Upper Volta - Cotton Road Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 3528 THE WORLD BANK PROJECT PERFORMANCE AUDIT REPORT UPPER VOLTA: COTTON ROADS PROJECT (CREDIT 316-UV) June 24, 1981 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of th e r o ff ica .-. u nm, n o. 1at sav a u a w i l b e ic u sl ue w i u t "U or Mi D aRIZ E U lU KI L M BEUII .  VflD nVWTCTI AT TV f%T V PROJECT PERFORMANCE AUDIT REPORT UPPER VOLTA: COTTON ROADS PROJECT (CREDIT 316-UV) TABLE OF CONTENTS Page No. Preface .......... .........................i......................i Project Performance Audit Basic Data Sheet ......................... ii Highlights ........................................................ iv PROJECT PERFORMANCE AUDIT MEMORANDUM Economic Justification ................................... 1 Excessive Delays in Project Completion ................... 2 ATTACHMENT: PROJECT COMPLETION REPORT I. INTRODUCTION ............................................. 5 II. PROJECT PREPARATION AND APPRAISAL......................... 5 III. PROJECT IMPLEMENTATION ................................... 9 IV. ECONOMIC RE-EVALUATION ................................... 14 V. CONCLUSION ....................................... 15 Tables 1. Actual and Appraisal Estimates of Project Costs 2. Actual and Expected Project Implementation 3. Schedule of Disbursements 4. Economic Analysis This document has a restricted distribution and may be used by recipients only in the performance of . their otticial duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT PERFORMANCE AUDIT REPORT UPPER VOLTA COTTON ROADS PROJECT /f"nv%-T' -31 r_TT1 \LMAA JUUV} PREFACE - C ~A%L %. LLU '. L. Lu "V<%U 0 K L UJ= L. in Upper Volta for which Credit 316-UV for US$3.5 million equivalent and a cnnlementar- reitA4 f TTCC 1.315 4w a T...,ve 107n TunA A10)an Akcoe. 1071 respectively and final disbursements were made in October 1980. The report consists of a s,mmar- of Highlights anA a brief Project Perfomance Audit Memorandum (PPAM) prepared by the Operations Evaluation Department (OED) and a Prniat- remmlatin van^wt 190rp pre- "n A h h West Africa Reglonal Office. No separate PCR mission was undertaken. The memorandum is based on the PCR and discussions with Bank staff; the transcript of the Executive Directors' meetings which consId Aered th project, have been read and project documents have been reviewed. On the hma4a mf thic 14m41-aA -d-in Ory) has --luded that the DPCI cowve he project experience in a comprehensive manner and found no reason to question its nnt fr tho rae of return. Thneraft rnl w to the Borower f comments in the normal course; however, none were received.  PROJECT PERFORMANCE AUDIT BASIC DATA SHEET UPPER VOLTA: COTTON ROADS PROJECT rCPpnTT 11A-HV KEY PROJECT DATA Item Appraisal Estimate Actual Total Project Cost 3.5 5.1 Overrun (% 46 Credit Amount (US$ million) 2.8 4.15.a Disbursed N - A. 1 Repaid ) December 31, 1980 Borrower'sObli-ation )41 Date Original Components Completed 12/74 06/75 Pro-ortion Completed b Above Date - - 7 /b Proportion of Time Overrun (%) 20 D ate- P r oj--c t Co -m pl 1eted nl)2/8qn Proportion of Time Overrun (%) - 207 Economic Rate of Return (.) 174 Cumula 1tiv Esti mate .a nd Actua.1 l Disbursements (US$ Millions) FY73 FY74 FY75 FY76 FY77 FY78 FY79 FY80 (i) Estimated 0.20 1.30 2.50 4.15 4.15 4.15 4.15 4.15 f(4 ) Actua ..M1 - QI ' 3.2 3A3 -33 3.8 415 % of (ii) to (i) - - 62 79 82 82 93 100 OTHER PRATWET DATA Item Original Plan Actual First Mention in File 08/06/69 Government's Application 08/71 Neeotiations 04/72 04/72 Board Approval 06/72 06/08/72 Loan A2reement Date 06/26/72 06/26/72 Effectiveness Date 09/25/72 09/25/72 Closing Date 12/75 12/79 Borrower Republic of Upper Volta Executing Agency Direction des Travaux Publics Fiscal Year of Upper Volta January 1 to December 31 Follow-on Proiect Third Highway Proiect - iii - MISSION DATA Month/ No. of No. of Man- Date of Item Year Weeks Persons weeksL. Report Identification 7/69 0.5 3 1.5 8/5/69 Preparation 8/71 1 1 1 8/27/71 Appraisal 12/71 - 2 2 4 5/9/72 Total 3.5 6.5 Supervision I 9/72 i i i 10/12/72 Supervision II 1/73 1 1 1 1/31/73 Supervision III 12/73 i i 1 2/1/74 Supervison IV 5/74 2 2 4 7/2/74 Supervision V 3/75 2 2 4 4/3/75 Supervision VI 10/76 2 3 6 10/24/76 Supervision VII 3/77 3 1 3 5/19/77 Supervision VIII 6/77 1 2 2 8/17/77 Supervision IX 9/77 2 1 2 4/6/78 Supervision X 3/78 2 1 2 4/6/78 Supervision XI 5/78 1 3 6/12/78 Supervision XII 1/79 2 1 2 2/16/79 Supervision XIII 6/79 I 2 2 7/23/79 Supervision XIV 10/79 2 1 2 11/21/79 COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) CFA Franc (CFAF) Year: Appraisal Report Year Average 1972 Exchange rate: US$1 = 250 I ntervening Years Avrg:1972-1979 US$t1 ='220 iUC ~. i- i AW-. 5C. . *I a *ý I iUaQ4. Completion Year Average: 1980 US$1 - 230 .2/ Inluding a- supplementary credit of: UVSY11.35 mi1llion equivalent. i,/ "ne 217 kmroad4 section was omited from the project. c/ Time spent on the supervision of Credit 36-UV was, n fact, less thZan mission time because missions had other duties to carry out. - iv - PROJECT PERFORMANCE AUDIT REPORT UPPER VOLTA COTTON ROAD PROJECT (CREDIT 316-UV) HIGHLIGHTS The purpose of the project was to reduce transport costs in the western part of Upper Volta where traffic was expected to increase as a result of a Cotton Development Project for which an IDA credit!/ had been granted one and a half years earlier. The transport cost savings were to be achieved through the reconstruction of two road sections totalling 97 km to engineered gravel road standards; project preparation studies for a next highway lending operation were also included. The total estimated project cost was US$3.5 million equivalent. The project was implemented in a modified form. One road section (27 kn) was eliminated since, as a result of several years of drought, cotton production did not increase as expected and the construction of a ginnery which was to be served by the road was postponed (PCR para. 3.04). The other road section (70 km) was built as planned but with a cost overrun of about 50% which resulted from design changes, underestimates of the construction cost, inflation and exchange rate fluctuations (PCR para. 3.13).4 The preinvestment studies were completed satisfactorily at approximately the estimated cost (PCR para. 3.11). In view of the higher contract cost the Government requested a supplementary credit for US$1.35 million equivalent which was approved by the Association before the construction contract was awarded. The above project components were implemented with a delay of six months or 20% longer than the original plan, but final disbursements were delayed for a considerably longer period as explained in paragraphs 5 and 6 of the Audit Memorandum. Mostly because of the cost overrun the reestimated economic return, according to the PCR, dropped to 12% from 17% calculated at the time of appraisal. The PCR, however, does not provide an explanation for a much higher percentage of sav- ings resulting from the road improvements than was estimated in the appraisal report. Using the original savings percentage the audit found an economic return of 4% (para. 2). Credit covenants which especially dealt with staff- ing, data collection and road maintenance were not fully implemented and the PCR (para. 3.20) concludes that they were overambitious in relation to the country s resources. At the time of completion of the above components about US/uu,uuu in credit funds were left. After a long period of indecision these funds were used, at the request of the Government, for reconstruction or drainage 1/ Upper Volta-Cotton Project (Credit 22 5-UV): PPAR Report No. SecM8u-13. -v - structures on part of the highway system and a small amount of training. While the work was apparently useful in itself, it delayed completion of disbursements under the project by five years or over 200%. The following points may be of particular interest: - the appraisal and PCR have omitted to estimate those benefits which actually accrue to Upper Volta (para. 3); - in order to improve the reliability of economic return esti- mates more comprehensive traffic counts are required (para. 4); and - cancellation of remaining funds under the credit, once the original project components were completed, would have avoided long delays in credit closing and probably saved staff time (paras. 5 - 6). PROJECT PERFORMANCE AUDIT MEMORANDUM UPPER VOLTA COTTON ROAD PROJECT (CREDIT 316-UV) 1. The attached PCR prepared by the West Africa Regional Office covers the project experience in a comprehensive manner and this audit memorandum will deal with only a few points where clarification is required. Economic justification 2. The audit questions the approach used in the economic reevaluation. The appraisal report does not include monetary values for vehicle operating costs and only states that savings in vehicle operating costs were estimated at about 40% due to the improved condition of the roads and to the use of heavier and more economical vehicles that this would allow, while there was also a distance saving of 7 km. The values used for vehicle operating costs in the PCR are taken from the latest consultants' studies on Upper Volta. Using the most recent estimates for vehicle operating costs on a road in equivalent condition to the project road after improvements were completed would seem to provide a satisfactory estimate. The percentage savings result- ing from the road improvement was estimated in the PCR at 57% or almost 50% higher than the 40% used in the appraisal. This implies an assumption in the PCR that the road before improvement was in worse condition than was assessed by the appraisal team. Since there is no evidence for this new assumption and the "old" condition no longer exists, it would seem more appropriate to maintain the same ratio i.e. that the new road condition has resulted in a 40% saving.±/ This estimate is also in line with the commonly accepted per- centage of savings used in recent Bank projects while some experts who were consulted by OED felt that 40% savings was according to their experience some- what on the high side.2/ Using the savings percentage from the appraisal report results in a reestimated economic return of about 4%.2. 1/ The Region is of the view that while the condition of the project roads prior to their improvement cannot be verified, the condition is still exhibited on similar old tracks and on this basis, it is appropriate to assume that the project road was in poor condition at the time of appraisal. 2/ The Region notes that 57% savings is based on actual VOC data prepared by consultants in Upper Volta and reflect the high percentage of medium and heavy trucks on the road. 3/ The Regions disagreement with the numerical values used in the computa- tion of the rate of return is reflected in footnotes 1 and 2 above. In addition, the Region and the Central Projects staff point out that the rate of return based on vehicle operating cost savings alone does not necessarily provide a realistic assessment of the economic justification, which should be seen in the context of the agricultural project it serves. This project would have an ERR estimated at 31% if the roads financed under Cr 316-UV are included as a component. - 2 - 3. Furthermore it should be noted that according to the appraisal report most of the benefits of the project were to accrue to the CFDT (Com- pagnie Francaise pour le Developpement des Fibres Textiles). At the time of appraisal CFDT was a French owned private company with, in Upper Volta, a minority participation by the Government. Apparently recently the Upper Volta Government share has increased. The appraisal report did not estimate how much of the benefits thus gained by a foreign owned company were recouped through taxes on profits or in other ways, but it appears that the economic return which Upper Volta was to derive from the roject would have been lower than the 17% estimated in the appraisal report./ Using the same reasoning, the audit concludes that only part of the 4% economic return generated by the project benefits Upper Volta. 4. Finally, where an economic return is the "best estimate", as above, the question arises as to how satisfactory such an estimate really is. The benefits are calculated by using actual traffic, anticipated traffic growth, estimated vehicle operating costs, distance savings and the estimated economic life of the road. Predicting traffic growth always involves a speculative element but the 5% rate used in the PCR seems reasonable. Vehicle operating costs used are based on recent consultants' studies and, even though the data remain approximations, further refinements may not be justified in view of the cost involved. However, information on traffic, according to the PCR (para. 4.03), is very limited with actual counts only made during the season of low agricultural activity; since no information on seasonal traffic variations is available and resources were lacking to generate these data, the average annual traffic figure used in the reevaluation is an assumption. This low level of accuracy in part of the basic data resulted in a reestimated economic return of low accuracy. This problem would merit special consideration in the preparation of future PCRs. Excessive delays in project completion 5. As shown in the PCR (Table 2), completion of the original project components (after the deletion of one road section because of reduced bene- fits) took only 20% longer than anticipated at the time of appraisal. How- ever, some US$0.7 million was left in the credit and about two years were needed to reach the decision that these funds were to be spent on reconstruc- tion of drainage structures on some highway sections. An additional two years were needed for preparation before construction started; as a result all components were finally completed about five years after the completion date mentioned in the original credit agreement. Some of the leftover funds were to be used for staff training but this effort was not very significant (PCR para. 3.07). 1/ The Region is of the view that computing the rate of return accruing to Upper Volta alone on an individual component of a joint product project is not feasible. - 3 - 6. Ine auait questions wnetner cis aelay and tne star tie involved in keeping the credit open and preparing a small project component, was justified. It appears that when Ene supplementary credit or US91. mllion was signed in March 1974 it should have already been clear that construction of the road section which was cancelled five months Later, was uncerain. It might therefore have been advisable to have granted the supplementary credit with a clear understanding that any remaining funas were to e cance U UPVLL completion-of the original project components.  -5- ATTACHMENT TTDDrD 1T7M TA COTTOM ROADn PRTVT - Cr31AUJTV PROTECT COMPTETTON REPORT I. Introduction 1.01 Uner Volta is among the poorest countries of Africa. with a per capita income below US$150. The basis of its economy is purely agricultural and 85% of the farms operate at a subsistence level. Livestock, cotton, and groundnuts are the only exports of this landlocked country. The Voltaic government recognized very early that the only means of improving the welfare of the people was to increase productivity in agriculture. The first two development plans were designed to concentrate public investment around that sector, in particular with a view to improving cotton growing which, in 1970, accounted for 30% of the country's foreign exchange earnings. 1.02 IDA's operations in Upper Volta in the late 60's and most of the 70's have also been aimed at supporting the agriculture sector, except for the first Credit (US$0.8 million) which was for telecommunications. The second Credit, in 1969, was for the West Volta Cotton Project (US$6.2 million); the third and fourth Credits were for the Rural Development Fund (US$2.2 million) and the Cotton Roads Project (US$2.8 million plus a supplementary 1.35 million), the subject of this completion report. 1.03 For the reevaluation of the project the following sources of infor- mation were used: the west Volta Coccon Project kur.L2-uv) files, project supervision reports, data on costs and actual physical implementation furnished by the consultant progress reports, and all correspounence Un fie perLd.auag to the project. HI. Project PrprtonadApasal A. Project Formulation 2.01 The Voltaic Government and the Bank started discussing the possibil- itv of a road transport project in late 1963, when the Government approached the Bank with a Transport Plan prepared by consultants and requested the Bank to finance those roads which the Transport Plan had identified as priorities. In March 1964, a Bank identification mission visited Upper Volta and concluded that since the European Development Fund (EDF) had pledged its support to these priority roads and the Fonds d'Aide et de Cooperation (FAC, France) was willing to finance road maintenance, road investment to be financed by the Bank should directly support agricultural development and should be preceded by the implementation of measures providing incentives to farmers for increasing their production. These measures were included in the West Volta Cotton Project and the Rural Development Fund. 2.02 The Upper Volta Cotton Road Project reflects this orientation and constitutes part of the effort to develop cotton production, the main source of Upper Volta foreign exchange. The cotton roads were located in the area covered by the West Volta Cotton Project but could not be financed under that project since the necessary engineering studies were not avail- able at the time of its appraisal in early 1970, and the project's funds were limited. 2.03 The study of four feeder roads in the West Volta Cotton Project area started in 1968, after the Government, FAC and the Bank agreed to the terms of reference of a summary feasibility study financed by FAC. The consultant used existing data in a desk exercise and concluded that improve- ments of these roads would have a low rate of return. In 1970 during the appraisal of the West Volta Cotton Project, at the request of the Voltaic Government, staff from the Bank Regional Mission in Western Africa (RMWA) inspected the feeder roads within the project area. In the light of new cotton production forecasts, PMWA reviewed the economic analysis done by consultant and concluded that the rate of return on two of the four roads, Solenzo-Koundougou and Hounde Bereba, would be high enough (respectively 19% and 14%) to warrant improvements works. It was therefore decided that the West Volta Cotton Project would finance the detailed engineering for these two roads with the intention of undertaking their construction in a proposed Cotton Roads project. 2*U4 in L U i W tHA e mHo LitswiULv UL LL ARVUIVmWWC L in ter maintenance program, FAC financed an update of the 1963 Transport Plan which was completed i n October I171 A-._--. . - th updated - ---- Pla ha -o been L 3 - released yet by the Government, a RMWA mission could only review the data reae "to traffic cont a.nd oigi detiato s4uv-LJ y. H-owever,J'tV e mission recommended including in the proposed Cotton Roads Project funds for the-reinvestment-st-di-s-of-- he-Bob-- Do---a- ----Ddo roadwhic also serves the West Volta Cotton Project area. B. Project Appraisal and Negotiations 2.05 In December 1971, as soon as the detailed engineering for the Solenzo-Koundougou and the Hounde-Bereba roads was ready. an appraisal mission left headquarters for Upper Volta. Since the internal rates of return for the two roads were not very high. the construction standards and cost had to be carefully reviewed as well as the forecasts of cotton production. In the case of the Hounde-Bereba road, forecasts of cotton production were particularly critical since the road was to serve a cotton ginnery (to be financed under the West Volta Cotton Project) which would be built only when cotton produc- tion reached a suitable level. RMWA's proposal to include in the proposed project preinvestment studies for the Bobo Dioulasso-Dedougou road was found premature. As the update of the Transport Plan was not available to the Bank, there was no assurance that the Bobo Dioulasso-Dedougou road would be one of the highest priority roads and the appraisal mission proposed to - 7 - defined when the Transport Plan would be released. The Bank requested a copy Government did not agree with all the consultant's recommendations and argued th'1&at 1 t could not make the consul-I ---bl before ft was approved. 2.06 Important issues were discussed during negotiations: (i) The problem of the justification of the Hounde-Bereba road: The Rank insisted that the road should not be built if cotton production was too low to justify construction of the cotton ginnery but the Government insisted on building the road as it believed that cotton production would increase. Finally the Bank aareed to the construction, but later as explained in para. 3.04, as cotton production decreased due to the 1972-1974 Sahelian drought, the road was not built. (ii) The use of labor intensive methods for the construction of the two feeder roads had been suggested at the Loan Committee Review. The Voltaic delegation refused this proposal on the grounds that they did not wish to pay for more studies, and that the works would take longer and might end up being more costly. (iii) The Voltaic delegation asked for the financing of 100% of local costs net of taxes, to which the Bank agreed. (iv) The Voltaic Government was also reluctant to pay for consultant supervision of the construction works since they considered that the Ministry of Public Works (MPW) had enough competent personnel.to undertake the supervision. It was finally decided that, provided the Bank found their qualifications and experience satisfactory, engineers from MPW would supervise the works. C. Project Description 2.07 The project was presented to the Board on June 6, 1972, signed on (a) Th a mf t-wn ara"a- grral road,q totiA174na 97 kem- The Solenzo-Koundougou road (70 km) connected the marketing center m n Ia -p- ttlvm, -n a nmusa rnsa loe-fm t-n Tknh n %i1=QQn~ wh ri (- nt-t- nn ------------ ---------o-- ---- -----.--- was ginned prior to being exported via the railroad. The Hounde- Rprpba road (27 km) would nrovide a link between the new 2innerv to be built in Hounde, and the railroad in Bereba. Supervision of the construction works was to be carried out by the Directorate of Public Works (DPW) but funds were provided by the project for consultant services in case DPW needed to be assisted. - 8 - (b) Feasibility studies for about 400 km of primary roads, followed by the detailed engineering of those sections found to be of the highest priority. The 400 km of roads would be selected once the updated Transport Plan would be made available. D. Credit Covenants 2.08 Particular covenants required that: (i) the Government employ consultants acceptable to the Bank for the supervision of road construction and the preparation of the preinvestment studies (Section 3.02 of the Credit Agreement); t11) tne Government make tne Transport rlan availaule to the Bank as soon as possible (Section 3.06 of the Credit Agreement); (iii) the Government collect the necessary data for the planning of Directorate of Public Works be staffed in adequate number with qualfied r and expe ice persnnel (Sectio 4.03 of the Credit Agreement); and (v) the Government properly maintain the roads built under the project and the other roads under the responsibility of the Directorate of Public Works (Section 4.04 of the Credit A2reement). E. Cost Estimates and Financing 2.09 At the time of appraisal, the total project costs (including taxes) were estimated at US$3.5 million equivalent (Table 1). These cost estimates were based on 1971 contract prices for EDF and FAC financed road projects in Upper Volta. Contingencies of 20.75% for construction and 15.5% for consultant services were added to the basic cost to cover quantity increases and price escalation over a two year period. 2.10 The project was financed by an IDA credit of US$2.8 million, the estimated amount needed to cover 100% of both the foreign exchange and the local costs components, net of taxes, of road construction and consultant services. The Government was to contribute US$0.7 million equivalent representing taxes and duties on goods and services acquired under the project. - 9 - A. Implementation Road Construction a) Selection of Contractor 3.01 The bidding documents for the construction of the Solenzo-Koundou- gou and Hounde-Bereba roads were completed in August 1972. and Prequalifica- tion of interested contractors followed. In November 1972, tenders were invited under Bank ICB provisions from the ten prequalified firms. 3.02 The bids were opened in March 1973 and the Tender Board recom- mended that the contract be awarded to LTPA (Ivory Coast) who had made the lowest offer. Since bids were considerably higher than estimated at appraisal, the Bank reviewed them and notified the Tender Board that it concurred with its choice in August 1973 (see para. 3.12). b) Construction of the Solenzo-Koundougou road 3.03 The contract for the Solenzo-Koundougou road was finally awarded in December 1973 after the Bank agreed to increase the amount of the Credit to cover the difference in costs between the appraisal estimates and the contractor s bid. LTPA started work in January 1974, approximately three months later than estimated at appraisal. A team from the MPW and the National Laboratory for Building and Public Works headed by a road engineer from consultant BCEOM (France) 1/ carried out the supervision. Construction works had to be slightly modified because some drainage had been under designed. The drainage was redesigned by the supervising team and required larger quantities. The contractor requested a two-month extension for comple- tion of the works, which was granted. Aside from this and the slow start of the works, the construction of Koundougou-Solenzo road and its supervision were satisfactory. Construction was completed on May 15th, 1975, within the 34 T.TPA had been nntfftd that they hadi how czaarrqA fer nnvQtru tion oZ the Solenzo-Koundougou road with provision to extend the contract to include at a later state construction of tha Honnda-RprPba rad. Th laftr represented only 28% of the estimated cost of constructing both roads. In July 1974, the Bank suggested cancelling construction of the Hounde-Bereba road since the ginnery at Hounde, on which the economic justification of the road was principally based, was not going to be built because of the low cotton production resulting from the severe 1972-1974 Sahelian drought. 1/ BCEOM was awarded the supervision contract in December 1973. - 10 - The Government agreed in August to delete construction of the road and proposed using the funds thus made available to construct either the Fada N'Gourma-Bifa road (on the Benin route) or two bridges, one on the Diebougou-Bobo Dioulasso road and the other on the Ouagadougou-Leo road. 3.05 These works were not found economically justified and the govern- ment prepared other proposals. In late 1976 it was agreed that the remaining funds would be used to repair and construct drainage structures on several existing roads. In October 1976, the Credit Agreement was amended to reflect the change in the use of funds. The Government and the Bank selected works on the following roads: Ouagadougou-Kaya, Tenkodogo-Garango, Batie-Kpuere, Ouagadougou-Leo, and Ouagadougou-Kongoussi. In September 1977, the Government and the Bank agreed to delete the drainage structures on the Ouagadougou-Kaya road from the bidding documents since the preinvestment study being carried out under the Third Highway Project (Credit 653-UV) showed that a full-fledged upgrading of the road to paved standard was justified and it was likely to be carried out in the next project (proposed Fourth Highway Project). Because of the small size of each lot, the Bank agreed that the invitation to bid would only be locally advertised, and that the Government would directly supervise execution of the works. The Bank also agreed to apply a 7.5% preference to domestic contractors, but this did not affect the award of the contract. .J~J~JA. 9& I U,1 61 %2 VkJ.U~1 aL LJ LUU ILi S3.V L . awarded works on the Tenkodogo-Garango and Batie-Kpuere roads to the UDEC for the Ouagadougou-Leo and Ouagadougou-Kongoussi roads. At the request of the Voltaic C-ove-rmn an given th e _- ount o f ava _l1able funds , th'-,e Ba nk agreed to finance 90% of the net of tax cost of the works. The contractors mnhilli,d In Ortnhar 197A and QAIQfartnV y1v rnm"latp tha nrk an March 1980, within the contract price. 3.07 In April 1980 it was decided to include rehabilitation by contractor of the Hounde-Bereba road in the nronosed Fourth Highway Proet. inis traffic is increasing following a hike in cotton production. d) Training 3.08 In the spring of 1977, the Bank and the Government started discus- sing the possible actions to remedy the lack of competent staff within MPW. The Government identified the need for three on-the-job training programs, one for three clerical staff for DPW, a second for five technicians in charge of data collecting in the Directorate of Transport (DOT), and a third for technical staff of the National Laboratory for Buildings and Public Works. The Bank agreed in August 1977 with the Government request and proposed that US$70,000 be made available from the unallocated category of the Credit for a training component. But in May 1978 in view the high bids received for the drainage works which required most of the unallocated funds, the training - 11 - component was reduced. Finally the project only financed the expenses of five high school graduates who were trained by the technical assistance provided to DOT under the Third Highway Project.1/ The Government promised that the trainees would be assigned to MPW after their graduation. The Credit Agreement was modified to include this training program in June 1978. 3.09 Between January and December 1978 the five trainees were in- structed in different subjects relating to transport including: transpor- tation modes, data collection techniques, traffic regulations, tax systems and pricing policies. After these formal courses, four of the trainees 2/ collected data and presented a paper on the trucking industry of a different region ou upper Vula. ZUUL LZ.L1t1 have ueen integrated in the Directorace of Transport and the fifth is on a librarian course in Dakar before taking 310i Tr Nn,1 N= n - durng ne-oiai4- se pra-207 tha the Government would make the updated Transport Plan available to the Bank Moith1n 01n -11 _111- -_ -_n ofte rJect. Throads to be studied were to be chosen on the basis of this plan. The Bank received the plan in November 1977 (5 monthq Afrr rho nrniApt- cianIn Ate)l tx4rli r 1l roiena' proposals for the feasibility studies: the Hounde-Bobo Dioulasso (105 km), Bobo Dioulasso-Dedougou (180 km). Bobo Dioulasso-Banfora (89 km) and Ranforn- Ivory Coast border (70 km) roads. It had then become obvious that the amount of project funds allocated to the studies would not be enoulh to cover 400 km of feasibility and detailed engineering studies but that about 280 km would be more realistic. Furthermore, a Preliminary assessment showed that only the first and third roads had a reasonable chance of being economically justifiable. The Bank, therefore, recommended that the feasi- bility studies be undertaken for these two roads only, and this was accepted by the Government. 3.11 The terms of reference prepared by the Government were approved by the Bank in May 1973 together with a short list of eight consultants. In October 1973 five proposals were received and BCEOM was selected for both roads. BCEOM started work in April 1974 and satisfactorily completed the feasibility studies and the detailed engineering for both roads in December 1974. To incorporate Governments comments, BCEOM produced an annex in June 1975. Construction of the Banfora-Bobo Dioulasso-Hounde road was financed under the Third Highway project (Cr 653-UV) and completed in May 1979. S A. Lus u. one DPW stf Jsan o cutinuu as a component of tne Secona Education Project (Cr. 956) which provides for improving the training drafters, storekeepers, shop supervisors and trainers. The implemen- 2/ One trainee was on qick pavp at tho tfim. - 12 - B. Cost and Disbursements 3.12 At the time of appraisal, the total project cost (Table 1) was estimated by the consultant based on costs of similar works carried out in 1970-71 under FAC financing. Construction costs of the two roads, Solenzo- Koundougou, to be started in the fall of 1973, and Hounde-Bereba, to be started in early 1974, were estimated at CFAF 712 million including contin- gencies. When the bids were opened in March 1973, the lowest bid for the two roads (from LTPA) was CFAF 733 million or about 25% more than the estimated CFAF 590 base cost. The bid and contract price for the Solenzo-Koundougou road was CFAF 530 million or about 25% more than estimated. In dollar terms, the bid was about 50% more than the base cost at appraisal. The exchange rate used for the estimates was CFAF 256 for a dollar and had dropped to CFAF 210 when the bids were evaluated. 3.13 In June-July 1973, the Bank reviewed the LTPA bid and decided that, given the unit costs of the recently completed Koupela-Togo border road and the clear appraisal underestimate, the LTPA offer was acceptable. Actual costs were 47% more than contract prices. This larger cost overrun was due to larger quantities than expected and to a high inflation rate. The even larger difference between appraisal and actual costs reflects the lower value of the dollar during project execution. Consultant services costs were in line with appraisal estimateS. asked the Bank for a supplementary Credit of $1.35 million. The amendment to the Credit Areemet wasc appro..ved i ct-oe _1973UIS. b-#N -- 4a-44 March 1974. Subsequently when construction of the Rounde-Bereba road was AeletA 4,in Augas 1074, ahntS TTit.71 milion were availhle for drainAgp structures and training. Bids for the drainage works came to $0.76 million, 90n of which ns to he financed by the Credit. The rest financed 100 of the subsistence expenses of the 5 trainees, while the Third Highway Project flnanepd thp onr of -he tachnfcal assistance in charge of the teaching. 3.15 Disbursements (Table 2) started two years later than estimated at appraisal as road building and studies were supposed to start in the fall of 1973. There were also no disbursement in the three years between January 1976 and January 1979 when the Government was deciding on which road to execute the drainage works and was preparing the technical studies. C. Institutional Performance 3.16 The project as appraised did not include any institution building objectives. Nevertheless, in the course of its execution, when changes in the allocation of funds were made, additional elements with an institution building objective were introduced in the project. The first element was the preparation, by MPW with the advice of the Bank, of engineering studies for the execution of the drainage works and of the economic justification of the improvement proposed. This was certainly a learning experience, espe- cially since it was done just after the FAC-financed technical assistance left MFW. The second element was the training of five technicians to strengthen the ministry. Thus, the project had some direct impact in strengthening the institutions responsible for transport in Upper Volta. - 13 - 3.17 The project covenants were most relevant to institution building and although little could be done to enforce these covenants during project implementation, the Bank and the Government started to discuss the issues of maintenance, organization, training at all levels, and planning and program- ming of road investments in the preparation of fUolLw-up pLjeC; 40 described in para. 3.20. D. Government Performance 3.18 As originally planned, the Government's input in the project was t o b e qui-4te small. As noted in -'-' r 01 f -l--~~ during negotiations that MPW's staff should supervise the road construction was soon apparent that MPW did not have enough personnel available and -nnoiitan t. ArnM nrnfeid aiinrviing am lpndpr. TTndr thiq arranm,t supervision by the BCEOM/Government team helped by the National Laboratory was satisfactorv. 3.19 When a surplus materialized (Dara. 3.04) with the cancellation of the Hounde-Bereba road the Government proposed using the funds for small projects. The Bank accepted the proposal. However, the lack of local staff to prepare these projects in a manner that they could be appraised by the Bank became a major bottleneck. It took more than three years, after the cancellation of the road construction component, for the Government to come up with a sound proposal. With regard to the training element, the Government selected five trainees rapidly and hired them at the end of their training period, as promised. 3.20 In general, the covenants (para 2.08) were too far reaching relative to the country's available means to implement them. - Covenant iii, requiring that the Government collect the data necessary for the planning of road works, is only now being complied with. UNDP-financed technical assistance has been provided since 1977 to the Bureau of Planning and Programming (BPP), created under the Rural Roads Project (Cr 579-UV). The terms of reference of this technical assistance include the systematic collection of data on, and planning of, the road sector. iork on data collection and analysis was started in 1979. - The Government has had difficulties umplying wIt ovenant iv, which requires that the DPW be adequately staffed. There is not enough qualifieu personnel Lu Upper vula. However, ou-tue-jou traiuug was oriented to enhancing the quality of staff. - As for Covenant v requiring the Government to maintain Upper Volta roans it wa o amitiou fo th financial resources of the cou-ntry. In order to alleviate this problem the Bank provided for the development of t-hp RuirAl Radq Proiect maintenance program. In narticular. the Solenzo-Koun- dougou road is in a good state of maintenance and was reshaped by the Rarvice d'pntratien des routes secondaire (SERS) in early 1980. The Bank is also financing under the Third Highway Project a study of Upper Volta maintenance needs, and the proposed Fourth Highway Project would finance a four-year highway maintenance program. E. Bank Performance 3.21 The Bank tried to appraise the project as quickly as possible so that implementation of the West Volta Cotton Project would not suffer from lack of road serving the area. This led to the underestimation of the project costs and to an overlapping of tasks by RMWA and the Western Africa Projects Highway Division, both involved in the preparation, appraisal and early supervision of this project. During project implementation the Bank adopted a flexible attitude towards project changes and helped the Government prepare the new project components. Nevertheless this preparation took a long time, because of slow Government procedures. This is one of the setbacks when developing Government capacity for project preparation. In retrospect, project completion delays could have been avoided by cancelling the funds remaining under the Credit. IV. ECONOMIC RE-EVALUATION A. The Solenzo-Koundougou Road 4.01 The appraisal report stated that construction of the Solenzo-Koun- increased production of cotton expected from the West Volta Cotton Project. consider the road construction as part of the overall cotton development proj ect . T- stead,it uses more tr_aditionsi mehdlg whc inc.ludes savings in vehicle operating costs (VOC) as the only benefits of the road methodology, the economic rate of return for the road was estimated at 17%. Tho arnnnn4 __r-lA n nmiq hth methodq nnp Iq h=Qnd on the rate of return for the overall agricultural/road development scheme, the other 'tuAntifi1aq VOn Savines only. 4.02 As mentioned in para 2.02. the road construction could not be included in the West Volta Cotton Project because of project preparation constraints, but clearly is a key element for agriculture development in the area. The West Volta Cotton Project has an economic rate of return (ERR) reevaluated at 37% in the PPAR of January 4, 1980 (SecM80-13). If the economic cost of construction and maintenance of the road is added to the cost of the investment for agricultural development, the ERR is still high at 31%, and justifies the investment decision. - 15C - 4.03 Using the more conservative methodology of the appraisal report, based on VOC savings, presents some difficulties since traffic data are scarce. The road is one of those included in the study currently carried out by consultant Halcrow Fox and Associates to determine the social impact of road development in Upper Volta. The only traffic count available to date, of about 35 vehicles per day, covers the month of March 1979 when traffic is at ebb because of the low agricultural activity of the period. Traffic peaks in the November-January period when cotton is harvested and evacuated. Thus, for the whole year, an average traffic of 45 vehicles per day has been assumed. Other data used in the re-evaluation are given in Table 4. In particular, traffic growth has been assumed at 5% based on the projected growth of cotton production during the 15-year life of the road, and regular maintenance costs, including two regravelling operations, have been introduced. The economic rate of return based on these assumptions is 12%. Assuming an 8% traffic growth as per the estimates put forth by consultants BCEOM in their maintenance study (June 1980), the economic rate of return is 15.5%, in spite of the higher construction costs of the road, and it is close to the appraisal estimate of 17%. 4.04 Based on either methodology, construction of the Solenzo-Koundou- gou road shows to have been a justified investment. The VOC savings metho- dology seems conservative; among others it overlooks benefits to the users of some 60 motorcycles and 75 bicycles a day. The study of the social impact of road development should provide a better understanding of the benefits deriving from an improved infrastructure, and help refZUe Lte economic analysis. B. The Drainage Works 4.05 The Government's proposal to finance drainage works was justified on the basis that t,he roads were----------------------------------,8an sometimes after, the rainy season. The works were completed in March 1980 and there i.s no economic dat to make an ex-post eval",Atione V. CONCLUSIONS 5.01 This project was the first in the road sector in Upper Volta, and constituted the preliminary step towards opening a dialogue with the Govern- ment and starting highway operations. 5.02 At the time, the only possible way for building the Solenzo-Koun- dougou road speedily was using consultants for the detailed engineering and a contractor for the construction works. This led to rather high geometric standards for a gravel road carrying less than 50 vehicles per day at its opening. But Government's capacity for preparation and execution of road works was limited: when the Government requested inclusion of simple drainage works on other roads, it took about three years to arrive at a satisfactory proposal. - 16 - 5.03 In the meantime, the Bank assisted in the strengthening and develop- ment of the Ministry of Public Works. The Rural Roads Project (Cr. 5/9-4V), approved in June 1975, provided the means for the Service d'Entretien des Routes Secondaires (SERS) to start a substantial improvement and maintenance program of rural roads which proved successful. Should a road similar to the Solenzo-Koundougou road need to be built now, SERS has the capability to prepare the study and carry out the works. As definition of the works could be uIe £.LLea.ue, LuIV wuIu auw uC LuUF.1U uuLu LL u L1u SCLUricstanuards and allow for savings on construction costs. 5.04 In spite of this success, road maintenance is still a weak point ben Ufits fVrOa, poj mcut s a te Ban orill& nov v fc t.s 1&e onethis benefits from past projects. The Bank will now focus its efforts on this UPEIIR VOLTA COTTON ROADS PRiO,ECT ~ CRIEIT 316-UV PRO-1PCT COMPLETION RIEPORT Actual and Appratsal Estimates of Project Costa Actual Coat as Actual Cout as . _-Appraial EatImates Coat--- ------- Actual Coat --------- % Appiraimal Contiract 1.0cal Foregn Tol Contract. Amount Locil Forein Total Costa cottm 1. Contiruction Å. Solenzo-Koundougou (in million FCFÅ) 254 253 507 3/ 530 389 388 777 153 147 (in Lbousand US$) 990 990 1.980 31 2,255 1,770 1.769 3,539 179 157 B. lounde-Ber-eba (in millioln FICYA) 103 1102 205 166 1/ 83 83 166 .a. 100 (In thoucand Us$) 399 398 797 760 384 385 769 n.a. 101 Ttal Conotr-uction Coat (in million FCFÅ) 356 356 712 696 472 471 942 n.a. n.a. (Lik tliounaid US$) 1,389 1,388 2,777 3,015 2,154 2,154 4,308 n.a. n.a. 11. Co1ultanit Servicem A. Supierviaton cf I (in million FCFA) 7 27 34 52 13 37 50 147 96 (in tlnuand US4) 26 106 132 217 58 174 232 176 107 t. Preinvedtinent Studies (in milltion FCFA) 30 120 150 105 31 90 121 79 115 (In tuuaiid US$) 115 462 577 437 132 395 527 91 121 Total ConsulLanti Servicia (Iin million I'CFA) 37 148 185 157 43 126 169 91 108 (In thouaunld US$) 142 568 710 654 190 569 759 107 116 (in million FCFA) n.a. n.a. ..a. n.a. 3.6 - 3.6 n.a. I. d. (in thc,udand US$) 1.a. n.a. 15 - 15 n.d. n.a. ToLal_ Prl,c Coai (inl million K;FA) 393 504 897 853 2/ 519 597 1,116 124 131 2/ (it i ,åoan,1 Uss) 1,531 1,956 3,487 3,669 3/ 2,359 2,723 5.082 146 139 2/ 1/ Covru tike Draikage Work 2/ Not includin8 lte Scholarslip 3/ Including conttinlgence.lt!. Table 2 UPPER VOLTA COTTON ROADS PROJECT - CREDIT 316-UV PROJECT COMPLETION REPORT Actual and Expected Prolect Implementation Project Contractor ---------Dates of Component Cnsulant/ -ILU r=F~.;=LL %A,)UU4dZ%L .7_UA.LA8 ."A. IJ. L'V" s MJ.LJ6 . WA & Nationality Actual Expected Actual Expected Actual Expected Actual Expected Road Constr. Lorraine de Solenzo- Travaux Pu- Koundougou blics Africains 3/73 2/73 12/73 7/73 1/74 10/74 5/75 3/75 Drainage UDEC (France) Tenkodogo- and Kanazoe Garango Rd. (Upper Volta) 2/78 7/77 8/78 4/78 10/78 9/78 2/80 7/79 Road Ouagadougou- Dragages 10/78 9/78 5/79 3/79 Lio Rd. (France) Ouagadougou- Kongoussi Rd. Feasibility BCEOM 10/73 3/74 2/74 4/74 4/74 8/75 6/75 Study Supervision BCEOM 4/74 1/74 10/74 6/75 4/75 of Constr. (France) Training OFERMAT 1/78 1/78 12/78 12/78 (France) - 19 - Table 3 UPPER VOLTA COTTON ROADS PROJECT - CREIT 316-VI PROJECT CObMPLETION REPORT Ockledule -of- Di-L..sbursements Revised Semester/ Estimates Appraisal FY Act uml 12/73 Estimates 1/73 2/73 200,on0 1/74 600,000 2/74 1,000,000 1,300,000 1175 2,800,000 1,900,000 2/75 1.548.505 3.700.000 2.500.000 1/76 2,668,561 4,150,000 2,800,000 2/76 3,257,461 1/77 3,257,461 2/77 3,385,161 1/78 3,385,161 2/78 3,385,161 1/79 3,385,161 2/79 3,862,467 1/80 4,077,957 2/80 4,150,000 - 20 - Table 4 UPPER VOLTA COTTON ROADS PROJECT - CREDIT 316-UV PROJECT COMPLETION REPORT Economic Analysis Element of Rate of Return Calculationl/ Investment: New road 70 km. zood gravel road Old road : 77 km, bad track Life : 15 years starting in 1975 Construction cost : FCFA 621 million (excluding taxes) Maintenance : - routine: FCFA 57.000 per km - periodic: FCFA 200,000 per km of regravelling every 7 years Benefits: Average daily traffic : 35 vehicles in 1975 Annual traffic growth : 5% Mean vehicle operating cost : - new road: FCFA 57 per km - old road: FCFA 132 per km Calculation Costs Vehicle Operating - - - - - (in million FCFA) - - - - - - Cost Savings Year Construction Maintenance Regravelling (in million FCFA) 1 1974 621 2 1975 4 7 3 1976 4 80 4 1977 5 1978 4 89 6 1979 4 94 7 1980 4 98 a 1oo C A A 1 o .LVO. J.-- 9 1982 4 108 ItA 10n97 A1R .&V .L7 U .9.--. - 11 1984 4 119 13 1986 4 132 14 19,87 4138A 15 1988 14 145 16 1989 4 192 Re-aqtfmatAd ERR = 127 1/ All prices are in 1975 prices. REPUBIC OF UPPER VOLTA - -iTomnbooOa- COTTON ROAD PROJECT ROAD NETWORKy PRIMARY ROADS, Po,ed I PR M R ADS Go 02 4 - SECONDARY ROADS, Earth Aribindo - , RO JE CTROADS D ® NATINAL APITAL N~- -NTERNAIQA BOUAR° OUAH GOUYA .N 5BAR ALOGHO s KAYA K Yoko Bog nd. 01mboon%~.?%," Ben BO LSSA DEDOUGOU N 0 T c ~ ib . G D -Xo z 0<eOl 'N . ":,^ SAFANE 00 KOUPEL A ukuoOuoyo -p oo e- Saboo pag r KOMBISSIR FADA NGOURMA N udoarou L Bi o Bom0 oukou.- N ooom o Dom0 0 X0 G-__~ o.g.j j Dono i-bLEO Sangha OROIDARA0~ . •- -•..B* DEEBOUGODO The mm w m n viry NiLGOEER ti 2O 5 50 75 00 25 SIEiNAbE" E o A l. z4 -4,.--j- ~ 4 . KL T '^ LER u-- AARH 172 BRD 381