Rwanda - Fourth Highway Project

A Document of The World Bank FOR OFFICIAL USE ONLY Report No. 9503 PROJECT PERFORMANCE AUDIT REPORT RWANDA FOURTH HIGHWAY PROJECT (CREDIT 769-RW) APRIL 19, 1991 Operations Evaluation Department This docu..ent 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 EQUIVALENTS (US$ to Rwandese Francs (FRW) ) Appraisal year 1977 US$1.00 - FRW 92.84 1978 92.84 1979 92.84 1980 92.84 1981 92.84 1982 92.84 1983 98.54 Completion yearl984 104.36 ABBREVIATIONS AfDB - African Development Bank BOR - Back-to-Office Report CPS - Central Projects Staff, World Bank EAR - Eastern and Southern Africa Regional Office, World Bank EDF - European Development Fund EEC - European Economic Community ERR - Economic Rate of Return FRG - Federal Republic of Germany ICB - International Competitive Bidding IDA - International Development Association IP - Issues Paper MPWI - Ministry of Public Works and Infrastructure OED - Operations Evaluation Department, World Bank p.a. - per annum PCR - Project Completion Report PPAR - Project Performance Audit Report PR - President's Report RB - Road Branch RMEA - Resident Mission in East Africa SAR - Staff Appraisal Report SR - Supervision Report WFP - World Food Programme FISCAL YEAR OF THE BORROWER January 1 - December 31 FOR OMCIAL USE ONLY THE WORLD BANK Washington. D.C. 20433 U.S.A. oikce of Dimctor-CeaI Operations Evaluatkan April 19, 1991 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report on Rwanda Fourth Highway Project (Credit (769-RW) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Rwanda Fourth Highway Project (Credit 769-RW)" prepared by the Operations Evaluation Department. Attachment 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 OFFICIAL USE ONLY PROJECT_ERFORMANCE JDIT REPORT RWANDA FOURTH HIGHWAY PROJECT (Credit 769-RW) TABLE OF CONTENTS Pa&2 No. Preface . . . . . . . . . . . . . . . . . . . . . . £ Basic Data Sheet . ....... ......... . i11 Evaluation Summary.. . . . . . .... . iv I. INTRODUCTION . . . . . . . . . . . . . . . . . . 1 II. THE FOURTH HIGHWAY PROJECT . . . . . . . . . . . 3 A. Project Inception and Preparation . . . . . 3 B. Project Implementation . . . . . . . . . . . 6 C. Project Results . . . . . . . . . . . . . . 7 III. FINDINGS AND CONCLUSIONS ......... . . 10 Overview . . . . . . . . . . . . ... . 10 Status and prospects in 1990 . . . . . . . . 12 Summary . . . . . . . . . . . . . . . . . . 16 Annex 1: Rwanda - Summary description of the First, Second and Third Highway Projects . . . . . . . . . 18 Annex 2: Rwanda -- Notes ot demography and economics . . . . 20 Attachment: Comments from the Borrower . . . . . . 23 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. PROJECT PERFORMANCE AUDIT REPORT RWANDA FOURTH HIGHWAY PROJECT (Credit 769-RW) PREFACE 1. This is the Project Performancn Audit Report (PPAR) on the Fourth Highway Project (Credit 769-RW) of US$15.0 million equivalent, signed on April 25, 1978, and declared effective on August 4, 1978. The Closing Date was extended from July 1, 1982 to June 30, 1984 and the Credit was fully disbursed. The project was co-financed by the aid agencies of the Federal Republic of Germany (US$2.32 million equivalent), and Belgium (US$0.46 million equivalent), by the UNDP (US$1.3 million equivalent) and by the WFP (US$1.94 million equivalent). 2. The Transportation Division of the Bank's Eastern and Southern Africa Regional Office prepared a Project Completion Report (PCR) which was published on May 21, 1986. The Bank's Operations Evaluation Department (OED) had audited the First, Second and Third Highway Projects (OED Report No. 2545 of June 13, 1979) and noted that IDA assistance was primarily focussed on strengthening road maintenance, an admittedly long-term objective. The initial intention was to keep the Audit on the Fourth Project narrowly focussed. The PCR 1986 stated that this was a successful project, and OED intended to verify that benefits continue to be sustainable with respect to the two main objectives: improving major road connections (between Rwanda and its neighbors and between Kigali and the regional centers), and strengthening institutional capability for efficient road maintenance. 3. File study and discussions with IDA staff in Washington, encouraged OED to consider the project against a broader background. Rwanda's demography and resources place it among those Member Countries where sectoral improvements are intimately connected to macroeconomic, institutional, and sock1 conditions. Specifically in the case of Rwanda, continuous dnor assistance is indispensable for both the investment and the recurrent budget. Staff training and career development are challenging, and extremely long- term, tasks in a country with growing unemployment and serious political problems. Finally, Rwanda's land-locked territory requires cheap and dependable access to the sea but, despite efforts over decades, transit traffic remains expensive and undependable. Have these parameters been taken into account during inception and preparation of IDA-financed transport projects? Did IDA-Government exchanges produce agreement or disagreement on essentials such as the use of technical assistance, the importance of staff training, and the burden placed upon the recurrent budget by new capital investments? What has IDA done to assist Rwanda with better access to the sea? 4. An OD mission visited Rwanda in May 1990, discussed the project with the Borrower and representatives of cofinanc'nS agencies. inspected roads improved under the project. and gratefully acknowledges the assistance and hospitality of the Ministry of Public Works. OED prepared the PPAR by studying the full set of project files, including the Staff Appraisal Report (SAR). the President's Report (PR), the Loan Agreemen-., and the PCR, trenscripts of the Executive Directors' meeting which considered the projects, and published material on the ecoamy and the country. Some basic data from the Project Completion Report are attached for the convenience of the reader. 5. According to standard procedure, OED sent copies of the draft PPAR to the Government and cofinanciers for comments. The comments received from the Ministry of Public Works are reproduced as an Attachment to the PPAR. iii PROJECT PERFORMANCE AUDIT REPORT RWANDA FOURTH HIGHWAY PROJECT (Credit 769-RW) BASIC DATA SHEET KEY PROJECT DATA Actual or Appraisal Current %ug Expectation Estimate Total Project Costs (US$ million) 23.7 27.8 Overrun (%) 17.0 Credit Amount (US$ mi1on) 15.0 15.0 Disbursed (12/31/81) 15.0 9.9 (06/30/84) 15.0 Cofinancing-Total (US$ million) UNDP ) FRG ) 6.02 Belgium ) WFP ) Date Physical Components Completed 12/81 12/83 Proporti- completed by that date (X) Appraisal Completion Date (%) 100 70 Actual Completion date (1) 95 Economic Rate of Return 100 155 a) Improvement Program 195 147 b) Rehabilitation and Upgrading 35 163 Overall (a + b) +100 155 Institutional Performance PROJECT DATES Original Plan Actml First Mention in Files 4/17/74 Government's Application Negotiations 11/14/77 12/22/77 Board Approval 2/21/78 Signing 4/25/78 Effectiveness 8/ 4/78 Closing Date 7/ 1/82 6/30/84 OTHER PROJECT DATA Borrower: Government of Rwanda Executing Agency: Follow-on Proects: Name Fifth Highway Project Credit Number Credit 1250-RW Amount (SDR. million)23.3 Approval Date V PROJECT ERFORMANCA AUDIT REMR RWANDA (Credit 769-RW) Project preparation (paras 6-14) 4. Four-Year (1978-1981) Periodic Road Maintenance Prgxu 1. Project preparation began - - T Betterment of about in early 1976. Appraisal took 500 km of gravel roads; place in February-March 1977, Board Rehabilitation of about 1,000 km of approval in February 1977, and the earth roads; Resealing of about 100 project became effective in August km of paved roads. Reau... 1978. Preparation was not easy Implementation took six years because of differences of opinion instead of four due to delays in between IDA and Government procurement of equipment. 100% of primarily on the importance of road the betterment works; 70Z of the maintenance, and on technical rehabilitation and upgrading, and assistance. 90% of the resealing works were completed (pars, 22). Project Lmplementation and results 5. Four-Year (1978-1981). RoUtine Road Maintenance Pro gXam, - 2. Project objectives -T-rS1e Routine maintenance of Intentions: To assist the 3,500 km of earth and gravel roads; implementation of an 8-year highway Routine maintenance of 350 km of maintenance program; and to replace paved roads. expatriates in responsible Quantitative results of the routine positions in MPW. Achiev-emelnts maintenance operation are not Maintenance operations were available. Labor gangs did and strengthened but expatriate staff continue to do satisfactory routine was not replaced because qualified maintenance work (para 23). national staff prefer to work in the private sector (para 20). 6. Strenthenina and E&oansion of Road Branch's CaRAM 3. The project was completed with Tggtm* Procurement of a two year delay (in July 1984), equipment, tools, etc.; Extension largely because of difficulties of workshops; Technical with equipment procurement and with assistance. &*Ul= Delivery of implementing the training compinent equipment took longer than (paras 14-19). Experience with anticipated because of delayed individual components is as preparation of bidding documents follows: and delay in contract awards. l- this summary, references to the PPAR are shown as (para ...) and references to the PCRs corresponding to the Ports Rehabilitation Project and to the Highway Rehabilitation Project are shown as (PC, parar ...e vi Workshops and warehouses were Findings and conclusions satisfactorily improved. Technical assistance performed well (para 9. The 1986 PCR examined the 24). project as a free-standing operation anA~ the tone of that 7. Four-Year Roja4 document is ustifiedly positive. Maintenance Training ProaraM The Audit has examined the project Targets: Gradual replacement of as part of a series and its expatriate personnel; Training of conclusions are more reserved road maintenance personnel. because of four main Results: Two year delay in the considerations. construction of the training facilities. The project did not (a) The project cost US$4.0 train higher level staff. About 410 million more than technicians were trained but the 54 estimated at appraisal best qualified left Government and a Member Country service for the private sector. perenially strapped for Expatriate staff will be needed for cash had to pay the quite some time to come (para 25). difference (para 32). (b) Rwanda's current economic Project costs. economic rate of situation precludes return. and sustainability of allocations for highway benefits (paras 26-28) maintenance without massive donor support 8. According to the PCR (PCR (para 33). paras 3.10-3.11 and Annex III), total estimated costs were US$23.7 (c) After more than twenty million equivalent and total actual years, donor coordination costs were about 17% higher remains an elusive goal (US$27.8 million equivalent). The (para 34). major cause for the increase was the two year delay in the execution (d) New credits are needed of the road maintenance and either to repair work training programs. The cost done under earlier ones, overrun was absorbed by Government, or to continue whose contribution increased from initiatives which were the original estimate of US$2.6 expected to be self- million equivalent to US$6.8 sustaining (para 35). million equivalent. The Economic Rate of Return (ERR) for the betterment, rehabilitation, and Recent IDA operations do not, in upgrading programs, was estimated the Audit's view, seem to have been at appraisal to be above 100%. The sufficiently informed by experience re-estimated ERR is also above (paras 36-37). Similarly, 100%. Even if benefits are reduced Government must assume part of the by 50%, the ERR would still be responsibility for results so far satisfactorily high. In the obtained (para 38). Audit's view, and data limitations notwithstanding, the project 10. Lessons for the future can be remains economically justified. summarized under the following three headings: vii (a) Financial -and technical general, and staffing of asistance. For the Rh in particular, will be foreseeable future, road extremely long-term maintenance and further propositions. Given that strengthening of road salaries are higher in transport will depend on the private sector, RB the volume of financial will not retain the assistance -- and on the necessary numbers of quality of technical experienced and competent assistance -- from engineers and managers. abroad. The Audit noted For the foreseeable that Rwanda is not, on future, expatriate staff the whole, favorably will have to remain in RB disposed towards in the same positions and expatriate staff and this in the same numbers as in generates unwelcome 1990. As few as possible friction, detrimental to conditions and/or efficient working covenants ought to be conditions. Apart from a attached to future change in local projects. Those that are attitudes, sound use of attached ought to reflect foreign assistance, be it essential concerns (e.g., financial or technical, scrupulous auditing of will require that project accounts; proper exchanges among Donors be awarding of contracts) more frank, and far more and ought to be strictly integrated than in the observed. past. It is an open question, who among the (c) Project implementation. Donors will play the part As much as possible of of coordinator. If IDA the physical work (both undertakes to perform in new construction and this task, it must in maintenance) ought to allocate considerably be turned over to the more staff resources than private sector. Most of it has done so far. the supervision will have to be done by technical (b) Project inception. assistance staff. Physical objectives ought Management, works to be simple and supervision, and training measurab 1 e . tasks must be kept Institutional objectives separate. Expatriate ought to be modest. The staff assigned to one civil service is too big activity must not be to be flexible, and any expected to engage in change is perceived as a another. IDA disturbance of supervision missions comfortable practices. ought to focus on Staff training is useful essentials. Failure to and must continue, but submit correctly audited its effects will not be accounts, and diversion visible for a long time. of project resources to Strengthening of non-project purposes, transport planning in must be regarded as vii sufficient cause for immediate suspension of disbursements. PROJECT PERFORHANC AUDIT REPORT RWANDA FOURTH HIGHWAY PROJECT (Credit 769-RW 1. 1=D_fIQN 1. Rvanda has been a member of the Bank Group since 1963. Government inexperience with Bank procedures and a decision not to borrow for studies led to the abardonment of two operations before the US$19.0 million First Highway Project was approved in 1970. The US$5.4 million Second Project for Highway Maintenance was approved in 1972, and the US$6.3 million Third Highway Project in 1974.1J/ 2. These were followed by the US$15.0 million Fourth Highway Project (the object of this Audit) in 1978, the US$25.9 million Fifth Highway Project in 1982,2/ the US$36.0 million Sixth Highway Project in 1985/88,3/ and the US$40.0 million Transport Sector Project in 1990.4/ 1/ The three projects were audited together. Please see OED Report No. 2545, of June 13, 1979. Annex 1 (Summary Descriotion of the First. Second and Third Highway Projects) is an abstract of that audit. 2/ The project convsfted of construction to paved standard of 53.5 km of the Butare-Cyangugu road; consulting services for studies and technical assistance; procurement of materials and supplies for two-year routine maintenance of the Kigali-Gatuna road; and procurement of weighing scales for enforcement of vehicle weight regulations. 3/ The project financed elements of the 1986-1989 Road Maintenance and Reconstruction Program: routine maintenance of earth and paved roads; regravelling of earth roads; part of the training program; consultant services; the MPW soils laboratory; and reconstruction of the Kigali-Gatuna road (which had been financed under the First Highway Project and failed prematurely). 4/ The project aims to protect capital investment in main roads through improved maintenance, to strengthen institutions through better planning and resource mobilization, to promote comunal development and better economic integration through improved communal roads, to improve access to the Kibuye prefecture, to support lake transport, and to improve vehicle safety. In addition, the project aims at policy dimensions such as streamlining the regulatory environment, financing of sectoral expenditures, and improving institutional arrangements and responsibilities for the sector. 2 3. The initial intention was to keep the Audit narrowly focussed. The PCR published in 1986 1/ stated that this was a successful project, and OED intended to verify that benefits continue to be sustainable with respect to the two main objectives: improving major road connections (between Rwanda and its neighbors and between Kigali and the regional centers), and strengthening institutional capability for efficient road maintenance. 4. File study and discussions with IDA staff in Washington, encouraged OED to consider the project against a broader background. Rwanda's demography and resources place it among those Member Countries where sectoral improvements are intimately connected to macroeconomic, institutional, and social conditions.&/ Specifically in the case of Rwanda, continuous donor assistance is indispensable for both the investment and the recurrent budget. Staff training and career development are challenging, and extremely long-term, tasks in a country with growing unemployment and serious political problems. Finally, Rwanda's land-locked territory requires cheap and dependable access to the sea but, despite efforts over decades, transit traffic remains expensive and vndependable. 5. Have these parameters been taken into account during inception and preparation of IDA-financed transport projects? Did IDA-Government exchanges produce agreement or disagreement on essentials such as the use of techaical assistance, the importance of staff training, and the burden placed upon the recurrent budget by new capital investments? What has IDA done to assist Rwanda with better access to the sea? In an attempt to explore these and related questions, the M&y 1990 Audit Mission met with Government and donor agency officials who were kind enough to express their views on the experience gained so far. 2/ Report No.6209 of May 21, 1986. / Annex 2 (Rwanda -- Notes on Demography and Economics, is a brief introduction to a complicated set of issues. 3 II. THE FOURTH blGHIAY PROJECT A. PROJECT INCEPTION AND PREPARATION 6. The first document in the file is dated April 17, 1974, and states that a Project Brief (PB) prepared by the Bank's Resident Mission in Eastern Africa (RMEA) was not up to date. Further preparation did not take place for two years. The second document in the file is dated June 24, 1976, and refers to a possible project which would continue assistance to road maintenance begun under the Second Project, and would focus on training of staff at all levels. The June 1976 note stressed that the success of any road maintenance program would depend on Government's ability to finance recurrent costs. 7. By the summer of 1976, the project was included in IDA's lending program but an August 4, 1976, note states that, in view of the scarcity of local resources, IDA ought to finance a h' )i proportion of total costs. However, discussions between IDA Management and Rwanda's Minister of Finance during the 1976 IMF/IBRD Annual Meeting revealed that the two sides were not in agreement concerning either the role of IDA in Rwanda transport, or the manner in which the policy dialogue was being conducted..Z/ The Minister did not think there was any need for IDA to prepare a Transport Sector Memorandum, and that enough studies were already available. He also felt there were too many "flying visits" by IDA staff who did not have time to discuss issues at leisure. 8. The February 10, 1977, Preappraisal Issues Paper (IP) recorded that the Ministry of Public Works and Infrastructure (MPWI) and various IDA preparation missions had identified a project which would consist of continuation of the highway maintenance program; training of the Road Branch's (RB) staff; assistance to domestic contractors; and road construction. Total estimated cost was US$54.5 rillion equivalent. This would be covered by US$15.0 million (provided by the IDA Lending Program), US$14.0 million (allocated by the Federal Republic of Germany for road construction), and US$13.5-17.6 million (to be contributed by Government), leaving a gap of US$9.0-13.0 million. The IP expressed serious doubts as to whether Government would in fact be able to contribute much. Under "Technical Issues", the IP noted that equipment procurement might be delayed because of Government procedures; that staff training had not so far been done because of lack of local funds; and that road construction experience with Credits 196 and 475 suggested that building and maintaining roads in Rwanda was both difficult and costly. 9. Appraisal took pla s in February-March 19, 1977 buw did not resolve a number of disagreements within Government and between Government and IDA. The March 17, 1977, Minutes of Meetings Held at the MPWI and the April 12, 1977, IP reveal that the Minister of Finance did not agree either Z/ RwandA -- Annual Meeting Discussions, Note to Files. October 21, 1976. 4 with the project composition, or with the amount of Government and IDA financing, or with the usefulness of road maintenance in general. He preferred construction of paved roads which, according to him, did not need maintenance. He considered maintenance a "wasteful intangible" and suggested that of the US$15 million available from IDA, only $5.0 million should be allocated to maintenance and US$10.0 million should go for the construction of the Kigali-Ruhengeri-Cyanika road which he believed could be approved and started without detailed engineering.&/ The I? also revealed that other unresolved issues confronting the project were not insignificant. There were problems with the Road Branch (RB) administration,2/ with the Central Workshop, and with long payment procedures.19/ 10. During the Decision Meeting of April 20, 1977, the Bank's Central Projects Staff (CPS) expressed concern that the project might be too large for the country's financial and administrative resources, and that the attitude of the Minister of Finance did not bode well for a smooth implementation. Other participants noted that Covenants regarding adequate highway maintenance expenditures had been included in previous projects 11/ and had stipulated that Government would provide funds, facilities, services, and other resources needed for road maintenance. Nevertheless, recurrent expenditures for maintenance had never exceeded US$1.0 million per year. Consequently, the meeting doubted that Government could agree -- as the Appraisal Mission assumed -- to triple this expenditure under the Fourth Highway Project. The Meeting also noted that project execution would require considerable cofinancing by other donors, especially since the proposed external financing (US$21.0 million, or about 70% of total costs net of taxes) appeared on the low side for a poor country like Rwanda. In this connection, the Meeting felt that the size of capital versus recurrent costs, and the cost sharing formula ought to be reconsidered. The Meeting concluded that a letter be addressed to I/ Issues Paer, April 12, 1977, para 7. 2/ Execution of the Second Project (Credit 299-RW) had been difficult because RB management was pressured by higher levels to divert equipment for purposes other than road maintenance. Lack of a proper accounting system prevented efficient cost control. RB did not have control over its drivers, who were hired and paid by the Ministry of Posts and Communications. RB had just acquired an inexperienced new manager who did not seem prepared to accept expert advice (IP, paras 12-13). IQ/ The system in force required that invoices be cleared by at least five different levels of management, resulting in delays of up to six months. Suppliers were therefore hesitant about accepting new orders. The same procedure applied to personnel payments, which were often delayed by two or more months. Morale was low and efficiency left much to be desired (IP, para 15). J1/ The Credit Agreements for the Highway Maintenance Project, for the Third Highway Project, and in the Amending Agreement for the First Highway Project. 5 Government regarding these concerns and that, pending Government's reaction, the SAR would be prepared on the assumption that Government would accept the maintenance and betterment program proposed by the Minister of Public Works during appraisal.12/ 11. By August 1977, when the draft SAR was circulated for comments within IDA, it was clear that implementation might be hindered because the different opinions held on maintenance by the Minister of Finance and by the Minister of Public Works did not arise solely out of economic policy considerations. Although the Minister of Finance was now prepared to support the project by proposing to the Council of Ministers a US$10.0 million road maintenance budget, he was firmly against the idea of hiring foreign experts because he believed that enough qualified Rwandese were available to run the project.1L/ This view was not shared by IDA staff and their reticence was strengthened by the fact that Government had used maintenance equipment financed by the Second Project for road construction near Kigali. Furthermore, Government counterpart funds for the First and Third Projects were used to finance part of roadworks for an OCAM Conference and that, as a result, Government payments to contractors for work under the IDA projects were substantially delayed. In addition, project equipment had been rented out during weekends to private persons, against payment, and the proceeds had been loosely administered.14/ 12. It was against this background that negotiations took place in December 1977. The project was approved on February 21, 1978, the Credit Agreement was signed in April 1978, and the project became effective in August 1978. Co-financing was provided by the Federal Republic of Germany, the UNDP, Belgium and the World Food Program. 13. According to the February 6, 1978, SAR (para 4.02), the US$23.7 million project consisted of: (a) a four-year-betterment program to (i) improve about 500 km of gravel roads, (ii) rehabilitate about 270 km of very old and 12/ Rwanda -- Fourth Highway Project, Decision Memorandum, May 16, 1977. 12/ In the course of discussions on the proposed project during his September visit to Washington for the 1977 IMF/IBRD Annual Meeting, the Minister of Finance reiterated to IDA staff his conviction that the project should give less emphasis to maintenance and more to construction. He also contended that the project provided for too many foreign experts to maintain earth roads. In the course of the discussions, it emerged that the reluctance of the Government to use technical assistance resulted from inapprorpiate conduct by some consultants working on IDA projects, including alleged bribery (Rwanda -- Annual Meeting Discussions, October 6, 1977). 14/ Please see Rwanda -- Fourth Highway Project, dated August 19, 1977, and summarizing discussions in Kigali during August 6-10, 1977. See also Rwanda -- Proposed Fourth Highway Project, dated September 1, 1977, and summarizing CPS views regarding the desirability of scaling the project down and strentching it over a longer period. 6 deteriorated roads to gravel standard, (iii) upgrade about 730 km of earth roads to gravel standard, and (iv) seal about 100 km of paved roads (SAR, paras 4.03-4.06); (b) a four-year routine maintenance program to strengthen operations on about 3,500 km per year of earth and gravel roads, and to initiate routine maintenance on about 350 km of paved roads (SAR, paras 4.07-4.09); (c) a strengthening and expansion of RB's capacity to implement both programs through (i) procurement of some additional equipment and vehicles, (ii) technical assistance, (iii) construction and expansion of workshops, (iv) procurement of machine and hand- tools, and (v) procurement, for the betterment program only, of spare parts, fuel and materials (SAR, para 4.10); and (d) a training program for maintenance personnel at all levels, including courses and scholarships abroad (SAR, paras 4.11- 4.12). B. PROJECT IMPLEMENTATION 14. The Credit became effective in August 1978 and was expected to be closed in July 1982. It was actually closed in June 1984, two years behind the planned date. Implementation experience is summarized below. 15. Procurement of equipment and spare parts -- All maintenance equipment was delivered by mid-1980. By the end of the project, and thanks to the German consultants, procurement of spare parts and the reorganization of the central store in Kigali had been satisfactorily implemented. A complete inventory of spare parts was carried out, necessary parts to maintain the equipment fleet had been identified, procurement had been rationalized, and waste and pilferage had been reduced. On the whole, the system put in place by consultants improved equipment maintenance. 16. Execution of road maintenance works -- Target volumes for road improvement and maintenance works were met. Quality of the work done was satisfactory. Work was performed by force account with four equipment- intensive and two labor-intensive brigades, all run by technical assistance staff. Towards the end of the project, and in order to train RB staff, three of the six brigades were assigned to Rwandese foremen. Two of these brigades were run properly but the third needed further assistance by the consultants. Government found it hard to recruit more foremen because pay was better in the private sector. 17. Improvement of workshops -- Expansion works in the Kigali workshop were completed in January 1981. Additional space was needed and the necessary work was done by December 1982. A satellite workshop, originally intended for Butare, was eventually built at Kibuye. 7 18. Cost accounting -- The German consultants introduced cost accounting procedures for road improvement and maintenance works. Counterparts were hard to find so that technical assistance could be phased out. 19. Staff training -- Implementation was carried out two years behind schedule, partly because of the late conitruction of the training center, and partly because of the difficulties in locating qualified candidates. Late construction of the training center was due to the lack of action by Governme .t to cancel the contract with the original contractor who discontinued works due to low prices, and to call for new bids, and award the contract to a new firm. As a result, in-class training provided by consultants was carried out at about 60% of the anticipated rate for more than two years. This, in turn, necessitated an extension of the consultants' contract from March 1981 until mid-1982 so that the training program targets could be met. Eventually, about 410 local staff were trained. Consultants also trained seven local instructors who required further consultant support, and this was provided under the Fifth Highway Project. The project had provided for fellowships (US$80,000) for further training of RB staff abroad but the Governments of France, Belgium and the Federal Republic of Germany offered similar scholarships, and IDA agreed that project funds be reallocated to road maintenance operations. Government was not able to utilize fully the scholarships provided by donor countries. C. PROJECT RESULTS Project objectives 20. Intentions: To assist the implementation of an 8-year highway maintenance program designed in 1971; and to replace expatriates in responsible positions in MPW. Achievements; Maintenance operations were been strengthened but expatriate staff was not replaced because qualified national staff preferred to work in the private sector. Project components 21. The project consisted of four major components whose implementation can be summarized as follows. 22. Four-Year (1978-1981) Periodic Road Maintenance Program -- Targets: Betterment of about 500 km of gravel roads; Rehabilitation of about 1,000 km of earth roads; Resealing of about 100 km of paved roads. Results: Implementation was carried out by force account and took six years instead of four due to delays in procurement of road maintenance equipment. Despite the delays, 100% of the betterment works were fully completed; 70% of the rehabilitation and upgrading and 90% of the resealing works were acceptably completed, and labor-intensive routine maintenance works (-cantonnage") were carried out and continue to be carried out satisfactorily. 8 23. Four-Year (1978-1981) Routine Road maintenance Program -- Targets: Routine maintenance of 3,500 km of earth and gravel roads; Routine ma!ntenance of 350 km of paved roads. Results: Quantitative results of the routine maintenance operation are not available. Labor gangs did and continue to do satisfactory routine maintenance work. 24. Strengthening and Expansion of Road Branch's Capacity -- Targets: Procurement of equipment, tools, etc.; Extension of workshops; Technical assistance. Results: Delivery of equipment took longer than anticipated because of delayed preparation of bidding documents and delay in contract awards. Workshops and warehouses were satisfactorily improved. Technical assistance performed well. 25. Four-Year Road Maintenance Training Program -- Targets: Gradual replacement of expatriate personnel; Training of road maintenance personnel. Results: Two year delay in the construction of the training facilities. The project did not train higher level staff I ause consu- ltants concentrated more on works execution, because of lack of qualified candidates, and because qualified local candidates prefer to work in the private sector. About 410 technicians were trained under the project5/ but the 54 best qualified left Government service for the private sector. Evidence that expatriate staff would be needed for quite some time was given by the fact that, when the consultant's contract ran out during the 1984/85 maintenance season, operation of the road maintenance brigades deteriorated. Project costs, financing, economic rate of return, and sustainability of project benefits 26. According to the PCR (paras 3.10-3.11 and Annex III), total estimated costs were US$23.7 million equivalent and total actual costs were 1/ Trained personnel includes: 40 foremen, 135 equipment operators, 5 workshop superintendents, 20 storekeepers, 15 313ctricians, and 195 mechanics. 9 about 17% higher (US$27.8 million equivalent).I/ A major cause for the increase was the two year delay in the execution of the road maintenance and training programs. Cost overruns were absorbed by Government, whose contribution increased from the original estimate of US$2.6 million equivalent to US$6.8 million equivalent. Disbursements corresponded to the delays in project execution. 27. The Economic Rate of Return (ERR) for the betterment and rehabilitation, and upgrading programs (excluding resealing), was estimated at appraisal to be above 100%. The re-estimated ERR at completion is also above 100%.11/ Lack of detailed data prevented the Audit from doing an independent re-calculation but even if benefits are reduced by 50%, the ERR would still be satisfactorily high. Regarding project timing, the first year returns of the three reevaluated components exceed the assumed opportunity cost of capital of 12%, which suggests that works were overdue. In the Audit's view, and data limitations notwithstanding, the project remains economically justified. Sustainability of project benefits depends on continued availability of concessional aid, and on the continued presence of expatriate staff. 1W/ There are discrepancies in project cost and financing as reported by the PCR and by Supervision Reports. Specifically, the March 31, 1982 and the January 26, 1983, Supervision Reports (Section 5) list the following breakdown of costs and financing: Cost estimated at appraisal: US$23.7 million Latest cost estimate: US$31.3 million Financing (US$ million): According to SR: According to PGR Original Revised IDA 15.0 15.0 15.0 UNDP 1.3 1.3 1.3 FRG 2.3 3.5 2.3 Belgium 0.5 0.7 0.4 WFP 2.0 2.8 1.9 Government2.6 8.0 6.8 Lack of detailed data prevented the May 1990 OED Mission from reconciling these differences. 12/ The PCR Basic Data Sheet notes that, for the project as a whole, the estimated economic rate of return was +100%. The section on economic reevaluation in the PCR (paras 5.01-5.15) does not explicitly say what the re-estimated figure for the project as a whole is. PCR para 8.04 does say, without indicating how the conclusion is reached, that "the overall ERR is about 155%". The lack of clarity (at least with respect to the rehabilitation and upgrading program), is possibly due to the fact that after appraisal many low traffic roads were excluded and higher traffic roads were included. This does not allow comparisons between the appraisal and the reevaluated ERR since the reevaluated program is substantially different from the one described at appraisal. 10 III. FINDINGS AND CONCLUSIONS Oyerviw 28. Evaluated as a free-standing operation, the Fourth Highway Project has proven to be economically justified and to have attained the major objectives set at appraisal. However, like a number of other operations audited by OED over the past few years, jf/ the project is part of a series aimed to improve highway sector performance over a long period of time. Consequently, experience is enhanced when sectoral status and prospects are juxteposed to the status and prospects of the national economy. 29. In theory, IDA assistance to individual sectors is extended after a thoughtful review of a member country's socio-economic environment. In anticipation of sectoral contributions to national development, lending operations are mounted, lessons are drawn, and subsequent projects benefit from preparation and implementation of preceding ones. The Audit believes that IDA assistance to the highway sector would have been more effective if fuller account had been taken of the country's meager resource endowment, and of the political, social, and economic problems arising out of unresolved political differences and unrestrained population growth.11/ 30. The PCR on the Fourth Highway Project2/ examined the project as a free-standing operation and the tone of that document is justifiedly positive. The Audit has examined the project as part of a series and its conclusions are more reserved because of four main considerations. 18./ For example, the Fourth Highway Project in Zaire (OED Report No. 6185 of May 12, 1986), the Fourth Highway Project in Madagascar (OED Report No. 6148 of April 18, 1986), and the Second Highway Project in Burundi (OED Report No 6255 of June 9, 1986. 1_/ For greater detail, please 6ee Rwanda:First. Second and Third Highway Projects (Credits 196, 196-1. 299 and 475-RW) (OED Report No. 2545 of June 13, 1979); A Report on the International Transportation Bottlenecks Affecting Rwanda and Burundi (World Bank, December 1980). Randa: Fifth-Highway Project -- President's Report, Staff Ap2raisal ReRor& (May 1982). Rianda Economic Memorandum Recent Economic and Sectoral Develoments and Current Policy Issues (May 1983). Rwanda: Sixth Highway Project -- President's Renort, Staff Appraisal Report (November 1985). Rwanda: Recent Economic Developments and Current Policy Issues (October 1986). Rwanda: Public Expenditure Program -- An Instrument of Economic Strategy (October 1989). Rwanda: Transport Sector Project -- President:s Report, Staff Appraisal Report (April 1990). 22/ OED Report No. 6209 of May 21, 1986. 11 31. The project cost USS4.0 million more than estimated at aDraisal and a Member Country Rerenially straRoed for cash had to pay the difference. Cost increases were due to delays in equipment procurement and in construction of the training center. Government is partly responsible for the former, and wholly responsible for the latter. Still, public funds are scarce and Rwanda would have benefitted if IDA tried somewhat harder to get maximum value per dollar. If IDA had concluded that Government was not showing due regard for economy and efficiency, it ought to have resorted to strict responses, but did not. This leads the Audit to believe that, sometimes, preservation of amicable relations between IDA and a Government may turn out to be financially costly for the Member Country. 32. Rwanda's current economic situation precludes allocations for highway maintenance without massive donor suoort.21/ The 1980s were not easy. Domestic product grew by 1.5% annually and real per capita consumption declined between 1982 and 1988. Government failed to realize that the high revenues of the 1979-80 coffee boom were temporary, did not cut back on expenditure, attempted to sustain growth through a public-led approach, saw that this could not be done, froze public wages at the 1982 levels but, even so, continued to act as if the crisis was temporary and could be weathered by even greater administrative intervention. Between 1982-88, private investment declined and private consumption increased. The overall economic situation deteriorated, financial difficulties became severe after 1986, and interest payments on government debt have been the fastest growing item in the budget, averaging 19% annually, as Government borrowed increasingly to finance its development expenditures. Meanwhile, intervention shows no sign of decreasing, and government procedures have been getting even more cumbersome. There is an expanding external gap and any new aid has to be highly concessional because the debt service ratio should not be allowed to increase. At the same time, local currency shortage is likely to remain the major constraint to the implementation of public expenditure programs. All things considered, road maintenance is not likely to be a priority item for a Government that, for the first time, is facing famine in certain districts and major political problems elsewhere. Consequently, road maintenance will be possible only if most of this capital and recurrent costs are financed from abroad. 33. After more than twenty years. donor coordination remains an elusive goal, Rwanda's financial situation, and its shortage of qualified professionals, make expatriate technical assistance and coordination of donor activities critical to the effective design and implementation of public expenditure programs. In this connection, para 131 of the October 1989 Rwanda: Public 21/ For a delicately phrased summary of the country's stark prospects, please see Rwanda: Public ExRendituXe Program -- An Instrument of Economic Strategy, two volumes (October 19, 1989). 12 Bxnenditure Program -- An InstrMent of Economic Strategy deserves to be quoted in full: "It is of little use to improve the technical capacity of sectoral ministries to prepare expenditure programs which are consistent with their likely resource availabilities, staff requirements, and development priorities, if these programs are not fully recognized by donors as providing the framework within which their own assistance should be given. Donors need to operate closely in support of technical ministries when these ministries are attempting to mount a public expenditure program. Donors must also be willing to modify their policies and procedures, in order to operate more fully within such a program. Unfortunately, this has not been generally the case in Rwanda, as donors have usually followed their own agenda (sectoral priorities, budgetary cycles, procurement procedures, etc.)." Notwithstanding these serious assertions, the Audit has not perceived any serious attempt to get the donors together. 34. New credits are needed either to repair work done under earlier ones. or to continue initiatives which were expected to be self-sustaining. The principal construction component of the Sixth Highway Project is the reconstruction of a road which was poorly built under the First Project and failed prematurely. The Fourth Highway Project led to the implementation of the third phase of road maintenance program under the Sixth Highway Project. The point is that when the Fourth Project was approved, there was not supposed to be a third phase. Still, the Sixth Project would assist in improving major road connections, and in strengthening institutional capabilities to carry out efficient road maintenance. All these were basic objectives of the Second and Fourth Projects. Furthermore, the Sixth Highway Project would put priority on training in general and, in particular, on the training of managers to replace the expatriate technical assistance at RB. Again, these were explicit objectives of the Second and Fourth Projects, and perhaps the only new element under the Sixth is that it would retain consultants whose sole job would be training. The Sixth Project envisaged, in addition to the training program itself, a number of accompanying measures to establish training and manpower development within RB. These would include the strengthening of RB's Personnel Bureau, the setting up of a Training Unit and of a Training Coordination Committee to provide a functional link between training and personnel management on the one hand and training and the technical services on the other. The question arises whether Rwanda has demonstrated a capacity for such institutional mechanisms and, in the Audit's opinion, it has not. Status and Drosoects in 1990 35. Rwanda is a small country that exports about 50,000 tons of coffee annually. Subsistence agriculture feeds (though with increaoing difficulty) 90% of a rapidly increasing population. The public service, which is the major employer in the "modern" sector, has grown disproportionately large but cannot run without financial and technical 13 assistance from abroad. Donors do not like to collaborate because they prefer their projects to be easily identifiable and fully subject to their procurement guidelines. The private sector is small and, in transport at least, dominated by companies enjoying strong political backing. Thus, "privatization" of road transport is likely to lead not so much to increased competition as to stronger oligopolies. 36. In 1990, the main road network is in place and, thanks to foreign aid, it is relatively well-maintained. The main tasks for the future are to keep highways maintained, to improve rural roads, to avoid unnecessary investments, and to emphasize maintenance instead of reconstruction. Maintenance expenditures ought to represent about a third of total road expenditure. The bulk of road revenues ought to come from user charges. Coordination between ministries would be desirable so that intersectoral linkages and priorities are reflected in the ongoing and future programs. Access to the sea remains a collection of thorny issues that, after more than twenty years of studies and debates, remain as intractable as ever.22/ Donor coordination is a sine qua no for effective strengthening and operation of the highway sub-sector but also remains elusive. 37. Have these parameters been taken into account during the inception and design of recent IDA-supported initiatives to strengthen transport? SAR para 1.03 of the 1990 Transport Sector Project reads as follows: "The prevailing economic and financial conditions in Rwanda have been taken into consideration in designing the proposed transport sector program and its financing. The program is seen as an integral part of the economic development process and therefore calls for appropriate policy actions within the transport sector. A letter of Sector Development Policy with an Action Plan has been prepared by the Government. Policy actions will aim at more emphasis on maintenance of infrastructure, more reliance on the private sector for road maintenance, better planning and improvement of cost recovery mechanisms as well as more liberalization of transport operations, in order to make exports more competitive and to reduce the costs of imports and internal transport." In the Audit's view, the scope and objectives of this project were not sufficiently informed by experience and convenient assumptions have been made regarding aspects such as the following: 2/ For an extensive discussion of the difficulties encountered by land- locked East African countries (Uganda, Rwanda, Burundi, Malawi and Zambia), please see the PPAR on Burundi -- Second Highwa Project (OED Report No. 6255 of June 9, 1986) and the PPAR on Burundi -- Third Highway Project (under preparation in October 1990). 14 (a) Government willingness to proceed with policy measures which Government officials either do not agree with,2/ or do not have the capacity to implement,2A/ or are constrained from implementing because of the many Ministries and agencies involved.W/ (b) Government willingness to sponsor deregulation of road transport issues, especially regarding the road transport industry, despite the fact that this will threaten comfortable sinecures and can therefore could be counted upon to generate strong opposition.2&/ (c) Government ability to resolve the problem of staff training and development. 38. If the above impression is correct, questions may be raised concerning the extent to which Rwandese top-level management is genuinely 212/ For Government reservations about the "Fonds Routier" (the first attempt for whose establishment dates back to an August 23, 1963 Law), see paras 4-8 of the May 15, 1989, Rwanda -- First Transport Sector_Dperation -- Preparation Mission -- Back-to-Offige Report. 2A/ On Government's difficulty to prepare (even with technical assistance) thb study of communal roads and of the road maintenance strategy, see paras 4-7 of the June 5, 1989, Rwanda and Burndi -- Operational Review Mission. 2.g/ According to para 1.05 of the May 1989, Project Brief on the proposed Transport Sector Project, two Ministries are directly involved in transport: (a) The Ministry of Public Works, Energy and Water (Minitrapee) for road construction and maintenance; and (b) The Ministry of Transport and Communications (Minitransce) for the development of other modes and for utilization of the transport infrastructure in general. This Ministry is also responsible for parastatal organizations such as Onatracom and STIR. In addition, three other Ministries are indirectly concerned with transport: (a) The Ministry of Planning, which determines sector priorities; (b) The Ministry of Finance, which influences the sector through taxation of inputs such as fuel; and (c) The Ministry of the Interior, which responsible for the communes and therefore the maintenance of the large network or rural, unclassified roads. Finally, the Central Bank influences sector performance through allocations of foreign exchange and the approval of import licences. 2&/ On the privileged position of STIR and CORWACO for the transport of extremely lucrative international cargo, see page 2 of the June 7, 1989, comments on Rwanda: Transport_Sector Project -- Project Brief. 15 committed to substantive strengthening of road operations. Such questions can be centered on five points: (a) Institutional capability -- Six Highway projects have provided training for large numbers of staff over a period of about 20 years. Even though the individual level may not be too high, it seems unlikely that final results ought to have been so low. Although some may have left public service for more money in the private sector -- which, after all, might be regarded as a benefit generated by the Bank-supported projects -- those who stayed would have produced more if only they had been given the increased responsibility they deserved. Good training has to be complemented by a sound personnel policy which, alone, ensures genuine sustainability of project benefits and reduces dependence on extriate assistance. In the Audit's opinion, it seems that the Rwandese top-level management continues to Implement personnel policies influenced by intuitive considerations and not by objective criteria, like performance. (b) Use of technical assistance -- At the initial stages, donot- do intervene in the selection of technical assistance staff by establishing size of the program, terms of reference and recruitment procedures. However, as soon as an expatriate team is in place, the Borrower takes over totally and it has proven very difficult for the Executing Agency to effect changes in the team or improvements in its use. In this connection, the Audit feels that unsatisfactory results have arisen out of the Borrower's top-level management attitudes. (c) Lack of coordination among donors -- This is a fact and probably more pronounced in Rwanda than in other countries. The Audit believes that the government has mastered the art of partitioning and has made it difficult for each donor to know what is being done, or what is intended, by the others. The net result is neither good for Rwanda nor encouraging for the donors. (d) Road maintenance versus road construction -- Since the Second Highway Project, the Bank has advocated better maintenance practices but the Borrower has shown little interest because he prefers rehabilitation or reconstruction works. If proof of this assertion is demanded, the Audit feels that ample evidence exists during project appraisals, and even project implementation, when road maintenance equipment was diverted to construction operations. The problem is exacerbated by the fact that road maintenance is done with rather poor management and control, generating unreliable output figures. The Audit feels that the Borrower remains unconvinced that maintenance is an economically worthwhile activity and that, for this reasons, inadequate funds are presently available for the maintenance of the country's road network. (e) Planning -- Poor planning results in implementation delays. The Audit feels that planning schemes are often help up for long 16 periods before the Rwandese top-level management approves them. Then, during implementation, plans are changed again. In this connection, it must be noted that, after 20 years of Bank- Government collaboration, the quality of bidding documents and procurement procedures remains unsatisfactory. (f) Cost recover -- It is true that traffic has increased more than expected. However, and despite repeated suggestions by the donor community, notning has really been done by the Rwandese top-level management to recover costs through axle-load taxation. The weighing station that was eventually installed continues to be inoperative. Summary 27/ 39. Lessons for the future can be summarized under three headings: 40. Financial and technical assistance. For the foreseeable future, road maintenance and further strengthening of road transport will depend on the volume of financial assistance -- and on the quality of technical assistance -- from abroad. The Audit noted that Rwandese attitudes were not in the past, and are not at present, favorably disposed towards expatriate staff. This generates unproductive friction, detrimental to efficient working conditions. Apart from a change in local attitudes, sound use of foreign assistance, be it financial or technical, will require that exchanges among Donors be more frank, more integrated than in the past, and more willing to re-examine the form of technical assistance they are willing to provide.2/ It is an open question, who among the Donors will play the part of coordinator. If IDA undertakes to perform this task, it must allocate consieerably more staff resources than it has done so far. 41. Project inception. Physical objectives ought to be simple and measurable. Institutional objectives ought to be modest. The civil service is too big to be flexible, and any ehange is perceived as a disturbance of comfortable practices. Stat training is useful and must continue, but its effects will not be visible for a long time. Strengthening of transport planning in genera-, and staffing of RB in particular, will be extremely long-term propo.-tions. Given that salaries are higher in the private sector, RB will not rektain the necessary numbers of experienced and competent engineers and ma.agers. For the foreseeable future, expatriate staff will have to remain in RB in the same positions and in the same numbers as in 1990. As few as possible conditions and/or 22/ The OED Mission to Ruanda followed a visit to Burundi where the Third Highway Project was audited. Lessons for the future are remarkably similar for the two countries. ZV For example: should one continue with the provision of fairly lage teams that stay in the country continuously, or should assistance take the form of numerous short missions to perform a specific task within a specific period and then to return for follow-up? 17 covenants ought to be attached to future projects. Those that are attached ought to reflect essential concerns (e.g., scrupulous auditing of project accounts; proper awarding of contracts) and ought to be strictly observed. 42. Project implementation. As much as possible of the physical work (both in new construction and in maintenance) ought to be turned over to the private sector. Most of the supervision will have to be done by technical assistance staff. Management, works supervision, and training tasks must be kept separate. IDA supervision missions ought to focus on essentials. Failure to submit correctly audited accounts, and diversion of project resources to non-project purposes, ought to be regarded as sufficient cause for immediate suspension of disbursements. Annex 1 SUMMARY DESCRIPTIONS OF THE FIRST. SECOND AND THIRD HIGHWAY PROJECTS 1/ 1. The First Project comprised construction to paved standard of the 79 km Kigali-Gatuna Road, and purchase of maintenance equipment. It was completed with a delay of 4.5 years. Construction work experienced difficulties, costs more than doubled, the project was reappraised, additional financing was provided and sections of the pavement failed. The road was not used (as originally intended) for international transit traffic because Uganda had not paved the section inside its own territory and would not allow heavy vehicles on it. The PCR concluded that the ERR was 16%, as against 13% estimated at appraisal but the Audit suggested that PCR conclusions on the ERR were optimistic. 2. The Second Project for Highway Maintenance helped finance the first phase of an 8-10 year highway maintenance program and comprised reorganization of maintenance operations and a training program; purchase of equipment and construction of a workshop; and improvement of routine and periodic maintenance of 2,200 km, and betterment and deferred maintenance of 600 km. The project was completed with a delay of 1.5 years and with a 41% cost increase. Neither maintenance (1,100 km actual) nor betterment targets were met (340 km actual). Traffic growth, however, was higher than expected and kept the reestimated rate of return at high levels (27% as against 28% estimated at appraisal). 3. The Third Project comprised construction of the Ruhengeri- Gisenyi road; feasibility and engineering studies; technical assistance; and training of maintenance staff. The project was completed with a delay of about one year. Some studies, the technical assistance and the training components were deleted when construction costs of the Ruhengeri-Gisenyi road escalated. The PCR found that the re-estimated ERR would be the same as that at appraisal because, in spite of higher than expected road construction costs, the value of vehicle operating cost savings had also increased. 4. These three projects were among the first Bank Group operations in Rwanda. Road works financed have been implemented with considerable delays and cost overruns. Reestimated ERRs range between 13%-27%, which are in line with appraisal estimates. Cost overruns were offset by gasoline price increases, by larger vehicle operating cost savings, and by much higher traffic than anticipated. The traffic increase on the major road built under the First Project was particularly high. The projects helped improve Rwanda's internal road network which at appraisal consisted largely of unengineered earth or gravel roads. 1/ The three projects were audited together. Please see OED Rep(ort No. 2545, of June 13, 1979. This Annex is an abstract of that audit. 19 5. Works execution experienced many difficulties due mainly to poor project preparation. In one instance, the quality of engineering was inadequate partly because it had been rushed. Consultant performance was unsatisfactory, indicating that they should have been replaced earlier than they eventually were. In another instance, engineering was only partially completed be,ore awarding a contract, and the delayed completion of engineering work led to delays in construction. Finally, the objectives of the maintenance project (Credit 299) were too ambitious in the light of local constraints, and were not achieved. Implementation schedules were over-optimistic. Credit effectiveness was usually delayed (in one instance by six months due to the time required for a formal agreement among Government, UNDP and IDA, and the appointment of a consulting firm for technical assistance). 6. In sum, the three projects were completed with time overruns of between 1 and 4.5 years and with cost overruns ranging between 16% to 144%. Although the Kigali-Gatuna Road was completed under the First Highway Project, its objective of serving as an import/export route was not attained. The road turned out to be justified by benefits not identified at appraisal, suggesting that the economic investigation preceding appraisal may not have been thorough. The road's catalytic role for regional economic development was not considered at appraisal. 7. Many of the difficulties arose because these were the first IDA operations in Rwanda. However, even after allowing for such difficulties, it is evident that the projects were inadequately prepared and IDA's control was not as effective as it might have been. The handling of engineering preparation of the First Project was haphazard, leading to increasing complications and cost increases. As a consequence, the project had to be reappraised when its cost almost tripled. Consultants responsible for poor engineering and supervision should have been relieved of responsibility far sooner than was the case. IDA's handling of the Third Project was also somewhat lax, appraising a project which had been only partially engineered, making several substantive revisions of the credit amount as a consequence, and finally not reacting decisively to the delays while a decision were being debated about the base material to be used. The target set for the Highway Maintenance Project was optimistic in view of local constraints with respect to manpower and budget and coordination with the delivery of equipment. NOTES 0N DENOGRAPHY AND ECONOICS 1. Rwanda, a small country (26.3 thousand km2)1/ with 6.7 million people in 1990, is surrounded by Zaire, Uganda, Tanzania and Burundi. It is poor (US$270 and US$310 per capita GNP in 1983 and 1988 respectively, among the lowest in the world), and backward: average life expectancy was 44 years in 1983 (the corresponding figure for *Low Income Africa, South of Sahara" and OMiddle Income Africa South of Sahara" was 48 and 52 respectively). Over 90% of the population subsists on exhausted and fragmented farmland. Coffee and tea are the main export crops and account for about half of the country's foreign exchange needs. The other half comes from concessionary aid. Throughout the 1970s, Government enjoyed large volumes of aid which began to dry up in the early 1980s. Beginning in 1983, balance of payments and budgetary difficulties reached such proportions that, given the extremely high rate of population growth,2/ little scope seemed possible for real growth. 2. Most of Rwanda's trade is not with its neighbors but with developed countries, and is highly imbalanced. Between 1985 and 1988, imports averaged about 250,000 tons annually, with only 50,000 tons of exports (out of which 30,000 tons was coffee). In the foreseeable future, trade volume is expected to increase by 5-6% annually, and to remain imbalanced. Under the circumstances, cheap and reliable accens to the sea is an important desideratum which, however, has not been attained for well over 20 years..3/ 1/ Figure given in Tables 3A of Annex 1 in the President's Report for both the Fifth and Sixth Highway Projects. The PR for the Transport Sector Project does not contain a Social Indicators Data Sheet. The SAR for the same project (para 1.01) says that "the country covers an area of 25,900 square kilometers". 2/ The SAR of the 1990 Transport Sector Project (para 1.r 1 says that the rate of growth of population is "over 3.0%". The %. try Data Sheet in the PR for the 1985 Sixth Highway Project says that the rate of growth is 3.7%. ~/ For a detailed discussion of sea access problems, please see the PPAR on Burundi -- Second.Highway Proiect (OED Report No. 6255 of June 9, 1986) and paras 1.06-1.09 of the SAR for Rwandese Republic -- Transport Sector Project (Report No.8386-RW of April 30, 1990). 21 3. The 1983 Rwanda Economic Memorandum: Recent Economic and Sectoral Deyelopments and Current Policy Issues summarized the country's prospects, and the summary remains valid in 1990. (a) Government's current expenditures exceed revenues; (b) programs and projects are started without much regard to their financial feasibility; (c) there is insufficient coordination and control of investment planning; (d) many agricultural programs have failed; (e) Government has long been regarded as the employer of first resort; (f) recurrent budget implications are divorced from investment planning decisions; (g) there has been a proliferation of money-losing parastatal enterprises; and (h) Government has paid little more than lip service to the population problem. 4. Rwanda is the most densely populated country in continental Afri,.a. Its 1983 population was estimated at about 5.7 million and population density in terms of agricultural land was almost 400 persons per km2. Birth rates had always been high but were offset by famines, communicable diseases and emigration to neighboring countries. There has not been a major famine since 1943, health conditions have reduced the impact of disease, and emigration has declined since the 1950s. The result was a steady increase of the population growth: late 1940s: 2.0% 1970-1978 : 3.3-3.5% 1978-1983 : 3.7% 5. In 1978, Rwanda's fertility was the highest in the world (8.3 children per woman) and population could reach 24 million in 2020. Given the country's small area and meager resources, the only hope for survival would be massive charity from abroad. The implications of population growth have been known for years but the steps taken to confront them have been purely cosmetic.4/ The problem is compounded by the fact that the vast majority belongs to the Roman Catholic Church and the general feeling is that secular exhortations regarding birth control fall on deaf ears as long as the Church remains unbending. / Para 1.35 of the 1983 Ryanda.Boonomic_Memorandum lists these steps: (a) creation of a Population Council in 1974 to study problems related to population growth; (b) establishment of a National Population Office in 1980 to plan, coordinate and monitor all population activities; (c) the start of a program to sensitize the population about the implications of population growth; and (d) the start of a few pilot family planning programs "to test the effectiveness of alternative service delivery methods and to determine the knowledge and use of family planning among the local population". 22 6. Over the past thirty years, foreign aid has been substantial.5/ Infrastructure has been laid down, electrification has spread and even earth roads are passable. However, Rwanda remains small, landlocked, with few natural resources, a dense population whose steady increase does not bode well for economic growth or political stability, and an absolute dependence on aid for budgetary support. Scarcity of work opportunities outside subsistence agriculture has inflated the public sector, but new recruits lack the rigorous vocational training received by the previous generation under the colonial system. Job security is what the public sector has to offer, but "privatization" has become the most recent panacea. 7. In order to achieve productivity and efficiency increases, so the argument runs, the private sector can use rewards and punishments which are denied to the public sector. The argument might have had more merit if there had been a clear distinction of ownership and decision-making between "private" and "public" sectors. In fact, the private sector contains large holdings of public figures and there is no mechanism to check conflicts of interest. A study of the real ownership of "international" transport companies would be a sobering experience for those who advocate privatization as the most promising solution for all social and economic evils. 5/ Aid has been massive but, also, indiscriminate and more attuned to donor preferences than to genuine needs. Rwanda's size has the advantage that a 100-km donor-financed road, or 50 donor-financed trucks, will be far more visible than 1,000 km, or 500 trucks, in a country like Zaire. Easy access to external aid has encouraged neglect of maintenance, and trucks that last 15 years in Europe will last no more than 4 in Rwanda. - 23 - ATTACHMENT Comments from the Borrower 91E0596 April 9, 1991 French (Rwanda) OED Dl PS:jl REV:EMcM Florin I. Vladescu Civil Engineer ILEX TO THE WORLD BANK, WASHINGTON (one page) Telex No. 1 202 477 7658 Kigali, April 4, 1991 Mr. Graham Donaldson, OEDD1 Chief, Agriculture, Infrastructure & Human Resources Division Operations Evaluation Department. Bef-: Fourth Highway Project (Credit 769 RW) Project Evaluation Report Your letter of January 31, 1991 Dear Mr. Donaldson: I regret to inform you that your letter took an extremely long time to reach me because it had been sent to the Ministry, where there is no "Direction G6n6rale AJS." I read the report with great interest and have no comments to add. There are, of course, certain minor details but these are not worth mentioning because they alter neither the spirit nor the overall accuracy of the report. Please be good enough to pass on my appreciation and congratulations to the authors of the report for their competent, clear and thorough work. Very truly yours, /s/ Florin Vladescu