Cameroon - Fifth Highway Project

Document ^ The Work HL, 1.t1t,,Itl!! ' C,, FOR OFF0 CIl t ktg-t No. 9866 PROJECT COMPLETION REPORT REPUBLIC OF CAMEROON FIFTH HIGHWAY PROJECT (LOAN' 2180-CM) AUGUST 28, 1991 Infrastructure Operations Division Occidental and Central Africa Department Africa Regional Office This document has a restricted distribution and may be used bv recipients only in the performance of S Ar *! I , * - I W, .I  COUNTRY EXCHANGE RATES Name of Currency: CFA France (CFAF) Rate at .Appraisal: 270 1982 .329 1983 .381 1984 437 1986 .449 1986 .346 1987 . 301 1988 .298 1989 .336 ABBREVIATIONS AND ACRONYMS ADT -- Average Daily Traffic AfDB/BAD -- African Development Bank/Banque africaine de developpement BADEA -- Arab dank for Economic Development in Africa/Banque arabe pour le d6veloppement hconomioue en Afrique CCCE -- French Development Fund/Caisse Centrals do Coop6ration Economiqua CIDA -- Canadian International Development Agency/Agence canadienne de developpement international CT -- Inter-Ministerial Technical Committee CTB -- Central Tender Board DH -- Department of Highways/Direction des Routes DSP -- Division of Studies and Planning in MINT EDF/FED -- European Development Fund/Fonds europ6en do d6veloppement FAC -- French Bilateral Assistance/Fonds d'Aide ot de Coop6ration (Franc.) Labog6nle -- National Civil Works Laboratory/Laboratoiro do G6nle Civil MINEP -- Ministry of Economic Affairs and Planning/Minlst&re du Plan *t do l'Am6nagement du Territoire MINEQ -- Ministry of Equipment/Ministere de l'Equipement MINT -- Ministry of Transport/Minist4re des Transports MINTPT -- Ministry of Public Works and Transport (croated in mid-1988 by combining MINEQ and MINT)/Minist&re des Travaux Publics et du Transport REGIFERCAM -- National Railroad Company of Cameroon/R69ie Nationale des Chemins do For du Cameroun SAR -- Staff Appraisal Report TOR -- Terms of Reference  T1 wouo oa.. . FOR OFFICIAL USE ONLY Was"O". 0 C. ao}} U.S A OGftbs of Dtw-Cm.ua August 28. 1091 MEMORANDUM TO THEECUTIVE DICO AND THE PRESIDENT SUBJECT: Project Completion Report on Cameroon Fifth Hi_hvav Prgiect (Lgan 2180-CM) Attached, for information, is a copy of a report entitled "Project Completion Report on Cameroon - Fifth Highway Project (Loan 2180-CM)" prepared by the Africa Regional Office. No aud-'t of this project has been made by the Operations Evaluation Department at this time. Attacbment Attachment  FOR OMCIAL US5 ONLY REPUBLIC OF CAMEROON FIFTH HIGHWAY PROJECT. LOAN 2180-CM PROJECT COMPLETION REPORT Table of Contents PaRe No. Preface ... . . . . . . . . . . . . . . . Basic Data Sheets . . . . . . . . . . . . . . . . . . . . ii Evaluation Suzmmary . . . . . . . . . . . . . . . . .v I. INTRODUCTION .... . . . . . . . . . . . . . . . . . 1 II. PROJECT IDENTIFICATION, PREPARATION AND APPRAISAL . . . . 2 III. PROJECT IMPLEMENTATION AND COST . . . . . . . . . . . . . 6 IV. ECONOMIC REEVALUATION . .1C V. PERFORMANCE OF CONSULTANTS ..12 VI. INSTITUTIONAL PERFORMANCE . . .13 VII. ROLE OF THE BANK ... .13 VIII. CONCLUSIONS AND LESSONS LEARNED . . . . . . . . . . . . . 15 Annex 1 Details of Economic Reevaluation . . . . . . . . . . . . . 17 Annex 2 Cash Flow Tables and Supporting Data . . . . . . . . . . . 20 TABLES 2-1 Critical Path of the Project .. .30 2-2 Cofinancing by Lots . . .31 3-1 Cost Estimates at Appraisal and Completion . . . . . . . . 32 3-2 Sources of Financing ..33 MAP: IBRD 15470R Fifth Highway 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 CAMEROON FIFTH HIGHWAY PROECT. LOAN 2180-CM PROJECT C!QPLETION REPORT PREPACE 1. This is the Project Completion Report for the Fifth Highway Project, for which Loan 2180-CM in the amount of US$70 million was approved on June 15, 1982. The loan was fully disbursed on February 27, 1989 and was closed June 30, 1989, two years behind schedule. Several co-lenders participated in the financing of the project. 2. The PCR was prepared by the Infrastructure Operations Division, Occidental and Central Africa Department, Africa Regional Office, and is based, inter alia, on the Staff Appraisal Report, the Loan Agreement, Bank files, interviews with Bank staff assigned to the project, and a field visit in June 1990. 3. The PCR was read by the Operations Evaluation Department (OED). The draft PCR was sent to the Borrower on June 14, 1991, for comments by July 26, 1991, but none were received.  - il - REPUBLIC OF CAhiEROOW FIFTH HIGHWAY PROJECT. LOAN 2180-CM PROJECT COMPLETION REPORT BASIC DATA SHEETS KEY PROJECT DATA Appraisal Item Expectation Actual Total Project Cost (US8 million) 216.8 16S.7 Cost Underrun (U) -- 16.9 IBRD Loan Amount (US$ million) - 70.0 Disbursed -- 7.6 La Cance lld -- o RepTid (as of 03/03) 11.5 Outstanding (as of 03/90) -- 68.6 Date Physical Components Coepleted 12/81/86 12/81/88 La Proportion Completed by Abovo Date (X) 100 1e Proportion of Tim Overrun (X) so Lo Economic Rate of Return (3) 24.5 16.7 Institutional Performance Fair CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS (USB million) item FY83 FY84 FY85 FY86 FY87 FY88 FY89 Appraisal Estimate 14.0 26.0 42.6 60.0 70.0 -- -- Actual 9.8 16.2 24.7 51.8 68.6 64.9 70.0 Actual as X of Estimate e6 64 69 86 84 98 10l OTHER PROJECT DATA Item Original Plan Revisions Actual First Mention In Filee 69/01/77 Government's Application N.A. Negotiations 08/8s ok -- 08/25-29/82 Board Approval of the Loan 16/86 Lb __ 08/15/82 Loan Agroement Date -- -- 07/26/82 Effectiveness Date -- -- 01/21/88 Closing Date 08/86/87 66/30/88 06/30/89 Borrowor Republic of Cameroon Executing Agency Ministry of Equipment (MINEQ) Fiscal Year of Borrower July 01--June 80 Follow-on Project Sixth Highway Project Loan Number Loan 2584-CM Amount (USS Million) USS12S Million Loan Agreemnt Date November 19, 1985 /a Original project works, except for the Labog6nie building, were completed by the end of 1987. Lb Schedule at tim of Project Brief dated O:tober 12, 1979.  - ltl - HISSXON DATA No. ot Mission man It.. Month/Year Persons Composition LO Weko Date of Report Identitication /b Preparation 67/78 1 H 6.5 08/02/78 Preparation 16/78 2 H, E 6.5 11/16/78 Preparation 02/79 2 H, E 1.0 64/12/79 Preparation 11/79 1 E 1.5 11/27/79 Preparation 12/79 1 H 0.6 61/11/86 Appraisal lO/80 8 214, E 6.0 11/14/66 /e Post Appraisal 08/S1 2 2E 1.0 65/84/81 H at Appraisal 06/81 2 2H 2.0 67/17/81 2osi. Appraisil 09/81 1 H 0.6 10/27/81 Subtotal 18.5 Supervision 1 04/82 2 H, E 2.0 67/28/82 Supervision 2 08/82 a SH 6.6 69/14/82 Supervision 8 1/82 1 H 0.2 12/68/82 Supervision 4 01/83 1 H 1.0 08j8O/88 Supervision 5 02/88 1 H 2.0 67/19/88 Supervision 6 16/88 1 H 1.0 01/19/84 Supervision 7 68/84 2 H, E 0.6 04/18/84 Supervision 8 1/84 1 H 0.6 12/28/84 Supervislon 9 03/85 2 H, E 1.0 64/10/85 Supervision i 64/85 1 C 2.0 07/09/85 Supervision 11 09/86 2 H, E 1.6 1/07/86 Supervision 12 11/85 2 H, E 2.0 12/18/85 Supervision 18 02/86 2 H, E 2.0 64/16/88 Supervision 14 065/88 1 E 0.2 06/20/88 Supervision 16 66/86 1 H 1.0 07/17/86 Supervision 16 10/86 2 H, E 2.0 11/20/86 Supervision 17 68/87 1 H 0.5 64/16/87 Supervision 18 11/87 2 H, E 1.6 12/21/87 Supervision 19 s6/88 2 H, E 1.O 08/08/88 Supervision 20 68/89 1 E 0.2 08/28/89 Subtotal 27.6 Total 41.1 am lo E a Economist; H - Highway Engineer; C a Consultant Lb Project was Identified In the courso of earlier project work In 1976 (see pars. 1.02) /c Issue Paper  - Iv - STAFF INPUT (Staff Weeks) Flscal Yert FY81 FY82 FY38 FY84 FY68 FY86 FY87 FY68 FYS9 FY06 Total Preppralul 16.4 -- -- -- -- -- -- -- -- 16.4 Appraisal 69.4 18.6 - - -- -- -- -- -- -- 83.0 Negotiatlons -- 19.9 -- -- -- -- -- -- -- 19.9 Supervision -- -- 6.9 16.5 14.0 12.0 6.9 2.9 4.2 -- 67.4 Other 0 _ , -- -- -- _ -- 0.4 0.1 0.5 Total 35.8 88.6 0.9 16.6 14.0 12.6 6.9 2.9 4.6 6.1 177.2 Date for FYo 78-80 not available. AFlIN, August 8, 1990 A:da*..pcr (MM)  - v - REPUBLIC OF CAMEROON FIFTH HIGHWAY PROJECT. LOAN 2180-CM PROJECT COMPLETION REPORT EVALUATION SUMHARY Introduction 1. The main component in the Fifth Highway Project was the construction of the Edda-Yaounde section of the Douala-Edda-Yaound6 Road (281 km). The new alignment proposed (245 km) was about 36 km shorter than the existing road. The existing paved section from Douala to Edda consisted of a low standard road which was falling apart and badly needed reconstruction. The existing laterite- surfaced section between Edda and Yaounde followed a poor alignment and was frequently closed to traffic during the rainy season. Even in the dry season, the road was in poor condition and, given the heavy traffic, was difficult to maintain (para. 1.02). 2. The 1976 rainy season (July-November) was particularly critical .or transport operations between Douala and Yaound4. The road was completely impassable for an extended period and rail traffic was dislocated by accidents, derailments, lack of motive power and a landslide. As this situation was becoming a national issue, the President announced in October 1976 that an expressway was to be built between Douala and Yaounde to provide a secure means of transport between the two cities, and that construction was to start almost immediately. In January 1977, a firm of consultants was awarded a contract under local financing for the technical feasibility study and detailed engineering of a new expressway (para. 1.03). Obiectives 3. The objectives of the project were to assist the Government in reconstructing the Edda-Yaoundd section of the Douala-Ed4a-Yaound4 Road, to improve the Government's capacity for road rehabilitation and transportation planning and coordination, gnd Lo lay the foundation of a maintenance and rehabilitation program for the country's paved road network (para. 2.20). Implementation Experience 4. By the end of 1987, all project components had been successfully completed with the exception of the Labog4nie headquarters building, and the additional works included :i November 1985. Further preparation work was carried out for MINT which provided a lasis for the Sixth Highway Project. The Closing Date was amended twize, lastly to June 30, 1989, to accormodate the delays in completing the project (para. 3.15). 5. The Bank agreed that some remaining disbursements for the Labogdnie headquarters building be covered under the Sixth Highway Project. Certain works relating to the Douala-Yaoznde Road are also being financed under that project. Disbursements were somewhat behind the appraisal estimate because of the delay in start-up of construction of the Bank-financed Lot 7, and the strength of the US dollar to the CFA franc in which the contract for Lot 7 was denominated (para. 3.16).  - vi - 6. The Bank-financed works were completed with a saving of about US$20 M (para. 3.11). These savings were used to carry out substantial additional works on the project road, as well as for continuation of technical assistance and training. The total cost of the completed Bank-financed project underran the original estimate by about 172 in dollar terms, but overran the estimate in CFAF terms by about 12? (para. 3.17). Results 7. The economic rate of return (ERR) for the Edea-Yaounde road section in the Bank-financed project is 10.72, or about one third lower than the rate estimated by the consultants. The lower ERR was the result of higher construc- tion costs and less than projected traffic. The ERR in the Staff Appraisal Report was 24.52, which in retrospect appears over-optimistic. A comparison with the ERR given in the SAR is not appropriate because of the unavailability of the working tables related to cash flows (paras. 4.05 and 4.06). 8. Consultants played a substantial role in carrying out the project. The many consultants, including firms and individuals, generally performed well on all components of the project (para. 5.01). 9. The Government's Central Tender Board was not particularly responsive to the needs of the individual ministries it was supposed to serve, nor was it inclined to follow Bank guidelines on procurement. This led to some difficulties and delays which were eventually resolved (para. 6.03). 10. The project was thoroughly prepared by the Bank, which also par- ticis ated in seeking and arranging the substantial amount of co-financing required for the Douala-Yaounde Road. Supervision by the Bank contributed substantially to the successful completion of the project (paras. 7.02 to 7.04). Sustainability 11. The project road section is sustainable both in terms of benefit generation and asset conservation, provided that the overall Douala-Yaoundt Road is adequately maintained. The Road is a direct beneficiary of the substantial involvement of the Bank in upgrading road maintenance operations and institutions both before the Fifth Highway Project, and followed-up even more intensively under the Sixth Highway Project. Further upgrading of maintenance is likely to be a major component in a transport sector project currently under preparation (para. 4.13). Findings and Lessons 12. In persuading the Government to lower the concept of the road from a 'our-lane expressway to a two-lane trunk highway, the Bank acted in the Borrower's best interests in containing costs and avoiding a very substantial debt service burden in coming years (para. 8.08). 13. The Bank played its role as "Chef de File" well concerning the securing of cofinancing but had to limit that role for the construction phase (para. 8.09).  - vii - 14. The lessons learned aret (a) the Bank's insistence on thorough project preparation, its efforts to carefully coordinate co-financing, and its close supervision of the project contributed considerably to successful project execution; (b) substantially qualified bids should be immediately rejected in line with Bank guidelines; (c) the appointment of an overall consultant to complete the design and supervise the construction of the Douala-Yaounde Road would have expedited the work, especially during its early stages of preparation and execution; this would have avoided some of the design problems which arose on the overall road after some of the contractors had started work; and (d) caution is indicated for the Bank assuming a "Chef de File* role for project elements which. it does not finance, as happened for the Douala.-Edda section; substantial cost overruns (532) on this section compare unfavorably with the minimal (32) overrun on the section with Bank financing. On future projects, more use should be made of counterparts to provide junior engineers and economists with on-the-job experience (para. 8.10).  REPUBLIC OF CAMEROON FIFTH HIGHWAY PROJECT, LOAN 2180-CM PROJECT COMPLETION REPORT I. INTRODUCTION 1.01 The first three Bank highway projects concentrated on the construction of trunk roads, while the fourth was directed at strengthening road maintenance capacity. The audit reports note that physical objectives of all four projects were satisfactorily met but, because of inflation, costs of the second project were nearly double. The PCR for the Second and Third Eighway Projects found that 'progress in developing transport related institutions had been slow .... The PCR for the Fourth Highway Project noted that institution building, planning and vehicle control did not promise sustainable benefits. 1.02 The main component in the Fifth Highway Project was the construction of the Edea-Yaounde section of the Douala-Ed4a-Yaound4 Road (281 km). The new alignmant proposed was 245 km long and 36 km 1/ shorter than the existing road. The existing section from Douala to Edda consisted of a low standard paved road which was falling apart and badly needed reconstruction. The existing laterite-surfaced section between Edda and Yaounde was little more than a track. It followed a poor alignment and was frequently closed to traffic during the rainy season. Even in the dry season, the road was in poor condition and given the healry traffic, was difficult to maintain. 1.03 The 1976 rainy season (July-November) was particularly critical for transport operations between Douala and Yaounde. The road was completely impassable for an extended period and rail traffic was dislocated by accidents, derailments, lack of motive power and a landslide. As this situation was becoming a national issue, the President of the Republic announced to the National Assembly in October 1976 that an expressway was to be built between Douala and Yaounde to provide a secure means of transport between the two cities, and that construction was to start almost immediately. 1.04 The Douala-Yaounde corridor was already served by the National Railway Company of Cameroon (REGIFERCAM) which by mid-1979 had recei%bd four loans from the Bank. Bank files indicate that from the time the project was first considered by the Bank, the effect of competition from the proposed road on the railway was a matter of concern. 1.05 In the past, resource allocations in the transport sector were influenced by the Government's regional development policy which was aimed at satisfying socio-political demands. The project included the resources needed to assess transport sector needs and priorities, and to strengthen its management. 1.06 The Ministry of Equipment (MINEQ) 2/ was given responsibility for the implementation of all project components, with the exception of the 1/ SAR, para. 4.03. 2/ In 1988, MINEQ was merged with MINT into a new Ministry of Public Works and Transport (MINTPT).  -2- transportation survey and technical assistance to the Division of Studies and Planning in the Ministry of Transport (MINT) 3/ for which the latter was responsible. The road component of the project was supervised by the Department of Highways (DH) in MINEQ. II. PROJECT IDENTIFICATION. PREPARATION AND APPRAISAL Identification and Preparation 2.01 Reconstruction of the Douala-Yaoundd Road had the highest priority in the Government's Fifth Development Program. Preliminary engineering of the road was started in 1977 under the Fourth Development Plan when the Government engaged a consultant to carry out the engineering for a four-lane divided expressway. However, the expected traffic volume did not justify these high standards at this time, and after extensive consultations between the Bank and the Government, the latter finally instructed the consultant to reduce the design to the first two lanes, with the option of adding two further lanes at a later stage. The Bank prepared the terms of reference for a preliminary engineering and economic feasibility study, and agreed to finance it under the Fourth Highway Project. 2.02 In mid-1978, prelluminary discussions with the Government included the location of the alignment, the selection of design standards, the estimated traffic and cost. The project cost was tentatively estimated at US$225 H equivalent, including price contingencies and construction supervision. As a result of differences between the Government and the Bank over the Douala-Yaounde corridor study, relations between the Bank and the Government were noted in the files as having been strained. This may account, in part, for the Bank's unusual acquiescence in November 1978, months before the feasibility study was scheduled to become available, to the Government's proposal to begin construction work on the road ir the dry season (November) of 1980. The detailed engineering and bidding documents were expected by August 1979. Appraisal was proposed in May 1979 which would allow ample time for mobilizing co-financiers and for bid opening before Board presentation for a start of construction in late 1980. 2.03 By mid-1979, two issues had been resolved. The Government had (a) accepted a two-lane instead of a four-lane expressway; and (b) given up the idea of a toll road. The Bank concluded that a toll would limit vehicle access to a few intersections along the road, thereby reducing the road's development impact. Six major issues remainedt (a) the overall efficiency of road maintenance operations; (b) overall investments in the corridor; (c) the impact of the road on the railway; (d) detailed design of the road; (e) the penetration of the road into Douala; and (f) project financing. The Project Brief listed the critical path for preparing and completing the project (Table 2-1). 2.04 Perhaps the most difficult issue was the impact of the road on the railway. The Bank decided that it should limit its objective to ensuring fair ces..petition between the two modes. The studies (para. 2.01) showed that if designed as a two-lane trunk road, traffic would divert to the road to an extent diminishing railway revenues by 12-15Z, and that a diversion of such an amount was economically justified. However, if built to expressway standards, even 3/ See footnote 2.  initially to two lanes, a larger diversion would result, as even at peak hour its capacity would be utilized no more t.an 202. Thus, resource allocation in the Cameroonian economy would be negatively affected, not only by a substantial investment in an uneconomic expressway, but through its causing the diversion of railway traffic to the road beyond the optimum. 2.05 The first meeting of the co-donors for the Douala-Yaounde Road was held in Yaounde in July 1979. The road works were divided into nine lots. The French Government and EDF picked a lot considered suitable for their financing, while other representatives, who participated as observers, could not make commitments. The Government agreed that the Bank would contact potential donors not present at the meeting to explore their interest in financing the road. It was also agreed that the meeting would reconvene in October. The Bank promptly contacted a number of countries and the African Development Bank (AfDB) to sound out their interest in the project. 2.06 In October 1979, the Bank learned that there had been some slippage on the part of the consultant in preparing cost estimates and draft bidding documents for the project in part because of non-payment of consulting fees. A further update of the Project Brief was then issued to reflect the last information and schedule. 2.07 A second co-donors meeting called by the Government was held in Paris in December 1979 but like the first, it was not successful in closing the financing gap. However, a third meeting in Brussels in May 1980, organized by the Government with Bank assistance was successful in obtaining financing from 10 co-donors for all of the nine lots comprising the Douala-Yaounde Road (Table 2-2). The Bank undertook to finance Lot 7, and possibly Lot 8 for which firm financing had not yet been secured. Later, the Bank was asked to act as "Chef de File' to oversee coordination among all co-financiers, and ensure smooth implementation of the Douala-Yaoundd Road. 2.08 Prior to the project, the Bank and the Government had been engaged in a dialogue on the need to strengthen transport sector management. The Government was quite responsive to the need for planning, but initial attempts to reinforce the organization and administrative machinery for planning and policy-making had met with limited success. Developments before project appraisal indicated a more positive attitude. Government reorganization in late 1979 placed MINT in charge of transport sector planning. MINT planned to enlarge the scope of its transport planning unit and to upgrade it to the Division of Studies and Planning (DSP). 2.09 The Government planning effort had, in the past, been based on a project-by-project approach which was adequate when the transport investment alternatives were clear cut. Now, however, more sophistication and rational procedures for sectoral allocations were needed because increasingly complex investment choices had to be made to sustain Cameroon's generally healthy economic growth. The evolution of a systematic, long-term plan had also been hampered by scarce technical expertise. Concerned by these shortcomings, the Government requested Bank financing for a national transport survey to provide it with a longer-term perspective of the sector's needs. This was in line with the Bank's sectoral objectives which continued to be two-fold: (a) to develop the Government's transport planning capacity; and (b) to strengthen the management of the transport sector, particularly the efficient utilization of existing infrastructure, coordination of transport modes, and improved project implementation and monitoring. The survey was expected to form the basis for the  -4- transport component of the next development plan and influence the actual investments made in the sector in the last two years of the Fifth Development Plan and subsequent plans. Appraisal 2.10 The project was appraised in October 1980. Discussions were h& d in Paris with the consultants who had designed the Douala-Yaounde Road, as they were no longer working in Cameroon. Thus, at a critical time in the preparation of the project, no technical staff was locally available to the Government. The Bank engaged an independent cost consultant to carry out a cost analysis of the project. The Government had already called for tenders or awarded contracts for Lots 1-5 from Douala to Edda (Section I -- 64.8 km). There was no supervision engineer on the sites, but the Government indicated that it would award contracts shortly for the supervision of the lots in Section I. 2.11 The Edda to Yaounde section (Section II -- 180.5 km) was divided into Lots 6-9, of which construction and supervision of Lot 7 (65.3 km) had been reserved for the Bank, as had supervision of Lots 6-9. A consortium of Arab funds, the Netherlands Government and CIDA had already indicated an interest in the remaining lots (6, 8 and 9, respectively). As work on Section I had already started, the project was limited to Section II, construction of buildings for the National Civil Works Laboratory (Labogenie), and technical assistance to MINEQ and MINT. As considerable uncertainty remained regarding the cost of Lot 7, especially concerning the quality of the quarries, the Issues Paper recommended that negotiations not be held until bids had been received. This proposal did not delay project processing; in fact, it allowed the Government to proceed with a tender call in parallel with the appraisal process. 2.12 Two main issues were noted in the Issues Paper: (a) the role the Bank had been asked to play by the Government as "Chef de File" (para. 2.07) which resulted in the scope of the project as appraised being reduced (para. 2.13); and (b) the impact of the Douala-Yaounde Road on the Railway (para. 2.14). 2.13 It was noted that the Government had repeatedly asked the Bank to be "Chef de File' for the Douala-Yaounde Road construction. Co-donors had also expressed interest in continued coordination efforts by the Bank beyond project preparation and establishment of the financing plan. The Decision Meeting agreed that the Bank should accept in principle the Government's request to assume a -wider advisory role for the construction of the entire Douala-Yaounde Road. However, caution was expressed as to how far the Bank should get involved in the execution of work over which it would have little control. For this reason, the project included only Section II. To avoid any possibility that one or more lots of Section I might not be completed, and thus isolate the Bank-financed lot, the Issues Paper recommended that the Bank should require as a condition of effectiveness, that contracts for the non-Bank-financed lots of Section I be awarded, and that they be completed within a reasonable time limit. 2.14 As regards the impact of the Douala-Yaounde Road on the railway, it was anticipated that the road would result in substantial journey time savings, increased passenger and cargo safety, and service convenience which would divert  some traffic away from the railroad. About a third of the total truck traffic estimated on the improved road was expected to be diverted away from the railway. This, combined with loss of some passenger traffic, was expected to reduce Regifercam's revenues by about 12-15Z, potentially a serious loss. 2.15 In February 1981, the Bank conveyed to the Government its concern that it appeared to be awarding construction contracts without fully completed detailed engineering. It once again urged the Government to promptly appoint consultants on the Edea-Yaounde section to ensure that all aspects of the design and documentation were completed at the time of tendering. 2.16 The files indicate that, for various reasons, including conflicts between the Central Tender Board (CTB) and MINEQ, several months were lost in appointing consultants for Section II and in calling for bids on Lot 7. MINEQ had only one expatriate engineer to assist in this work and the Bank considered appointing a highway engineer to its Yaoundd Resident Mission to assist the Government. It was noted that work on Section I had started but serious problems had developed because of inadequate engineering; hence, the Bank's effort to expedite the appointm$t of consultants on Section II to avoid a similar situation. The visit of a senior Bank official in June 1981 was successful in securing Government action on a number of these matters of concern to the Bank. 2.17 In August 1981, the Minister of Equipment visited the Bank to discuss the procurement procedure problems the Bank was having with Cameroon. The main issue was the country's rule to open bids behind closed doors which contradicted the Bank's requirement for public bid opening. The Minister suggested that a representative of the Bank attend bid openings for projects in which its financing was involved. The Minister promised a proces-verbal of bid openings which spelled out the bids received, which the Bank could make public if it desired. Bank management later reluctantly accepted this arrangement for the Fifth Highway Project only. 2.18 Negotiations took place late in March 1982. The bid of the lowest bidder for Lot 7 was accepted, provided that the bidder waived certain reserva- tions related to the construction period and conditions of payment. The Bank agreed to the further strengthening of Labogenie's capacity, and to include fellowships for MINEQ to further enhance the training aspect of the project. The loan amount remained at US$70 H as proposed earlier. Retroactive financing was provided to finance urgently needed consultant services. The loan was approved by the Board on June 15, 1982, and became effective on January 21, 1983. 2.19 The objectives of the project were to assist the Government in constructing a section of the Douala-Edea-Yaounde Road, to improve the Government's capacity for road rehabilitation and transportation planning and coordination, and to lay the foundation of a maintenance and rehabilitation program for the country's paved road network. 2.20 To achieve these objectives, the project as finally defined consisted of: (a) civil works: (i) construction of the Edea-Yaound6 section (180.5 km) of the Douala-Yaounde Road; and  -6- (ii) construction of new buildings for Labogenie. (b) consultant services: (i) to supervise construction of the Edea-Yaoundb section; (ii) to prepare a national transport survey; and (iii) to evaluate the existing paved road network. (c) technical assistance to strengthen: (i) the DH in the MINEQ; (ii) the Planning Department in the MINT; and (iii) Labogenie. The project was expected to be completed by December 31, 1986 at a total cost of US$218.8 M equivalent. The total cost of the Bank-financed works was US$86.8 H. III. PROJECT IMPLEMENTATION AND COST Proiect Start-Up 3.01 By mid-1982, the contractors for the remaining works on Section I had been selected, and works were either underway or about to start on Section II. As the contractor for Lot 7 had been selected and the supervision consultants for Lots 6 to 9 chosen by the time of negotiations, execution of the Bank-financed civil works seemingly got off to a good start. However, as the low bidder for Lot 7 refused to waive some of the qualifications he had submitted with his bid (para. 2.18), his bid was eventually disqualified. The lot was re-tendered, the list being restricted to earlier bidders, except for the disqualified bidder. Further difficulties arose in adjudicating the award. Thus, unfortunately, a contractor was not selected until early in 1983 which at that time placed the Bank-financed Lot 7 on the critical path to the opening of the Road. 3.02 Terms of reference (TOR) for the national transport survey and evaluation of the paved road network were agreed at negotiations and by mid-1982 short lists of consultants were being prepared by the Government. A contract to assist MINEQ had been awarded in mid-April 1982. By mid-1982, MINT had agreed on the TOR for two experts and for short-term specialists and was pursuing their recruitment through the CTB. The same applied to Labogdnie for which TOR for four experts had been agreed with the Bank. The Government was preparing a list of equipment for Labogenie to be procured under the project and a list of possible candidates to receive fellowships. 3.03 HINT, for some unknown reason, invited proposals for the above two studies from short lists of consultants, and under TORs other than those which had previously been agreed with the Bank. Thus, late in 1982, MJTT had to ask the consultants to withdraw their proposals, prepare new short 13 ,, and follow the TOR agreed on earlier. The remaining components were proc .ng satisfac- torily and on schedule.  -7- Progress during 1983 to 1984 3.04 The February 1983 supervision mission reported that there were no major problems with the project's implementation. Work on all lots of the Douala-Yaounde Road was underway. Lot 7 was still the critical section for the overall road, although it was expected to be completed by the project completion date. The Bank had agreed to the short lists for the transport survey and the technical assistance to MINT, but the short list for the pavement study had not yet been prepared. However, technical assistance for MINEQ was in place. The consultant for the technical assistance to Labogdnie had been selected and bids for the equipment would be invited once the consultant was in the field. 3.05 By mid-1983, construction of both sections of the Douala-Yaound4 Road was generally well in hand, but some problems had emerged on Section I which was outside the Bank-financed project. Bids were in for the construction of the Labogenie buildings, ani the technical assistance team for Labogdnie had started to mobilize. Documents for inviting proposals for the two studies and technical assistance to MINT were in the process of being approved by the Bank. By late 1983, the construction of the Douala-Yaounde Road was making good progress, and proposals for the two studies were being evaluated. 3.06 By early 1984, about 45Z of the total construction of the Douala-Yaound4 Road had been completed. Contracts for the two studies had been awarded after a delay of about one year. The Bank was particularly concerned over the delay in launching the paved road study because it provided the basis for the Sixth Highway Project, which, at that time, was under preparation. 3.07 Towards the end of 1984, the Douala-Yaound6 Road was about 65Z completed, with the Bank-financed lot 75Z completed. The standard of workmanship of the Edea-Yaounde section was considered satisfactory except for the quality of the paving of Lot 8, which was later corrected. Construction of the buildings for Labog6nie had been delayed, but training for laboratory staff was proceeding well. Work had started on the studies, and technical assistance to MINT was satisfactory. Progress from 1985 to Completion 3.08 The March 1985 mission noted that the technical assistanie (coordina- ting consultant) financed under the Loan to assist MINEQ in managing the Douala- Yaound4 highway construction, was an important element in the overall progress of the project. The management information provided by the consultant's periodic reports had enabled MINEQ to monitor construction progress in a manner that would ensure "on time" completion of all sections. The mission noted that it was only unfortunate that there had been no adequate counterpart involvement with the technical assistance team so that the control techniques used could be carried on by the Ministry for future projects. 3.09 The pavement study was behind schedule but was now progressing satisfactorily. The transport survey was progressing well but technical assistance to MINT had to be temporarily halted due to the departure of the team leader for health reasons. Construction of the Labogenie building in Douala was nearing completion, but property acquisition problems had so far prevented start of construction of the new headquarters building in Yaounde. Training of Labog4nie staff was being satisfactorily carried out on the job and through fellowships in France. Some of the laboratory equipment had already been  ordered. No progress had been made in designating suitable fellowship candidates for the DH in MINEQ because political aspects had been introduced in the selection process. 3.10 In April 1985, a thorough supervision of the Douala-Yaounde Road, including the Bank-financed project, was made by a consultant. He reported that construction on the Douala-Yaounde Road was on schedule and it was expected to be open for traffic at mid year. Overall quality of the work was good, and the construction supervision consultants were maintaining reasonable control over the work. 4/ A substantial claim from the contractor on Lot 7 was under review. The Bank informed MINEQ that any settlement involving the use of loan funds would require Bank concurrence, and that the Bank, prior to any such concurrence, would require detailed analysis by the supervision consultants in support of the settlement. 3.11 The September 1985 mission found that the paved roads study was progressing well and the transport study was in its more intensive phase with the draft final report due in early 1986. The Bank subsequently approved the use of about US$20 H in anticipated savings for the additional work to improve traffic safety and a new bridge at Edda to replace an existing bridge. The savings were due to the favorable rate of exchange of the US dollar to the CFA franc. A minor amendment of the Loan Agreement was made to accommodate the above changes. 3.12 By the end of 1987, most of the road construction works had been completed. The contractor's claim on Lot 7 had been settled satisfactorily. The Douala building for Labogenie was expected to be completed by mid year, while a contract for the headquarters building was expected to be awarded shortly. The paved roads study was proceeding without major problems. The draft transport study had been reviewed in the Bank. 3.13 A supervision mission of July 1986 found that training of 30 MINEQ technicians in France had been completed. The level of proficiency of the trainees was generally good. Preparation of other training activities in Cameroon was progressing satisfactorily. 3.14 The October 1986 mission reported that all loan funds were committed. The Douala building for Labogenie was completed, but start of the Yaounde building had been further delayed because of an expropriation problem. The paved road study had been delayed because more road sections had been added. Technical assistance to MINT was progressing well and the transport study had been accepted. The report noted that as the value of the CFAF had recently increased in relation to t'.e US dollar, the loan funds would not cover all of the additional work epproved earlier. These were deferred to the Sixth Highway Project. 3.15 By the end of 1988, all project components had been successfully completed with the exception of the Labogenie headquarters building, and the additional works included in November 1985. Further preparation work was carried out for MINT which strengthened the basis for the Sixth Highway Project. The Closing Date was amended twice, lastly to June 30, 1989, to accommodate the 4/ The consultant noted the deficiencies in the pavement for Lot 8 (para. 3.07) and added that there had not been any other lapses in supervision.  - 9 - delays in completing the project. The execution of Section I, vhich was not included in the Bank project, was less problem free, and work was still going on to stabilize the road over a swampy area near Douala. Disbursements 3.16 The Loan account was closed on February 27, 1989 when the Loan was fully disbursed. The Bank agreed that some remaining disbursements, particularly relative to the Labogdnie headquarters building be covered under the Sixth Highway Project. Certain complementary works relating to the Douala-laounid4 Road are also being financed under that project. Disbursements were somewhat behind the appraisal estimate because of the delay in start-up of construction of Lot 7 (para. 3.01), and the strength in 1983-86 of the US dollar to the CFA franc in which the contract was denominated. Project Cost 3.17 The cost estimates of the project at appraisal and completion are shown in Table 3-1. Both the appraisal and completion costs are shown in US dollars and CFAF. The table indicates that while the cost of the project increased by 12.3Z in CFAF terms, the dollar costs were 16.9Z lower because of the strength of the US dollar during project implementation, as compared to the rate at appraisal. 3.18 The sources of financing for the project are shown in Table 3-2. It should be noted that as the actual disbursements of each co-donor are not known, their contributions were deduced from the Borrower's PCR on the assumption that the co-donors financed only the foreign costs. Project Maintenance 3.19 A drive over the Douala-Yaound6 Road shortly after it was completed indicated that the road was in reasonably good condition, except for some settlement and pavement failures on Section I (para. 3.15). A more recent inspection indicated that the road was not being adequately maintained and needed grass cutting on the side slopes, repair of potholes and slides, and further failures in the swampy area referred to above. The Bank has been endeavoring to persuade the Government to focus more attention on maintenance, and recently reallocated funds under the Sixth Highway Project to repair the failures noted above. 3.20 A drive over the Douala-Yaounde Road in June 1990 indicated that three separate sections totaling three km of pavement had failed on Section I. Repair work was being performed under a second amendment to the original construction contract. A third contract amendment was to be let out shortly, and the expected completion date for this section is December 1990. A landslide had occurred about km 65 from Yaounde on Section II; a contract to remove the slide was to be awarded shortly.  - 10 - IV. ECONOMIC REEVALUATION Road Works 4.01 As mentioned in para. 2.11, the Bank project was confined to the Edea-Yaounde section. However, for economic evaluation purposes, the Douala-Edda and Edea-Yaounde sections form a single road and were treated as such in the SAR. They were evaluated first as a single road and then as two separate sections. The same procedure was followed in this report. Traffic forecasts and construction cost estimates represent the two key elements that determined the outcome of the revised project ERRs as discussed below. Traffic Forecasts 4.02 For the Douala-Edda section, MINTPT counts in 1987 showed an ADT of 2,352 compared to the consultant's 1978 forecast of ADT 3,192. For the Edea-Yaounde section, the consultant's forecasts turned out to be remarkably accurate with a 1987 count of ADT 2,219 compared to a forecasted ADT of 2,231 for the same year. In traffic composition, however, there were considerable differences. In particular, actual normal traffic volumes were much lower than forecasted amounts, while induced traffic volumes were much higher. The net result of these variations was a reduction in project benefits because per vehicle km benefits for induced traffic count only half as much as those for normal traffic. 4.03 Also, the diversion from rail to road as a result of the project turned out to be larger than forecasted. However, the share of this traffic in the total is small, and so is the benefit coefficient of the diverted traffic units, so that the impact of this particular traffic volume variation in the revised ERRs was not significant. Proiect Cost Estimates 4.04 Project costs are discussed in Annex 1, while the cash flow tables are given in Annex 2. Economic Rates of Return (ERRs) 4.05 The following is a summary of the ERRs as they were calculated at various times and by various concerned parties. ERR. In X As Calculated by Y.ar of Calculation Douala-Ed6a E. a-Yaound6 Douala-Yaound6 (Section I) (Section TI) The Consultant. 1978 16.7 16.2 16.9 SAR (1981/82) 1981 28.4 24.5 28.4 Covtrneent's PCR 1989 9.1 14.8 11.7 Prosent PCR 1990 8.0 10.7 9.4 (See Annex 2 for the cash flow tablso.)  - 11 - 4.06 The ERRs calculated by the feasibility consultants are included in this report because details of the calculation of the ERRs in the Staff Appraisal Report are not available. The censultants' ERRs are thus considered a better basis for comparison. 4.07 The differences in ERR results between those arrived at in the Government's PCR and the Bank's PCR on one hand, and those in the consultants' study estimates are considerable. About two thirds of the differences can be explained by higher construction costs mainly on Section I, compared to the consultants' original estimates, and the remaining third by shortfalls in traffic projections compared to actuals. 4.08 There are slight divergences between the ERRs in the present report and those arrived at in the Government's PCR. They can mostly be explained by differences in vehicle operating costs and cost differentials. The values used in the present report are those obtained from the consultant's original report adjusted to the 1990 price level. These are lower than those used in the Government's PCR and result, therefore, in lower benefit aggregates. Also, the Government's PCR assumed residual values at the end of a fifteen year period for earth and concrete works at higher values than warranted. On the other hand, the Government's PCR did not include the value of time saved in its benefit estimates, whereas the present report includes them. However, the gain on account of passenger time savings is smaller in value terms than the VOC factor, thus the ERRs in the Bank's PCR are lower than those in the Government's PCR. Studies and Technical Assistance 4.09 The transport survey provided Cameroon with a first overall assessment of the sector, identified and screened future investment needs and proposed institutional policy improvements or modifications. The survey was guided by an Inter-ministerial Technical Committee (CT) which, on the basis of the consultants' reports, prepared its own report and reccamendations. The survey provided a significant input to the Sixth Development Plan, even if the delays in the start of the survey meant that the full survey report and detailed investment analysis were not available by the time the Plan had to be finalized. Also, the Government tended to take a more optimistic view of its project implementation capacity and placed somewhat less emphasis on economic criteria than suggested by the transport survey. As a result of the changed economic situation since the transport survey, the Goverranent has, as part of its structural adjustment program, stopped or deferred a number of projects included in its Sixth Development Plan. A revision of the investment program proposed in the transport survey may now be needed to reflect the change of economic circumstances. 4.10 A number of institutional and sector policy recommendations made by the transport survey were approved by MINT and submitted to the Presidency. Although no general approval was given by the Presidency at the time, a number of the recommendations have been acted upon or incorporated in the Government's recent Declaration of Development Strategy and more detailed Policy Matrix for the recently approved Structural Adjustment Loan (Loan 3089-CM). 4.11 The technical assistance to MINT helped the Government follow the transport survey and acted in some respect as the secretariat and advisor for the CT. It also helped prepare proposals to reorganize MINT, including the creation of a Studies and Planning Directorate to train counterpart staff and to carry out  - 12 - a number of transport policy and planning act2vities in addition to helping the senior officials with day-to-day tasks. The impact of the technical assistance, although clearly positive, was less than hoped for due to delays in the reorganization of MINT, subsequent delays in nominations of senior Cameroonian staff, and some changes in counterpart staff. More use of young economists and engineers at MINT and MINEQ would have given them wider exposure to practical experience. The technical assistance was therefore continued under the Sixth Highway Project. 4.12 The study to evaluate the existing paved road network provided a much needed assessment of the paved road network and relative priorities for its strengthening. The initial standards used in the studies were somewhat higher than strictly necessary and were subsequently reduced to reflect increasing budget consLraints and changing priorities. This resulted in delays in the studies and the subsequent start of these works under the Sixth Highway Project. Sustainability 4.13 The benefits of the main component of the project are sustainable provided that the overall road is properly maintained. This should be the case as the Douala-Yaounde Road is one of the most heavily travelled roads in the country. The road is a direct beneficiary of the Bank's past involvement in the road subsector, as well as its participatioia in the Sixth Highway Project. As noted in the SAR for that project (para. 3.01), 'the main justifications for Bank participation in the project are the focus on road maintenance and upgrading and the strengthening of institutions, as well as the broadening of the financial sources available for the development of this priority sector.' The objectives of the Sixth Highway Project included maintenance planning and execution, and traiaing begun under previous highway projects and strengthening of the management of the National Equipment Pool, the main provider of equipment to MINEQ. The road is a direct beneficiary of the substantial involvement of the Bank in upgrading road maintenance operations and institutions both before the Fifth Highway Project, and followed-up even more intensively under the Sixth Highway Project. Furthermore, the Government is currently reevaluating its highway maintenance needs, and this is likely to be a major component in a Transport Sector Project currently under preparation. V. PERFORMANCE OF CONSULTANTS 5.01 Consultants played a major role in the project, not only in completing the detailed engineering, but in the two major studies, and in the provision of technical assistance. The consultants generally performed well on all components of the project. Given the long period of execution and the delays in the award of contracts, a considerable number of changes had to be made in consultants personnel. Also, resignations and sickness necessitated further personnel changes during the execution of the work. In the case of the Transport Survey, the Government established the CT (para. 4.04) which assured prompt and coordinated guidance to the consultants and facilitated their access to data sources.  - 13 - VI. INSTITUTIONAL PERFORMANCE 6.01 olnce the Government decided to construct the Douala-Yaounde Road, it awarded a contract for the preliminary engineering of the road to four-lane divided expressway standards. Following the involvement of the Bank, these standards were reduced to a high standard two-lane paved road. In its haste to provide reliable road service between the two cities, the Government tended to overlook the need for thorough technical investigations and study of the project before committing itself. The subsequent break in relations between the Government and its design consultants created considerable difficulty for the Bank in securing sufficient information on which to base an appraisal. 6.02 In proceeding with the award of some construction contracts before completing detailed engineering, or even before engaging consultants to supervise the works on site, the Government took substantial risks. In proceeding in this manner, particularly on Section I, the Government discouraged the Bank from taking a more active role as *Chef de File" (para. 2.13). However, the Bank did fulfill this role on Section II, although its advice to the Government was not always heeded. For example, the Bank recommended that a single supervisory consultant be appointed on Section II to ensure coordination of the work throughout the Section, particularly as some lots were not readily accessible, and there was a concern that scarce quarry sites might have to be shared by the contractors. The Government, however, decided to appoint a separate consultant on Lot 8, and one for Lots 6, 7, and 9. Potential problems that might have arisen from this arrangement were mitigated by the technical assistance staff appointed to strengthen the DH which oversaw the overall project. 6.03 A particularly difficult problem for the Bank, which not only affected the Fifth Highway Project but other Bank-financed projects as well, was presented by the CTB, which was originally located in the Presidency. During the latter stages of the project, the CTB was converted into a separate Ministry for "Information et Marches Publics." This system was, in late 1988, replaced by a General Directorate of Works at the Presidency. The CTB was not particularly responsive to the needs of the individual Ministries it was supposed to serve, ncr did it always follow Bank guidelines on procurement. These matters involved a number of Bank projects, and as the problems created by the CTB and their resolution were well documented elsewhere, they are not discussed in this report, other than to note that they frequently delayed contract awards (see para. 7.05). VII. ROLE OF THE BANK 7.01 The Government put considerable pressure on the Bank to expedite its approval of the project and to approach co-donors for financing. The Bank's role became one of trying to keep up with Government expectations on the one hand, and give due attention to proper project preparation on the other. By the beginning of 1980, the Bank had fielded five preparation missions and by late 1980, was ready to appraise the project. 7.02 The project was carefully and thoroughly prepared by Bank staff. Good use was made of the Project Brief and its subsequent revisions, as a vehicle for planning and scheduling purposes, for identifying the issues and the critical path, and providing a list of the major steps to be taken to complete preparation  - 14 - (para. 2.03). The Bank worked closely with the Government in helping to select appropriate design standards and prepare contract documents; it also prepared the terms of reference for the engineering and assisted with the financing of this work under an existing project. The Bank oversaw contract award procedures and pressed for adequate on-site supervision. The Bank's efforts contributed substantially to the successful completion of Section II. 7.03 The Bank also performed well the task of seeking and coordinating the co-financing for the Douala-Yaounde Road, and keeping the co-donors involved adequately informed. The financing arrangements were innovative because of the participation of private banks as well as Government institutions. The co-financing was arranged despite the fact that the overall road was not engineered in detail, and because the preliminary design consultants were not present at critical times in Cameroon to provide Bank missions with technical data and costs. The Bank was wise to have an independent specialist carry out a cost analysis of the project, inasmuch as this information was not available to the appraisal mission. 7.04 Supervision of project execution by the Bank was painstaking and despite the changes in staff over the 10 year life of the project from inception to completion, the transition was smooth because the staff assigned were fully experienced personnel. An average of three to four supervision visits per year were made. Missions worked closely with Government ministries and their consultants which undoubtedly contributed to the successful conclusion of the project. 7.05 One of the most serious problems created by the CTB was the handling of the award of Lot 7 for construction (para. 3.01). As the apparent low bidder for this work had substantially qualified his bid, it would appear that the bid should have been promptly rejected both by the Government and the Bank, and the award mL,de to the next lowest evaluated bidder. The ensuing delay in asking the lowest bidder to drop his qualifications, and hia subsequent refusal necessitated a re-bidding, and resulted in higher prices and further delays. This placed Lot 7 on the critical path of the overall road, but fortunately, opening of the Douala-Yaoundt Road was not delayed. 7.06 The studies and technical assistance components of the project established a oasis for further Bank lending for highways to strengthen the paved road network, to establish priorities across the transport sector. It provided much needed training of local staff in the conduct of transport studies, administration and management of the highway sector, and the strengthening and expansion of Labogdnie, which participated fully in providing laboratory services for the road component. 7.07 The Bank played its role of "Chef de File" well in arranging the financing for the overall Douala-Yaound6 Road. As far as construction was concerned, the Government placed the Bank in an untenable position by failing to follow its advise by starting work prematurely (para. 2.10), by awarding contracts before the completion of detailed engineering (para. 2.15), and by appointing separate consultants to supervise Section II (para. 6.02). The Bank thus acted appropriately in limiting its role in the construction phase to Section I (para. 2.13) in the light of the Government's unwillingness to take its advice.  _ 15 - VIII. CONCLUSIONS AND LESSONS LEARNED 8.01 The project, including certain additional works, was successfully completed with a 16.92 cost underrun in dollar terms, but with a two-year delay. The delay is mainly attributed to slow award of contracts for the studies, and the time needed to complete the additional works. However, the road component as originally conceived was completed substantially within the original estimate and time schedule. The quality of the work on all project components was good, except for the deficiencies in the pavement of Lot 8 (para. 3.07). 8.02 The rate of return for the road component is substantially lower than the ERR estimated by the feasibility consultants (paras. 4.05-4.06). The studies provided a sound basis for the follow-on project, while the technical assistance helped to strengthen local institutions (paras. 4.09-4.12). 8.03 The many consultants, including firms and individuals, generally performed well on all aspects of the project. It is unfortunate that more use of counterparts was not made during the execution of the road works (para. 3.08) to take advantage of the opportunity for providing junior engineers with invaluable on-the-job experience. The same applies to the transport survey and, to a lesser extent, to the study which evaluated the existing paved road network (para. 4.11). 8.04 The Government, in its haste to complete the overall road, tended to overlook the need for thorough technical investigations and detailed design of the project before committing itself to the award of construction contracts. In so doing, the Government took substantial risks (para. 6.02). 8.05 The CTB was not particularly responsive to the needs of the individual ministries it was supposed to serve, nor was it inclined to follow Bank guidelines on procurement. This led to some difficulties and delays which were eventually resolved (para. 6.03). 8.06 The project was thoroughly prepared by the Bank, which also contributed to seeking and arranging the substantial amount of co-financing required for the Douala-YaoundE Road. Supervision by the Bank contributed to the successful completion of the project (paras. 7.02-7.04). 8.07 The studies and technical assistance provided a good basis for preparing the Sixth Highway Project, and contributed to strengthening technical staff of the ministries involved in project execution (para. 7.06). 8.08 The Bank acted in the Borrower's best interests in pressing the Government to limit the Douala-Yaounde Road to a two-lane facility. Whether, in the light of the reduced ERR, due in part to high costs, the design standards should have been further lowered is a more difficult issue to resolve. However, given the pressure to proceed, the Bank exercised a strong influence in containing the standards, and hence, the costs, of the Douala-Yaounde Road (para. 7.02). 8.09 As regards its role as "Chef de File," the Bank provided invaluable assistance to the Government in its effort to arrange cofinancing for the Douala-Yaounde Road. The Bank had little choice but to limit its role to Section II of the Road during construction in the light of the Government's haste  -16- to expedite construction more rapidly than appropriate under the circumstances (para. 7.07). 8.10 The lessons learned include: (a) the Bank's insistence on thorough project preparation, its efforts to carefully coordinate co-financing, and its close supervision of the project contributed considerably to successful project execution; (b) substantially qualified bids should be immediately rejected in line with Bank guidelines; (c) the appointment of an overall consultant to complete the design and supervise the construction of the Douala-Yaounde Road would have expedited the work, especially during its early stages of preparation and execution; this would have avoided some of the design problems which arose on the overall road after some of the contractors had started work; and (d) caution is indicated for the Bank assuming a "Chef de File" role for project elements which it does not finance, as happened for the Douala-Edda section; substantial :ost overruns (53Z) on this section compare unfavorably with the minimal (3Z) overrun on the section with Bank financing. On future projects, more use should be made of counterparts to provide junior engineers and economists with on-the-job experience.  - 17 - ANNEX 1 Page 1 of 3 REPUBLIC OF CAMEROON FEFTH HIGHWAY PROJECT. LOAN 2180-CM PROJECT COMPLETION REPORT Details of Economic Reevaluation 1. Traffic forecasts and construction cost estimates represent the two key elements that determined the outcome of project ERRs. Traffic Forecasts 2. The following is a comparison between the consultant's forecasts for the year 1987 and actual ADTs for the same year as established by the field survey carried out by MINTPT. 1987 ADTs Consultant$s Forecasts Actuals Ratio Douala-Ed6a Normal Traffic 2,418 1,085 0.46 Diverted from Rail 185 260 1.41 Diverted from Air 189 95 0.60 Induced Traffic 400 912 2.28 Total ADT 8.192 2.362 0.74 Ed6a-YaoundA Normal Traffic 1,661 892 0.84 Diverted from Rail 171 260 1.52 Diverted from Air 189 96 0.60 Induc-d Traffic 8101 829 4.29 Total ADT 2.231 2.219 0.99 Proipct Cost Estimates 3. The actual total project cost based on Government's PCR data was US$583.4 M in 1990 prices and at the rate of exchange of US$1.00 = FCFA 265. On May 1982, the SAR date, the rate of exchange was FCFA 270. For its total length of 245 km, the per km project cost was US$2.29 M. The average per km cost for the Douala-Edda section was US$4.6 M and for the Edda-Yaounde section US$1.6 M.  - 18 - ANNEX 1 Page 2 of 3 4. The following is a project cost comparison summary presented in the form of ratios. Ratio Doumla-Ed&& Ed6s-Ysound6 Douala-Ysound6 SAR/Consultant'e 1.17 1.67 1.36 Actual/Consultant's 1.79 1.61 1.7e Actual/SAR 1.58 1.03 1.23 5. The appraisal mission identified undercosting in the Edea-Yaound6 section of the project in the consultant's study and made .'iitable adjustments to the cost estimates. Actual costs turned out to be only 3? nigher than the SAR estimates inclusive of the quantity contingency allowance. However, for the Douala-Edda section, actual costs turned out to be 53? higher than appraisal estimates, mainly due to design modifications, high unit prices, volume overruns and weak cost control. Economic Rates of Return Government PCR ERRs 6. The best estimate in the Government's PCR was 18Z. A number of adjustments were introducad to compensate for some oversights found in the calculations: the principle ones of which were the following. (a) Vehicle operating savings were adjusted to their values net of taxes. (b) Diverted traffic from rail and air travel were taken out of estimates for induced traffic, and accounted for separately in the economic analysis. (c) Vehicle operating costs and construction costs were both adjusted to 1990 prices to compensate for their presentation at different price levels reflecting their monetary values only. The Government's PCR gave only a single ERR for the total project. The calculation on a rough basis of separate ERRs for each one of the two sections is based on cash flow information the Government supplied. ERRs in the Bank's PCR 7. ERRs of the Bank PCR are based on the consultant's original 1978 estimates suitably adjusted to reflect actual construction costs, 1987 traffic levels, their composition as established by the traffic survey, and lastly, to reflect changes in the general price level between 1978 and 1990.  -19 - ANNEX 1 Page 3 of 3 Comparisons between ERRs as Arrived at by Lifferent Sources 8. The following is a summary as they were calculated at various times and by various concerned parties. ERR. In X As Calculated byt Year ot Calculation Douala-Ed6s Ed&a-Yaound6 Dousla-Yaound6 The Consultants 19T8 16.7 15.2 16.9 SAR (1991/82) 1981 28.4 24.5 26.4 Cov-rnmnt 's PCR 1989 9.1 14.8 11.7 Present PCR 1998 0.8 10.7 9.4 9. There are slight divergences between the ERRs in the present report and those presented in the Government's PCR. They can be explained by differences in vehicle operating costs and cost differentials and by time values. The consultants took into account the maintenance costs of the old Edea-Yaounde road that was retained as a secondary road serving local traffic, whereas the Government PCR did not. 10. The Government PCR shows in the cash flow tables a residual value of 662 for the cost of earth works and 80 of structures. Showing such high residual values at the end of a 15-year period from opening year is indefensible. Also, in the cash flow tables, the Government PCR showed that the total project was executed over one year and showed full project benefits during the second year, whereas it took five years to execute the bulk of the project. 11. The benefit coefficients used in the present report are those obtained from the consultant's original report adjusted to the 1990 price level. These are lower than those used in the Government PCR and result, therefore, in lower benefit aggregates. By contrast, the Government PCR does not include the value of time saved in its benefit estimates whereas the present report includes them. However, the gain for the ERRs in the present PCR on account of the latter is much smaller than the loss on account of lower vehicle operating cost savings. The result of the factors described in this and the previous paragraph resulted in somewhat lower ERRs obtained in the present report compared to those appearing in the Government PCR. 12. Both the Government's PCR and the consultants underestimated savings in road maintenance costs due to the new road. One last point to note is that in neither the consultant's nor in the Government ERR estimates was an attempt made to estimate decongestion benefits for the traffic remaining in the Douala urban network and in those sections of the old Edea-Yaounde Road that was retained. These benefits may have added one or two percentage points to ERRs for the core investment. 13. For further details on the economic analysis, including cash flow tables and supporting data, reference is made to Annex 2. AP1IN August 30, 1990 AsAnxl.par (14)  -20- ANNEX 2 Page 1 of 10 REPUBLIC OF CAMEROON FIFTH HIGHWAY PROJECT, LOAN 2180-CM PROJECT COMPLETION REPORT Cash Flow Tables and Supporting Data 1. Cash flows are presented in three tables. The first pair present the discounted cash flow and ERR analysis for the Douala-Edda section, the second for the Edea-Yaounde section, and the third for the overall Douala-Yaounde Road. Further detail is provided in Table 7 on construction costs. 2. From feasibility study through appraisal and implementation, the project spanned a period of over 12 years between 1978 and 1990. This period went through a major inflationary bout not only in third world countries but in the world's industrial countries as well. The question of how to convert project-related cash flows based on monetary prices to cash prices at constant or real prices, is central to recalculating ERRs to enable a meaningful comparison between costs and benefits calculated at different times by different factors. 3. Two sets of comparisons are required. The first and detailed one is between the consultant's 1978 estimates and actual implementation results for the near completion date of 1990. Comparisons between the consultant's 1978 estimates and the 1981 appraisal estimates, and between the 1981 appraisal estimates and actual results were confined to construction costs only. For the benefits side of the equation, the comparison is between the consultant's 1978 estimates and actual results. 4. At the end of 1986, about 902 of the works were executed and accordingly 1987 was adopted as the first year in which full project benefits were realized for the actual traffic volume that was counted on the road during that year. 5. To bring all projected and actual cash flows to the same price level for both construction costs and benefits, the index chosen for the local currency component was the local cost of living index which is the only one available in Cameroon. However, local currency component represents only 232 of construction costs and about half of benefits. The index chosen for the foreign currency component was the French cost of living index. This is because the price level in the CFAF zone closely follows French prices and construction contracts are often signed in CFAF or in European currencies.  -21 - ANNEX 2 Page 2 of 10 6. The following is a statement of the two indices obtained from the IMF International Statistics Publication of 1989. Year Cameroon Index French Index 1978 51.9 50.5 1979 55.3 55.9 1980 60.6 63.3 1981 67.1 71.8 1982 76.0 80.3 1983 88.7 88.0 1984 98.7 94.6 1985 100.0 100.0 1986 107.7 102.5 1987 114.2 105.9 1988 124.0 108.7 1989 N/A 112.5 1990 N/A 114.9 The two indices do not diverge significantly. For simplicity's sake, the French index was used to do all the conversions to the 1990 price level. Using a combined index reflecting weights of foreign and local currency components would have produced higher construction costs at 1990 prices by a few percentage points than those arrived at. For purposes of converting CFAF values to US$, a rate of exchanige of US$1.00 = CFAF 265 was used prevailing on June 1990. The rate of exchange appearing in the SAR of May 1982 was US$1.00 = CFAF 270. ATTACHMENTS: Tables 1-7 APiIN August 3, 1990 A:n32.pcr (MM)  0 C4 Wh qn x pioJ q POj Pa a j (a) N~~~~~~~~~an eu ^sai eu si: (m 35010 CVWU- 6019 M OW00 PMON AMtlad 661) g*e013 s*au- $VIII oneO11IU gnatI'K ge- #,$M gat son; ult- fla sil 1is- out $ "|l 1 Ul- ,gun I'e i1t- Lai st 1San- s ilS lsn sot- ml $arn SUe- no st "t- t I'MlA M U- slim I's ci- 1i I'n $'1o- s1g s&l *Ml- ga *~~~~~~ oau giK- t UHaS-U t-SZ- out *110 rtl- 5192 rLU ot:- mu en:: *i:S- *,15 ea o1 t- W1 CIA I'~~~~~~~te:: Iod- stat $,a gig-le n Ie Ot Ot- 00I06 o'n $- ml otan- o'w- glg *- sin- ml I'Ms- g'ttm- ml os sis- t rig::- go ge es s-w- utl (e) (O () "u 1l%Ut PMD no ML Bl" 10 *o} q d * .10os lp. P te me 8 P alU P 1R03 ps q "101 13 ^olqa& 30t1U3UJ 85SR uUgl #ll P1131101314 3303 U3" p 3q apTq "'-"qu. 131WhUl IINOII  tauL 2 Ieosie ate of btan tolid U- fot m 1.IDE Mllilliosm Cmebuotl1 & For111 ntrnw & tou i ehicle t katc 1 hitume out for ltemmc uaip costa huhek Cost uts md lof forel roN W o1ld ha sum er tie fot the uith fear hula-eU *tif w*o et0 ead inrject aue cal traffic 1a) lb) lc) 1113 -1631.8 . 0.6 0.0 -161.6 114 -U16M.2 e.g *. -268.3 -15111.5 13 -15221.2 0.S 6.0 -2U.3 -150.5 1to -1611.6 12.4 386.4 -1315.4 -5014.1 1in -51.1 115. 461.1 -43.1 410.91 Il -51.1 15.0 44.4 -4.1 4414.1 116 -51.1 16.1 4i1.1 -122.1 412.6 1110 -3.6 MI5.6 41.5 -1545.4 648.1 110 -53. 1S.0 526. -01.4 5260. 1M2 -2256.4 l1S.I 511.0 -4.4 4181.0 13 -1406.0 1.1 $.1 -12. 4411.3 1114 -64.6 26.3 2.1 -134.1 501M. m 5-66.1 1182.6 660.5 -4.4 21.0 1116 -66.1 15.6 1610.3 -41.4 108.0 u1 -22.0 II. 1451.2 -I.6 5214. 1 -86.1 1111.6 1115.1 -1340.1 5263. 1o -441.6 1.T 646.2 -4.4 1IS1.? 26 -11.? 11.1 1t25.4 -41.4 1310.3 M5.05 -10.0 (a) lb yew MD h m t211e0 w lh of 1t im"t rood. (c) e d W n roa to FE" Om to mmmal 1ca traffl  44 0 x t-tsi n -f mseiox i- egg,, si- e:s em ci- ou em,tS el-- Zone; *ia m,"- mg 01061 0 11- 01515 gift 011- 560 011.5 Czl- 05505 COft 011- 50l 0'6s1 011- *0M CM 611- 1"i I'll t sin- l'etsu *n *stm- tot 01.111 011:- I'M1 0111 0'5- 1661 63111t C-It- 6116 0-m 036- 1cu 0136 -011- 0nu *m o *- ml 00113 016- 05'111 I'M 015- 606t Clug 011- *'zm IeM I's- Mt e Mis $- O'"" u*ti #'n- Mt6