Senegal - Third Railway Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 6822 PROJECT COMPLETION REPORT REPUBLIC OF SENEGAL * THIRD RAILWAY PROJECT (LOAN 1518-SE) June 4, 1987 Western Africa Region Projectt Department Transportation Division 1 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 EXCHANGE RATES Name of Currency: CFA Franc (CFAF) Year: Appraisal Year Average: 1977 Exchange Rate: US$1 - 235 Intervening Years Average: 1978-1984 US$1 = 295 Current Year Average: 1985 US$1 - 449  FOR OFVICAL US ONLY THE WORLD BANK Wdshington. DC 20433 USA Offce i Du.ttni.GAWq Ope*atml Et. wgt,n June 4, 1987 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Senegal Third Railway Project (Loan 1518-SE) Attached, for information, is a copy of a report entitled "Project Completion Report on Senegal Third Railway Project (Loan 1518-SE)" prepared by the Western Africa Regional Office. Further evaluation of this project by the Operations Evaluation Department has not been made. Attachment This document has a restricted distribution and may bf used by recipients only in the pertormance of their official duties. Its contents may not otherwise be disclosed without World Bank authoriation.  REPUBLIC OF SENEGAL FOR o ciAL USR oNLY THIRD RAILWAY PROJECT, LOAN 1518-SE PROJECT COMPLETION REPORT Table of Contents Page No. Basic Data Sheet ............................ i Highlights ..... .o.... .o.o.... o....... o...o......00...........0. iv I* INTRODUCTICNo ................................o ......... 1 II. PROJECT IDENTIFICATION, PREPARATION AND APPRAISAL .......................................... 1 III. PROJECT IMPLEMENTATION AND COST ..........o ..........o 5 IV. TRAFFIC AND OPERATIONS ....... o......... o.... .... 12 V. FINANCIAL PERFORMANCE. . . o . . . .. o-.ooo. .oo. o o. oo 17 VI. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT.......... 22 VII. ECONOMIC REEVALUATION... ..T.. 25 VIII. ROLE OF THE BANK.................o.....oo.o.......... 28 IX. CONCLUSIONS..... o..o.....oo 31 TABLES 3.1 Actual and Expected Physical Completiono ............. 33 3.2 Actual and Appraisal Estimates of Project Costs...o.... 34 4.1 Actual and Appraisal Forecast of Passenger and Freight Traffic, 1977-1985....................... 35 4.2 Actual Efficiency Indicators vs. Operational Targets.. 36 4.3 Selected Operating Statistics, 1977-1985.............. 37 5.1 Appraisal Forecasts and Actual Income Statement of RCFS, 1977-85. .. . *. . .. .. ... ......... * 38 5.2 Appraisal Forecasts and Actual Balance Sheets of RCFS, 1977-85 .o.. .. . . . . . . . . . . .. . .. . 39 7.1 Economic Reevaluation: Track Rehebilitation -- Dakar-Thieso................ 40 7.2 Economic Reevaluation: Track Maintenance............. 41 7.3 Economic Reevaluation: Dakar Marshalling Yard........ 42 7.4 Economic Reevaluation: Spare Parts and Workshop Equipment...... ....... ....................o.. 43 ANNEX I Compliance with LoanCovenants...................... 45 ANNEX 2 Borrower Comments ..... ........ . .. ..... ............... ............. 49 GRAPH Taux de Disponibilite des Locomotives................. 51 MAP IBRD 13063R -- Senegal, Third Railway Project 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.  REPUBLIC OF SENEGAL THIRD RAILWAY PROJECT, LOAN 1518-SE PROJECT COMPLETION REPORT PREFACE This Project Completion Report (PCR) has been prepared for the Third Railway Project for which Loan 1518-SE, in the amount of US$11 million, was signed on March 17, 1978. The Board approved the project on February 7, 1978. The closing date was extended from December 31. 1981, to Jine 28, 1985. The Loan was disbursed up to a total of US$10.71 million, and the balance of US$0.29 million was cancelled. The PCR was prepa- ed by WAPT1 and is based on Appraisal Report No. 1691a-SE dated January 23, 1978, supervision reports, legal docu- ments, project correspondence files, and consultant reports. Following a review of this PCR, the Government of Senegal (GOS) wanted the post-PCR developments -- primarily the agreements reached between the GOS and the IBRD -- integrated into this PCR. Accordingly, a separate Annex (Annex 2) entitled "Borrower Comments" is now attached. In accordance with the revised procedures for project perfor- mance audit reporting, this Project Completion Report was read by the Operations Evaluation Department (OED), but the project was not audited by OED staff.  - Li - REPUBLIC OF SENEGAL THIRD RAILWAY PROJECT. LOAN 1518-SE PROJECT COMPLEIION REPORT BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Item Espectation current Estimte Total Project Cost (USS million) 14.26 U."6 Over/Underrun (S) (CFAF million) - 20 Loan/Credit Amount (US$ million) - 21.0 Disbursed 10.71 Cancelled 0 Q.29 Repaid - 1.83 Outstanding - 8.88 Date Physical Components Completed June 30, 1981 December 31, 1985 Proportion Completed by above date (%) 100 50 Proportion of Time Underrun or Overrun (M) 100 Economic Rate of Return (%) 22 13.5 Financial Performance Fair Poor Institutional Performance Fair Poor CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENlS (US$ Thousand) FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY86 Appraisal Estimate 5,965 9,120 10,360 11,000 11,000 31,000 11,000 11,000 Actual 822 4,500 6,830 8,455 9,800 10,050 10,290 10,710 Actual or % of Estimate 14 49 6s 83 89 91 93 97 OTHER PROJECT DATA Actual or Item Original Amended Current Eatliate First Mention in Files or Timetable 01/09/75 Government's Application 11/19/76 Negotiations 11/77 12/12/77 Board Approval 12/77 02/17/78 Loan/Credit Agreement Date 02/15/78 03/1h7/78 Effectiveness Date 07/19/78 08/17/78 Closing Date 12/31/81 12/31/82 06/28/85 Borrower Regie des Chemins de Fer du S6negal (CFS) Executing Agency Chemins de Fer du Sndgal (CFS) Ficcal Year of Borrower July 1 - June 30  - iii - MISSION DATA No.of No. of Date of Item Mo. lYear Weeks Persons Manweeks Identification - - - - - Preparation 02/75 1 3 3 03/75 Preappraisal 06/76 / 1.5 5 7.5 07/76 AppraisAl 02/77 % 3 4 12 01/78 Total 55 22.5 Supervision 1 03/78 1.5 2 3 05/78 Supervision 2 07/78 1 3 3 11/78 Supervision 3 11/78 1.5 3 4.S 02/79 Supervision 4 06/79 1 5 5 08/79 Supervision S 12/79 2 3 6 01/80 h Supervision 6 05/80 2 2 4 07/80 Supervision. 7 U/80 1.5 3 4.5 02/81 Supervision 8 03/81 0.5 2 1 04/81 Supervision 9 05/81 2 2 4 -06/81 Supervision 10 U/81 1.5 3 4.5 12/81 Supervision U 06/82 1 3 3 07/82 Supervision 12 12/82 1 1 1 01/83 Supervision 13 03/84 1 1 1 04/84 Supervision 14 1/84 0.5 1 0.5 12/84 Supervision 15 05/85 0.5 1 1 06/85 18.5 46 /a Called preparation mission, but in effect a preappraisal mission. k lack to Office Report only, no supervision report in files. August U, 1986  - iv - REPUBLIC OF SENEGAL THIRD RAILWAY PROJECT, LOAN 1518-SE PROJECT COMPLETION REPORT HIGHLIGHTS i. The objectives of the project were (a) to strengthen the Regie des Chemins de Fer du Senegal (CFS) as an institution, so that opera- tional efficiency could improve; and (b) to improve the cooperation between the Senegal and Mali Railways, so that more international traffic could be attracted to the railway. Within one year of Loan signature, the project was revised and the track renewal work beyond Thies was replaced by track rehabilitation between Dakar and Thies, the most heavily trafficked section, which was considered essential and urgent (3.04). Revision of other components included (a) the addition of auditing of CFS accounts to consultant activities; (b) a reduction in the training component because the emphasis on on-the-job training and management training eliminated the need for the training school building Dlanned under the project (3.11); and (c) a slight reduction in the ,.ocomotive rehabilitation program as some locomotives were damaged in accidents and were scrapped during project implcmentation (3.09 and 3.10). ii. Overall progress of the project was poor: there was more than a four-year delay in implementation, mainly because local financing was not available in time for the track rehabilitation component (3.05 and 3.06); there were very few improvements in the railway's operational efficiency. In addition, the financial situation of CFS deteriorated (5.02 to 5.04). However, substantial gains were made in such important areas as improved coordination between the Senegal and Mali railways (8.06 and 8.07), and the demonstration of the unprofitability of passenger services. Besides, an exhaustive exercise in the conversion of CFS into a Societe Nationale with a well-defined Contrat Plan was carried out but it was not implemented by the Government, as it had no solution to the over-staffing problem. But the exercise served to demonstrate to the Government the need for improved cost recovery from CFS operations through cost reduction, notably staff costs. As a result of teis sensitization, CFS has gradually reduced its staff numbers (6.02). iii. There were some limited improvements in the operational efficiency of CFS during the last three years of the project, namely, increased locomotive availability, increased locomotive-km per year, wagon turnaround time, and staff efficiency. The improved locomotive availability was partly due to the substitution of seven new locomotives during project implementation in lieu of older ones which had to be scrapped or were damaged in accidents (4.05 and 4.06).  - v - iv. The reestimated rate of return for the project is 13.5%, as compared to 22% at appraisal(7.10). The appraisal estimate includes track renewal which was subsequently dropped from the project in favor of the more urgent rehabilitation of the track between Dakar and Thies which carries practically all of CFS's traffic. v. The following points are of special interest: -- competition for international traffic to Mali has in- creased from other transport routes, which calls for maintaining the coordination between the Senegal and Mali railways at least the level gained during the project, if existing traffic is to be retained and possibly increased (4.01 to 4.03); -- progress in such areas as improved coordination of the two railways was very slow despite promises of Government action on these issues embodied in covenants in the Loan Agreement; and conversion of CFS into a Societe Nationale never came to pass (4.07 and 6.09); -- .CFS made very limited improvements in operational effi- ciency, primarily because management lacked the will and the freedom to steer clear of the political implications of key issues like staff costs, staff discipline and better operational understanding with Mali Railways (6.06 and 6.07); -- The Bank played a key role in improving coordination between the two railways and in dissuading the Government from some major misinvestments. Recently, however, the Government has been negotiating with Donors the financing of new investments which the Bank considers uneconomic. The Bank is discussing these proposals with the Government and Donors in order to forestall any misinvestment. It is evident that the Bank has an important role in advising the Government on complex railway issues, and is probably the only international agency that can achieve results (8.05 to 8.07). -- some parts of the project were revised in the light of new information received within one year of loan signing; and -- the Study on the Future of the Senegal Railways, complet- ed irn April 1986, should li come the basis for restructur- ing the railway's management, finances and traffic; including abandoning the unprofitable national services and focussing on the international service, particularly freight.  - vi - Postcriit: Government's comment on the PCR: The Government's covment on this completion Report (Annex 2) consists basically of a request to reflect the developments in the discussion between the Government and the Bank. As these developments are more of a policy nature rather than of a narrow project focus, it was considered expedient to include the agreements themselves as attachements to Annex 2 and make a brief reference to them here. In brief the agreements covered: (i) limitation of new investments to an agreed investment plan, notably, choosing intensive maintenance of the main line track to Mali in preference to complete renewal and thus reducing the cost of this component from US$88.0m to US$36.0m. (il) decision to finalize and implement a Contract plan for RCFS by January 1988; (iii) a rigorous review of the viability of the Tivavouane-St. Louis section with an 4treement in principle to close down this branch line, if the review did not establish its viability.  REPUBLIC OF SENEGAL THIRD RAILWAY PROJECT, LOAN 1518-SE PROJEC1 COMPLETION REPORT I. INTRODUCTION 1.01 The Senegal railway system consists of a main line from Dakar to the Mali border, a major branch connecting Dakar with the port of St. Louis, and other minor branches. When the Third Railway Project was arpraised, the Regie des Chemins de Fer du Senegal (CFS) carried most of the country's phosphate production, part of the groundrut production and Mali international traffic. The Senegal-Mali railway was less costly (financially and economically) than the alternative C^te d'Ivoire route for most of Mali's imports and exports but, despite the cost advantage, traffic to and from Mali only grew by 48,000 tons between 1972-1975 on the Senegal railway route, while increasing by 156,000 tons during the same period on the C8te d'Tvoire route. 1.02 The purpose of the Bank's first two railway projects in Senegal (Credit 96-SE, US$9.0 million, 1966 and Credit 314-SE, US$3.2 million/Loan 835-SE, US$6.4 million, 1972) was to improve the railway's carrying capacity and level of service so that it would offer a competitive service for Mali international traffic. While the two projects (together with parallel projects in Mali) helped to prevent a collapse of operations on the Mali/Senegal railway system, Implementa- tion was disappointing. Both projects were several years behind sched- ule: deterioration of the railway was not fully arrested and operating performance either declined or remained at the same low level. Lack of coordination of operations between the two railways was also a serious obstacle to improving performance. Consequently, the Third Railway Project concentrated on improving operational performance, traffic and international cooperation, giving emphasis to institution building, technical assistance and training programs and a structural and organi- zational study for CFS. In order to achieve an improvement of Senegal-Mali railway coordination, the Bank prepared a third project for the Regie des Chemilns de Fer du Mali (CFM) to be implemented at the same time as the Third Railway Project in Senegal. II. PROJECT IDENTIFICATION, PREPARATION AND APPRAISAL Identification and Preparation 2.01 During supervision of the Second Railway Project, the need for a follow-on project was identified. Following project preparation missions in 1975, the Bank expressed concern about the future role of  -2- the railway because of (a) the decline in groundnut and other domestic traffic; (b) the decline in international traffic to 1970 levels; and (c) very little growth of phosphate traffic because of capacity con- straints. The decline in traffic was largely because of the railway's poor operational performance, vhich Bank missions attributed to the lack of competence and authority of management, the lack of motivation of the staff, poor utilization of technical assistance, lack of management-union cooperation, and an inability to strike a rapport at operating levels with Mali Railways An additional problem at the preparation stage of the project was the Government's proposal to transport phosphate over a new railway line to be built from the phosphate mines to the port of Cayar, thereby seriously threatening the financial viability of CFS (this project was later dropped for lack of external funding). Given its concern, the Bank recommended that appraisal of the project be postponed until traffic increased, CPS improved its efficiency and the Government reached a decision on the transport of phosphate. However, an increasing need to move the appraisal of this project upstream was felt not too long after appraisal of the Third Railway Project in Mali (in July 1976), since the two projects were regarded as complementary and essential for the coordinated movement of Mali traffic. 2.02 The Bank decided to place emphasis on the technical assistance and training components because of CFS's staffing and managerial prob- lems, and to keep the hardware components (equipment and civil works) to the minimum necessary. The manpover and training needs of the railway were identified during preparation missions in 1976, and a training program was drawn up by the Bank and the Government for implementation during the project. 2.03 In order to put pressure on the Government to strengthen CPS as an institution, the Bank decided to make appraisal contingent on a set of measures aimed at reorganizing and strengthening CFS management. To make the railway more effective as an institution, the first condi- tion was the completion of the reorganization plan for CFS by the Senegalese Bureau d'Organisation et Methodes (BOX) and a start of imple- mentation of the recommendations. The second condition required the Government to issue decrees reestablishing the authority of the General Manager and giving him the required managerial freedom, since in the past, too many decisions which should have been made by CFS were made by Government. tiie other two conditions covered the appointment of two assistant General Managers, one expatriate for technical a 'tters and a local for administration and finance: the Bank considered that new and more experienced management would help to overcome the managerial weaknesses of the organization. The Government complied with all the above conditions by February 1977, except for approval and implementa- tion of the new organigramme. In order not to delay appraisal, the Bank made this a condition of negotiations.  -3- Project Appraisal 2.04 The project was appraised in February 1977 with the Bank proposing to finance 100% of foreign costs and 68% of local costs. However, by Board Presentation the estimated cost of locomotive spares had nearly doubled together with smaller increases in other foreign costs, so that the remaining loan funds could only cover 35% of local costs. This decrease in local funding later proved to be a significant obstacle to project implementation (para 3.05). Software components (training and technical assistance) finally cDmprised about 20% of project costs (Table 3.2), as opposed to thewA0% originally envisaged, because it was considered inopportune to spend on a new training center building, which, in any case, was found unnecessary, while funds were required to meet the higher cost of locomotive spare parts necessary for improving operations. In addition, the allocation for technical assist- ance was reduced, as the Government preferred the technical assistance financed from bilateral aid, although eventually it proved to be a mixed blessing (para 6.05). 2.05 Most of the issues raised at appraisal and post-appraisal (July 1977) centered around the institutional and financial problems of CPS: measures to resolve some of those issues eventually became condi- tions of negotiation. These conditions included: (a) the promulgation and implementation of the new organigramme for CFS; (b) the filling of 12 key positions at the department director and assistant levels; (c) preparation of a training program fir the gradual replacemer.t of expatriates by Senegalese engineers and technicians; (d) restructuring of CFS's balance sheet mainly through the Government waiving about CFAF 2 billion of debt; (e) preparation of transport plans for major commodities and passenger services; and (f) agreement with the Taiba Phosphate Company to improve terminal operations at the mine and port of Dakar. All of these conditions were met by the Government and negotiations were held in December 1977. 2.06 Other issues for which acceptable plans were agreed at negoti- ations were: (a) the control of staff growth; (b) appointment of expatriates to positions for. which qualified nationals were not available;  -4 - (c) agreement between Government and the railway union to cooper- ate on the CFS reorganization scheme; (d) improved cooperation between CFS and CFM; and (e) improvement of track renewed (but poorly laid) under the Second Railway Project to official CFS track-maintenance standards. As a result of Bank pressure during appraisal, the Malian and Senegalese Governments agreed to suspend a cumbersome international railway transit agreement which had been disrupting traffic between the two countries. The appraisal mission also recomended selective tariff increases which were implemented in May 1977. 2.07 Conditions of Board Presentation. In order to make the CFS focus on the problems of low locomotive availability, the Bank set as conditions of Board Presentation: (a) the preparation of locomotive maintenance programs, including spare part requirements; (b) the prepa- ration of daily utilization plans for each locomotive; and (c) the setting up of a wagon rotation information system. All of these condi- tions were met before Board Presentation. *2.08 The uncooperative attitude of the railway union had long been a major obstacle to staff motivation and productivity. Therefore, another condition of Board Presentation was a written agreement between. the Government and the trade union, promising full staff cooperation for smooth implementation of the railway's reorganization and improvement in staff productivity. However, because of an illegal strike in June 1977 and subsequent dismissal of union officials, Government and union talks did nuc resume until October 1977. Consequently, the Pank dropped the requirement for a written agreement after the Government issued a decree in December 1977 allowing productivity-linked bonuses for public corpo- ration staff. Later events (paras 5.03, 6.07) proved the continuing poor relations between the union and CFS to be a major problem. Proiect Description 2.09 The primary objective of the project was to improve the efficiency of CFS by rehabilitating existing infrastructure and strength- ening CFSts operational and managerial effectiveness. Physical compo- nents of the project (Table 3.1) included: (a) track renewal (60 km); (b) track maintenance equipment and material (switches, gang cars and other equipment); (c) marshalling yard and telecommunication improvements in the Dakar area; and  (d) spare parts for motive power and ro. ing stock and workshop equipment. Software and institution-building components included: (a) training, including construction of a training school, train- ing equipment and materials, technical assistance and training abroad; and (b) consulting services for studies on groundnut transport and the ownership, structure and organization of CFS. 2.10 Major covenants included in the legal documents are discussed in Chapter Vl and in Annex 1. III. PROJECT IMPLEMENTATION AND COST Effectiveness and Start-up 3.01 Loan effectiveness was delayed by four months due to the Government's cumbersome administrative procedures, slowing down submis- sion of the required standard documentation to make the loan effective. There were no specific conditions of effectiveness. Preparation of bidding documents was affected by the delay in effectiveness. Revisions 3.02 There were four revisions to the legal documents for the project. The first was in November 1979, when track rehabilitation of the Dakar-Thies line was included in the project instead of renewal of track on the international line (para 3.04). The second was in May 1980, when the financing of consultants to audit the railway's accounts was added to the project (para 3.12). The third was in June 1983, when the Bank agreed to finance 1002 of the ballast costs instead of 35% of total track rehabilitation costs (para 3.06). The fourth was in March 1985 when CFS was allowed to purchase ballast on the spot market (para 3.06). Implementation of the Project 3.03 Completion of the project was delayed by more than four years, not only because of delays in loan effectiveness, but also becausQ of slow preparation of bidding documents, delays in the supply of equipment, lack of local funds to carry out track works, repeated failures of ballast contractors, the railway's weak management, and Government's delay over the structure and ownership study. At one time, implementation of the project and compliance with loan covenants were so poor that the Bank seriously considered suspending disbursements. In addition, the impact of the project on railway operations was below appraisal expectations during the early stages of the project, but during the last two years, operations improved significantly.  -6 - 3.04 Track Rehabilitation. This component included the renewal of 60 km of track between km 271 and the Mali border (to Kounghuel). However, immediately after loan signing, the Bank inspected the finished track renewal works financed under the First and Second Railway Pro- jects, since improvements of the renewed track had been a condition of disbursement for the track renewal component. An alarming number of defects were found, particularly on the heavily-trafficked Dakar-Thies section, apparently due to the use of soft limestone for ballast. A consultant, financed by French bilateral technical assistance, quickly completed a comprehensive tfack rehabilitation study by May 1978 and found the Dakar-Thies section to be of the highest priority. Therefore, the project was subsequently amended to include rehabilitation of the 70 km of double track between Dakar and Thies and additional equipment for the track maintenance component instead of the original track renewal component. Increased costs were financed out of unallocated funds and savings from other categories. 3.05 Track works were scheduled to start in July 1979, but most of the equipment was not received until May 1980 because of delays in bidding and supplies: ballast wagons were delivered even later (para 3.16). Track works did not start until early-1980 but progressed qulte well until October 1980 when, with only 25X of works completed, operations were halted because of a shortage of local funds. There were intermittent works between December 1981 and October 1982, but supervi- sion missions found that inferior ballast was once again being accepted by CFS and imposed stringent quality control measures as a condition for further disbursements. As a last resort, in mid-1982, the Bank proposed to provide Government with the necessary local funds for the remainder of the track rehabilitation works through the Structural Adjustment Loan if the Government approved the Contrat Plan (para 6.09). This would have effectively released the Government from its obligations under Sec- tion 2.02(a) and (e) of the Guarantee Agreement (Annex 1), but this proposal was not implemented, as the Government agreed to make a special effort to find the requisite counterpart funds. The major reason for the shortage of local funds was the increase in CPS salaries, which amounted to 66% of total operating expenses by 1982 (para 6.07), and the general financial difficulties of the Government. 3.06 Track works were to resume in early 1983, but did not start because of CFS's inability to find reliable contractors to supply ballast due to the railway's reputation for delaying payments. Conse- quently, in June 1983, the Bank agreed to provide 100% financing for ballast works as the only recourse available to attract reliable con- tractors on the condition that CFS meet all other track rehabilitation costs. The work was to have been completed by June 1984, but CFS only signed the contract for ballast in March 1984 because of the lengthy and difficult procurement process. However, the new ballast contract also failed in mid-1984 because of the supplier's impending bankruptcy. In order to obtain an immediate supply of ballast, the Bank advised CFS to alter its procurement procedures and purchase from the spot market, where small suppliers offered.ballast at prices equal to or even less than the contract price. The Government and CFS agreed and the loan  7- document was revised 4ccordingly. The track rehabilitation works were finally completed in December 1985. However, unless measures are taken to prevent trespassing by pedestrians who disturb the ballast bed and to remove blown sand and garbage dumped by city dwellers, the tracks in the Dakar area will quickly deteriorate again. As with track renewal works under the Second Railway Project, the quality of ballast used for track rehabilitation proved to be a serious problem. In October 1980, the Bank became so concerned that test certificates certifying good quality of ballast were required for disbursement against ballast works. CFS wanted to purchase low quality ballast on the plea that ballast of good quality was four to five times more expensive. But, later on, the spot market disproved this contention. CPS found its hands tied in choosing good ballast contractors and consequently several contracts failed. Eventually, when management freedom was granted, reliable contractors came forward to supply good quality ballast at reasonable prices, provided they were assured of prompt payment. 3.07 Track Maintenance. Delivery of equipment for track mainte- nance was delayed until early 1980, when a systematic track maintenance program was begun. CFS concentrated on the maintenance of 130 km of international track from km 271 to the Halian border. given the extreme- ly poor condition of the line. Work on this section progressed satis- factorily dutring 1980-1982, using tamping and other equipment purchased under the project and track material purchased under the Second Railway Project. At the Bank's insistence, supervision by technical assistance was increased and the expatriate experts were used more effectively in the field. Consequently, derailments were reduced by over half in 1984. 3.08 Improvements in the Dakar Area. Bid documents for equipment for the improvement of track and telecommunications in the Dakar area were combined with equipment for track work components to take advantage of economies of scale, and, as a result, the bidding process took four more months. Bidding for other civil works (e.g., repair workshop in Bel-Air) was 17 months behind schedule because CFS and its consultants were very slow in preparing the design and bidding documents. There- fore, those civil works did not start until November 1980 instead of August 1978 as originally scheduled. Construction of a repair workshop at the marshalling yard was satisfactorily completed by contractors by December 1982, while track work undertaken by force account was not completed until early 1984 because of problems with ballast supply. 3.09 Spare Parts and Workshop Equipment. CFS did not carry out advance preparation for bidding of spare part supplies as expected at appraisal, due to the technical and managerial weaknesses of the rail- way. Consequently, bids were called eight months later than scheduled. However, all the spares, except the critical ones, were delivered within appraisal estimates (June 1981), while the critical spare parts were delivered about 18 months late. As a result, the locomotive rehabilita- tion program could not begin until December 1980, over two years behind schedule. The program was satisfactorily completed by December 1982, eighteen months behind schedule. At appraisal it was estimated that 28 out of 36 CFS main line locomotives would be rehabilitated under the  -8- project, but only 14 were actually rehabilitated. This reduced program was because (a) three locomotives were lost in accidents; (b) seven locomotives were scrapped because they were all over 20 years of age and had deteriorated faster than anticipated; and (c) the four CC 2400 locomotives, which were not suitable for the local conditions in Sene- gal, 1/ proved too problematical and expensive to be completely rehabil- itated. Only a few spare parts were provided for these locomotives (see paras 4.09 and 4.10 for a detailed analysis of this component of the project). 3.10 Of the six BB llOOs and five BB 12009 rehabilitated, one BB 1200 has either been cannibalized or scrapped and one BE 1100 had been awaiting repairs for 16 months as of the end of FY85 after an accident in May 1984. Three BB 500s were repaired, but due to their advanced age of over 25 years, certain parts, which had not been covered under the rehabilitation, were found to be excessively worn. These locomotives were deemed to be not reliable enough for main line opera- tion and instead of sinking more money on those old locomotives, they were transferred to shunting service. Two of these locomotives are still in shunting operations, while a third one was lost in an accident. The availability of the BP llOOs averaged 63% in FY85 compared to 65Z target and 54% in FY76 and the BB 1200s 50% in FY85 compared to 80% target and 79% in FY76. 3.11 Training. During preparation of the project,. a Bank training expert drew up training programs for CFS together with a French techni- cal assistance advisor. The Bank was concerned about the quality of the French technical assistance staff since few had actual training experi- ence. Consequently, on the advice of the Bank, an experienced pedagogic expert was assigned to head the Training School. The recruitment of this expert and his start-up delayed the commencement of rail-specific training courses. During supervision, an urgent need for mznagement training was perceived. A different consultancy group was retained for this purpose, and training programs for middle and senior managemeut officials were successfully carried out in collaboration with the Senegalese Bureau d'Organisation et Methodes (BOM). Nevertheless, this component was also delayed because the first consultants selected were rejected by the National Tender Board on the grounds.of being too expensive, and rebidding was necessary. Courses were completed in December 1981, six months behind schedule, as the program was found beneficial and, hence, enlarged during implementation. Construction of the training center was deleted from the project because on-the-job training was considered more important than classroom training. Therefore, classroom training was re-designed to consist of short courses with well-focussed objectives for which existing buildings were utilized. Some didactic equipment and books for a library were financed under the 1/ See PPAR for Senegal Second Railway Project (Credit 314-SE/Loan 835-SE).  -9 - project. Besides, the project financed scholarships for training of senior staff and executive level staff in the track training facility in Abidjan and the workshop training facility in Douala,. both of which were established and organized with the Bank's help. 3.12 Consulting Services. Given the poor state of CFS accounts and the fact that there had been no audits since 1976, the Bank agreed to revise the project in 1980 to include financing of consultants to carry out audits of CFS accounts for 1979-81. This work Vas satisfactorily carried out (para 5.11). The Groundnut Transportation Study was carried out on schedule. However, the consultants' recommendations regarding the transportation of groundnuts by the railways never had much chance of being implemented because groundnut production fell very sharply after the study. Also, the Government closed down the agency responsi- ble for distribution of groundnuts because of allegations of corruption. 3.13 There were considerable delays with the Structure anu Owner- ship Study because: (a) Government was reluctant to let consultants study the owner- ship of CFS outside the basic framework of a Societe Nation- ale, thereby. delaying preparation of TOR; (b) Government was slow to approve consultants; and (c) the study period was longer than anticipated. Phase 1 of the study considered seven alternative organizational ar- rangements and finally recommended the division of the railway into two separate companies, one for international and the other for national traffic. The latter was to be the responsibility of a Societe Nationale and this was accepted by the Government. However, the idea of a bi-national company for Mali international traffic.was not accepted, but the Government agreed that the consultants could study the possibility of an international service run by a separate unit under the Societe Nationale. So far, neither of these recommendations has been implement- ed, due to lack of political will necessary to take such a radical step in the face of a turbulent labor force. Nevertheless, these are desir- able changes to be pursued in an overall sectoral context. 3.14 The second phase of the study, completed in January 1981 (two years after scheduled completion), concentrated on the structure and organization of the Societe Nationale and the setting up of a joint rail operations service. The Bank did not agree with all of the findings of the study, particularly: (a) over-optimistic traffic forecasts which encouraged over-ambitious investment programs; (b) advice to continue and even increase the scope of passenger operations, despite overwhelming evidence that these services caused major operating losses to CFS;  - 10 - (c) recommendations that the Government assume all the long-term debt of CFS by turning it into contributed equity In the new Societe Nationale; (d) omission of a study to determine the effect of proposed tariff increases on traffic demand; and (e) weak proposals for the joint rail operations service. Phase 3 of the study was tiot carried out due to the divergence between the study recommendations, Government's own priorities, and what the Bank projected as realistic. Nevertheless* this study paved the way for the Government to treat the transport requirements of the Industries Chimiques du Senegal (ICS) project as an enclave operation and as a first step in the restructurinj of railway operations. In view of the railway's limited managerial aud technical staff resources to satisfac- torily handle the delicate ICS traffic (such as corrosive acid and other chemicals), the Government decided on creation of a new entity, Societe d'Exploitation Ferroviaire des ICS (SEFICS), for this purpose. SEFICS was also to take over transport of the short-haul phosphate traffic from CPS, thereby fulfilling one of the objectives in the reorganization of CPS, i.e., streamlining of railway operations. A second spin-off advan- tage from this study was the urge it gave to CFS to coordinate better its operations with Mali. As a result, operations between the Senegal and Mali Railways improved in 1984 and 1985 with a consequent increase in traffic (para 4.01). Reporting 3.15 Reporting requirements were included in the Loan Agreement (Schedule 6); further details on the monthly progress of each project component were requested by the Bank in mid-1979, when the project was falling way behind schedule. Adherence to the Bank's reporting require- ments has been fair, although information was sometimes incomplete and usually late. Reporting requirements included preparation of future project implementation schedules, but these invariably proved to be unrealistic and unreliable. Procurement 3.16 There were a number of procurement problems during the course of the project, including delays averaging about six to eight months in awarding contracts, which seems to be inherent in Senegal's bureaucratic procedures. CFS failed to appreciate that railway equipment usually has a long procurement lead, and, hence, needed advance planning and stream- lining of procedures. The bankruptcy of the Belgian company which had been awarded the contract for ballast wagons posed a problem: rebidding took place in December 1980 and the wagons were delivered only in Octo- ber 1982, over three years later than scheduled. Tha Government incor- rectly evaluated the second set of tenders for ballast wagons because of the way price escalation was applied, and this was rectified after bids were re-submitted without applying cost-escalation. There were also |~~~~~~~~~~~~~~~~~~~~~~ M .  - 11 - problems with both the pre-qualification of consultants for the Struc- ture and Organization and Training studies, since the Government claimed that the Bank's procedure did not comply with their regulations: however, this was eventually resolved, although the problem did cause a two month delay in approval of the Structure and Organization Study. However, the most serious and persistent problems were in the area of ballast procurement. There were continual delays and problems in awarding contracts to ballast suppliers. As noted earlier, ballast contracting was an area with a high degree of external interference. The contractors were repeatedly found supplying sub-standard material; and when it was not accepted and payments were stopped or delayed, the contractors 5topped supplying. The procedure of buying from the spot-market, where prices were reasonable and any desired quality was available, saved the track rehabilitation component from failing com- pletely. In 1983, two contracts, one for ballast and the other for maintenance equipment, were completed without the "no objection" from the Bank. However, since the bidding procedures and evaluation did not violate Bank guidelines, the Bank agreed to disburse against those items. Costs 3.17 Table 3.2 summarizes actual costs vs. appraisal estimates for project components, both in local currency and in US$ equivalent, including contingencies. Variations between comparisons made in local currency and US$ are the result of fluctuations in the exchange rates during project implementation, ranging from CFAF 245 per US$ in 1976 to 211 in 1980, to 449 in 1985. However, nearly 75% of expenditures took place before 1982 when the CFAF declined far below the appraisal rate of exchange against the US$. Overall actual project costs of US$11.46 were 20% less than the initial US$ appraisal cost estimate of US$14.26 million, and 28% below the revised US$ cost (estimated after the track works components were changed in 1978) of U