Document of The World Bank FOR OFFICIAL USE ONLY Report No. 3373 PROJECT PERFORMANCE AUDIT REPORT CHAD: LIVESTOCK PROJECT (Credit 309-CD) March 12, 1981 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only In the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS 1971 (at Appraisal) US$1.00 = CFAF 256 CFAF 1,000 = US$3.91 1974-78 (Estimated Average US$1.00 = CFAF 218 of Disbursements) CFAF 1,000 = US$4.59 1979 (at Completion) US$1.00 = CFAF 220 CFAF 1,000 = US$4.55 WEIGHTS AND MEASURES 1 kilometer (km) 0.62 miles 1 hectare (ha) 2.47 acres 1 kilogram (kg) 2.2 pounds 1 metric ton (ton) 2,204 pounds ABBREVIATIONS CP Food and Agriculture Organization/World Bank Cooperative Program FAC Fonds d'Aide et de Cooperation Fund for Aid and Cooperation (French) FAO Food and Agriculture Organization FED Fonds Europeen de Developpement European Development Fund IEMVT Institut d'Elevage et de Medecine Veterinaire Tropicale Tropical Veterinary Institute (French) LDU Livestock Development Unit (in Ministry of Livestock) SERARHY Service des Amenagements Ruraux d'Hydraulique Rural Water Supply Service (established in 1972) SOTERA Societe Tchadienne d'Exploitation des Ressources Animales Chadian Cattle Company (established in 1977) UNDP United Nations Development Programme FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT CHAD: LIVESTOCK PROJECT (Credit 309-CD) TABLE OF CONTENTS Page No. Preface ..................................... .............. i Basic Data Sheet ... ..................... ................ . 11 Highlights ..................................... iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. Summary...................... 1 II. Main Issues ......................... ,....... 3 A. Project Design and Objectives ......... ....... 3 B. Delays in Project Process.......,. .......... 3 C. The Project Scope and Project Impact ............. 5 D. Procurement ............... .................... 6 Annex: Delays From Project Identification to Project Completion of Works ........ .. ............. 8 PROJECT COMPLETION REPORT I. BACKGROUND ... ...................... ............. 9 II. FORMULATION ............................................. 9 A. Chronology ........... ........ -.................. 9 B. Project Concepts, Design and Delays ...... ...... 15 C. Targets and Goals ......................... ........ 18 III. IMPLEMENTATION --... ......... .. ...... 21 A. Effectiveness and Start Up ..-..................... 21 B. Revisions ....................... 23 C. Physical Progress ...............*.............. 23 D. Procurement .......... ............ ... ..... ....... 24 E. Costs and Disbursements .......................... .24 Cost per Well ......................, 25 F. Covenants .................. .................... 27 IV. LIVESTOCK PRODUCTION IMPACT ............................ 27 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. TABLE OF CONTENTS (Cont'd) -2- Page No. V. FINANCIAL AND ECONOMIC RESULTS ........................ 30 A. Producer Incomes .................................... 30 B. Financial Implications for Government ................ 30 C. Economic Returns ........................ ........... 31 Significance of Maintenance ........................ 32 VI. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT ................ 32 A. Livestock Planning and Projects ................... 33 Changes in Follow-up Project .......................... 34 B. Pastoral Water Supplies ........................... 35 VII. GOVERNMENT AND BANK PERFORMANCE ............................ 35 Annexes 1 - 3 Map PROJECT PERFORMANCE AUDIT REPORT CHAD: LIVESTOCK PROJECT (Credit 309-CD) PREFACE This is a performance audit of the Livestock Project in Chad for which Credit 309-CD was approved in February 1972 in the ,sum of US$2.2 mil- lion. The credit account was closed fully disburs rin September 1978. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department and a Project Completion Report dated June 1980 prepared by the Western Africa Regional Office. The audit memorandum is based on a review of the Appraisal Report (No. PA-5a) of January 24, 1972, the Credit Agreement dated May 31, 1972, the PCR, and a review of the project supervision reports; correspondence with the Borrower and internal Bank memoranda on project issues as contained in relevant Bank files have also been consulted and Bank staff associated with the project have been interviewed. Due to political unrest in the country, the draft report was not sent to the Borrower. On the basis of this review, OED supports the PCR's findings con- cerning the project's principal achievements and shortcomings. - ii - PROJECT PERFORMANCE AUDIT REPORT BASIC DATA SHEET CHAD: LIVESTOCK PROJECT (CREDIT 309-CD) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Estimate Total Project Cost (US$ million) 3.4 2.7 Overrun - none/a Credit Amount (US$ million) 2.2 2.2 Disbursed )- 2.2 Cancelled ) June 30, 1980 - none Repaid to ) - none Outstanding )- 2.2 Date for Completion of Physical Components 12/74 12/76 Proportion Completed by Appraisal Target Date (%) 100 100.7 Proportion Time Overrun (%) - 80% Incremental Economic Rate of Return (%) 11 -/b Cumulative Estimated and Actual Disbursements (US$ million) FY72 FY73 FY74 FY75 FY76 FY77 FY78 Estimated 0.4 1.7 2.2 - - - - Actual - - - 0.8 1.6 2.0 2.2 Actual/Estimated % 0 0 0 36.4 72.7 90.9 100 OTHER PROJECT DATA Original Actual or Item' Plan Revisions Current Estimate First Mention in Files 02/23/65 Government-s Application - 01/31/68 Negotiations 04/69 07/22/69 Board Approval 02/70 02/15/72 Credit Agreement Date 05/31/72 Effectiveness Date 09/28/72 11/28/72 Closing Date 06/30/75 06/30/76, 06/30/77, 12/31/77, 09/30/78 11/09/78 Borrower Republic of Chad Executing Agencies Ministry of Livestock, Ministry of Rural Development Fiscal Year of Borrower Jan. 1 - Dec. 31 Follow-up Project Name Second Livestock Project Credit Number 783-CD Amount (US$ million) 11.5 Credit Agreement Date 04/14/78 - iii - MISSION DATA Month/ No. of No. of Man- Date of Mission Sent by Year Weeks Persons Weeks Report Identification FAO/CP 05/66 2 2 4 06/17/66 Preparation I FAO/CP 11/66 3 4 12 02/24/67 II FAO/CP 06/67 2 3 6 07/13/67 Appraisal I HQ 09/68 4 5 20 12/31/68.c Follow-up HQ 09/69 1 3 3 11/12/69/d II HQ 09/71 2 2 4 01/24/72Ze: TOTAL 14 49 Supervision I HQ 05/72 2 1 2 06/14/72 Supervision II HQ 11/72 1 3 3 01/22/73 Supervision III HQ 05/73 1 2 2 06/08/73 Supervision IV HQ 02/74 1 1 1 04/05/74 Supervision V HQ 06/74 1 2 2 07/18/74 Supervision VI HQ 12/74 1 1 1 01/03/75 Supervision VII HQ 07/75 1 1 1 07/28/75 Supervision VIII RMWA 12/75 0.5 1 0.5 01/28/76 Supervision IX RMWA 02/76 0.5 1 0.5 03/06/76 Supervision X HQ 04/76 0.5 1 0.5 04/29/76. Supervision XI HQ 12/76 0.5 1 0.5 12/23/76 Supervision XII HQ 06/77 0.5 2 1 07/06/77 Supervision XIII HQ 10/78 0.5 1 0.5 10/20/78. TOTAL 11.0 15.5 COUNTRY EXCHANGE RATES Name of Currency -- Franc Communaute Financiere Africaine (CFAF) Year: Appraisal Year Average Exchange Rate = US$1 = CFAF 256 Disbursement Average (1974-77) US$1 = CFAF 218 Completion Year Average (1978) US$1 = CFAF 220 /a Project was scaled down. 7 Assuming maintenance of wells is assured, ERR can be estimated at about 11%. Since maintenance stopped in 1979 because of civil war, ERR is expected to be minimal or negative. /c Green Cover. /d Back-to-Office Report. /e Grey Cover Report No. PA-5a. /f Progress Review memoranda, not full supervision reports. - iv - PROJECT PERFORMANCE AUDIT REPORT CHAD: LIVESTOCK PROJECT (Credit 309-CD) HIGHLIGHTS The project was expected to support the construction of new wells, the rehabilitation of existing wells and the establishment of two units to maintain wells in the main livestock-producing areas of the Sahel region. The project justification was that the construction, rehabilitation and mainten- ance of wells would bring into productive use about 600,000 ha of grazing land, and prevent decline and ultimate collapse of waterpoints serving an area of over 2 million ha. The project was expected to be completed in three years and its ERR was estimated at about 11%. The inordinately long period of ten years and seven months between identification and completion of the project is the most striking feature of the project since project design was simple and did not substantially change after appraisal. Extraneous factors (security problems and drought), Govern- ment's financial problems and Bank-s rigid procedures equally contributed to the slowdown of the project process. Three appraisals and three credit agreements were needed before the start of works. The project scope as defined in the first appraisal was scaled down by about 60% to stay within available funds. In contrast with the initial delays and despite continuous security problems, project implementation proceeded remarkably well. The project succeeded in achieving its objective to improve watering facilities but its impact was much smaller than initially planned. The grazing area covered by the project can be estimated at about 30% of the appraisal projections. Since increases of civil works were entirely offset by similar increases in meat prices, the project's ERR could have come close to appraisal estimates had the civil war not broken out and hampered well mainte- nance. In view of the latter, the ERR is re-estimated to be minimal or negative. The following points may be of special interest: - The project concept was dictated by the consideration that the Bank was entering a new field of lending in the country and that it was desirable to start with a simple project (PPAM paras. 8 - 10 and PCR paras. 2.15 - 2.17). - The Bank's project preparation procedures proved cumbersome and costly for the simple project concept. By contrast, when the project design was finally amended, the Bank showed flexibility and full cooperation (PPAM paras. 14 - 16 and PCR paras. 2.18, 2.22 and 3.08). - For small contracts and dispersed works strict adherence to civil works procurement procedures proved similarly cumbersome when flexi- bility in execution was important - (PPAM paras. 20 - 21 and PCR para. 2.19). PROJECT PERFORMANCE AUDIT MEMORANDUM CHAD: LIVESTOCK PROJECT (Credit 309-CD) I. SUMMARY 1/ 1. The livestock industry accounts for about 15% of the Chadian gross domestic product and for about 40% of exports. Livestock raising provides the main source of living for a third of the rural population. About 80% of the national herd is raised in the Sahelian zone where pastoralists are not nomadic but practice seasonal migration. The importance of livestock was identified as one of the promising sectors for Bank Group financing during the first economic mission to Chad in 1965. By mid 1966, an FAO/CP mission identified the Livestock Development Project (Cr. 309-CD) which became the Bank-s first operation in the agricultural sector. 2. The exceptionally long period of 10 years and 7 months between project identification and completion of works was the main feature of this project. Although project works were implemented in only 22 months (March 1975-December 1976), almost nine years, three appraisals (1968, 1969 and 1971) and three Credit Agreements, were needed before the start of works. The project scope, as defined at the first appraisal, was scaled down by about 60%. Moreover, the civil war, which broke out soon after project completion, renders the impact and benefits of the project uncertain. Since maintenance stopped in early 1979 and will not be easily resumed when the political situation improves, the actual project return is likely to be minimal or negative. 3. As originally identified, the project was to be an integrated live- stock development project in the Sahel region, including the construction and rehabilitation of watering points, marketing infrastructure, ranches, roads and training of pastoralists. During preparation and the first appraisal, the project-s broad objectives appeared to be over-ambitious and were reduced to a single component including the construction and rehabilitation of wells. The final scope of the project was defined during the second (1971) appraisal as follows: (i) construction of 38 new wells; (ii) repair of 102 existing wells; (iii) establishment of two units to maintain 821 wells in the main livestock producing areas; (iv) expansion of the Livestock Development Unit (LDU) within the Ministry of Agriculture, and (v) carrying out by LDU of four studies mainly aimed at the preparation of further and more comprehensive livestock development projects. 1/ Adapted from the PCR. - 2 - 4. The project was expected to be completed in three years. Its total cost was estimated at US$3.4 million, of which $2.2 million (65%) was to be financed by IDA, $0.5 million (15%) by UNDP to cover technical assistance to LDU, and $0.7 million (20%) by Government. The project justification was based on the simple assumption that (i) the construction of wells would bring into productive use about 600,000 ha of grazing land with a carrying capacity of 80,000 cattle and 70,000 sheep and goats, and (ii) the repair of wells would prevent decline and ultimate collapse of 102 waterpoints serving an area of over 2 million ha and supporting about 210,000 cattle and 180,000 sheep and goats. The well maintenance component of the project was considered essential to prevent a 5% to 10% per year decline in water extraction efficiency of the well network. Based on those assumptions the ERR was estimated at about 11%. 5. Although simplification of the project concept at appraisal could have facilitated prompt implementation, an additional 6-1/2 years, two updated appraisals and two amendments to the original Credit Agreement were needed before the start of project works. The reasons for these unusual delays, which are detailed and commented in paras. 11-16, mainly include security problems in Chad, Government's financial difficulties and Bank's rigid proce- dures. Consequences of the delays were a large increase in project costs and subsequently a 60% reduction in project scope to stay within available funds: only 18 new wells were constructed and 20 existing wells were repaired. The two maintenance units were established, staffed and equipped in 1975. The program of studies, although slightly different from the one originally planned, was carried out by consultants and led to a second livestock project which was appraised in 1977. 6. In contrast with the initial delays and despite continuous security problems, implementation proceeded remarkably well; the project was sub- stantially completed within 22 months compared with 30 months estimated at appraisal. By and large, the project succeeded in achieving its objective to improve watering facilities although its impact is much smaller than initially planned. The grazing area concerned by the project can be estimated at about 30% of the appraisal projections. Increases in cost of civil works were entirely offset by similar increases in meat price. Therefore, the project-s ERR could have come close to appraisal estimates had the civil war not broken out and hampered well maintenance. 7. The project also had an unexpected impact on institution building through the improvement, reinforcement and training of Livestock Depart- ment staff, and through the establishment of the rural water supply agency (SERARHY) which turned out to be one of the most effective government agencies working in rural areas. - 3 - II. MAIN ISSUES A. Project Design and Objectives 8. During project identification and preparation, project design and objectives were controversial. Several approaches were considered: In 1966, the terms of reference of the FAO/Bank Cooperative Program identification mission focused on marketing which was perceived to be the main area for possible investment in the livestock sector. The identification mission, however, advocated a broad integrated approach, including wells, training of and extension to pastoralists, health program, construction of roads and a possible cattle marketing organization. The Bank-s Abidjan office recommended that the project be oriented mostly towards ranching, deemphasizing the provision of extension services and water points to traditional herdsmen. 9. At appraisal, in 1968, the decision was finally made to split the project into two phases and to limit the first phase to the improvement of water resources. This approach was dictated by the consideration that the Bank was entering a new field of lending in Chad and that it was desirable to start with a simple project. The project was not defined as a comprehensive type of livestock development but would lead to an expansion of grazing areas because of the greater availability of stock water. Thus the project would result in increasing stock numbers and offtake. Given the realities and limitations of the country this decision was sound. The Bank was also correct to introduce the concept of maintenance of existing wells in the whole project area . The type of well to be constructed also became a matter of contro- versy; it was finally decided to adopt open wells with construction costs higher than boreholes but with operation and maintenance much easier and significantly less expensive. 10. As defined at appraisal, the project was therefore extremely simple and not particularly innovative since FED had already financed the construc- tion of 318 wells in the country. The merit of this simple project concept was to facilitate quick implementation and to leave more sophisticated invest- ments for a second project. However, project implementation was delayed for several years after appraisal. B. Delays in Project Process 11. The inordinately long period of ten years and seven months between identification and completion of the project is the most striking feature of the project since project design was simple and did not substantially change after the first (1968) appraisal. The PCR (paras. 2.01-2.22 and 3.01-3.08) accurately reflects the causes of delays as well as the continuous misunder- standings that arose between Bank and Government. Extraneous factors, Govern- ment and Bank were equally responsible for the delays, the length of and reasons for are summarized in Annex. 12. Three extraneous factors contributed to the slowdown of the project process: (a) Security problems in the project area started in 1969 and hampered the inventory of wells and well maintenance; they also resulted in a reduction of the project area on its eastern border and made necessary changes in the location of new wells and wells to be repaired. Lack of security certainly contributed to the high cost of well construction and repair con- tracts. Finally, improvement of security was made a condition for Board presentation that took long to be fulfilled. (b) High rates of international inflation combined with the depre- ciation of the dollar entailed a 60% increase of project costs, leading to reductions in project scope. (c) The 1972-1973 drought, which struck all Sahelian countries, aggravated the public finance situation and diverted Government efforts towards drought relief operations at the expense of the project. 13. Government-s financial problems and the complexity of some proce- dures (17 signatures needed for approval of a simple contract) have been also among the causes of delays. Unwillingness - and later inability - to make advance payments to contractors resulted in the withdrawal of the successful bidder and in the need for renegotiating the contract with the second lowest bidder. Later the budgetary situation deteriorated so much that Government (i) had to request 100% IDA financing and the provision of a revolving fund and (ii) hindered project start-up through procedural procrastination such as no clearance of legal documents and non-ratification of the second well contract until the project became effective. Finally, the lack of experienced Government officials made it necessary to rely entirely on expatriate staff and this resulted in long delays in recruiting qualified staff. 14. In such an environment, the Bank's rules and procedures proved extremely cumbersome and sharply contrasted with the simple project concept. The Bank prescribed an unduly long list of actions to be taken prior to Board presentation. The appraisal mission found that the sites and the costs of new wells and required repairs were not sufficiently determined and therefore recommended that an inventory of wells be taken by consultants in order to assess more accurately the costs and the economic viability of the project. This survey, which took four months and required the services of consultants, was probably justified although in the audit-s view it was unnecessary to wait for its completion to prepare the tender documents. During negotiations, in July 1969, the Bank unnecessarily drew up a new list of required actions aiming at establishing sounder cost estimates of well repairs and the benefits of each new well. The Bank also insisted on a field survey to determine the carrying capacity of the grazing areas to be opened by the new wells. Later, the Bank recalculated the rate of return each time new information became available. In retrospect, the Bank-s insistence on having the Credit Agree- ment signed by the National Assembly rather than by the President also is of - 5 - doubtful justification and caused further delays. The Bank-s attitude was obviously not understood by Government which was unfamiliar with Bank rules and which was better used to deal with other more flexible external financing agencies. 15. In 1972, the project finally became effective, but Government's deteriorating financial situation, the withdrawal of the successful bidder, and the need to renegotiate the contract, as well as the need to carry out a further well inspection tour, led to another delay. It became apparent that available funds would not permit implementing more than 40% of the initial program. Government requested the Bank to finance the cost overrun in order to maintain project scope as initially planned, but the Bank refused on the grounds that a change of the Credit Agreement would require six-month addi- tional procedures. In conclusion, it can be said that the Bank was faced with a simple project, which could have been treated in a simple manner. The Bank, however, was continuously entangled in its own procedures and was unable to react quickly to the changing situations in the country.1/ 16. By contrast, when the project was finally amended to reflect its reduced scope, the Bank showed great flexibility and full cooperation which undoubtedly contributed to the quick implementation of the project after the extended delays prior to start-up. C. The Project Scope and Project Impact 17. Project scope, as defined by the 1968 appraisal, was adversely affected by the increase of project costs and the lack of Government counter- part funds. From 1968 to 1976 the evolution of project scope, project cost and financing was the following: Repairs of Financing Total New Wells Wells IDA UNDP Government Cost --u-----units-------- ---------US$ million---------- 1st Appraisal (1968) 45 126 1.3 - 0.3 1.6 Last Appraisal (1971) 38 102 2.2 0.5 0.7 3.4 Actual (1975-76) 18 20 2.2 0.5 - 2.7 18. When it became apparent that the volume of work undertaken had to be tailored to available IDA and UNDP funds, surprisingly the decision was taken to reduce the well repair program by about 80% while the well construction 1/ Central Projects staff states that the PPAR and the PCR are both critical of the Bank and its procedures as a cause of the delays between appraisal and Board approval. However, one could argue that the Bank had good reason to be cautious at that time because of the serious security situation, the uncertainties regarding costs and the Government's imple- mentation capacity and the marginal rate of return. - 6 - program was curtailed only by about 50%. The numerous calculations of the rate of return (see para. 14 above) had always shown that (i) repair of wells had a higher return than construction of new wells, and (ii) repair costs amounted to only 20% of new construction costs. In the audit-s view, when the second well contract was negotiated in October 1974, it would have been more appropriate to have the available funds entirely devoted to the implementation of the repair program, thus ensuring a better return from the project and reaching a larger number of potential beneficiaries. No indication was found in project files that such an alternative had been contemplated by the Bank and based on its own ERR calculationsl. This shortcoming was, however, slightly reduced in its impact because costs turned out to be significantly higher than anticipated (about 60% of the construction cost). It was also found that (i) well deterioration was generally greater than originally anticipated, and (ii) the real nature and accurate extent of well repair could not be determined until repair works started. 19. The impact of the project has finally been more important in terms of the Bank gaining more experience and better knowledge of the country-s livestock subsector than in terms of physical implementation. A second and more comprehensive livestock project (Credit 783-CD) which was signed in April 1978, made full use of studies carried out under the first project and of lessons learnt from the first project's experience. The following innovations were introduced in the second livestock project: (a) construction and mainte- nance of wells should be closely related to the traditional land and water use and made contingent on the pastoralists' participation. Each group of pastor- alists should obtain recognized grazing and water rights in its area, which should be carefully delineated. In return, the group would financially contribute to well construction, repair and maintenance operations; (b) the provision of a technical package to groups of pastoralists, including animal health, protection of rangelands and extension service; (c) the provision of credit to some pastoralists. D. Procurement 20. The project demonstrated the problems arising due to small contracts for dispersed works, the nature of which is not apparent until men and equip- ment have been moved to the work site. In this project, extensive field surveys were carried out to determine the location of the wells to be con- structed and repaired and to make cost estimates more accurate. Because of the security situation, the nature and sites of civil works were finally quite different from expectations. The Bank-s insistence for following a strict and orthodox civil works procurement procedure, when flexibility in execution was important, proved therefore cumbersome and unnecessary. A sample would have sufficed to establish reasonable cost estimates and to draft tender documents. 1/ The Region comments: "We agree that the available funds could, in principle, have been more appropriately allocated to maintenance but Lthe visibility and conceptual-interest factors (PCR, para. 5.07) were fairly compelling arguments to proceed with some new wells". -7- 21. The project also faced the difficulty that foreign firms were not interested in this kind of contract and that only few local contractors were able to implement the works. One of the merits of the project is to have contributed building an effective institution (SERHARHY) able to construct, repair and maintain wells. Under the second project, well construction and maintenance were expected to be carried out under force account by SERHARHY. ANNEX PROJECT PERFORMANCE AUDIT REPORT CHAD LIVESTOCK PROJECT (CREDIT 309-CD) DELAYS FROM PROJECT IDENTIFICATION TO PROJECT COMPLETION OF WORKS Stage Date Length of Delay Main Reasons for Delay FA0/CP Identification Mission 05/66 Preparation mission I 11/66 Conflicting views on project objectives and scope. Preparation mission II 06/67 2 years, Transmission of preparation report 5 months Need for inventory by consultants of new wells and wells to Government 11/67 to be repaired. Government application 01/68 Appraisal Mission 09/68 . Inventory of,wells,by consultants. Inventory of wells by consultant Deterioration of the security situation. engineering firm 12/68-4/69 10 months Recalculation of ERR following new cost estimates by consultants. IDA loan committee 02/69 Negotiations 07/69 Bank request of a field survey to assess the carrying capacity of the grazing land. Follow-up appraisal mission 10/69 Delay in preparing tender documents. Preparation of draft tender documents 02-07/70 Conditions set for Board presentation: improvement of security Final tender documents 12/70 situation and receipt of bids for civil works contracts. Invitation to tender 01/71 Bank-s insistence for locating wells to be repaired and site of new wells. Contract awarded to lowest bidder for 3 years, Government has not cleared legal documents. 38 new wells and 102 repairs 05/71 4 months Need for reviewing the economic viability of the project. IDA updating appraisal mission 07/71 Because of high rate of inflation, and increased project costs, need for Revised draft credit agreement proposed revision of the Credit Agreement. to Government 01/72 Start of works hampered by non-payment by Government of the start-up funds. Board presentation 02/72 Insistence of Bank on having the Credit Agreement signed by the National Supervision mission I 05/72 Assembly. Signature of credit agreement 05/72 Project Effectiveness 11/72 100% financing requested by Government 11/72 Government decision not to ratify the well contract until Credit Government request for 100% financing effectiveness. accepted by IDA 05/73 Public finance situation aggravated by drought and internal security Amendment to Credit Agreement 02/74 costs. Withdrawal of the successful bidder 02/74 2 years, Need for increasing IDA financing up to 100% of net-of-tax project Effectiveness of the amended Credit Agreement 04/74 4 months costs and for establishment of a revolving fund. End of new well inspection 07/74 New well inspection, carried out by Government, the second lowest Agreement on a negotiated contract with the bidder, and a consultant. second lowest bidder for 18 new wells and Project Authority preoccupied with programming drought relief and 18 repaired 10/74 assistance. Start of Works (new wells and repairs) 03/75 Well maintenance team becomes operational mid-75 Livestock Department carries out studies 1 year, and prepares second project 75-76 8 months Completion of project works (including two 12/76 additional repairs) TOTAL 10 years, 7 months -9- CHAD LIVESTOCK PROJECT - COMPLETION REPORT I. BACKGROUND 1.01 The Bank's first economic mission to Chad found that the country was unusually poorly endowed with natural resources and that distances to ocean ports as well as poor internal transport facilities severely hampered trade. One of the few sectors identified as promising for Bank Group investments was livestock which, in the mid-1960s, accounted for about 15% of GDP and about 40% of exports. A livestock project was identified in mid 1966, prepared in 1967 and appraised twice, in September 1968 and September 1971. The resulting Livestock Development Project, for which an IDA credit of US$2.2 million (Credit 309-CD) was approved on February 15, 1972, became the Bank Group's first operation in the agricultural sector. 1.02 The project's broad objective was to provide for the planning of the integrated development of the livestock industry and to prepare projects for external financing. The project's main specific objectives were to improve stock watering facilities with a consequent expansion of available grazing, "a fundamental requirement for development of the livestock industry," and to ensure that all wells in the main cattle areas would be properly maintained by Government. 1.03 The project was to be executed by Government's Rural Engineering Department in the Ministry of Agriculture, and the Livestock Development Unit which had been set up in 1970 in the Ministry of Livetock; the Unit was to be funded by UNDP for three years as part of the project. These agencies took the lead in responsibility for execution but, increasingly, and particularly after mid-1974 when the bulk of the work was undertaken, the Livestock Department became the key agency in project execution. In February 1977, IDA appraised a follow-up project, Chad Livestock II, which was approved on March 30, 1978. This report is based on a desk study of the voluminous project files and documents accumulated over the period 1965-78. The Basic Data Sheet is at page ii, and basic documents consulted are at Annex 1. II. FORMULATION A. Chronology 2.01 The first mention in Bank files, in February 1965, referred to a possible investment in a cattle route, from Abeche in the east central part of the country near the Sudan border to Sarh in the southeast. This project idea inspired the terms of reference dated January 1966 for a FAO/Bank Cooperative Program (CP) livestock project identification mission, for which a Government request had been made the previous fall. The terms of reference - 10 - indicated that a project "might include not only the equipment of cattle trails in water points and other facilities, but also the equipment of holding grounds and the building up of slaughter houses and marketing facilities;" marketing, not production, was perceived to be the main area for possible investment. 2.02 On arrival in Chad in May 1966, Government presented the two-man mission with an aide-memoire asking that attention be directed to Batha Province in the center of the country, and to an investment package including wells, training of and extension to pastoralists, cattle routes from Batha to N'Djamena in the west, and a possible cattle marketing organization. The mission's report, noting that the initial Abeche-Sarh cattle route was being considered by FED, proposed a $3.9 million project along the lines of Government's aide-memoire. It included provision of 85 wells, both along cattle routes and in stock-raising areas, a training and extension program, and a cattle staging ranch close to the N'Djamena abattoir, from which annual beef exports by air had already reached 5,000 tons by the mid-1960s; the report envisaged that the project would be executed by a semi-autonomous agency to be managed by a six-man expatriate team supplied under contract by a consulting firm. The report cautioned that little was known about cattle marketing and that, while it was clear that over-valuation of the Nigerian pound and customs duties and other trade barriers strongly influenced cattle trading, the subject should be further studied during preparation in order to ascertain whether there would be effective demand for cattle movement and marketing facilities. 2.03 The identification report provoked much comment, including strong advocacy from the Bank's Abidjan office that the terms of reference for a preparation mission "be directed towards a project of one or several ranches under foreign management deemphasizing completely the provision of extension service and water points to traditional herdsmen"; development through ranching, or through improving traditional pastoralism, was being heatedly debated at the time. The actual terms of reference did not pre-empt the mission's judgment on this question stressing, instead, the need for better information on the physical characteristics of the project area, and on the marketing channels, and for a detailed proposal for project organization and management. In December 1966, the five-man preparation mission returned with a two-phase proposal. The first phase was to include building of 45 new wells in Batha Province, repair of the 110 existing wells, and establishment of a project management unit in the Livestock Department which would undertake studies on (a) the exact location and cost of an additional 40 wells in Batha Province, (b) the problems and cost of setting up a staging ranch near N'Djamena, and (c) the setting up of 6-7 central ranches in the area where the 40 additional wells were envisaged. The Back-to-Office report included a further proposed first-phase activity, an ill-defined hybrid: "the provi- sional establishment of 3-4 central ranches in the area. These would func- tion as centers for water control, extension services and marketing." The - 11 - second-phase was to include provision of the additional 40 wells and estab- lishment of 10-12 ranches, including the definitive establishment of the first batch of 3-4 central ranches. The report, issued in February 1967, removed the ambiguous "provisional establishment" of ranches and referred, instead, to studies for the establishment of 9 central farms; farms or ranches, their purposes remained unclear, beyond providing the project authority with some physical livestock production facilities. Total project cost for the first phase was estimated at US$1.8 million. 2.04 The report also raised two broad issues: it considered Government's livestock tax policy as both punitive and regressive, a disincentive to development, and it raised the question of whether concrete-lined open wells, while nonetheless acceptable in terms of the rate of return, were the best solution to the water extraction problem. The Bank's comments focussed on these two broad issues but also expressed a reluctance to go ahead with a first phase, and its implicit commitment to a second phase, without a closer look at the feasibility of the follow-up investments. A three-man follow-up mission was accordingly arranged for June 1967, but the terms of reference were limited to the choice of open wells versus mechanical pumping from tubewells. No mention was made of the tax problem nor of the purposes of the central ranches. The report. issued in July, conclusively showed that open wells were preferable to tubewells where the aquifer was at 30-50m; this was the case in the area where the 45 new wells were envisaged. Based on new evidence from well contracts let in early 1967, project cost estimates were revised to US$1.3 million. The report was transmitted to Government in November 1967 with a covering letter inviting Government to state its inten- tions with respect to livestock tax policy and to the amount of the Chadian contribution to project costs. Government replied in January 1968 welcoming appraisal in March, remaining silent on tax policy and project financing but pointing out that (a) wells should be sited not only on the basis of agrosto- logical and hydrogeological data but also on the basis of ethnological data, in order to avoid problems between ethnic groups by arranging that each well serve a grazing area traditionally exploited by a socially-homogeneous group; and (b) alternative surface installations at wells should be tested. The Bank welcomed these suggestions but, because the official Government request had not been received until February, regretted that appraisal could not now take place in March and would have to be postponed to the fall. 2.05 Arriving in Chad in September 1968, a five-man appraisal mission found that no work had been done to identify sites of the proposed 45 wells nor to identify wells to be repaired, the nature of the repairs and their costs. The mission cabled Bank headquarters with a proposal to recruit, on the appraisal budget, a consultant engineering firm then working on super- vision of well construction in the country to undertake an inventory of wells to be repaired. This was accepted and the mission arranged, prior to its departure, that the inventory to establish detailed well repair cost estimates be started, with a target completion date of March 31, 1969. Apart from rejection of Government's proposal to rehabilitate 15 diesel-electric powered tubewells near N'Djamena, on the ground that such works should be studied in - 12 - the wider context of water requirements of the area, the appraisal mission's Back-to-Office report found the project "basically sound". Total project cost was estimated at US$1.56 million subject to the inventory of well repairs for which a tentative estimate of US$100,000 was included in project costs. As appraised in 1968, the project concept was simple: a limited well construction, repair and maintenance program would be implemented and, alongside these economically justified components, studies leading to a more comprehensive second-phase project would be undertaken by a Livestock Development Unit to be set up in the Ministry of Livestock. That the first phase was not more comprehensive was regretted by the Bank's Livestock Division but the design of the proposed project "is not dictated by our choice of approach...but rather by the reality that it is all that the Chad Government is in a position to carry out in the next two or three-year period in addition to the emphasis that it has been putting and expects to continue to put on livestock disease control." 2.06 In December 1968, Government replied to the Bank's appraisal follow-up letter by (a) confirming its agreement to the well repair inventory; (b) outlining a procedure for specifying the sites and costs of the new wells, for which it was estimated that sufficient work would have been done to permit negotiations in June 1969; and (c) proposing that, while the terms of ref- erence of the Livestock Development Unit be retained, financing be sought from grant money. The Bank followed up on (c) by proposing that Government approach UNDP, and took the initiative in briefing UNDP on the background as well as the advisability of the Bank being appointed as executing agency. As pre- sented to the Bank's Loan Committee in February 1969, the project had two uncertainties: well repair costs might be anywhere between US$55,000 and US$500,000, and it was not yet clear whether UNDP would finance the Livestock Development Unit. An IDA contribution of 80% of total project costs was proposed, or about 90% of net-of-tax costs. Following Loan Committee approval, the Bank informed Government in March and pointed out that (a) the two uncertainties would have to be cleared up, during negotiations at the latest, but (b) the well siting survey did not need to be completed before negotiations. 2.07 Field work for the well repair inventory was completed on April 1, about a month late, and agreement in principle was reached in April with UNDP on their financing of the Livestock Development Unit. In contrast to these steps being made toward mid-year negotiations, the Bank was becoming concerned about reports of a deteriorating security situation in Chad. The appraisal mission had been alert to such problems but had been impressed by the fact that construction of FED-financed wells had never been interrupted for lack of security; in addition, the mission considered that if the well repair inventory could be completed without incident, it would be a good indication that subsequent execution could be undertaken. In March 1969, the Bank learned that construction of FED-financed wells had been stopped after a site foreman had been taken hostage and subsequently released after payment of a ransom. In the Bank's letter of May 8 to Government, the security question was - 13 - for the first time addressed explicitly. The Bank pointed out that reason- able bids from civil works contractors were unlikely to be received if security were uncertain and asked, as a condition for negotiations, for information on the status of the FED program and when work would be resumed. Government replied on May 31 stating that work had been disrupted for technical reasons but would resume next October after the rains, but that the well repair inventory had been carried to a successful conclusion. Government also pointed out that bids for the IDA-financed wells were unlikely to be as low as for the FED contract since the former involved 45 rather than 165 wells and they were more dispersed. The significance of this observation for the project cost estimate was not followed up, neither by the Bank nor by Government. 2.08 Following contacts with other donors, the Bank became convinced that the security situation was indeed deteriorating. Meantime, the Bank was informed that the well repair inventory would be available early in July (it was received on July 10) and negotiations had been scheduled for July 21. The Bank decided to proceed with negotiations and to make improvement in the security situation a condition of Board Presentation to be manifested by resumption of work on the FED-financed well program and receipt of reasonable bids for civil works contracts. These points were put to Govern- ment in the Bank's letter of July 7. It is uncertain whether Government actually received that letter prior to the delegation's departure from N'Djamena; it is certain on the other hand that the delegation did not receive the draft Credit Agreement until their arrival in Washington. 2.09 Although substantial agreement was reached on the Credit Agree- ment, negotiations were inconclusive because of new information from the consultant engineer (para 2.05) that cast doubts on the economic justifi- cation of the project. With new data on the depths and costs of new wells, a revised calculation gave an economic rate of return to new wells of 8.5% (down from 11%). While the appraisal mission leader argued that the project should nonetheless go ahead, the Bank decided that the major assumptions in the economic analysis should be checked in the field, in October, after further consultant studies on the precise location, depth, and cost of the new wells and the carrying capacity of the rangeland to be opened up. Since it had been agreed that Board presentation would be conditional on calling for bids and receipt of reasonably-priced submissions, the additional studies and checks were not considered a further possible cause of delay. With one member of the Chadian delegation who stayed on after forma- negotiations were concluded, an aide-memoire was drawn up specifyin, , actions to be taken prior to Board presentation, now scheduled for February 1970, with work expected to begin after the 1970 rains in October. 2.10 On leaving Washington, the two senior members of the Chadian delegation visited New York to request UNDP financing for an advisor to the Director of Livestczk. The UNDP was willing subject to enquiring whether the Bank considered such a procedure might interfere with the planned Lirestock - 14 - Development Unit. The Bank replied that appointment of an advisor would, if anything, assist establishment of the Unit. UNDP then gave FAO authority to recruit and the only candidate, to whom the Bank had no objection, was approved by Government in January 1970 not as Livestock Adviser but as Head of the Livestock Development Unit. It is unclear whether this was a deliberate distortion of intentions or a genuine misunderstanding arising from lack of continuity in senior personnel of the Chadian Livestock Depart- ment. In any event, from that time on Government regarded the Unit as having been established. 2.11 Meantime, a three-man follow-up appraisal mission visited Chad in mid-October and found that the security situation had worsened so that no field work was possible. The mission considered, however, that further field studies were not warranted, other than a detailed site survey to be carried out by a hydrogeological consultant in the company of a representa- tive of the successful bidder. The timetable of actions to be taken (para. 2.09) was revised, with Board presentation rescheduled for July 1970 but with work still scheduled to begin in October. The mission reviewed the findings of the well repair inventory, adjusted the project benefit assump- tions, and reworked the economic rate of return; their new estimate was 11% for new wells. 2.12 Government's substantive though incomplete response to the revised timetable of actions was received in the Bank in early February 1970. Government alerted the Bank of further probable cost increases: a revised rate of return calculation for new wells gave a return "approaching 10%". In April, the Bank advised Government to proceed with calling for bids but, in June, the Bank was informed that the tender documents had not been completely drafted. End-July, complete drafts were received by the Bank from the consultant enLineers. In early August, the Bank received from Government 12 pages of comments on the draft tender documents. The Bank then found itself, however inadvertently, the arbitrator between consultant engineer and client. The outstanding matters were, however, resolved fairly quickly and revised draft tender documents were received in the Bank in mid-October. These were approved and final documents, ready for distribu- tion, were received in December. The invitation to tender was issued on January 15, 1971, with bid opening set for May 3. However, as one Bank staff commentator noted An a memo of October 1970: "there is still a feeling in Chad that, like FAC and FED, we will do all that is necessary to have the project executed... .it will be difficult to get the Chad Government to issue the bids, evaluate them and then supervise execution of the project." 2.13 Three bids were received for the well construction and repair contract, all from firms already operating in Chad. Government proposed to award the contract to the lowest bidder, the smallest of the three firms. Still another recalculation of the return to new wells gave 11.4%. The Bank then decided, in August 1971, to send an updating appraisal mission to (a) assess the security situation, even though reasonably-priced bids had been received; (b) assess the Livestock Development Unit, even though it had been established and recruitment of a livestock and a grassland specialist was underway; (c) clear with Government the substance of the legal documents agreed to during negotiations two years earlier; and (d) review the economic - 15 - viability of the project, even though all calculations of economic rates of return, from the outset, had yielded between 8% and 11% for new wells. The Zwo-man mission ound no surprises and, even, that the security situation had improved as manifested by the resumption of activities on the FED-financed well program. 2.14 During updating of the December 1968 Green Cover appraisal report in late 1971, the extent of the required changes in the draft Credit Agreement became evident and the question arose of whether to renegotiate. The widely- shared reluctance of staff to go through negotiations again was given support b- Government's letter of November 4 to the Bank; this stated that unless work cn wells could begin in February 1972, it would not really get started until November and inflation at 8% would jeopardize the project cost estimates. It also became apparent that expansion of the existing one-man Livestock Develop- ment Unit through IDA/UNDP cofinancing would require a complex administrative procedure and that it would be preferable to seek an increase in UNDP financing, appointment of FAO as sole executing agency, and provision of IDA funding for specific studies to be undertaken by or through the Unit. A Bank staff member visited Chad in mid-December to discuss the revised draft Credit Agreement and reached agreement on all points except a Government proposal to increase project costs in view of local inflation. Government afterwards furnished satisfactory evidence on the inflation rate, and an increase in project costs, along with a corresponding increase in the IDA credit amount, were accepted by the Bank. A further revised Credit Agreement was sent to Chad in January 1972. Agreement was finally reached and a credit of US$2.2 million was approved by the Board on February 15. Prior to signing, (a) Government proposed and the Bank accepted that the well construction and repair contract be exempted from taxes and that the IDA disbursement per- centage be increased accordingly; (b) a Bank lawyer visited Chad to clear up two outstanding legal issues (para 3.01) concerning ratification of Credit Agreements; and (c) a one-man supervision mission visited the country for a week in May. The mission discovered that the successful bidder on the well contract had not yet started work, and would not therefore be able to start until the end of the rains, because Government had not advanced him the start-up funds. The Credit was finally signed on May 31, 1972. B. Project Concepts, Design and Delays 2.15 It took exactly six years from issue of the initial and quite detailed identification report to the signing of the Credit Agreement. It took two and a half years from identification through to the Green Cover appraisal report. This was not exceptionally long and the project design as of December 1968 was not later changed in substance, yet there was a subsequent delay of three and a half years between Green cover and signing: this was exceptional and all the more intriguing in the retrospect since project design was not at issue. Project design had started in 1965 with an exclusive focus on marketing. At Government initiative, the project concept was broadened in mid 1966 to production and marketing, with production to be obtained through provision of extension and services to pastoralists. This concept was chal- lenged by some Bank staff who advocated the ranching approach that was being followed at that time in the design of the Kenyan and Madagascar livestock - 16 - projects.l/ This challenge had some initial success in that the preparation report of February 1967 relegated extension and services (other than watering points) to pastoralists to a possible second-phase project, to be realized as a subsidiary function of central ranches which were to be studied in the first phase. While an important study component was subsequently retained by the 1968 appraisal mission, the scope of the studies was broadened and the concept of ranches as focal points for provision of services to pastoralists or as demonstrative devices was not mentioned. The project as designed in December 1968 was simple and made no pretense at being a conceptually-exciting exercise. 2.16 Based on the observation that there were unexploited grazing resources to which access was limited by lack of watering points, the project was to include construction of new wells, repair and maintenance of existing wells, and studies leading to a livestock development plan and preparation of further projects. That there were considerable unexploited grazing resources was beyond doubt, but there were doubts about the adequacy of the detailed local information required to permit the exact siting of new wells. An issue in project design was that exploitation of hitherto unexploited resources might lead to overgrazing; this concern led to the selection of open wells with manual or animal water extraction rather than mechanical pumping from tubewells, quite apart from the cost advantage of the first alternative. Potential overgrazing was, on the other hand, not treated with the reverence that the issue attracted in the mid 1970s. During preparation and appraisal, it was observed that the main grazing species were annual grasses which set their seeds in September, during the rainy season; subsequent overgrazing around watering points in the dry season could not therefore affect the next rainy season's pasture production. This botanical point was and remains valid but was overlooked in the arguments, in the mid-1970s, about watering points as a cause of overgrazing and, thereby, advance of the desert. 2.17 The main reason for the limited scope of the project was the judgment by the preparation and appraisal missions that Government did not have the executive capabilities to carry out any more demanding activities. Government Departments comprised a few senior staff, sometimes only one technically-qualified man, and there was no prospect of any significant change in the availabiity of qualified staff in the next few years. Nor were francophone consultants experienced in livestock production readily available, either in ranching or in improved pastoralism, other than in the narrower fields of water development and disease control. In the latter field at least, Chad was fortunate in that the regional headquarters of IEMVT had been set up in N'Djamena in 1953; first-class veterinarians were always on call. The project therefore consisted of an uninspiring though worthwhile well program to be executed by contractors and engineering consultants but also, paving the way ahead, a study component also to be undertaken by consultants. There was, however, a contradiction between the importance the Bank attached to the study component with the explicit expectation of a more conceptually exciting follow-up project within three years, and the paucity of Government's 1/ For which Project Performance Audit Reports have been issued: Kenya (Credit 129-KE) October 19, 1976; Madagascar (Loan 585-MAG), April 11, 1977. - 17 - executive capability which was unlikely to change quickly. It is unclear, moreover, whether much thought was given to the utility of a livestock development plan other than as a respectable framework for a future project. There were not then, nor are there now, many alternate courses of action. A simple and less costly approach would have been to examine, during supervision missions, possible follow-up project ideas as well as sectoral issues like taxation, and then to undertake a specific preparation study. 2.18 The delay between the Green Cover appraisal report and the signing of the Credit Agreement was remarkable, possibly a record. Although difficult to gauge their relative importance, four factors contributed to the 42-month delay. First, the authority of the central Government was not clearly established and was increasingly challenged in the center and north of the country, including the proposed project area. These challenges took the form of sporadic skirmishes and kidnapping but the September 1968 appraisal mission found the situation calm and learned that well construction work in the center and north had not so far been affected. Beginning in March 1969, news reached the Bank of occasional incidents prejudicial to effective execution, and therefore the costs to be expected, of well construction and other works. To what extent did the Bank, in response to these uncertainties, drag its feet in processing the project? The record is clear in terms of tests set up by the Bank (paras 2.07-08) to establish whether project works could be effectively carried out. It is unclear, however, whether the Bank's awareness of the security problem and consequent doubts about the practicabil- ity of efficient execution did not find expression in ways other than the specific tests. These ways are suggested by the next two factors that con- tributed to the 42-month delay. 2.19 Second, in negotiations in August 1969, the Bank drew up a list of actions to be undertaken prior to Board presentation. The main purposes of these actions were to establish a sound cost estimate of well repairs, and to site, cost, and assess the benefits of each of the new wells. Since the initial cost estimate for well repairs was between US$50,000 and US$450,000 and the appraisal report had provisionally retained US$100,000, greater accuracy was certainly desirable. But the Bank's insistence that each of the wells to be repaired, scattered over 3,200km2, be inspected was unwarranted; a sample would have sufficed to narrow down the costs, quite apart from the practical problem that the extent of major repairs cannot be readily specified until work is underway. Equally for the new wells, accurate siting was not necessary to establish reasonable cost estimates nor, even, to draft tender documents. The Bank also insisted unnecessarily on a field survey to establish the carrying capacity of the grazing land to be opened up by the new wells, this ostensibly to permit a more accurate assessment of the benefits; on this point, the Bank subsequently withdrew its insistence. Even allowing for the desirability of lending later rather early in the project cycle, this list of actions was unduly long; perhaps, having discarded the ranching approach, the Bank simply did not know what to - 18 - do; prudence in a relatively new kind of investment would have played some part, as did an unduly orthodox civil works approach to discrete, dispersed rural works for which flexibility in execution was important; perhaps, too, uncertainties arising from the security situation made for much caution. 2.20 Third, the economic rate of return to new wells, initially esti- mated at 11%, was recalculated each time new information became available. Because benefits arose primarily from long-term herd growth, the rate of return was fairly insensitive; the recalculation done in September 1969 gave the lowest return, of 8.5%. This finding caused increased concern in the Bank about the project's economic justification, the project having been earlier characterized as "marginal", and thereby gave added impetus to the insistence for more information on costs and, particularly, benefits. The main reason for the importance attached to the rate of return findings was that the insentivity of the return was, in 1969, overlooked; perhaps, too, a 10% rate of return was a more rigid decision criterion at that time than later. 2.21 Fourth, just as Government had little executive capability so too did it have little planning and preparation capacity. Government's substan- tive response to the list of actions stipulated in August 1969, did not come until January 1970. Moreover, Government had little familiarity with Bank procedures and the differences between the Bank's role in project planning and execution as compared to the more directly participative role played by other external financiers. But the Bank, too, was not faultless in the rapidity of its reactions; Government's letter of January 28, 1970 was not answered, with the key authorization to proceed to invite bids, until April 22. Perhaps, too, the Bank did not make sufficient effort to explain its own methods and procedures to Government officials although, while there was good continuity in the Bank staff involved in project processing, such efforts would have been hampered by frequent changes in Government depart- ment directors. 2.22 In sum, the four factors that contributed to the delay were: (a) internal security problems, (b) Bank prescription of an unduly long list of actions to be taken prior to Board presentation, (c) Bank concern about the economic justification of new wells, and (d) Bank and Government's lack of familiarity with each other's methods and procedures. While it is difficult to gauge the relative importance of these factors, it would be surprising if the Bank did not in fact exercise more caution in respect of (b) and (c) because of (a). C. Targets and Goals 2.23 The main links between objectives and means as specified in the appraisal report are shown in the chart at page 19. Pending effective means of delivering services to increase animal productivity, increasing animal numbers was considered to be the only practicable means of increasing Obecties and e-ans ab Specified in the Appr"is eprt Spe wciu P i'rovide dry-seaon accens to n Setain and improve dry season acces. to 2 million ha of rangeland and thereby Plan development of the liveetock Objcties dditional 600,0h. of ran geln«a pre.ent a decline of 42%oin herd and flock numbers and (b) permit an induetry and pr.er. fUture ivsmna and thereby permit a 12N incree i ncre.ae of 1?% 1n herd and flock nme in herd and flock nurber-. blaem Con.truct "i open welle in south- Repair 102 open welle built in the mid- Maintain 821 open welle in the £our Expand the Livestck Developent unit went Batha Provence where (a) water 1950'a in western Batha Province and cattle-raising provineG, includineg (from ene man) and reorient it. "ork, is at len than 60, (b, carrying thereby increae efficiency, .esured by Btte Province, in the Sahelian n., by adopting September 1968 Terme of Refer- capacity of the rangeland is better effective over thoretical water extraction, and tuereby increase efficiency fre nce, toard. liveøtok pl.aing and than i adul t cattle nit per loh.. tro_ 75% to 100%; without-the-project, 90% to 10()%; without-the-project, project preparation. efficiency would decline by 5 percentage efficiency wuld decline by 5 pont per ysar to 50% after 5 year. and pereentage points per year to 50%1dertak. 8 pecific atui. reeain unchanged thereafter. efter 8 yere and remmin unchanged Locate new ..ell o the ba-ic,among thereafter. other criteriu, that the reu of grar.g reana.bly acceanibe to ay tinge to b* c ed bY th. well is within a radius uf 7 1wtkm, £c .nt of L t to exafina the Unit,* that there would be 15,0Oha per wel Eoc well to be visited by e work and progran with the Mini.tries ef aufficient for 1500 adult atttle unite. maintennce unit every 3 yeara; Finane, Planning, ad Agriculture (Rural capacity of a unit: 140 welle/year. Engineering Department). Provide wsll uf 1.e diam.ter equipped with parapet end 6 pulleye se that 6 -eanin can drw water i.ulteaneuly, to extract i lit/sec or 36,000 liters Per lO-hour day, .ufficient for 1500 adult eattle unita at 25 itere/unit/ Winnýing bidder elected but contract not yet awaded; wok to beg.nassona Two .aintennc unit. to be .et up fincainog secured aupervision tu be provIded by the Rkra Engineering Departet. and run a f - unt operation strengthened by additional technical aqeistance, or by consultant engineere. by the R.ural Eng.neering Department; ID opoie100% of oPerating cot nYe.r 1, 501% in Y-a 2. Effect of eew welln, repair. fed meintenea would be an ncrease in livestock number, with no change in animal productivity: -attle of take 11.5%, mallet.ck offtake 25%, it being a~eu,ed in er, 0 that il exinting welle are being fully utilized; without-the-project. offtake would reain unehneged exept that additi.ane stress nales would oc.c an effeetve water extraction declined; with-the-project. ,fftake for consumption would decline during the perjod of herd and flock huildup to full uti)lation of the repaired, mairtined on n eella to 8.5% for cattle and 20% for small atock, the balance Ieing put into herd growth. A,-mal prod.utlvity would not encrease unt1i further inveeteet. were made. __a - _ _ _ . ._..,. Prepare at leat one lies.tock develop.e. t prject - 20 - livestock production. Animal numbers were considered to be limited by access to dry-season grazing resources; hence the focus on repair and maintenance of existing wells as well as construction of new wells. New wells were, however, to be constructed so that the number of animals was consistent with carrying capacity, the key elements being: (a) discharge obtainable from an open well with extraction by six animals working simul- taneously, estimated at one liter per second or 36,000 liters per 10 hour work day (work being done at night in the hottest part of the dry season); (b) daily water consumption per adult cattle unit estimated at 25 liters, so that 1,500 adult cattle units could be watered per well per day; and (c) spacing of wells so that the number that could be watered could also be fed: with a carrying capacity of one adult cattle unit per 10 ha, for instance, 15,000 ha would be required and the radius of action of each well would be 7 km. It was also considered that 7 km was the maximum reasonable walking distance; it followed that wells should only be constructed where the carrying capacity was at least 1 per 10 ha if full use was to be made of the well. This model was, at the time, innovative in its precision and clarity but, reasoning "on the average", it did not take account of the variability of the environment. Where a well was capable of such discharge, water was in extreme cases extracted for 24 hours per day, not 10; similarly, walking distances in extreme cases were 15 km or more, not seven. The model's built-in control of animal numbers in relation to pasture resources would therefore be broken in conditions other-than-average. 2.24 The production effect was based on two key related assumptions: (a) the pre-project number of animals was the maximum that could be watered from the existing well network at its existing water-extraction efficiency; and (b) animal numbers would be increased as soon as new wells were constructed and efficiencies of existing wells improved. There was no sound evidence for assumption (a) but it was not seriously challenged and subsequent events suggest that it was valid. Assumption (b) required that, in the absence of increased animal productivity, livestock producers would be willing to allocate some of their pre-project offtake to herd growth, i.e. that they would be willing to accumulate capital in their herds and flocks at the expense of consumption or, indeed, investment elsewhere. It was assumed that cattle offtake would remain unchanged at 11.5% but with 3% now allocated to herd growth and only 8.5% available for other purposes. There was no sound evidence for this assumption and it, too, went unchal- lenged; indeed it reflected the then conventional wisdom that traditional pastoralists had few consumption wants and that what savings they were able to realize went into increasing herd size. The appraisal report did not contain any analysis of producers' capacity to forego offtake during herd and flock buildup to the maximum that could be watered from the expanded, repaired and maintained well network. That maximum, expected to be reached in Year 10, was 30% higher than pre-project animal numbers. By contrast, instead of assuming constant animal productivity, a case could have been made that easier access to water would have improved the calving rate and the calf survival rate, at least while pasture was relatively abundant. - 21 - 2.25 The specific objective of preparing a plan for the livestock industry and at least one project was to be reached by increasing technical assis- tance beyond the one advisor who had been supplied to the Ministry of Livestock by UNDP in 1970. A four-man unit was envisaged, to operate under the original terms of reference drawn up in September 1968. The project was to include eight specific studies that were considered, prior to any work having been undertaken on a livestock development plan, to be relevant to a possible follow-up project. Quarterly meetings with the three other mini- stries directly concerned were to review the Unit's work and program. III. IMPLEMENTATION A. Effectiveness and Start Up 3.01 The project was approved on February 15, 1972 but did not become effective until November 28, 1972. No special conditions of effectiveness had been set. This project was, however, used by the Bank to clear up two outstanding issues concerning effectiveness in Chad in general. One issue concerned the validity of the President's signature, prior to approval by the National Assembly, of the decree authorizing the Government to contract a loan. The Legal Department over-ruled other Bank staff and insisted on National Assembly approval. The Assembly was in recess during the summer, did not reconvene until October, and approved on November 24 the decree that had been signed by the President on July 4. The delay in effectiveness arising from this legal nicety had consequences greater than the mere five months would suggest. In Chad, as in the other Sahelian countries, field work takes place between November and June, in the dry season. Planning for each year's activities takes place in September - October and any activities that are still uncertain at that time are not accorded high priority. This was the price of the delay in effectiveness until end November, and it came on top of a long series of frustration and delays. 3.02 Nor did the project start promptly thereafter. Three quite separate factors made for further delays. First, ratification of the well contract, which had been awarded after Bank approval in 1971, required seventeen signa- tures; Government decided not to start the ratification process until it was assured of IDA financing, i.e. until after effectiveness. The contract was still not ratified at the beginning of 1974, when the successful bidder withdrew. It is unlikely that this delay was due solely to administrative ineptitude; rather, since the contract provided for an advance payment to the contractor, Government may have stalled the ratification process because it had not made, and was later unwilling to make, budgetary provision to finance the advance payment. Second, the two well maintenance units were to be managed by expatriate well-diggers. The recruitment process was initiated early in 1973 but no prospective candidates had been found by the end of the year. Third, the studies which were to be commissioned by the LDU did not, with one exception, get off the ground in 1973 due, initially, to the in- sistence of the Head of LDU that he work to the broad, advisory terms of reference under which he had been appointed rather than to the specific, - 22 - investment-oriented terms of reference accepted by Government and IDA. These differences were resolved by May 1973, at about the time that the severity of the incidence of the drought was becoming apparent. For the following year and a half, LDU was preoccupied with programming the abundant drought-relief assistance that was then forthcoming and the original program of studies was, necessarily, given low priority. 3.03 A further effect of the drought was that the public finance situa- tion, already aggravated by internal security costs, deteriorated further. This event, together with a 20% decline in the value of the dollar since appraisal, prompted the May 1973 supervision mission to propose 100% financing of net-of-tax project costs and establishment of a US$50,000 revolving fund to prefinance project expenditures under those categories for which direct payment was inappropriate. These innovative proposals were accepted by the Senior Vice President, despite the reservations of the Controller's and Legal Departments to the revolving fund, and approved by the Executive Directors in December. Effectiveness of these changes was made conditional on an appropriate amendment to the Credit Agreement, which was signed on February 1, 1974, but also on ratification of the well contract that had been awarded in 1971. At this time, however, the successful bidder withdrew. The February 1974 supervision mission then agreed that a contract should be negotiated with the original second-listed bidder, whose bid had been only 5% higher than that of the lowest bidder. But it also became apparent that a new inspection of the wells to be repaired was warranted, since the original inspection had taken place in 1969. It was further considered, rightly, that the new inspection could entail changes in the locations of new wells to be constructed. Since all this preparatory work would take some months, the Bank proposed to drop the second condition of effectiveness of the amendment to the Credit Agreement; the amendment became effective on April 4, 1974, the special account for the revolving fund was opened in June, and the first disbursement was made in July, 29 months after Board approval. 3.04 Nor did work start promptly thereafter. The new well inspection was carried out by a team consisting of a representative of each of the Livestock Department, the Rural Engineering Department, the Provincial Governor, the second-lowest bidder, and a consultant engineer. Their work was to have been completed in March so that, provided a reasonable contract price could be negotiated, work would begin in October. The inspection was not, however, finalized until July and construction did not begin until April 1975. The two reasons for this further delay were illustrative of the structural problems in getting things done in Chad. First, the Livestock Department representative became indisposed after the first phase of the inspection that had entailed 2,000 km of rough riding in a four-wheel drive vehicle. He could not be promptly replaced because of the lack of availability, owing to other priority commitments, of the few senior staff of the Livestock Department. As soon as one man was indisposed, work ground to a halt. Second, the rainy season from July to September was a dead season for all kinds of work, including planning and preparatory activities. Many of the senior technical staff were bilateral technical assistants who took their two-month vacations each year at this time. Likewise, although the well inspection was completed in uly, follow-up leading to a negotiated contract did not resume until October. The only project activity that actually started in 1974 was the procurement of vehicles and equipment for the two well maintenance bases, but deliveries did not take - 23 - place until mid-1975. Throughout 1974, the LDU was still preoccupied with programming drought-relief assistance and initiated only one of the studies that had been foreseen, namely the Ati-N'Djamena stock route. Following the aerial photography that had been done in Batha in late 1973, the study was to assess the costs of existing cattle trekking in order to determine what new physical works, primarily wells, were justifiable. Before field work could begin, however, financing for the stock route had been obtained and the study became redundant. This was illustrative of a further structural problem that had already come to light during the establishment of LDU and that, in the aftermath of the drought, was clearly in evidence: foreign financing was abundant relative to the paucity of project ideas and the executive capacity of the few senior technical staff. B. Revisions 3.05 The Amendment to the Credit Agreement that became effective in April 1974 (summarized at Annex 2, Table 1) provided for 100% financing of net-of-tax project costs and a revolving fund of US$50,000 to prefinance expenditures under Categories III, VA and VB. The increased share of financing was estimated to require an additional US$400,000 which was to be covered from the original contingency provision of US$500,000. The decline in the value of the dollar was also estimated to require an additional US$400,000, but the amount of the credit was not increased to compensate for the shortfall. 3.06 A further amendment was made in June 1978 when it became apparent that there would be a small unspent balance after all work, as redefined (para. 3.07), had been completed. A further category was added for procure- ment of vaccines and a corresponding payment of US$43,000 in November 1978 completed disbursement of the US$2.2 million credit. 3.07 Following the mid-1974 well inventory, which confirmed that the condition of the wells had indeed deteriorated, new cost estimates were made by the original second-lowest bidder for new wells and renovation of the more-deteriorated existing wells. The cost changes are examined at para. 3.12. A negotiated contract was subsequently agreed, and approved by the Bank, for the construction of 18 new wells and the repair of 18 wells for the amount that had originally been provided in the credit, namely US$925,000, being Categories I and II and the corresponding amount from the unallocated. The original provision had been for 38 new wells and 102 repairs; counting a repair as one-fifth of a new well, the volume of work was reduced from 58 new-well equivalents to 22, i.e. a scaling down of 62%, after a three-year delay between the award of the first and second contracts. C. Physical Progress 3.08 Project activities really began in early 1975 and in view of all the previous delays, and despite continuing security problems, proceeded remarkably well. The well construction and repair works began in March 1975, even though the contract was not ratified until April, and works were sub- stantially completed by June 1976. Works were completed at the beginning of - 24 - the following dry season, by December, and included two additional repairs. Works actually undertaken are compared to the original intentions at ap- praisal on Map 3626R. All works were supervised by a consulting engineer, with considerable assistance from the Rural Engineering Department which happened to be well-staffed at the time. The finishing touches at the beginning of the following dry season were supervised by the Rural Engi- neering Department directly. The well maintenance bases became operational during 1975 and were fully staffed by the end of the year. UNDP assistance to the LDU was terminated during 1975 whereupon the Unit ceased to exist. At the same time, and with considerable assistance from RMWA, four major studies were commissioned by the Livestock Department; all field work was completed during 1976 and the last of the reports was issued in July 1977 (Annex 1, Table 2). A second livestock project, appraised in March 1977, drew heavily on these and other studies that had been done in 1976. As originally appraised, the first livestock project was to have been completed in 30 months; in effect, once it got started, it was substantially completed in about 24 months. D. Procurement 3.09 The contract for well construction and repair had been awarded under international competitive bidding prior to Board approval of the credit. The works were ultimately carried out through a negotiated contract with the second-lowest bidder on the original tender; this contract covered 50% of the credit. Vehicles and ancillary equipment, amounting to 15% of the credit, were to be procured through local competitive bidding and this was in fact done. The services of technical assistants and consultants, amounting to about 20% of the credit, were obtained following acceptable procedures. The remaining 15% of the credit financed: (a) operating costs of the two well maintenance bases, for which cement, steel reinforcing bars and other materials were procured through local competitive bidding while other items, mainly labor and vehicle operating costs, were not suitable for competitive bidding, and (b) vaccines, which were procured through a negotiated contract with the local vaccine-production laboratory. E. Costs and Disbursements 3.10 The appraisal report estimated project costs at CFAF 876 million, equivalent to US$3.4 million at the prevailing rate of CFAF 256 per US dollar. The cost estimate included 10% for physical contingencies on new wells and well repairs and price contingencies at 10% per year. The tax element in project costs was estimated at CFAF 105 million (US$0.4 million). IDA was to finance US$2.2 million or 73% of net-of-tax costs; UNDP was to finance US$530,000 or 16% of net-of-tax costs, exclusively for the LDU; and Government was to finance the balance, amounting to 11% of net-of-tax costs. At completion, actual costs and financing are best examined excluding the - 25 - UNDP-financed LDU activities since, in practice, there was not 'one' project; what happened is better characterized as 'contiguous financing' rather than cofinancing. By main project activity, the net-of-tax comparison between actual costs and appraisal estimates is as follows: Appraisal Actual Actual/ Estimates Costs Appraisal ---(US$ '000)------ % New wells 729 652 89 Well repairs 389 459 118 Supervision 179 181 101 Vehicles & equipment 371 326 88 Well maintenance 579 355 61 Studies 234 184 79 Vaccines -- 43 - Total 2,481 2,200 89 Source: Annex 2, table 2. No account taken of the decline in CFAF: USS rate from 256 at appraisal to 218, the weighted average of disburse- ments. With minor modifications between project activities, the breakdown of actual costs was remarkably close to appraisal estimates. Yet these similarities disguise more than they reveal since, after the 1974 amendment to the Credit Agreement, expenditures were determined by reference to the disbursement schedule and the volume of work undertaken was fitted to the available funds. 3.11 IDA disbursements had not even begun before the date of completion of disbursements that had been estimated at appraisal. Yet once the project effectively started, in early 1975, disbursements proceeded at much the same pace as had been envisaged. The appraisal estimate was that disburse- ments would be completed over seven quarters; in the seven quarters ending December 31, 1976, over 80% of the credit was in fact disbursed. Disburse- ments dragged on for another two years due partly to slow requests for payment from consultants that had undertaken the studies but mainly due to indecision, of both parties, on how to wind down the revolving fund. 3.12 Costs per Well. - (Annex 2, Table 3) The appraisal estimate was based on an awarded, but unratified, contract for which the bids had been - 26 - opened in May 1971. Unit costsl/ were CFAF 4.69 million per new well, for which the average depth to water table was 44m with an additional 5m for the filter column, and CFAF 0.92 million per well repair, for which the average effective water-drawing capacity was estimated to be 75%. These unit costs included a 24% tax element. The ratio of prices for new wells to well repairs was 5:1; the awarded contract, for 38 new wells and 102 repairs, can be considered to have been for 58.4 new-well equivalents. The negotiated con- tract, for which base prices were net-of-tax as of September 1974, was for 18 new wells and 18 well repairs, or 21.6 new-well equivalents at a total net- of-tax cost of CFAF 225 million; specific unit costs were CFAF 7.35 million and CFAF 1.44 million respectively, excluding a common overhead of CFAF 67 million. Allocation of the overhead by the value of the work yielded unit costs of CFAF 10.46 million and CFAF 2.07 million; the 5:1 price ratio was found again but both unit costs had increased by 120%. Of this increase, 49% was due to inflation but a 24% decrease should have arisen from exclusion of taxes from the negotiated contract. The real increase in costs was therefore 95%. A further 35% was due to the reduction in the volume of work over which the heavy overhead costs of establishment and operation of a supply base were to be spread. In the negotiated contract, these costs represented CFAF 3.1 million per new-well equivalent in contrast to CFAF 1.1 million had the original volume of work been maintained. The real increase in unit costs consequent on negotiation of the contract was therefore 60%. Before agreeing to the contract, the Bank reworked the economic rate of return and found, primarily due to increased cattle prices, a return of 12%, but see para. 5.05. 3.13 Although unit costs had increased by 60%, there is absolutely no evidence in the files of any malpractice in the procurement of these works. The likely explanation of the increase in cost is threefold: (a) it was increasingly necessary to pay staff special allowances to induce then to move into harsh country of questionable security; (b) there was plenty of construc- tion work available in Chad in 1974-75 as a result of drought-relief assist- ance; and (c) there were few construction firms anyway. The initial contract, which had been let after international competitive bidding, had attracted only three bidders; all were firms already established in the country. In view of this experience, recourse to a new tender through international competitive bidding was not a reasonable alternative to proceeding with the negotiated contract. Since the investment still seemed economically worthwhile, the best course of action was that which was followed. 3.14 The actual cost of the well contract was CFAF 240 million, the 7% increase over the negotiated price arising from the proper application of the price adjustment clauses. The total cost of engineering supervision was CFAF 37 million. Expressing these items pro rata with the base costs, the actual unit costs were CFAF 12.91 million for new wells and CFAF 2.56 million per well repair or, at CFAF 218 per dollar, about US$60,000 and US$12,000 respectively. The cost per linear meter of new wells was about US$1,200. During the same period, the cost per meter in West African coastal countries was about US$800; given the circumstances, the costs in Chad were not unacceptably high. Such costs do nonetheless raise the issue 1/ To avoid ambiguities arising from changes in the CFAF: US$ rate, this discussion refers to CFAF prices only. - 27 - of whether tubewells ought not to have been drilled instead. Yet although probably cheaper, tubewells would not have been a practicable alternative: pump technology in the early 1970s could not cope with pumping from 50m using manual or animal power; and even though such pumps are now available, maintenance of the pumps would have been, and would still be, a particularly difficult problem to master in Batha Province. F. Covenants 3.15 Government complied with all but three covenants. Two of these related to LDU, which was not caused "to (i) function under terms and condi- tions to be agreed upon from time to time by the Borrower and the Associ- ation and (ii) submit progress reports every three months to UNDP and the Association", as per Section 3.07 (a) of the Credit Agreement. One of the reasons for non-compliance is to be found in the wording of the covenant itself; since LDU was financed by UNDP, the terms and conditions under which it was to function were a matter, primarily, for the Borrower and UNDP. And one of the terms agreed by the Borrower and UNDP, prior to the signing of the Credit Agreement, was that progress reports should be made six-monthly, not quarterly. From the outset, LDU was subject to two different terms of reference and the Head of the Unit chose to work to those set by his employer, the FAO. This matter was resolved in principle in May 1973, in favor of the terms of reference proposed in the appraisal report, but then the unforeseen task of programming drought-relief assistance necessarily became the first priority for the Unit. Section 3.07 (b) provided that the work of LDU was to be reviewed quarterly by an interministerial committee. Such meetings were never held. The third case of non-compliance concerned audits: audit reports were not furnished within four months of the end of each fiscal year, as per Section 4.01 (b) (iii) of the Credit Agreement. Only one audit was carried out, in March 1976, which covered all project expenditures up to that time, when the credit was about 60% disbursed. The audit included on-site inspection of physical facilities, stocks and methods of drawing up statements of expenditures, and reported favorably. Although no subsequent audit was carried out despite repeated Bank requests, most of the disburse- ments made after March 1976 were direct payments in respect of contracts approved by the Bank. In addition, the operation of the revolving fund was reviewed and found satisfactory by a member of the Bank's Controller's Department during the eleventh supervision mission in December 1976. IV. LIVESTOCK PRODUCTION IMPACT 4.01 As appraised, the project's new well component was to bring into productive use some additional 600,000 ha of grazing land (15,000 ha per well) with a carrying capacity of 80,000 cattle and 70,000 smallstock (1,500 adult cattle units per well). To be made accessible by the construc- tion of 38 wells, grazing resources were estimated to become fully stocked through natural herd growth in 10 years, it being assumed that offtake for consumption and sale would decline to 8.5% for cattle and 20% for smallstock - 28 - during the period of herd and flock build-up, and then revert to the pre- project rates of 11.5% and 25%. The project's well renovation component was to make fully effective 102 existing wells which were estimated to be operating at only 75% of their water-extraction capacity. The increased water availability was to permit a pro rata increase in animal numbers, to be attained in 10 years as for the new wells. The project's well mainte- nance component was to sustain 100% water-extraction capacity of the new and renovated wells as well as to make fully effective all the other existing wells, which were estimated to be operating at 90% of capacity. Of these latter wells, 258 were in the area from which livestock were expected to be drawn to the new and renovated wells. A key effect of the well maintenance component was to prevent deterioration of the 360 existing wells (102 plus 258): the appraisal report estimated without-project (with-deterioration) output in Year 10 at only 58% of the pre-project output. This estimate derived from the ungenerous assumption that, without the project, water extraction capacity would not deteriorate below 50%, yet this level was expected to be reached as early as Year 5 for the 102 wells that were already defective in Year 0. The relative importance of the three well components as contributors to incremental output is detailed at Annex 3, Table 1 and summarized below, where cattle offtake is used as a proxy for all livestock output: Maintained Renovated New Wells Total wells(258) wells (102) (38) (398) -----------------(head)--------------------------- Incremental cattle offtake arising from: 1/ Prevention of deterioration - 23,400 7,700 - 31,100 59 Expansion of resources 2/ 6,200 6,100 9,200 21,500 41 Total 29,600 13,800 9,200 52,600 100 Relative contribution to incremental cattle offtake 56% 26% 18% 100% 1/ Difference between Year 0 output and Year 10 without-project (with-deterio- ration) output. 2/ Difference between Year 10 with-project output and Year 0 output. 4.02 Just as the appraisal estimates were based on best informed guesses in the absence of any data, actual output has to be estimated in the same way. A first approach suggests that since actual execution reached only - 29 - 50% for new wells and only 20% for renovated wells (or 38% of new-well equivalents), these two components together could not have generated more than about 40% of the output expected from expansion of accessibility to grazing resources. Other things being equal, this incremental output could not have been attained unless herd growth took place at the pace envisaged (para 2.24). But other things were not equal: whereas a key assumption underlying the appraisal estimates of output was that the available grazing resources were fully exploited in 1970, the extended drought of 1972 and 1973 caused severe livestock losses. While post-drought herd growth was subject to biological constraints, the rangeland was quickly rejuvenated and by 1975 already, the available grazing resources were less than fully exploited. Comparison of the results of the livestock census conducted in 1975 with 1970 Livestock Department estimates of the animal population suggests that 30% of the national cattle herd was lost during the drought. Batha Province was considered, even at the time of the drought, to have suffered less seriously than neighboring provinces; losses in Batha were perhaps 20%. Between 1970 and 1975, the water-extraction capacity of existing wells wld have deteriorated; comparison of the 1969 and 19;4 wvil inven- tories shows that the weighted average deterioration was about 10%. The net change was, therefore, that in 1975 there was a 10% 'surplus' in available grazing resources. Investments made in new wells and renovations in 1975 and 1976 have certainly expanded accessible grazing resources, but it is likely that only now, in 1980, is any genuinely incremental output beginning to be produced, as a net result of the two trends; (a) natural herd growth, but also (b) further deterioration of existing wells. 4.03 Nearly 60% of the incremental output was to be obtained through the prevention of deterioration of existing wells. The two well maintenance units started working in the 1975/76 season and their operation was financed under this project through the first quarter of the 1977/78 season. There- after, the unit in Batha Province was financed under the Rural Projects Fund (Credit 664-CD) and the unit in Ouaddai Province was financed by several external sources. Field work was intermittently disrupted in 1977/78 as the security situation deteriorated and was stopped in early 1978/79 when the civil war broke out. Field work has not yet resumed. It has not, therefore, been possible to assure regular maintenance of existing wells as had been envisaged, namely dredging and minor repair of each well every three years. In addition, once minor repair work was started on a well, it often became apparent that further work was required. In the one, complete, uninter- rupted season of 1976/77, the unit in Batha Province maintained only 30 wells but works required at six sites were so extensive that they consti- tuted major repairs rather than routine maintenance. In two further cases, it proved impossible to effect the repairs and new, replacement wells were constructed. 4.04 Appraisal estimates were based on a growth of output beginning in Year 1 and rising through Year 10 to a plateau, determined by full exploita- tion of accessible grazing resources with all wells maintained at 100% waterextraction capacity. Project expenditures began in earnest in 1975 and, through the first five years, it is unlikely that there was any genuinely incremental output. Beginning in 1980, incremental output will be generated but not sustained unless the wells are maintained, yet the current situation - 30 - in the country prohibits resumption of maintenance work. While it is not yet possible to foresee when this situation will improve, the new wells and the renovated wells are providing and will provide for the next five years, even without maintenance, access to more pasture than if the project had not been undertaken. But unless maintenance is resumed, accessible pasture will decline and, with it, incremental livestock output. 4.05 The appraisal report did not foresee any technological changes as a direct result of the project, even though it might reasonably have been argued that easier access to water would have improved technical coeffi- cients, at least while pasture was relatively abundant. Herd composition and carcass weights were nonetheless assumed to remain unchanged. Such changes as have occurred through 1979 arose from exogenous factors like the drought, not from the project. V. FINANCIAL AND ECONOMIC RESULTS A. Producer Incomes 5.01 The appraisal report contained no analysis of producer incomes. The report ignored milk, the key output of extensive pastoralism, and could not have foreseen what happened to beef. Over the last decade, producer incomes have changed dramatically, first with the drought, and then with the post-drought scarcity of slaughter stock in the face of buoyant demand in Nigeria. Liveweight cattle prices have increased five-fold, from CFAF 40/kg in 1970 to CFAF 200/kg in 1979. Output is probably still less than it was in 1970, perhaps 15% lower, since the post-drought herd recon- stitution process has not yet caught up with pre-drought numbers. Livestock producers' incomes have therefore increased four-fold in current prices between 1970 and 1979, or about 50% in constant prices. In this respect, they have fared better than cotton farmers in the south of the country whose real incomes have declined over this period. These changes have arisen quite independently of the project but, since it is not yet possible to assess what, if any, incremental output has been generated, it is impos- sible to take them into account in assessing the project's effect on producer incomes. These price and income changes do suggest, however, that there is scope for exacting beneficiary contributions to the costs of any future investments (para 6.03). This question was not addressed at appraisal when the preoccupation was almost the opposite, namely the concern that livestock head taxes as then practiced were regressive and unduly high. B. Financial Implications for Government 5.02 At appraisal, it was envisaged that Government would finance 22% of project costs, of which about half would be recovered through taxes payable by the contractor for new wells and well repairs; Government's contribution to net-of-tax project costs was to be 11%. In the event, Government made no - 31 - direct contribution to project costs, although some staff were seconded to the well maintenance bases, and there was much participation of staff, without formal secondment, in supervision of well construction and repair, initial establishment and operation of well maintenance bases, and in commissioning and supervising studies. 5.03 The appraisal report did not examine the project's longer-term financial Implications for Government, presumably because the context was that Government revenue from the livestock sector was more than double Government expenditure on the sector. In the late 1960s, expenditure was about CFAF 160 million (US$0.6 million); revenue rose to a peak of CFAF 600 million (US$2.3 million) net of collection costs in 1966 and then declined due to increasing evasion of livestock head taxes to CFAF 340 million (US$1.3 million) in 1970. Subsequently, the livestock head tax, which accounted for half the sectoral tax proceeds in 1970, became increasingly difficult to collect and was in fact abandoned during the 1973 drought. Exports of chilled meat, which accounted for most of the rest of the proceeds, fell from 12,000 tons in 1970 to 3,000 tons in 1975. Revenre P- of ccllactior. costs had fallen to CFAF 150 million (US$0.7 million) in 1975, while expenditures had risen to CFAF 180 million (US$0.8 million). During this period, the overall public finance situation deteriorated; the current account changed from a small surplus to increasing deficits. 5.04 The continuing costs to be borne by Government related to the operating and capital replacement costs of the two well maintenance units. During the project, Government was to have financed 50% of the operating costs in Year 2 and 100% in Year 3. The subsequent costs were estimated to be CFAF 41 million (US$190,000) per unit per year in 1973 prices. Year 3 was, in fact, the 1977/78 season, when the original obligation had attained CFAF 66 million (US$300,000) in current prices. At that time, it was simply not possible for Government to finance these costs and, in view of the high profitability of well maintenance (para 5.07), the continued financing of the unit in Batha Province was assured under the Rural Projects Fund (Credit 664-CD). C. Economic Returns 5.05 Economic rates of return at appraisal and as reassessed in October 1974 are detailed at Annex 3, Table 3. These returns, and the corresponding net present values at a discount rate of 10% over 30 years, are compared at Annex 3, Table 4 to the results obtained from analysis of two revisions to the October 1974 data: (a) a five-year lag in benefits to reflect the impact of the drought (para. 4.02), and (b) a five-year benefit lag together with valuation of benefits at 1979 prices and adjustment of costs to 1979 terms. For the four cases examined, the economic rates of return are similar: 1. Appraisal 11.0% 2. Reassessment in October 1974 12.4% 3. As at 2, with benefits lagged 5 years 10.6% 4. As at 3, with benefits at 1979 prices, costs adjusted 11.0% - 32 - The similarities between cases (1) and (2), and between cases (3) and (4), arise from the slow growth of benefits, the substantial impact of the residual value obtained from the growth in herd size, and the unchanged basic assumptions from which the benefits are derived. That there is not a larger difference between cases (2) and (3) arises from the unusual benefit stream; without the project, stress sales increase as the well network deteriorates and these negative benefits are not offset by increased with- project offtake until Year 5. A five-year benefit lag (Case 3) entails zero instead of negative benefits in Years 1 through 5, whereupon the five years of negative benefits begin. Two of-the above assumptions were, and remain, questionable; although little was known at the time of appraisal and under- standing is still far from complete, recent improvements in information increasingly cast doubts on the assumptions: (a) that pastoralists would be willing to reduce offtake sufficient to reach maximum exploitation of newly-accessible pasture resources after 10 years; and (b) that an aggregate herd model is perfectly divisible so that all wells and pasture would be 100% exploited after 10 years. Inspection of Map 3626 shows that, during execution, some new wells had to be sited at less than the minimum distance between wells (15 km) precisely because of indivisibilities arising from, for instance, acquired grazing rights of different tribal groups. Although the herd growth and divisibility assumptions can therefore be questioned, it is not possible to select alternatives that could be defended against the charge of arbitrariness. It is clear, on the other hand, that the rates of return shown above are maxima. 5.06 Those rates of return are maxima in another sense, namely that they all presuppose that maintenance will be assured. This has not happened (para. 4.03). Maintenance stopped in early 1979 and will not be easily resumed even if the general political situation improves quickly. Given this uncertainty, it is not possible to calculate a realistic rate of return at completion beyond the 11% (maximum) shown above. The other extreme can be illustrated by assuming no maintenance over 30 years and a 5% per year linear decline in water extraction efficiency of the well network; the rate of return would be less than 0%. 5.07 Significance of Maintenance. Economic rates of return are shown at Annex 3, Table 4, broken down by the three kinds of works: maintained wells, renovated wells, and new wells. In each of the four cases examined, the economic rate of return to the maintained wells is about 1.4 points above the return to the renovated wells, which is about 2.6 points above the return to new wells. Expressed as rates of return, these differences are small; the return to maintenance emerges more clearly by reference to net present values (also at Annex 3, Table 4). While the numbers do not imply that the renovations and new wells should not have been undertaken, they do serve to highlight the significance of maintenance and the priority that it should be accorded. This confirmation of a common-sense point can also be shown by a common-sense approach, by examining the relative costs per unit of grazing made accessible: % of Accessible Grazing Cost Cost/1% of grazing Before After (1979 $) made Accessible New Wells 0 100 60,000 600 Renovated Wells 60 100 12,000 300 Maintained Wells 90 100 1,500 1/ 150 1/ 3 years maintenance to move from 90% to 100% accessibility. Thereafter all wells (of all kinds) maintained at the same cost to sustain 100% accessibility. - 33 - The cost of making more grazing accessible is indeed lowest in the case of maintained wells. Despite the striking truth of this observation, there is in practice a strong tendency to overlook the significance of maintenance. In this project, the new wells were the largest single cost item, the most visible of the project's outputs, and were conceptually interesting in respect of the criteria established to determine specific sites. Much of the effort of project management and of Bank supervision was put into the new well component. The wells were eventually constructed but, even before maintenance was disrupted by the political situation, no watertight system of assuring regular maintenance had been set up, despite provisions made in later projects. Future investments in pastoral water supplies will have to give first priority to maintenance. VI. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT A. Livestock Planning and Projects 6.01 The planning objective was to be reached through a four-man tech- nical assistance team, the Livestock Development Unit. It was expected, though it was not explicitly stated, that LDU would outlast the project. In the event (paras 3.02, 3.04 and 3.08), the Unit became preoccupied with programming the abundant drought-relief assistance that became available in the wake of the 1973 drought. By the time the UNDP technical assistance project was terminated in 1975, a national livestock development plan had not, even in outline, begun to be designed. This was, however, not an important failure since the utility of a national livestock development plan had never been clearly demonstrated (para 2.17), quite apart from the genuine urgency of the programming work that was undertaken in 1973-1975. With the termination of UNDP assistance, LDU ceased to exist as there was no one available to staff it. Its functions were taken over by the few senior staff of the Livestock Department. Given the circumstances, this was the best outcome. 6.02 The studies that were to have been undertaken through LDU were, with some modifications, commissioned by the Livestock Department in late 1975. Three important studies were undertaken in 1976. Two of these were valuable inputs to the design of the follow-up project, Livestock II (Credit 783-CD), namely the identification of pastoral units in Batha Province, and health and productivity of sheep and goats. The third study, on livestock taxation, clearly traced the recent history of the sector as, first, a large net contributor to the Treasury and, later, a modest net recipient of Treasury funds. The study made two key recommendations, of which one was promptly adopted. This was that, in view of widespread evasion of livestock export taxes and the cost of better enforcement, Government should enter into a partnership with the private exporters of live animals by setting a mutually- beneficial export fee. The setting and collection of this fee became one of the responsibilities of the Chadian Livestock Company (SOTERA) that was established in mid-1977. The SOTERA system has worked well. The other main recommendation was that Government should introduce fees for vaccinations. Government agreed in principle but considered that charges for vaccinations - 34 - were premature, arguing that not all pastoralists were convinced of the benefits of vaccinations, so that the universal coverage required to protect against contagious diseases and ultimately to eradicate some of them would be jeopardized. The Bank shared this view. A good case can nonetheless be made that since livestock vaccinations are, in practice, seldom free, it would be preferable to levy explicit charges in order to get part of the proceeds out of the pockets of vaccination staff and into the Treasury. 6.03 Changes in Follow-up Project. 1/ A follow-up project was appraised in March 1977, when the results of the three main studies were available as well as the results of other separately-financed studies, notably agrosto- logical maps of the Sahelian zone, a livestock census, and herd reconstitu- tion in Kanem Province. The follow-up project (Cr. 783-CD) is quite different from its predecessor; in the sense that it is broad-based, it is less different from the original project proposed by Government, endorsed by the first identification mission, but rejected by the Bank in 1966 (para 2.02). The Bank, after having also rejected the ranching approach, consi- dered that the objective of increased livestock .production was to be reached by providing animal health services, stock watering points and marketing facilities. It was recognized that the thrust of health, water and marketing is in the direction of increasing the rate of exploitation of grassland. It was not, however, recognized that provision of new wells by Government establishes a public good to which all pastoralists have access and that existing arrangements governing access to pasture are thereby broken. The follow-up project was designed to tackle this critical oversight, primarily by allocating stock watering points and thereby pasture to re- cognized associations of pastoralists. Water, and pasture, would once again be privatized so that the incentive system would be switched away from exploitation of public goods and towards management of limited resources. This would create the scope for introducing improved animal husbandry measures and, although much less is known of the subject, in improved range management. Based mainly on the pastoral unit identification study, this was the key innovation in design of the follow-up project. It has not yet been put to the test, however, for although signed in April 1978, Credit effectiveness has been held up by the security situation. A second impor- tant innovation was the introduction of payment by pastoralists for drugs, minerals, vitamins and food supplements, and the introduction of cost recovery provisions for the construction of new wells as well as maintenance. These measures, too, have not yet been put to the test. 6.04 The follow-up project was to be executed by the existing Livestock Department, strengthened only in those two areas where existing staff were not fully competent, namely financial management and animal production. Indeed, the project was designed to build on the strengths of existing staff and in-country consultants. That this approach could have been contemplated 1/ The Rural Projects Fund (Credit 664-CD) and the Sahelian Zone Project (Credit 739-CD) included components dealing with pastoral water supplies but were not, strictly follow-up projects. - 35 - in 1977 is a measure of the institutional development that had taken place over the previous decade. While such changes occurred independently of the first livestock project, there was one positive link in that the project was, after mid-1974 when the bulk of the work was undertaken, increasingly executed by the Livestock Department. The few senior staff of the Department thereby acquired a somewhat broader experience and they were obliged to come to grips with questions of sectoral strategy. B. Pastoral Water Supplies 6.05 At appraisal, it had been envisaged that the two well maintenance units would be set up within Government's Rural Engineering Department. Before the units were established, however, Government had set up a semi.- autonomous agency, within the Rural Engineering Department, with overall responsibility for execution of rural water supplies (SERARHY). This agency was to establish well maintenance, repair and construction bases in each of the main regions of the country. The first two such bases were the mainte- nance units provided under the project. By 1975, when the two units were operating satisfactorily, SERARHY had established a deserved reputation as one of the most effective Government agencies working in rural areas. It attracted much technical assistance, operated a sound training program, and quickly built up a good complement of middle-level staff. It did not succeed in providing a regular well maintenance program; succumbing to the pressures of visibility (para 5.07), it concentrated on major well repairs and, even, construction of new wells, although some of the latter were really reconstructions of existing wells that proved to be too difficult or dangerous to repair. Nonetheless, by 1978 SERARHY was performing satisfac- torily. SERARHY was, in 1979, one of the agencies hardest-hit by the civil war, in terms of material and staff losses, both local and expatriate. Resumption of activites will entail virtually starting from scratch. Like the Livestock Department (para 6.04), SERARHY's institutional development in the 1970s was largely independent of the project; yet the fact that the project provided for the establishment of two well maintenance bases, which constituted the nucleus of SERARHY's initial activities, gave the agency the means to prove its competence. However unexpected, and even though SERARMY did not solve the maintenance problem, the development of a competent pastoral water supply agency was an important indirect effect of the project. VII. GOVERNMENT AND BANK PERFORMANCE 7.01 The inordinately long delays between the first appraisal mission and signing, and between signing and effective start-up, were examined in detail in Chapter II. Although both parties were responsible, the persis- tent theme of Chapter II is that many of the delays would have been avoid- able if the Bank had wanted to go faster. Government as, at worst, intent on avoiding spending its own money (of which it had very little) and on substituting outright grant money for IDA funds. The Bank was, at best, cautious in the face of an uncertain security situation and unimaginative - 36 - when confronted with an unusual civil works program as the most visible part of an unexciting project. Perhaps the crux of the matter was that neither partner really understood the other, the Bank being ready to lend according to its own rules and procedures even though disappointed with the unambitious content of the project, with Government standing ready to receive funds provided that that entailed little effort, financial or otherwise, on its part. 7.02 In view of the history of delays and misunderstandings, the pace of project implementation, once it did get started, was quite remarkable. Here, too, both partners were responsible, the Bank for having unblocked the financial bottleneck and subsequently manifesting flexibility in super- vision, Government agencies for seizing the opportunity to show that they were capable of effective execution. This spirit of fruitful collaboration continued through preparation and appraisal of the follow-up project and was not disrupted until the civil war. ANNEX 1 - 37 - CHAD LIVESTOCK PROJECT - COMPLETION REPORT GENERAL INFORMATION Basic Documents 1. Report of the Livestock Project Identification Mission to Chad, 17 June 1966, FAO/CP Report No. 8/66 Chad 1. 2. Report of the Chad Livestock Development Project, 24 February 1967, FAO/CP Report No. 4/67 Chad 2. 3. Report of the Chad Livestock Development Project, 13 July 1967, FAO/CP Report No. 4/67 Chad ?. 4. Appraisal of Livestock Development Project, Chad; Bank Appraisal Report No. PA - 5a of January 24, 1972. 5. Pathologie Aviaire, May 1975; a consultant report on poultry disease prevention. 6. Fiscalitg de l'Elevage, August 1976; a consultant report on fiscal policy towards the livestock sector; 7. Moyens d'Exhaure, March 1977; a consultant report on methods of water extraction from depths exceeding 30m; 8. Dglimitation d'Unit6s Pastorales au Batha April 1977; a consultant report on pastoral unit identification in Batha Province; 9. Petits Ruminants, July 1977; a consultant report on animal health and productivity measures to be undertaken for sheep and goats; 10. Rapport sur la Wrification de Cr6dit 309-CD, April 1976; report on the audit undertaken by the Government Financial Controller. - 38 - ANNEX 2 Table 1 CHAD LIVESTOCK PROJECT - COMPLETION REPORT COSTS AND FINANCING Amendments to the Credit Agreement Original % of expenditures Equivalent Amended (04/04/74) Category allocation to be financed net-of-tax % of expenditures financing 1/ to be financed ------------------------------US$'00--------------------------------------- I - New wells 490 702 of total 92% 100% of total 11 - Well repair and renovation 255 70% of total 92% 100% of total III - Engineering Consultants fees 170 75% of total 99% 100% of total IV - Vehicles, equipment and spare parts for well maintenance 300 80% of total 80% 100% of total V- Well maintenance units (a) Technical Assistance 130 100% of total 100% 100% of tctal (b) Operating expenses from first month through twelfth month inclusive 110 100% of total 100% 100% of total (c) Operating expenses from 13th month to 24th month inclusive 55 50% of total 50% 100% of total VI - Technical Assistance for Studies 190 76% of total 76% 100% of total VII - Unallocated 500 Total 2,200 1/ Tax component in project costs: 24% of total costs of new wells, well repair and renovation, and engineering consultants' zees. - 39 - ble 2 LIVESTOCK PROJECT - COKPLETIONM REPORT COSTS AND FINANCING ApPralsaL Costs and Financino and Actual Costs AppraIsaL Total Appratsl Costa Aopra.tel Finrancing Actal Costs Project Costo, Ntof-Taxes DA 1 C0T 1NDP '.00 A F .nancn (CFAF Uillon (USS000) (CFAF m lIono (US5-000) --------U'009------- xS' 00)' Bas. Cost 17r.3 700 123.7 :22 490 210 - Phy.tial Contingencies 17.3 ' .4 £2 49 21 - Eopeoted PrIc. Imm nrese 23.0 20p 79.' 15 1A '2 - Total 249.1 97? l09.4 729 2, · IW - Well Repajr. BSe Cast 93.9 369 71.7 2" 253 114 Phy.ical Conringencies 9.4 '7 '.1 23 2' i - Expected Price Incrta*ss 27.9 110 21.2 ,3 - 14 Total 131.2 99.6 3A9 357 159 459 I[L - Supervison L-p Su- 1.1 224 4.7 179 1"0 CA1 TV - Vehitles and Equipsent Be Cost !. 07 NO chango from total 00 Lpeored Prlot Icreass 1.- .ro.jec.o.ts ?0 4 Total 94.- ~1 0 -l26 V - A Technical Assistance Sese Cost 2".0 LKt -- .Mraotd Pri. llor ...ot 5.1 102 Expected Prict Incese~.~" Total 35. 14- - - v - O perating Costs Year 1 Bae Cot 29.0 11 - - Etpectod Prico Incr.aae.. -." !4 Total ",7 14' 14' - - 92 V - C Oporating Cos: Yar Bas. cost 29.0 11- ExpectedPrice Increaset .2 .4 - 7 - Total 37. 14",2- 5 - 2 ortin Coett :Ooer 3 Saco Cot 29.0 11L copetedt Priot Inoreases 2.~ 4 04 - Total 7 141447 VI- Livetock Development Jnt Bas Cott 190.0 '42 190 4 53 Expected PrLc Inreats : 22 - Total 9. ~4 190 - 0 184 TOTAL nate lott O. 2.1' -42. 2,¯0 1.-00 -0 -00 Phyttocal Conttngencies 2'.2 107 20.1 -2 )2 ) - Prie129. ~ 109. -2,00 Total 0.0 0 .432 " . 1.211 2.222 22 -2 Total. Lncloding onty st-dt L from LiS U Base Cost :9'.0 2.325 A37 2.011 1720 302 Pysttl Contingencies 27.2 12 20.2 22 09- E-pCted Poss In reases 123.' 46 Lo0.7 C0 Total "4N.9 2.91 CI1.9 2.497 2.200 '02 - 2,200 Sett cots at per disbseent ateot.t; phyical contingencies (10) allocated dirto ly; .pected pric ;ncreaseallocated at 200 provison al An .n.s. 5, 6. and -, t.. 272 n% new ellt and noll repatts (hih was to cover 'additional co.t for precise well oiting' and "roat of experimnta for variou s fode-ctstr parapet.. nator drsoIng equ-pment, otc. as noll ss tto nonmal m fltton contngency. . 101 on ehielte and tq0opment. 30% on tehotoIal aslsstance and oporatlog cost - vn .hongh tho tsta ste, that "price contingencies have been estimtaed an the bass of a rat. of oscglatLOn Of Cost of 10a per annua". This 4llocation rIelds a total physical contgnor of USS 5.000 ond a totl pri. contingatcy of USS 314.000. for a total of USS 89.000. -ha unallocated cacegoy was set at jSS 520.000. Sv dofftrence. 1otal ohostal onting.vn was USS x2.000 and total price ontingene- US 151.000..for a total of USS 23002. The appraisal report implies a total conineny of US 124.300. As r ohe ftoal .s.tmeno of ditbursemeots o 01 03 79. UNDP disotrsmnt ateot kown bur otto ores-bly f the order of USS 50.000 whon one LDU technical Assli:ras pro c . teminat.d in Septesber 1972. Goveret made 'no direct eoenditures on the piroot ot there os some teconoment of stoff ta the well maintenance bae. and atuch partcpation of staff. without foral secondment. in sopervision of well construction and repair. .ntial establiahent and operation of aellatnanot natt, and ln coesstioning end sparvisag studiea. 4 15% of too stIl and tell topal costs includineg contngncies. 2 Of whith, S$ 3.000 di.eurtod on ~accins. Shon enolu ivn o rhe UNDP-Etanced LDU a.livittes sinne. in practic., there as& Contguous - or .ca t : . - 40 - ANNEX 2 Table 3 CHAD LIVESTOCK PROJECT - COMPLETION REPORT COST AND FINANCING Costs per Well New Wells Repairs 1. Appraisal -------------(CFAF '000)------------ 1.1 Unit costs (May 1971) as per awarded contract 4,693 921 1.2 Less: tax element (24%) (1,126) (221) 1.3 Net-of-tax unit costs 3,567 700 1.4 Inflated at 12% per year to October 1974 (1.49) 5,315 1,043 2. Negotiated Contract 2.1 Unit costs (October 1974) as per contract 7,354 1,444 2.2 Supply base overhead 1/allocated pro rata to value 3,111 622 2.3 Effective unit cost 10,465 2,066 ----------------------------- ------------ )--- -------------- 2.3 as % increase over 1.1 + 120% of which: inflation 49% comparable tax basis 24% so: 2.3 as % increase over 1.4 + 95% of which: comparable volumeof work 35% so: increase in costs consequent on negotiated contract + 60% -------------------------------------------- (CFAF '000)-------------- 3. Actual Costs 3.1 Per contract 10,465 2,066 3.2 Price escalation clauses 733 145 Subtotal 11,198 2,211 3.3 Supervision costs 2/ allocated pro rats to value 1,713 2!344 Total 12,911 2,555 +CFAF 67.2 million /bCFAF 37 million 3/ 15.3% of actual construction costs; all estimates had assessed supervision at 15% of construction costs. ANNEX 3 41- Table 1 CHAD LIVESTOCK PROJECT - COMPLETION REPORT OUTPUT AND FINANCIAL AND ECONOMIC RESULTS Livestock Output as per Appraisal Cattle Small stock No. Offtake No. Offtake -----------------('000 head)-------------- New Wells (38) Pre-project - PY10 without project - - - with project 79.8 9.2 68.4 17.1 increment 79.8 9.2 68.4 17.1 Renovated Wells (102) Pre-project 1/ 160.8 18.5 137.8 34.4 PY10 without project 93.7 10.8 80.4 20.1 with project 213.9 24.6 183.4 45.8 increment 120.2 13.8 91.7 25.7 Maintened Wells (258) Pre-project -/ 487.5 56.1 417.9 104.5 PY10 without projeq 284.3 32.7 243.6 60.9 with project - 542.1 62.3 464.6 116.2 increment 257.8 29.6 232.3 55.3 Total (398 wells) Pre-project - 648.3 74.6 555.7 138.9 PY10 without project- 378.0 43.5 324.0 81.0 with project 5/ 835.8 96.1 716.4 179.1 increment 457.8 52.6 392.4 98.1 Relative Shares Maintained Renovated New Wells Total % Wells (258) Wells (102) (38) (398) -------(incremental cattle offtake in head)----.-- 6/ Prevention of deterioration - 23,400 7,700 - 31,100 59% Expansion of resources 7/ 6,200 6,100 9,200 21,500 41% Total 29,600 13,800 9,200 52,600 100% Relative contribution to incremental cattle offtake 56.3% 26.2% 17.5% 100% 1/ Assumed to be 75% effective; 76.5 new-well equivalents. 2/ Assumed to be 90% effective; 232.2 new-well equivalents. 3/ All 258 wells assumed to be improved to 100%. 4/ Allocated between renovated and maintained wells in same proportion as pre-project new-well equivalents (76.5:232.2). 5/ Allocation between the three types of wells: (a) new wells calculated explicitly, (b) breakdown between renovated and maintained wells in same proportion as post-project new-well equivalents (102:258). 6/ Difference between Year 0 output and Year 10 without-project output 7/ Difference between Year 10 with-project output and Year 0 output. 42 - ANNEX 3 Table 2 CHAD LIVESTOCK PROJECT COMTLETION REPORT OUTPUT AND FINANCIAL AND ECONOMIC RESULTS Livestock Output as per October 1974 Reassessment Cattle Number Offtake ------ ('000 head) ------- New Wells (18) Pre-project PY10 without project with project 37.8 4.4 increment 37.8 4.4 Renovated Wells (18) Pre-project 22.8 2.6 PY10 without project 14.7 1.7 with project 38.0 4.4 increment 23.3 2.7 Maintained Wells (342) Pre-project 561.8 64.6 PY1O without project 363.3 41.8 with project 3/ 594.9 68.3 increment 231.6 26.5 Total (378 wells) Pre-project 584.6 67.2 PY1O without project - 378.0 43.5 with project 5' 670.7 77.1 increment 292.7 33.6 Relative Shares Maintained Renovated New Wells Total Wells (342) Wells (18) (18) (378) ------ (incremental cattle offtake in head) --------- Prevention of deterioration - 22800 900 - 23700 71' -7 Exoansion of resources - 3700 1800 -400 9900 29% Total 26500 2700 4400 33600 100% Relative contribution to incremental cattle offtake 78.9% 8.0% 13.1% 100% 1/ Assumed to be 60% effective (down from 75% at appraisal); 10.8 new-well equivalents. 2/ 158 wells assumed to be 90% effective, 100 at 75%, and 84 at 60% (appraisal assumed that all were 90% effective); 267.6 new-well equivalents. 3/ Pre-project wells operating at 75% and 60% assumed to be not improvable, the other 158 assumed to be improvable to 100% (appraisal assumed that all pre-project wells were improvable); 283.. new-well equivalents in Year 10. 4/ Allocated between renovated and maintained wells in same proportion as pre-project new-well equivalents (10.8 : 267.6). 5/ Allocation between the three types of wells: (a) new wells calculated explicitly, (b) break-down between renovated and maintained wells in same proportion as post- pro4ect new-well equivalents (18 : 283.4). 6/ Difference between Year 0 output and Year 10 without-project Jutout. 7/ Difference between Year 10 with-Project output and Year 0 output. - 43 - L17WTOCK PROJECT - 00'TION REPZCT rPUT AND FINASCIAL k" SCONOmIC RESULT3 sreakdn of Appraisal .nd Re.ssssen (otober 1974) Ecooi Raturn. Yar1 2 3 4 5 6 7 8 9 10-29 30 E(R - --.- ------------- CFAF illion ) ---.---------------------------------------- ---- oreakdown of Amaratsal Return Costs l/ ( 70.0) ( 20.0) (20.0) (20.0) (70.0) (13.0) (13.0) (13.0) (13.0) (19.4) (19.4) incre~*nt. Benefits (300.0) (260.4) (220.1) (181.2) (141.1) (37.9) ( 3.8) 26.6 241.9 334.8 2.2L9.7 Nat B..ftc (370.0) (280.4) (240.1) (201.2) (211.1) (50.9) (16.8) 13.6 228.9 315.4 2,230.7 12.6 Renovatea «e'l! costa ( 79.1) (71.8> - - - (5.1> (5.1> (5.1) (5.1> ((7-7) (7.7> Incremena1 Ben*fits (139.6> (121.2) (102.4) (84.3) (65.7) (17.6) (1.8) 12.4 112.6 155.8 .1,46.< Ntt Bnenfits (218.7) (193.0) (102.4) (84.3) (65.7) (22.7) (6.9) 7.3 107.5 148.1 1,39.2 11.2 New.,ollo Costa (229.2) (57.3) - - - (1.9) (1.9) (1.9) ( 1.9) (2.9) (2.9) Incre..tal Bne.fits (93.3) (80.9) (68.4) (56.3) (43.9) (11.8) (1.2) 8.2 75.2 104.1 699.3 Nat Oecefits (312.5) (138.2) (68.4) (56.3 (43.9) (13.7> (3.2) 6.3 73.3 8.1.2 696.4 8,7 - Coat. 378.3< (149.1) (20.0) (20.0 . (20.0 (20.0) (20.0 (0.0< (3D-0) (30.0) ,ncremental 'ene1it4 (532.9) (462.5) (390.9) (321.8) (20.6) (67.3) (6.8) 47.2 429.7 594.7 395.9 'let Eer.fts (911.2) (611.6) (410.9) (341.8) (ý2o.6) (87.3) (26.8) 27.2 409.7 564.7 3,965.9 11.0 532 Oreaan of Reassessea Rsturn 6/ Cosa 4/ (70.0) (20.0) (20.0) (20.0) (70.0) (18.0) (18.0) (18.0) (18.0) (27.0) (27.0) Incraent. Oa.netita (662.8) (578.3) (377.3) (308.3> (143.8) 223.4 256.9 285.2 499.6 499.6 3,379.5 Net Bnfits (732.8) (598.3) (397.3) (328.3) (213.8) 205.4 238.9 267.2 481.6 472.6 3,352.5 13.2 698 Renovate.amells co.ta (32.0) (13.7) - - - (1.0) (1.0) (1.0) (1.0) (1.5) (1.5) "ncremsntal. Enefita (67.2) (58.6) (38.3) (31.5) (14.6) 22.6 26.0 28.9 50.7 50.7 342.7 at B.oefits (99.2) (72.3) (38.3) (31.3) (14.6) 21.6 25.0 27.9 49.7 49.2 341.2 :1.8 52 Har .ellc coata (158.3) (67.8) - - - (1.0) (1.0) (1.) (1.3) (1.5' C..5 Inreental Beftfl (110.0) (96.0) (62.6) (51.2) (23.9) 37.1 42.7 47.4 82.9 S2,9 56:..l Net Ben.flt. (263.3) (163.8) (62.6) (51.2) (23.9) 36.1 41.7 46.4 81.9 81.4 559.6 9.1 -61 Cost. (260.3) (101.5) (20.0) (20.0) (70.0) (20.0) (20.0) (20.0) (20.0) (30.0) (31.0> Incrosantal Bønefits (840.0) (732.9) (478.2) (390.8) (182.1) 283.1 325.6 361.5 633.2 633.2 4,283.' Oet Bata (1,104, (834.4) (498.2) ý410.8) (252.3) 263.1 305.6 341.5 613.2 603.2 4,253. 12.o 686 va¯~ysar; in thes original calculation, al.l y~a re identical with the sal. except.. ,ion f rpla t of capita te' s x- 30 YW lio for ths call ma.nteace units every tifth yeør, in yeart 10, 15 ect. includ*s residual Value of lucremsntal herd and flock Inrem-ntal benefitn per camponant assued to grow pro rata with each aomponcnt's shar, tf yar 10 incremental benefits; maintained oalsl 56.3%; reno-sted -l. 26.2% n.w .118 17.5%(Anne, Table 1). I Firat five y~eatcst wholly attributed to maintaeast cll.; subsaquen-: cost allocatd prorata between the th.e types ef call.. / A er Appatsial Raprt, Annex 2, page 1, ith th. sale exteptio notd at 1/. A. the Oto-ar 1974 ramssøsnt of the cono ic ju.t.fication .otd:" At appraisal the ate f rtun af the proe3et (11%) had besn calcultad with pric ocentingnciesa and as included in th. Cost of civil sckA sad tair superviaion (total cast: CA 437.4 alllion). This ot isreuoed to CPAF 293.3 million wsa Pce cootingenciøs and an 18% tax cospent a deducted. Bowee, thm mate f return, when calculatd oan the basis af this lcoeracest, incrasaesnly to 11.5%". 5/ Benefits allocatad . at 2/ but ths share ol er 10 icm-ntal benefita cbass to: afiatained wells 78.9%; enoVastd wsll& 8.5%; ne~ ~elle 13.1% (Anex 3, 2sble 2). Costa exclude taxe. and contiagncis. Joint suprvsion co-ta split in sam praportion (5:1) se unit csatas af nw wellø and ranovsations. / As p*r sa l.blanc/vanGigeh ef 10/17/74 and -orking papers, with tha sole exception noted at 2. -44- ANNEX 3 Table 4. CHAD LIVESTOCK PROJECT - COMPLETION REPORT OUTPUT AND FINANCIAL AND ECONOMIC RESULTS Economic Rates of Return and Net Present Values Maintained Renovated New Total Wells Wells Wells Economic Rates of Return (30 years) -------------------------(ERR %)------------------------------ Appraisal 1/ 12.6 11.2 8.7 11.0 rFeassessed (October 1974) 2/ 13.2 11.8 9.1 12.4 Reassessed, benefits lagged 5 years 3/ 11.6 10.2 7.5 10.6 Reassessed, 5-year lag, 1979 prices 4! 12.0 10.9 8.2 11.0 Net Present Values (10%, 30 years) -------------------------(CFAF million)------------------------- Appraisal 1/ Investment in Years 1-3 5/ 110 151 287 548 NPV 333 98 -99 332 Reassessed (October 1974) 2/ Investment in Years 1-3 5/ 110 46 226 382 NPV 698 52 -61 686 Reassessed, benefits lagged 5 years 3/ Investment in Years 1-3 6/ 110 46 226 382 NPV 256 4 -136 124 Reassessed, 5-year lag, 1979 prices 4/ Investment in Years 1-3 7/ 152 64 313 530 NPV 484 26 -155 344 1/ As at Annex 3, Table 3, for 258 maintained wells, 102 renovated wells and 38 new wells. 2/ As at Annex 3, Table 3, for 342 maintained wells, 18 renovated wells and 18 new wells. 3/ As at 2/, with incremental benefits lagged 5 years to reflect the impact of the drought on the relation between herd numbers and grazing resources, i.e. maximum utilization of without-project grazing resources is reached in 1980 instead of at the outset (1975). 4/ As at 3/, with benefits revalued pro rata with the change in liveweight cattle prices, from CFAF 120/kg in 1974 to CFAF 200/kg in 1979; costs inflated by the International Index of Inflation, EPDIT of February 1979: 1975: 71.7; 1976: 73.0; 1979: 100. 5/ Sum of costs in years 1-3 at Annex 3, Table 3. 6/ No change from Reassessed as at 5/. 7/ Sum of inflated costs in years 1-3. - i ___________________ ___ -- I;4 L- 41- * 0< * 4 UJ *48< ( k O 0.. * I I.@ 41 i 8 0 -'& O + i I+.+oj
Группа Всемирного банка · Project Performance Assessment Report
Chad - Livestock Development Project
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Project Performance Assessment Report
Дата
Страна
Чад
Источник
worldbank_document