proposed project to be financed by the Bank. Nevertheless, the World Bank requested additional information on the parameters used in the assessment of works, as well as specific authorization from the Urban Development Department for some projects, as communicated in a letter dated May 24, 1984, and in other subsequent documents. 2.04 The studies showed a program of construction works on 244 roads for the period 1984-88, at a cost of Mex$19,070 million -- in constant 1984 prices -- to construct 2,790 km. 2.05 However, with the economic crisis faced by the couutry there was a progressive loss of interest in the infrastructure construction program as policies to allocate resources changed -- higher priority was given to the maintenance and rehabilitation of existing infrastructure. Only Mex$1,434.1 million at constant 1984 prices were earmarked for the project, allowing the construction of only 444.3 km of the total planned, i.e., 7.5% in financial terms, and 15.9% in physical terms of the total planned, respectively.  - 9 - 2.06 Compounding the already mentioned problems, the process undertaken during each loan phase was slow and difficult. This is obvious in the long negotiation and identification period (at least three years), contracting, loan effectiveness (nine months) and the execution period (two and a half years), during which it was possible to disburse only 4.8% from the loan. 2.07 It is important to note the above as lessons to avoid future project failures. In this regard, the Federal Government should authorize pluriannual investment programs when funded with foreign loans. It would also be desirable that the commitments to be entered into adhere more to the ways in which the project implementing agency works, or that, in any case, before entering into a commitment in which extraordinary work, are required, both parties ensure that the necessary infrastructure to adequately comply with the commitment has been implemented. 2.08 We are aware of the World Bank's intentions in optimizing planning and programming of the resources and needs of the beneficiaries of its loans. However, to undertake each program the required planning should be consistent with existing budgetary limitations. Therefore, it is necessary for the Bank to be more flexible, especially with this type of loans which have economic and social implications and whose main objective is the development of the poorer areas of the country. Their main goal, thus, is social benefit. 4.09 It is important to note that, as the requirements set by the World Bank become more consistent with the implementing agency's ability to respond, the projects will be more realistic. III. Project Description 3.01 The objective of the project was to promote the development of the State of Chiapas through the improvement of a rural roads network. The subprojects were selected on the basis of technical, economic, and social criteria, and were supplemented by various general guidelines. The parts encompassed by the project are detailed as follows: 3.02 PART A: Subprojects for: (i) the construction of approximately 1,000 km of rural roads; (ii) rehabilitation or improvement of approximately 2,000 km of rural roads; and (iii) engineering studies. PART B: Technical support to strengthen the implementing capabilities of the Secretariat of Communications and Transport and COPLADE: (i) with the objective of carrying out planning studies on transport investments, including, for instance, the development of a road master plan; and (ii) to carry out construction and maintenance works, and the preparation and updating of manuals related to rural roads construction and maintenance. 3.03 In addition, the following actions were agreed with the World Bank: (i) all road construction would be authorized by the Urban Development and Ecology Department, to avoid, as much as possible,  - 10 - ecological damages; (ii) yearly updating of the rural roads inventory and SCT's programs in order to prioritize their reconstruction and periodic maintenance; (iii) to prepare a study on passenger and cargo transport in the rural environment -- the main conclusions of which would serve as basis for implementing actions to improve rural transport; (iv) to present to the World Bank a pilot project on road sections on which bunds and surfacing be compacted, and including purchase of equipment when made by force account, and a second alternative considering the execution of this test by contract. Data obtained on these trial sections would be made available to other States in order to analyze the feasibility and cinvenience of implementation in the road programs. 3.04 However, the lack of resources made it impossible to carry out any of the above-mentioned studies. 3.05 Along these lines, and with the objective of achieving the goals of the project, it was estimated that an annual investment equivalent to US$6 million was required, exclusively for the program financed with foreign credits. This goal was never reached. Glcbal investment during the period 1984-1988 averaged US$2.21 million per year, which obviously was not enough to reach the goals set in the project. 3.06 Described below are the estimated investments and goals, according to the Appraisal Report, for the program financed with Loan 2525-ME' TYPE OF INVESTMENT ESTI4ATED INVESTMENTS PROGRAM 1984 1985 1986 1987 1988 1989 TOTAL Rural Roads Development Cost (Mex$ million) 129 464 1242 1242 1025 1025 5127 Goal (km) (40) (80) (240) (240) (200) (200) (1000) Rural Road Rehabilitation Cost (Mex$ million) 42 168 242 305 305 348 1410 Goal (km) (50) (200) (350) (450) (450) (500) (2000) Consultants Technical Assistance and Price (Mex$ million) Contingencies Services 42 144 287 431 491 615 2010 TOTAL MEX$ (millions) 213 776 1771 1978 1821 1888 8547 US$ (millions) 1/0 3.8 8.6 9.7 8.9 9.7 41.7 3.07 According to the table above, the Mexican Government should have invested, up to 1988, the equivalent of US$32 million in the externally funded program. However, only US$11.8 million was allocated for the rural road program as a whole. 3.08 From the appraisal expected total project cost, i.e., US$41.7 million, the World Bank would finance the equivalent of US$22 million, approximately 50% of the cost. The following disbursements were estimated.  v 11 - ESTIMATED DISBURSEMENTS Year 1986 1987 1988 1989 1990 1991 TOTAL Disbursement (US$Million) 2.0 3.9 5.1 4.8 3.8 2.4 22.0 3.09 According to the schedule, US$11.0 million should have been disbursed by 1988. However, only US$1.06 million was reached, i.e., 9.6% of the estimated disbursements. IV. Project ImRlementation 4.01 The Loan and Guarantee Agreements for the Project were signed on July 16, 1985, and the Loan became effective on April 4, 1986 when the Coordination Agreement between the Secretariat of Budgeting and Programming, the Secretariat of Agriculture and Hydraulic Resources, the Secretariat of Communicatiouis and Transport, and the State Government of Chiapas was signed. 4.02 The amount contracted with the Bank to finance this project reached US$22.0 million, allocated as follows: US$16.5 million to finance 50% of the cost of civil works, US$1.5 million to finance 100% of the cost of technical assistance and consulting services, and US$4.0 million for contingencies. 4.03 During project expcution there were different problems, notably the budgetary shortfall, which affected the attainment of the estimated goals. Additionally, other operational problems that had a negative impact on adequate loan execution were, among others: (a) Inadequate coordination among those involved during project execution. (b) Confusion as to the requirements for making loan disbursements. (c) Lack of training by the project executors, especially on the adequate presentation of documentation to review project eligibility and verification of expenses. (d) Lack of coordination with the Urban Development and Ecology Department to obtain the authorization requested by the Bank in the case of rural roads in areas where the environment was going to suffer an adverse impact. 4.04 It is clear that the main obstacle for using the loan was the low budget. Thus, the Government's decision to cancel part of the loan was considered right. That cancellation was done in two stages: the first  - 12 - (US$10 million dollars) on July 27, 1987. A series of actions were proposed at this first stage, with a view to optimizing the use of the remainder of the loan and are described below: (a) To increase the Bank's financing share from 50% to 70% or 80%. (b) To include works with higher design standards of construction than those eligible for financing under thec loan, as well as high-cost roads executed in the State of Chiapas. (c) To include construction and rehabilitation of rural roads in other States of the Country. 4.05 Bank missions opposed these proposals, especially that of section "c", because they felt that it was not within the objectives of the project. 4.06 Thus, facing the imminent impossibility of allocating enough resources to the individual project in the short term, a decision was made to disburse the maximum amount possible, using the existing documentation for eligible expenditures incurred up to that moment, and to cancel the remainder that could not be disbursed. In this sense, only US$1.06 million was used, approximately 50% of what could have been financed. 4.07 The above implies that the original project did iLot consider the possibility that the project's implementing agency would have enough resources available. Also, approval was granted to a project for which there was still the need to comply with many information requirements, as was made known in a letter dated May 21, 1984, and subsequent documents; for instance, SEDUE's authorization for construction works, the presentation of adequate economic evaluations, and confirmation on the economic potential of the project's area of influence. 4.08 As to the execution of the loan, in the presentation (March-April 1987) of the initially agreed documentation to disburse funds, which consisted of preparing SOE's endorsed by the Chief of the SCT Center in Chiapas, certain inconsistencies were found in the correspondence received by the financial agent such as double reporting of expenses, presentation of expenses for projects that were not accepted to be financed, and amounts that included the Value Added Tax. As a consequence, it was agreed to suspend withdrawals from the Special Account until that information was clarified. 4.9 In view of the situation, implementation of new mechanisms, with the intention of defining and supporting the report of incurred expenses was attempted. One of these mechanisms was to coordinate among the agencies involved, but it did not reach the desired objectives. 4.10 As a second option it was decided to compile all the supporting documentation of the expenses report, under the responsibility of the executing agency and the SCT-Chiapas Center. Once the documentation was compiled, problems still persisted -- there were some cases of incomplete documentation and, in the judgment of the financial agent, did not meet the minimur requirements to be eligible for financing.  - 13 - 4.11 Finally, a decision was made to form a working group consisting of IBRD, SCT's Directorate General of Planning, SCT's Chiapas Center and NAFIN, which travelled to the project area. There, the available information was analyzed and clarified, and a determination of what could be funded by the Bank was made. 4.12 Given the latter experience, it is desirable that for future loans, enough time and care should be exercised to svoid these types of problems, which apparently are of no great importance, but that without adequate training and minimum care, can evolve into serious situations. By the same token, it is important to consider the possibility of implementing periodic training seminars, at least up to the implementation of each specific project, which are aimed at direct project executors, so that each loan's specific guidelines and key points are known and can be given greater emphasis. 4.13 External loans should be executed with greater efficiency. To that end, periodic coordination among the institutions involved should be increased and the need to implement an adequate structure to handle foreign credit in the executing institutions should be analyzed. V. Economic Reevaluation 5.01 The economic analysis of a road project is focused on quantifying its inherent benefits and costs. Traditionally, such quantification was related to monetary savings, brought by the project to the user in the form of reduced operational and travel time. However, that type of emphasis is inadequate to analyze rural roads on which traffic volumes are very low. Therefore, the analysis should be based on the impact each road has on its area of influence, especially in promoting agricultural production development thus increasing the level of income of the population directly involved with it. 5.02 The road's area of influence is defined as a function of its localization, as follows: (a) From 3 to 5 km on each side of the road axis, on flat terrain. (b) From 2 to 4 km on each side of the road axis, on hilly terrain. (c) From 1 to 3 km on each side of the road axis, on mountainous terrain. 5.03 The economic evaluation of each project is based mainly on its impact on agricultural production, considering that it is the most sensitive sector and easier to quantify. The methodology employed encompasses three variations, as follows: 1. First year benefit. If it was higher than or equal to 12%, the project was considered to be profitable. If not, the second method, as follows, was used.  - 14 - 2. Net present value. Under this method, incremental agricultural output was considered for the 10 years of the project's useful life. If the result was negative, the third method was used. 3. Cost per inhabitant served - Total Road Cost less US$450 Number of inhabitants than served 5.04 The evaluation was done using a sample encompassing 72% of the executed projects. The result was that the projects were economically viable, according to the method first defined. Values obtained demonstrate, for the most part, that the projects were executed with delays. VI. Conclusions and Recommendations 6.01 It is evident that, for both the Mexican Government and the World Bank, the lessons learned from this Loan have highlighted several issues that will have to be addressed soon, to improve the performance of World Bank loans. 6.02 As a result, the following findings and recommendations, considered to be the most important are presented: 1. Even when it was impossible to complete the project according to the initial estimations, it remained fully justified, as demonstrated by the results of the economic reevaluation. 2. The main problem hampering project execution was lack of resources due to the economic crisis affecting the country during the 1984-1988 period. 3. For future loans, it is recommended that the Government makes sure to have enough national resources available on a timely basis. To that end, prioritized Pluriannual Investment Programs should be agreed upon among participating State Secretariats with a view to guaranteeing the timely disbursement of external loans, thus strengthening the inflow of hard currency into the country. 4. It is necessary to provide training for staff in charge of external loans, especially in those agencies initiating the execution of a loan. Periodic seminars should be carried out to provide established guidelines for the execution of external loans. 5. Projects with external financing should be provided with timely local funds, so that the Institution having the responsibility of implementing them will be able to comply with the established minimum periods for bids and would greatly improve the programming of civil works.  - 15 - PART II STATISTICAL INFORMATION 1. Related Bank Loans YEAR OF AMOUNT NUMBER TITLE PURPOSE APPROVAL US$ MILLION 2526-ME Chiapas Agricultural Agricultural 1985 58.0 A/ Development Project Development a/ Original Loan Amount was US$90.0 million - US$32.0 million were cancelled. 2. Project Timetable Date Date Item Planne Actual Identification 1/18/83 1/18/83 Preparation 3/83 3/83 Appraisal 2/84 2/84 Negotiations 10/84 2/25/1985 - 3/20/1985 Board Approval 12/84 4/30/85 Loan Agreement Date 3/84 7/16/85 Effectiveness Date 10/16/85 4/4/86 Closing Date 6/30/91 6/30/91 3. Loan Disbursements Cumulative Estimated and Actual Disbursements (US$million) F X86 PX8Z FY88 EY FY29 FY91 Estimated 2.0 5.9 11.0 15.8 19.6 22.0 Actual 2.0 -/ 2.0 S/ 2.0 A/ 1.06 h/ 1.06 1.06 Actual/Estimated (%) 100 34 18 7 5 5 a/ Initial Disbursement to the Special Account. k/ US$.94 million were reimbursed to the Bank.  - 16 - 4. Project Implementation Appraisal Indicators Estimate Actual Rural Roads Construction (km) 1000 443 Rural Road Improvement (km) 2000 1000 5. Proiect Costs ani Financing (US$ million) A. Proiect Costs Appraisal Item Estimatea Acrual Rural Road Construction 32.0 4.94 Rural Road Improvement 8.2 2.12 Consultant Services 1.5 0.10 Total 41.7 7.16 B. Proiect Financing Source Planned Actual (US$'000) (US$'000) IBRD 22.0 1.06 Domestic 19.76 Total 41.7 7.16  17 - 6. Project Results Economic Impact Ex-Ante and Ex-Post Economic Evaluations EX-ANTE EVAL. EX-POST EVAL. ESTIMATED ERR FIRST YEAR &STIMATED PRO