Upper Volta - Bougouriba Agricultural Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 4547 PROJECT PERFORMANCE AUDIT REPORT UPPER VOLTA: BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 496-UV) June 15, 1983 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. WEIGHTS AND MEASURES The Metric System ABBREVIATIONS CCCE Central Economic Cooperation Agency (Caisse Centrale de Coopération Economique) EEC European Economic Community CIDR International Rural Development Company (Compagnie Internationale de Développement Rural) ORD Regional Development Organization (Organisme Régional de Développement) SOFITEX Voltaic Textile Fibre Company (Société Voltaique des Fibres Textiles) FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT UPPER VOLTA: BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT (Credit 496-UV) TABLE OF CONTENTS Page No. Preface ............................................................ i Basic Data Sheet ...................................................1... . ii Highlights ............................................................ iii PROJECT PERFORMANCE AUDIT MEMORANDUM I. SUMMARY .................................................. 1 Project Design ............................................. 1 Project Implementation .................................... 2 Project Impact ............................................. 2 II. MAIN ISSUES ............................................... 3 Slow Pace and Cost of Development ...................... 3 Cotton Policy and Project Justification ................ 5 Annex 1 Comments from SOFITEX .................................... 9 PROJECT COMPLETION REPORT I. Background .................................................. 13 II. Costs and Financial Arrangements ........................... 17 III. Implementation .......................................... 19 IV. Agricultural Impact........................................... 31 V. Institutional Performance ................................... 39 VI. Economic Evaluation ......................................... 42 VII. Bank Performance ............................................. 43 VIII. Conclusion .................................................. 44 Annexes .................................................. ........ 45 Map: IBRD No. 17821 (PCR) This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT PERFORMANCE AUDIT REPORT UPPER VOLTA: BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 496-UV) PREFACE This is a performance audit of the Bougouriba Agricultural Develop- ment project in Upper Volta, for which Credit 496-UV was approved in May 1974 in the sum of US$8.0 million and closed fully disbursed on December 31, 1981. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department (OED) and a project completion report (PCR) dated July 1982. The PCR was prepared by the Western Africa Regional Office, following a mission in May 1982 and on the basis of a completion report prepared by the Regional Development Organization (ORD). The memorandum is based on the Staff Appraisal Report (No. 297a-UV) dated May 10, 1974, the President's Report (No. P-1359a-UV) of May 15, 1974, the Credit Agreement of July-19, 1974, the Staff Appraisal Report of the Second Bougouriba Agricul- tural Deveopment Project (No. 3126-UV), and the PCR; correspondence with the Borrower and internal Bank memoranda on project issues as contained in relevant Bank files have been consulted. An OED mission visited Upper Volta in February 1983. Discussions were held with officials of the Ministry of Rural Development, the Diebougou ORD and SOFITEX, the Voltaic Textile Fibre Company. The information obtained during the mission was used to test the validity of the conclusions of the PCR. A copy of the draft report was sent to the Borrower and to SOFITEX on April 1, 1983. Comments received from SOFITEX are in Annex 1 of the Pro- ject Performance Audit Memorandum (PPAM). Suggested changes have been introduced. The audit finds that the PCR covers adequately the project's salient features, and the PPAM generally agrees with its conclusions. In addition to summarizing the objectives and results of the project, the PPAM expands upon certain aspects, particularly the slow pace of development and the cotton policy, because of their importance to this and other agricultural and rural development projects in the same region. The valuable assistance provided by the Government of Upper Volta and the project staff met during the preparation of this report is gratefully acknowledged.  - ii - PROJECT PERFORMANCE AUDIT BASIC DATA SHEET UPPER VOLTA: BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT (Credit 496-UV) KEY PROJECT DATA Appraisal Actual or Actual as % of Item Estimate Estimated Actual Appraisal Estimate Total Project Costs (US$ million) 10.2 9.7L 95 Credit Amount (US$ million) 8.0 8.0 100 Date Board Approval 05/30/74 Date Effectiveness 10/17/74 07/18/75 28b Date Physical Components Completed 9/78 10/80 12L Proportion then completed (%) 100 Closing Date 03/31/79 12/31/81/c 158Lb Economic Rate of Return (%) 68 17 Institutional Performance good Agronomic Performance average Number of Direct Beneficiaries 16,000 10,00.14 CUMULATIVE DISBURSEMENTS FY75 FY76 FY77 FY78 FY79 FY80 Appraisal estimate (US$ million) 0.7 2.5 4.6 6.8 8.0 - Actual (US$ million) 0.6 2.4 4.2 6.1 7.8 0.2 Actual as % of estimate 86 96 91 90 98 100 Date of final disbursement : 12/04/80 Principal repaid to date (US$ million): None MISSION DATA Date No. of Mandays Specializations Performance Types of Mission (mo.Iyr.) Persons in Field Represented/f Rating-/ Trend/h Problems/i Identification Preparation Appraisal up to 1974 N/A 476 NGB up to 1974 N/A 89 Subtotal 565 Supervision 1 11/74 1 1o a 3 1 MP Supervision 2 4/75 1 13 a 2 1 T Supervision 3 9-10/75 1 13 a 2 1 MT Supervision 4 2/76 3 19 a,a,b 1 1 MT Supervision 5 10/76 2 5 a,a 1 1 MT Supervision 6 4/77 1 21 a 2 2 MT Supervision 7 9-10/77 2 14 a,c 2 2 MT Supervision 8 2/78 1 4 a 2 1 M Supervision 9 9/78 2 7 a,c 2 2 MT Supervision 10 2-3/79 2 7 a,c 2 1 MT Supervision 11 10/79 2 5 a,c 2 1 T Supervision 12 5/80 2 5 a,c 1 2 - Supervision 13 9/80 1 6 c 1 2 0 Supervision 14 2/81 3 6 a,a,c 1 2 0 Supervision 15 10/81 2 6 a,c 1 2 0 Total 706 OTHER PROJECT DATA Borrower Republic of Upper Volta Executing Agency Bougouriba ORD (Regional Development Organization) Fiscal Year April 1-March 31 Name of Currency (abbreviation) CFAF Currency Exchange Rate: Appraisal Year Average US$ 1.00 - CFAF 250.00 Intervening Years Average US$ 1.00 = CFAF 231.00 Completion Year Average US$ 1.00 = CFAF 211.3 Follow-on Project: Name Second Bougouriba Agricultural Development Project Credit Number Credit 1097-UV Credit Amount SDR 12.4 million (US$16.0 million equivalent) Date Board Approval 01/28/81 /a Or US$11.7 million if additional EEC loan of US$1.5 million to bridge the gap between Bougouriba I and II is taken into account. Appraisal estimates in CFAF 2,550 million equivalent to US$10.2 million at the exchange rate of US$1.0 - CFAF 250; total project expenditure was CFAF 2,204 million, at an average disbursement date of CFAF 227.50. /b, Calculated from Board approval date. 7- The closing date was extended to allow disbursing EEC additional loan. Td 10,000 is the estimated number of farm families reached by the extension service; the whole farming population (33,000) of the project area benefited from water supply and feeder roads constructed and rehabilitated under the project. /e Mandays in the field for Spn. No. 1-4, 6 and 7 are for the supervision of several projects jointly (3 or more). There is no breakdown for time spent on this particular project. /f a - agriculturist; b - agricultural economist; c - financial analyst. T I = problem-free or minor problems; 2 - moderate problems; and 3 - major problems. /h I - improving; 2 - stationary; and 3 = deteriorating. 71 M = managerial; T - technical; P = political; and 0 - other. a 4 # - 111 - PROJECT PERFORMANCE AUDIT REPORT UPPER VOLTA: BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 496-UV) HIGHLIGHTS The project aimed at developing the relatively isolated and undevel- oped southwestern region of the country. Its objective was to achieve self- sufficiency in food for the local population and to increase farmer incomes through expansion of cotton, the cash crop the best suited to the region. The project's main components were the provision of buildings, equipment, vehicles and staff to the project authority, establishment of supporting services to farmers and the construction of roads and village wells. Due to the difficult conditions of this backward region, the first three years of the project were spent mainly on constructing housing, build- ings and road infrastructure. Agricultural development started later than anticipated and the project had to be extended for two years, with additional financing from the EEC. At project completion, the construction program had been completed as planned and the extension and farmer support services were operating adequately. The project costs were slightly (15%) higher than appraisal estimates. A follow-on project, approved by the Bank in 1981, is expected to reinforce and extend the results obtained by the first project. The agricultural achievements of the project were significant but substan- tially lower than appraisal expectations. The number of farmers reached by the project was about 50% of appraisal estimates for cotton and 30% for foodcrops. The project ERR is now estimated at about 17%, well below the appraisal estimate of 68%. On the other hand, the institutional impact of the project was positive. The project authority is now a dynamic and efficient organization; its experience is expected to be fully utilized during the second project. In retrospect, the project consisted more of laying the foundations for future development than comprising in itself a lasting developmen7 of the region. The project illustrated (i) the difficulty and relatively high cost of a rural development project when it is first introduced in a backward area and (ii) the need, unforeseen at appraisal, to consolidate the results obtained by one or more follow-on projects. The following points may be of particular interest: - the implication of a succession of projects for regional developmeat is a cost per beneficiary that is generally higher than envisagez (PPAM, paras. 12-14); - the cotton price policy offered little incentive to coton groWSEe during project implementation (PPAM, paras. 15-17); - iv - - the project-s high capital and recurrent costs can only be justified by a clear emphasis on the crop offering the best return both to farmers and Government but not by self-sufficiency in food (PPAM, paras. 18-21); and - the project was instrumental in developing successful farmers- groups (PPAM, para. 11 and PCR, para. 5.06). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM UPPER VOLTA: BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 496-UV) SUMMARY Project Design 1. The project was designed to develop the Bougouriba region, located in the southwest of the country and covering about 17,000 km2, or 6% of Upper Volta's total area. In 1973 this relatively thinly populated region (about 360,000 inhabitants, or 21/km2) with an annual rainfall of 1,000 to 1,300 mm, i.e., more than the average for the country, was considered one of the more favorable areas for crop production because of its relatively fertile soils. 2. The project, which was to be implemented over four years, aimed at (i) increasing cotton production fivefold through an increase in areas and yields, and (ii) raising foodcrop production by about 5% through better farming practices. The ultimate objective was to achieve self-sufficiency in food for the local population and to increase farmer,incomes through expansion of cotton, the cash crop considered the best suited to the region. The number of cotton growers was to be increased from 10,000 to 16,000 over four years. 3. The main project components for achieving these objectives were the following: (a) Provision of the staff, buildings, vehicles and equipment to the Bougouriba Regional Development Organization (ORD), which was responsible for project execution; (b) Establishment of extension, input supply and credit services for the farmers; (c) Construction of 660 km of secondary and tertiary roads, 120 wells and an applied research and seed multiplication center. The total cost of the project was estimated at US$10.2 million, of which US$8.0 million (79%) was to be financed by the Bank and the rest by the Government (12%), the National Development Bank (6%) and the farmers them- selves (3%). The project rate of return was estimated at 68%. 4. In order to enable the Government to provide its contribution to regional development, a Development Fund was to be set up that would be financed through a fee levied on cotton sales. SOFITEX, the agency respon- sible for the processing and sale of cotton, was to act as collecting agent for the Development Fund. -2- Project Implementation 5. The effectiveness date of the project was postponed three times, mainly owing to problems in recruiting the extension staff. It also proved difficult to retain this staff in the harsh conditions of this backward region. The first three years of the project were spent mainly on construct- ing housing, buildings and infrastructure and reorganizing the extension service. Agricultural development did not really start until the third year. As a result of this delay, the closing date for the project was postponed by one year. Eventually, it became necessary to extend the project for a sixth year (1981), the financing for which (US$1.8 million) was provided largely (US$1.5 million) by the EEC. 6. At the final project closing date (December 31, 1981) the building, road and well construction had been completed as planned. The ORD staff were in post, and the extension and farmer support services were operating as expected. A program of applied research was under way. Input supplies and agricultural credit for the farmers, however, remained well below estimates because of a low fertilizer adoption rate and the financing of animal-traction programs by other institutions. The project costs remained close to appraisal estimates, except for inputs (20% of initial estimates). The actual total cost of ,the project was US$11.7 million, i.e., 15% more than the appraisal estimate if the additional year financed by the EEC is included, or US$9.7 million (95% of appraisal) for the five years financed by the Bank. The IDA credit (US$8 million) represented 68% of the total investments in the first case and 82% in the latter. Project Impact 7. In the absence of monitoring and evaluation, which were not intro- duced until the start of the follow-on project, the impact of the project on agricultural production and farmer incomes is difficult to measure. Neverthe- less, it became apparent during project implementation that the estimates made at appraisal were grossly exaggerated as regards both the situation before the project and the projections with the project. It was envisaged that total cotton production would increase fivefold as a result of expansion of the area under cotton from 5,000 ha to 12,000 ha and an increase in yields from 400 kg to 813 kg per hectare on average. Cereals were estimated on the basis of 25,000 ha before the project with a production of 700 kg per ha, which was expected to rise by 5% with the project. However, these figures were over- optimistic in terms of,areas and yields, for both cotton and foodcrops. 8. The appraisal assumptions were therefore corrected during project implementation and more realistic estimates made for areas and yields by the end of the project. From these figures, it is apparent that as regards cotton: (i) the project had no more than a slight impact on production prior to the third year;. (ii) areas expanded by about 40% (from 4,850 ha to about 6,800 ha) between 1975/76 and 1980/81; (iii) yields increased by about 56% (from 378 kg to 592 kg/ha) over the same period, and (iv) total cotton production increased by about 2.2 times. The number of cotton growers reached - 3 - by the project was 7,822 instead of the 16,000 estimated at appraisal. The higher production is due primarily to a certain improvement in yields and also because the farmers have slightly expanded the areas under cotton. Although these figures are lower than appraisal estimates, they show that the project had a significant impact on cotton production which, however, has stabilized since 1979/8C. Yields are also stagnant, partly because of the underutiliza- tion of fertilizer by the growers. 9. As regards cereals and groundnuts, production of which was also overestimated, the increase has been 5%, as envisaged at appraisal. Areas and yields with and without the project remain, however, well below projections. The number of farmers benefiting from the foodcrop component of the project has not exceeded 10,000 instead of the 33,000 initially estimated. Livestock owners have also benefited from the vaccination programs undertaken during the project. 10. On the basis of these figures, the economic rate of return is now estimated at 16.7%, i.e., well below the appraisal estimate of 68%, which reflects the relatively modest impact of the project on agricultural produc- tion in the Bougouriba region. A follow-on project, the Second Bougouriba Agricultural Development Project, was approved in January 1981 for an amount of US$16 million equivalent, to help reinforce and extend the results obtained by the first project. 11. The institutional impact of the project has been significant. The Bougouriba ORD is now an effective and dynamic organization able to provide the support services needed by the farmers. The experience gained due to the difficulties encountered during the first project is expected to be fully utilized during the second and facilitate its execution. The formation of farmer groups responsible for collecting the cotton for SOFITEX can be con- sidered a success of the project. These self-managed groups are already handling the collection of about 20% of the cotton produced in the region and have begun to market their own cereals production. The Development Fund called for at appraisal has been set up; it receives as income a levy of CFAF 4,500 (about US$13) per ton of seedcotton marketed through SOFITEX. MAIN ISSUES Slow Pace and Cost of Development 12. The Bougouriba project illustrates the difficulty, slow progress and relatively high cost of rural development when it is first introduced in a backward area. Prior to the project, little development effort had been undertaken in Bougouriba with the exception of some cotton programs carried out by SOFITEX. Unlike the regions with basic infrastructure on which a project can easily be grafted, in 1973 the Bougouriba region was still rela- tively isolated and undeveloped. By its nature and notwithstanding the fact that cotton was its target crop, the Bank-assisted project was more an inte- grated regional development project than a cotton project; this meant that the infrastructure had to be constructed, ORD staff recruited and the rural roads opened before any significant extension work could be started. In retrospect, a shortcoming of the appraisal was the failure to recognize hiis constraint by assuming a rapid agricultural development. In the circumsLances, three years were needed to build the infrastructure required. In the cn the project cor- sisted more of laying the foundations for future developman_ than comprising in itself a lasting development of the Bougouriba region- Extension of the project from four to six years, one of which was an addit_Icnal year financed by the EEC, followed by the approval of a second five-yea: project, showed clearly that: (i) the time needed for achieving a significant impact on agricultural production is particularly long for this type of project, and (ii) the results obtained very often have to be consolidated by a second project, whose achievements frequently need to be sustained by a third. OED has found1/ that with rare exceptions rural development pojects have been multiple-phase projects but that each of these phases could be economically viable, as is demonstrated by the present project which sows a small but positive rate of return. This phenomenon is not typical cnly of Africa. In other regions, the Bank also found that insufficient knowledge about the area targeted and the inexperience of the project personnel almost invariably led to a first rural development phase where tangible achievements proved minimal. 13. The implication of a succession of projects for regional development is a cost per beneficiary that is generally higher than envisaged. On the basis of the 11,000 families presently supported by the project, Bougouriba I actually cost US$1,063 per beneficiary family instead of the US$293 originally estimated. The second project aims at raising the number of families reached by the extension and support services from 11,000 to 22,000 at an additional cost of US$13.6 million, which will increase the final cost per beneficiary to around US$1,150. A possible third project would increase this cost further. Although this figure is within the limits of the averages observed for agri- cultural development projects, it is, however, higher than the average cost per beneficiary for West African countries.!/ It should also be noted that a significant part of the investment costs (33% in Bougouriba I and 57% in Bougouriba II) is represented by costs inherent in the administration of 1/ A special OED study (Report No. 2242) on the Bank's experience in Sub- Saharan Africa concludes that follow-on projects have been necessary in many cases to sustain projects that were expected to be self-sufficient by the end of the disbursement period. 2/ A study made in 1980 by the Western Africa Region of the cost of 72 agricultural development projects financed by the Bank between 1977 and 1979 arrived at an average of US$1,575 (1980) -er beneficiary family. However, this average did not exceed US$597 for the nine projects actually located in the Western Africa Region, -5- the project (buildings, housing, vehicles, staff salaries, etc.) while only a relatively small part of the cost goes toward productive investments. Here too, the Bougouriba project is no exception, but like other projects of the same type, carries the risk of benefiting primarily the administrative machin- ery if its production goals are not achieved (see para. 19). 14. In the present situation, the project-s high recurrent costs also pose a problem that can only be overcome by considerably increasing production in the years ahead or drastically reducing these costs. The Completion Report estimates the ORD-s annual operating costs for the first project at US$323,500 and those of the fertilizer subsidies at US$220,000. Against this total of US$543,500 per year, the budgetary revenues for the State can presently be estimated at around US$62,000 from cotton (US$13/ton) for the Development Fund, to which can be added a small amount in tax revenue from the few cereal products sold. Altogether, the revenue from the project for the State can hardly exceed 20% of the recurrent costs in the present situation. The net deficit to the State per beneficiary family (11,000 at present) therefore amounts to around US$49 per family per year. Although this figure is not excessive compared with other projects of this type,1l the Bougouriba region is one of the few from which Upper Volta can hope to derive any benefit whereas the other regions of the country require permanent subsidizing.l/ Cotton Policy and Project Justification 15. One of the features of the project is that it was based initially on rapid growth in cotton production and a small increase in foodcrop production. However, whereas foodcrop production is tending to rise steadily (and even faster than expected over the past two years), cotton production now tends to be static. After improving markedly in the 1976-79 period, areas under cotton and cotton yields have since remained stagnant and are far from reaching their optimum. In the Hauts-Bassins region, where climatic conditions are similar to those of the Bougouriba region, but where cotton is a more traditional crop, yields increased progressively from an average of 864 kg/ha in 1976/77 to 1,276 kg/ha in 1981/82. 16. One of the main reasons for this stagnation of cotton production was that farmers are concerned first and foremost with assuring their food supply, so that they sow their foodcrops before they move on cashcrops. Since in Upper Volta early sowing is a determining factor for the success of the crop, the result will be a mediocre yield of cotton. Another reason was a national 1/ The study referred to in para. 13 shows that of the nine projects in Africa, six had cash flows representing deficits for the State, with a per-family average of US$125. Three projects, on the other hand, brought in a surplus for the State of around US$120 per family. The average for the nine projects came to a negative cash flow of US$11 per family. 2/ See Upper Volta: Rural Development Fund, OED Report in preparation. -6- price policy that offered little incentive for cotton during project imple- mentation. Whereas the economic farm-gate price of seedcotton was estimated at CFAF 93/kg in 1981 by the appraisal report on the Second Project, the price paid to the grower, which was CFAF 40 in 1976, was raised to CFAF 55 in 1977 and held at that level for four years until 1980, then raised again to CFAF 62 in 1981 and 1982. If to this is added the CFAF 13 per kilogram of cotton represented by the fertilizer subsidy plus CFAF 4.50 for the Development Fund, i.e., CFAF 17.50 in all, the real producer price remains about 15% below what it should be. In contrast, the prices for cereals (sorghum, millet, maize and rice) fixed by the Office National des Cereales (OFNACER) were raised regularly by 12-15% each year from 1979 to 1982. Although OFNACER purchases only a small percentage of the cereals, which are mostly bought by the private sector (and often below the government support price) a continuous increase in producer price for cereals was noted during project implementation while that of cotton remained stagnant for four years. 17. Undoubtedly this policy of holding prices unchanged for a long period and the low prices paid to growers throughout the duration of the proje-t have had an adverse effect on the expansion of the cotton area and a negative impact on cotton production.!' This underproduction of cotton compared with the optimum obtainable leads in turn to undersupply to the ginnery which is currently working at only two-thirds of its capacity and is, therefore, operating at a deficit. The snowball effect of low yields and low price did not induce farmers to use more fertilizers and to significantly increase cotton production.21 It is hoped that price increase in 1981/82 and a continuous extension effort under the Third Bougouriba Project will help increase both area and yields of cotton. 18. With the exception of yams, grown in the southern part of the region, cotton is acknowledged to be the most profitable crop in the Bougou- riba region provided it is properly cultivated. The problems of high capital and recurrent costs mentioned above (paras. 13-14) show that a project of this nature can only be justified by a clear emphasis on the crop offering the best return. A goal limited to self-sufficiency in food would not require the present substantial infrastructure of Bougouriba; an approach similar to that of the Rural Development Fund in the Mossi plateau, with a low cost per beneficiary and per hectare, would be more appropriate for promoting only foodcrops and cereals production. 1/ The negative impact of low prices paid to growers for their cash crops has been observed by OED in a large number of agricultural projects in Sub-Saharan Africa, and in particular for West Africa, in Senegal (Report Nos. 2056, 3514 and 3662), in Benin (Report No. 2034), in Niger (Report No. 2055), in Ghana (Report No. 3526) and in Sierra Leone (Report No. 2066). 2/ SOFITEX comments that poor climatic conditions also adversely affected cotton production in the whole country since 1980. 19. Moreover, experience has shown that in West Africa, cash crops, when they are possible, are generally likely to generate greater earnings than foodcrops for both the farmers and the Government, which is able to collect its share at the time the crops are marketed. Price policy should therefore favor optimum output rather than acting as a barrier. However, by underpaying for cotton, the Government deterred farmers from planting it and indirectly contributed to the present low yields and limited areas of cotton. 20. It is very likely that by paying growers a fair economic price and progressively doing away with the fertilizer subsidy (the effects of which are acknowledged to be negative), the areas planted and cotton yields could be doubled in the Bougouriba region. Moreover, higher productivity per hectare and per working day would permit increasing the levy per ton of cotton, thus increasing the revenue of the Development Fund. The project and its longer- term return prospects will continue to be threatened as long as the Government is unable to derive from the project itself the resources necessary for financing the recurrent development cost;. 21. The Government-s concern to ae self-sufficiency in food a priority for the country is probably the reason why the price paid for cotton for export has been kept artificially low. Experience shows, however, that the careful cultivation of cotton as the lead crop in a rotation system is bene- ficial for the foodcrop that follows it, thanks to the mineral fertilizer residue. The problem is not therefore so much a matter of choosing between a cash crop or a foodcrop, but of raising the two together by means of a rational rotation system, appropriate fertilization and a motivational price policy.  Annex 1 Page 1 May 13, 1983 E-1042/83 French (Upper Volta) OED ENMcM:pa Soci6t6 Voltaique des Fibres Textiles - SOFITEX Bobo-Dioulasso, April 19, 1983 No. 043663 TF/efs The Director Operations Evaluation Department The World Bank 1818 H Street, N.W. Washington, D.C. 20433 U.S.A. Dear Sir: We acknowledge receipt of the Project Performance Audit Report for the Bougouriba Agricultural Development Project and thank you for involving us in this evaluation. Our comments will relate to the chapter that affects us directly, namely policy with regard to cotton. Without denying the impact of price as an incentive for increased production, we feel that in this specific case account should be taken of other factors to explain the failure of production to increase in the Bougouriba ORD over the period 1977-1980. Examination of naional statistics will show sizable fluctuations in output: 38,043 tons in 1977/78, 59,956 tons in 78/79, 77,520 tons in 79/80, 62,538 tons in 80/81, and 57,534 tons in 81/82. There was a general drop in all the ORDs starting in 1980, due essentially to poor rainfall at the start of the crop year. Performance in the Bougouriba ORD can in fact be viewed as relatively satisfactory, since it held steady (4,679 tons in 79/80, 4,050 tons in 80/81, and 4,178 tons in 81/82). We are also in agreement with you in recognizing that early sowing is a decisive factor in obtaining high yields and a successful cotton crop. This point is made repeatedly to the producers, who are pretty much aware of it. But regardless of conditions, the farmers are legitimately concerned first and foremost with assuring their food supply, so that they sow their grain crops before they move on to cotton. - 10 - Annex 1 Page 2 On the issue of fertilizer subsidies, the new government policy, approved by the Bank, seeks to gradually reduce them in light of producers' reactions. It may therefore be somewhat premature to state that the effects of such subsidies are negative, and it should be noted that in Ivory Coast, where yields are high, fertilizers and insecticides are supplied to cotton producers free of charge, while Mali has witnessed a sizable and significant reduction in output following a decision in that country to eliminate subsidies (we have since heard that the question was being reconsidered there). In short, we fully share your conclusion on the need to step up cotton production without at the same time neglecting foodcrop production, as the two are complementary and should be pursued simultaneously. The problem is to ensure that both activities are done properly in order to benefit both the producers and the country. Yours etc. /s/ Fulgance TOE Director General - 11 - UPPER VOLTA Bougouriba Agricultural Development Project (Cr. 496-UV) Project Completion Report July 16, 1982 Western Africa Projects Department Agricultural Division 4 s - 13 - UPPER VOLTA Bougouriba Agricultural Development Project Project Completion Report I. BACKGROUND Sector Background 1.01 Economic growth in Upper Volta is necessarily closely linked to progress in the agricultural sector which employs 80-90 percent of the population and comprises almost 40 percent of GDP and almost all exports. However, soils are poor, rainfall is low and uncertain, and the potential for irrigation is economically limited, requiring costly investments to develop. Improving agricultural productivity as is necessary for maintaining food security and increasing national income is thus a challenging task. 1.02 Although the country is relatively flat, there are pronounced regional differences in climate and soils. The low and erratic rainfall (600 mm)in the north results in unreliable harvests and greater reliance on livestock for survival, and consequently low population densities. On the central plateau, crop farming is at a subsistence level because of population pressure, over-exploited soils and unreliable rainfall. In the southwest soil and rainfall conditions (reaching 1,200 mm) are more favorable, population density lower, and production more market-oriented. 1.03 Agricultural production over 1961-79 has performed as well as could have been expected given the severe constraints. Foodcrop areas, which account for almost 90 percent of cultivated land, have increased less than the net annual growth in rural population. In contrast to most other food crops, sorghum production has grown more rapidly than population reflecting in part the residual effects of cotton fertilizer in crop rotation, as well as direct fertilization. Overall, the country has continued to satisfy all but a small share (5 percent) of its consumption of food crops from local production in years of normal rainfall. But this has only been achieved through increasing dependence on the southwest as a source of food surplus, which tends to limit the resources available in that region for growing the country's major export crop, cotton. This dependence must continue, however, in the short term, whilst adaptive research attempts to develop improved approaches for increasing productivity over wide areas of the center. 1.04 Cotton yields have risen markedly in the last decade and output rose from 36,000 tons in 1970 (from yields of about 380 kg/ha) to over 78,000 tons (from yields of about 960 kg/ha) by 1979, due to cultivation shifting to more fertile areas and to the use of fertilizers and insecticides, coupled with extension programs. Since 1979, cotton area has fallen, indicating that close attention needs to be paid to the relationship of cotton and cereals prices, including reviewing whether current taxes on cotton which are not high should nevertheless be reduced. - 14 - Sector Strategy and Bank Group Lending 1.05 The Government's strategy in agriculture has given priority to: (a) improving the productivity of rainfed agriculture; (b) promoting migration from the densely populated and relatively infertile central plateau to the more fertile and less densely populated areas in the southwest; (c) expanding irrigation and swampland production schemes; and (d) ensuring national food security. In line with this strategy, Bank Group lending for Upper Volta has emphasized development of rainfed agriuclture for food and cash crops in the southwest, small-scale land development schemes to expand the agricultural production base in the central plateau, and testing a larger scale swampland production scheme in the southwest. 1.06 Since 1970, IDA has financed eleven projects in the rural sector of which five have been completed. The Cotton Project (Cr. 225-UV), US$6.2 million, 1970) demonstrated that an improved technology, within the means of small farmers, could be spread widely and rapidly. The project's Audit Report noted substantial improvements in farming practices and yields. The First Bougouriba Agricultural Development Project (Cr. 496-UV, US$8.0 million 1974, object of this report). The Drought Relief Project (Cr. 442-UV, US$2.0 million, 1973) and the First and Second Rural Development Fund (RDF) Projects (Cr. 317-UV, US$2.2 million, 1972 and Cr. 640-UV, US$9.4 million, 1976) aimed at financing small discrete works with community participation. The Project Audits described the Drought Relief and First RDF Projects as highly innova- tive and successful in implementing diversified and scattered subprojects. The second RDF project, completed in December 1981, continued the development of bottomlands, wells and anti-erosion works on the central plateau. A third RDF project was approved by the Board on March 30, 1982, while the two Volta Noire and Hauts Bassins Agricultural Development Projects (follow-up of the ongoing West Volta ADP) were negotiated in June 1982. 1.07 The ongoing IDA-assisted agricultural projects are: the West Volta Agricultural Development Project (Cr. 706-UV, US$3.6 million, 1977, Cr. 706-5-UV for Can.$3.0 million and Special Action Credit 49-UV for US$3.0 million equivalent), the Livestock Development Project (Cr. 557-UV, US$6.0 million, 1975); the Forestry Project (Cr. 982-UV; US$14.5 million, 1980); the Niena Dionkele Rice Development Project (Cr. 1013-UV, US$6.5 million, 1980); and the Second Bougouriba Agricultural Development Project (Cr. 1097-UV1 US$16.0 million equivalent, 1981). They are being implemented satisfactorily, although the swamp drainage Niena Dionkele Pilot Project has run into tech- nical problems of water availability and rice varieties, and the Livestock Project had difficulties with its group ranch component which necessitated a redesign and a reduction in the credit by US$3.0 million. Project Background 1.08 Project Area. Located in the southwest of the country, between the Black Volta and Bougouriba Rivers, the project area covers the territory of the Bougouriba ORD (Organisation Regionale de Developpement - Regional Development Organization). This territory is divided into the administrative districts of Diebougou (7,090 km2) and Gaoua (10,360 km2) covering about 6% of the total area of Upper Volta. - 15 - 1.09 At the time of appraisal, population was estimated at 360,000 inhabitants, about 6.5% of Upper Volta's total population (divided into 190,000 people in the Diebougou District, and 170,000 in Gaoua), and was expected tc -row at 2% yearly. Recent figures put the population at 390,000, of which 85% (about 33,000 family units) are engaged in agriculture. The Dagari tribe is predominant in Diegoubou, and the Lobi in Gaoua; people from the densely populated Mossi plateau are steadily migrating into Diebougou, reaching between 15% and 20% of the population. 1.10 The project area shares with the adjacent Banfora ORD the best climatic conditions for agricultural development in Upper Volta. Total annual rainfall varies from 1,000 mm in the North to 1,200-1,300 mm in the South, water is in surplus for plant growth from June through September, this is adequate for the successful cultivation of a wide range of annual crops. 1.11 There were about 35,000 farms and 750 villages in the project area. Altogether some 200,000 ha of land are cultivated and this is equivalent to 10% of the total land cultivated in Uper Volta. Average population pressure is low at about 24 inhabitants per km , but in the onchocerciasis free, or relatively free zones, population is much more dense and more than 50% of the land is under cultivation. About 80% of cultivated land is under cereals, mostly sorghum, and, the rest under a variety of crops of which groundnuts, yams, sesame, cowpeas and cotton are the most important. The area's cattle herd was estimated at about 106,000. 1.12 The area was well served by main roads, 375 km of primary roads adequately maintained by PWD. However, most of the secondary roads (about 200 km) and all of the rural tracks had been neglected for years; communica- tions were also adversely affected by the collapse of culverts and small bridges. 1.13 Project Design. The project was conceived as the first major effort to develop natural and human resources of the Bougouriba area. Previous efforts, mainly by CIDR (Compagine Internationale de Developpement Rural), a private management consultancy firm, achieved limited results, the lack of funds being the main reason. The creation of the ORD, while a step in the right direction, was also constrained by limited budgetary allocations. In view of enlisting IDA's support, the Government of Upper Volta commissioned CIDR to prepare the project, indicating highest priority to development of agriculture in the better rainfall areas of the country with the triple objective of: (i) self-sufficiency in food production; (ii) higher farm incomes; and (iii) increasing livestock resources The preparation report was financed by funds provided under Credit 225-UV, an ongoing West Volta Cotton Project. 1.14 The project was appraised in April/May 1973; as designed by the appraisal mission, the project contained the following specific elements: (i) providing technical assistance to the Ministry of Plan, Rural Development, Environment and Tourism; - 16 - (ii) providing Bougouriba ORD with the staff, housing, offices, equipment, transportation, and finance to implement an effective development program in the project area; (iii) providing about 16,000 farm families with the extension and other support services needed to increase their production and incomes; (iv) providing efficient veterinary services for the ORD; (v) creating a training center with teaching facilities for project staff and farmers; (vi) creating a research center to carry out adaptive research demon- stration and operate a seed multiplication unit; (vii) improving domestic water supplies through a digging and rehabi- litation program covering about 120 wells; (viii) improving about 660 km of secondary and tertiary roads; (ix) establishing a topographic unit to map and prepare for the development of land for settlement by migrants; (x) providing participating farmers with seasonal credit for fertilizer and insecticides and medium-term credit for farm implements; and (xi) carrying out the studies and surveys needed for the improvement of the road network in the project area and in the Leo district. 1.15 The project was to be carried out over a period of four years, from 1974/75 to 1977/78. Being essentially directed at increasing incomes, the project under this first phase would not include measures to improve social infrastructure or services other than water supplies. The Government intended to continue to support existing social services and to develop, with its own or foreign charitable resources, a farmers' cooperative program; cooperative leaders would, however, be trained in the project training center. The improvement of social facilities, especially health and education, and the development of cooperatives would eventually be part of subsequent phases of development in the project ORD. - 17 - II. COSTS AND FINANCIAL ARRANGEMENTS 2.01 Total costs at appraisal had been estimated at CFAF 2,550 million (US$10.2 million equivalent at the exchange rate of US$1=250 CFAF), includ- ing contingencies. Foreign exchange costs represented 53% of that total (US$5.4 million). Financing was to be provided by: IDA, US$8.0 million (79%); UPV Government, US$1.2 million (12%); the National Development Bank, for agricultural credit, US$0.7 million (6%); and the farmers, for the down- payment part of credit, US$0.3 million (3%). 2.02 Although Government's contribution was modest, it was essential to ensure a regular flow of such funds to the project. For this purpose the Appraisal Report recommended, and Government agreed to, the setting-up of a Development Fund, which would receive as income a fee levied on cotton products sales. These funds would serve to finance agricultural development activities in cotton-producing areas; the project area's ORD, together with the adjoining ORDs of Bobo-Dioulassu and Dedougou yield 75% of Upper Volta's total cotton production. Withdrawals from the Fund account would be in the context of the approved ORDs annual budgets. As the first payments into the Fund were not expected until the following cotton season, Government agreed to advance CFAF 150'million to finance start-up operations. 2.03 The main cash crop of the area, cotton, would continue to be marketed by CFDT (Compagine Francaise pour le Developpement des Fibres Textiles), which is also running the ginneries. CFDT would be responsible for payment of the levies to the Development Fund. The contract between CFDT and Government, setting an "Association en Participation" (A.P.) which was due to expire by project start-up time, was to be extended, and amended to include provision for the above payments. 2.04 The EEC Special Credit. As the project was being implemented, and despite the limited impact on agricultural production during the first years (substantially below appraisal estimates), it became obvious that the permanent features developed, i.e., infrastructure and the build-up of a motivated extension services, would in the long-run permit an efficient exploitation of the area's agricultural potential. A second phase project was therefore introduced into the pipeline; the timetable schedule showed however, that there was a strong possibility of a one-year gap between the time that funds under the first project were exhausted, and the time when the follow-up project would become effective. When that possibility was confirmed in 1978/79, a special credit from EEC funds was processed, to secure interim financing for the 1980/81 campaign. 2.05 The EEC Special Credit was in the amount of US$1.5 million; together with Government funds of about US$0.3 million, it was to be used to finance essential items such as agricultural extens ion, headquarters support services, feeder roads and technical assistance. The following table shows the original financing arrangements, and the changes after the allocation of the EEC credit: - 18 - Table 2.1: FINANCING ARRANGEMENTS Original Project Interim Revised Financing at Appraisal Financing Project Costs ---------------------US$ million-------------- IDA 8.0 - 8.0 EEC - 1.5 1.5 Government 1.2 0.3 1.5 Farmers and BND 1.0 - 1.0 TOTAL: 10.2 1.8 12.0 The provision of this interim financing extended the project by a sixth year, covering the 1980/81 Agricultural Campaign. The second phase project became effective in December 1981. - 19 - III. IMPLEMENTATION A. Effectiveness and start-up 3.01 The Credit Agreement stated four conditions of effectiveness: (a) Establishment of a Development Fund (para 2.02 above); (b) Payment by Government of the CFAF 150 million advance; (c) Two senior officials (Deputy-Director and Accountant) had been appointed; and (d) A revised contract with CFDT (Compagnie Francaise pour le Developpement des Fibres Textiles) had been signed. 3.02 The original date of effectiveness was postponed three times: (i) from the original October 17, 1974 to December 31, 1974; (ii) then to May 31, 1975; and (iii) to July 31, 1975. In fact, all the conditions were met on time, except the.recruitment of key personnel under (c) above, which took place early in July 1975; the credit became effective on July 18, 1975. PY1 became 1975/76. 3.03 A Manager for the Bougouriba ORD (who is ex-officio the Project Manager) was appointed in May 1974; he was previously ORD Manager at Yatenga. Pending effectiveness, the Project Manager together with the supervision mission of October 1974, drew-up a start-up program for the immediate execu- tion of most essential items such as housing and office space, utilities, vehicles and appointment of key personnel. This would be followed-up by the preparation for agricultural activities. 3.04 The Project Manager left very shortly after the project became effective (going overseas for a two-year course); he was replaced by another Voltaic. With the arrival of the expatriates and local hirings, the project was staffed by end of 1975, albeit sketchily; the recruitment of local staff in adequate numbers at a reasonable level has been a constant problem, and has seriously hampered the expected agricultural development. In 1978, a new ORD Manager (the third in three years) was appointed. His performance has been reasonably good although he is more of an administrator than an agricul- tural specialist. The Technical Projet Manager (an expatriate Agriculturist recruited by APMU) has been on the project since its inception. 3.05 The project was originally planned to be carried-out during a four-year period, covering the agricultural campaign of 1974/75, 1975/76, 1976/77 and 1977/78. Due to the delay in effectiveness, and the difficulties in staffing and building the infrastructure PY1 was actually 1975/6; the agri- culture activities did not really start before the follow year and as a result of savings made in the slow start-up, the project could be extended to a fifth year. Eventually, with the EEC interim financing (paras 2.04 and 2.05) the project lasted 6 years. - 20 - B. Development of Agricultural and Livestock Services General 3.06 Development of trained extension workers, and subsequently their impact on farming practices, took place more slowly than anticipated at appraisal. But the creation of agricultural extension services was part of the larger institution building effort of the project: establishment of the ORD in Diebougou (the HQ) and in the field, organization and staffing of its different services, such as accounts, credit, livestock and ox-drawn equipment distribution, seed mnltiplication, fertilizer distribution. In this context, the project has achieved its expectations. 3.07 The agricultural achievements of the project were substantially lower than appraisal expectations which, however, project management found to be grossly over estimated. Total production by PY5 of cotton at 4,700 tons and cereals at 77,000 tons were respectively 48% and 64% of estimates. Nevertheless, cotton production increased significantly and the project probably helped maintain food self-sufficiency in the ORD since there were periods in the past when imports were needed. The evaluation of the agricul- tural impact of the project is dealt with in Section IV. Agricultural Extension and Staff Training 3.08 By PY5 some 10,000 farmers were estimated to have benefitted from the extension services or 60% of appraisal estimates; see Section V. The slow build-up in trained extension staff was partially due to a decision taken early in the project not to build an extension training center at Bobo- Dioulasso and instead to rely on short-term courses at an existing school. After this school stopped offering these courses in Year 2 of the project, the ORD Director decided, with the Association's approval, to organize in-house courses. 3.09 Progress has been slow and the need for on-the-project facilities became imperative. The follow-up project has provided funds for the con- struction of a 125 m2 classroom with audio-visual and other equipment to house a training section for recycling groups of up to 30 extension agents. Funds have also been allocated for a Training Specialist to be on the project for two years. Veterinary Services 3.10 The project has reinforced the veterinary services in the region, and the Regional Head of Livestock Services in the region was made responsible for the Livestock component of the project, working with staff hired under the project. Two Livestock Sectors have been set-up, at Diebougou covering the north and the centre, and at Gaoua, covering the south. The Diebougou Sector has three outposts, at Dano, Babora and Dissin; the Gaoua Sector has five - 21 - outposts, at Batie, Nako, Kampti, Lorepeni and Gaoua. Veterinary supplies were procured by the project and distributed at cost price. Results of the veterinary services are evaluated in Section V. Applied Research 3.11 At appraisal an applied research station of 65 ha was recommended which additionally to research activities would be involved in seed production on a large scale. This was not considered justified by the project which proposed to replace it by three substations; a supervision mission, late in 1976, agreed to that change. The three substations, each over 2 ha, were set up at Tiankoura (centre), Batie (south). and Oronkua (north), with the relevant Research Institutes (IRAT, IRHO, IRCT) providing seeds and research planning, and also analyzing the results at Bobo-Dioulasso. Seed production, which remained very limited throughout the project life, was entrusted to selected farmers. Useful applied research was done on cotton varieties and fertilizer needs, on improved sorghum varieties and on rice varieties. In conjunction with FAO an important demonstration program of fertilizer use on upland crops was conducted from 1978 on. Agricultural Credit and other Inputs 3.12 Animal traction cultivation is one of the technological improvements encouraged by the project with a medium term credit program; progress has been slow but shows that there is an increasing interest in adopting this technique. The following table indicates the evolution under the project (figures cumula- tive): Table 3.1: ANIMAL TRACTION - NUMBERS Number of Number of Participating Pairs of Equipment (sets) PY farmers Oxen Plows Maintenance Transport Other 1/2 n.a. n.a. 242 224 127 8 3 328 346 315 333 186 5 4 363 431 398 529 308 16 5 559 499 543 887 401 18 6 705 704 768 1,271 474 29 The relevant figures is the number of plows under animal traction, which has increased by 6% from PY3 to PY4, by 36% in PY5 and 41% in PY6. 3.13 The funds provided by the project for credit to finance the procure- ment by farmers of animal traction equipment were not used since lines of credit were made available by USAID, the Conseil de l'Entente and the French CCCE as follows: - 22 - Table 3.2: ANIMAL TRACTION: CREDIT Donor Campaign No. of Borrowers Cost of Equipment Interest Rate (CFAF million) USAID 1976/77 38 3.4 5.5% 1977/78 152 8.4 " C.E. 1978/79 122 5.2 " 1979/80 146 7.1 " CCCE 1980/81 238 12.1 8.5% Terms were usually one year of grace, and principal plus interest payable over 4 years. 3.14 From statistics collected on the project, the collection rate is estimated to be as follows, starting in 1978 when first repayments were due: Campaign Amounts due Payments Collected % (in CFAF) 1978 877,830 642,015 79.13 1979 2,935,630 2,151,100 73.03 Further development of ox-drawn cultivation has been included under the Second Bougouriba Agricultural Development Project, which also provided funds for related animal-traction agricultural implements (US$1.1 million for a six year period). 3.15 Uptake of other inputs was less than expected, due not only to the lower absolute increase in crop areas, but also below optimum use of fertilizer per hectare on cotton. The expected uptake of fertilizers of foodcrops did not materialize; farmers' interest developing late in the project with the advent of the FAO fertilizer demonstration program. C. Civil Works Buildings 3.16 Successful project implementation required that a large building program be completed early in the project period. The only realistic way to complete such a program was to undertake the work using existing local building contractors. Unfortunately, such contractors were not available in the project area, and some outsiders who bidded on the building tenders proved to be unable to finish the work. Some of the sites had to be completed on - 23 - force account by the Project. This involved considerable delays and the building program stretched throughout the project life, with the last units delivered at the end of FY5. 3.17 The following table indicates the scope of the building program: Table 3.3: BUILDING PROGRAM CFAF m Total CFAF CFAF m Total Units Unit Million Units Unit CFAF Million Offices: HQ 1 9.4 9.4 1 7.8 7.8 Diebougou 2 4.1 8.2 2 4.8 9.6 Sectors 7 2.0 14.0 7 3.5 24.5 Houses: Senior 9 4.7 42.3 ) Middle 11 4.1 45.1 ) 14 - 61.5 Junior 2 2.9 5.8 ) Stores: 500 m3 1 10.0 10.0 1 7.2 7.2 200 m3 9 4.7 42.3 6 4.5 26.0 50 m3 36 1.4 50.4 30 2.0 60.0 Training Center 1 11.3 11.3 - - Garages 2 3.0 6.0 1 8.2 8.2 Miscellaneous 1/ 37.3 3.8 TOTAL: 282.1 208.6 1/ Utilities (15% of houses and offices costs), vaccinations yards, oxen stables, etc. 3.18 It will be seen that the building program closely met the targets set for it, except that less staff housing was constructed than projected. This was caused by a project decision to pay a living out allowance to many of the junior staff assigned to field locations. The cost of buildings constructed was generally in line with the appraisal estimate. The training center was not built as facilities were available at Bobo-Dioulasso, and the vaccination yards were postponed to second phase. Costs would have been higher if it had not been for close supervision of contractors by project management. - 24 - Roads 3.19 Except for the main highways the very few rural roads in the area before the start of the project were in poor condition and suitable for dry season use only. The appraisal proposed the rehabilitation of 660 km of unsurfaced road and about 200 km of laterite surfaced road to bring most rural farmers within walking distance of a road. This was considered vital to enable extension work to be carried out and for fertilizer and other inputs to be made available to the farmer. 3.20 Construction techniques recommended at appraisal were the same as those for other areas in Upper Volta. The road construction unit adopted the concept proposed by appraisal of using the light motor grader as the principal construction machine. However, the project soon realized that these motor graders could not handle the work alone, and acquired a crawler tractor (D-6 type) as well as two heavier graders to carry out the heavier earthmoving and reshaping. These worked well, and the unit was completed with water and tipper trucks, and a self contained concrete production unit. 3.21 The rehabilitation work on the deteriorated feeder roads included the reopening of the tracks, reshaping, regravelling and the manufacturing of concrete elements (bridges, culverts and concrete slabs). The feeder roads unit improved about 646 km of roads (against 660 km scheduled at appraisal) of which 539 km were fully regravelled and 107 km partially. It was one of the most successful project components. Progress of work followed the appraisal calendar and the technical standards throughout were 5 m width for roads and bridges, and regravelling with between 2 and 3 inches of surfacing. 3.22 In addition to the road building and rehabilitation, the feeder road unit carried-out maintenance totalling 1,492 km during the 5-year period. The list of roads built, rehabilitated and maintained is at Annex 1. Analyzing the data available at the project, and from entries in the accounts department (which did not determine unit costs), the outlay for road building and/or maintenance can be summed-up as follows: - 25 - Table 3.4: ROAD CONSTRUCTION COSTS PYl PY2 PY3 PY4 PY5 Interim Year Total km constructed 646 km maintained (1,492) Maintenance/Construction 1/ (no breakdown available) 8:1 186 Total Unit km 832 Capital Costs 170.8 - 23.3 9.3 - 203.4 Operating Costs: Salaries/Allows. 0.8 12.8 21.7 35.1 23.2 17.6 101.2 Equipment Oper. 3.4 25.7 49.1 70.8 73.8 23.7 246.5 Euip. rental 2/ 12.9 76.5 7.6 - - - 97.0 Transport 0.1 7.0 - - - - 7.1 Sub-total: Cement - materials 12.6 27.5 24.9 4.2 - 69.2 Misc. 0.1 0.3 0.1 0.1 - 0.6 Total Op. 17.2 134.7 106.2 120.9 101.3 41.3 521.6 Oper. plus Capital 17.2 305.5 106.2 144.2 110.6 41.3 725.0 Without Capital Investment (CFAF '000) Average cost of km road built (US$2,322) 626.9 Average cost of km road maintained (626.9:8) (US$290) 78.4 With Capital Investment Average cost of km road built (US$3,227) 871.4 Average cost of km road maintained (871.4:8) (US$403) 108.9 1/ Assuming cost of building/rehabilitating a km of road costs eight times the cost of maintenance (a ratio obtained by examining the little data available, taking in consideration acceptable criteria). 2/ In view of late delivery of equipment procured, plant had to be rented to start the programme. - 26 - 3.23 These costs are more than double the appraisal figures, which have evidently been grossly underestimated; the nature of the terrain required much heavier equipment than originally forecast, which also increased substantially operating costs. The delay in procuring the heavy equipment made it necessary to rent engines to start the program, thus incurring an additional CFAF 97.0 million in unforseen expenditure. Even thus, the estimated real costs are within the West-African range of outlay for feeder roads. Two of the three Northern Nigeria ADP's spent an average N 10,000 per km built (approx. US$16,000) which is equivalent to about CFAF 4,000,000, i.e., six times the Bougouriba average; even taking into consideration the higher standard of the Nigerian roads, the Bougouriba costs are still acceptable. Wells 3.24 The appraisal target was 120 wells to be built at an estimated cost of CFAF 280 million (including contingencies). The project drilled 119 wells, of which 103 were completely successful, 13 had to be deepened, and 3 to be abandoned; total actual costs for the well program was CFAF 111.3 million, an average of CFAF 935,000 per well, as compared to CFAF 2,333,300 in the appraisal estimates. The technical characteristics were: Internal diameter 160 cm External diameter 180 cm Depth recommended 15 m Average depth achieved 15.31 m Rate of sucess: (no drying-up at the end 77% of the rainy season) 3.25 This component was considered by the project as a very satisfactory achievement. However, it is very possible that because of the way accounts were kept on the project, some of the wells program expenditure may have been charged to the feeder roads unit, although it has not been possible to ascer- tain this fact (this may partly explain the higher costs of the roads). Notwithstanding this possibility, it can be said that the project management carried out the wells program reasonably well. Technical Assistance to the Ministry of Rural Development 3.26 At the time of appraisal, the Ministry having jurisdiction over the project was the Ministry of Planning, Rural Development, Environment and Tourism, which was recently set-up and took over the former Ministry of Agriculture and all its technical services; development, planning and agri- cultural activities were thus placed under one roof. The project provided for a senior expatriate technical assistant to the Ministry, to assist in supervizing ongoing activities, planning future projects and coordinating activities with other Ministries and Agencies. A short time after his arrival (end 1975), there was another change and Planning was set up as a Ministry, as were Environment and Tourism, and the usefulness of this technical assistance - 27 - was reduced during the ensuing period of reorganization. In spite of verbal assurances, it does not seem that the Ministry knew how to use these services in a position where the expatriate could have an impact (for obvious political reasons), and the technical assistant himself could not impose himself in duties of ecine influence, a difficult task under any circumstances. APMU feels that not enough support, or attention, was given to this problem by supervision missions; the contract was not renewed after its original two-year period. Project Staff 3.27 Three expatriates were employed under the project, two recruited by APMU under secondment status, the Technical Director of the Project (also Deputy-Manager), an Agriculturist, and a Chief Accountant. The third was a Cotton Specialist, seconded by CFDT. Their performance was adequate. Studies 3.28 Two studies were carried out by the project, both relating to the roads program: the engineering of the Pa-Dano-Djipologo road (73 km), and a road survey in Leo district to establish priorities for improvement and extension of the district's road network. E. Project Costs 3.29 The table comparing actual project costs with appraisal estimates is at Annex 2, Table 1 and is summarized below: - 28 - Table 3.5: PROJECT COSTS EEC Apr. Actual Total Apr. Act. Inter. Total Est. Costs Est. Expen. Finan. Costs in US$ in US$ 1/ (4 .) (5 yr.) E/A (6th yr.) Phase I (@ 250 CFA) (@ 227.50) ----------------CFAF million------------- --------US$ '000-------- A. Operating Expenditures Ministry Support 80.9 10.7 13% - 10.7 324 47 ORD Administration 317.4 375.6 118% 90.0 465.6 1,270 2,047 (including training) Agricultural Develop. 811.4 606.0 75% 186.6 792.6 3,246 3,484 (incl. livestock, research) Wells Program 202.3 108.4 54% 2.8 111.2 809 489 Roads Program 251.7 480.3 191% 41.3 521.6 1,001 2,293 (incl. studies) Topographical Unit 16.1 48.7 302% 9.0 57.7 64 254 Input Supplies 330.8 66.1 20% - 66.1 1,323 291 Sub-total: 2,010.6 1 695.8 84% 329.7 2,025.5 8,037 8 905 B. Investments Buildings 267.5 188.5 70% 20.1 208.6 1,075 917 Vehicles 104.7 88.7 85% - 88.7 419 390 Road Bld. Equip. 62.2 203.4 327% - 203.4 249 894 Other equipment 105.0 27.6 26% - 27.6 420 121 Sub-total: 539.4 508.2 94% 20.1 528.3 2,163 _21322 Total Project Costs: 2,550.0 2,204.0 86% 369.8 2,553.8 10,200 11,227 plus EEC Special Credit 1,500 11,700 1/ Average rate of disbursements up to March 1981. - 29 - 3.30 Input supplies have reached only 20% of appraisal estimates for the following reasons: (i) about CFAF 78 million allocated for procurement of animal traction were not used since other donors had provided for lines of credit for the same purpose (para. 3.13); (ii) the expected take-off in fer- tilizer use by food crop farmers did not materialize - the project estimates that as little as 150 tons were used against a 1,900 tons forecast; and (iii) fertilizers for cotton growers is now distributed on seasonal credit by the cotton company, SOFITEX, which receives financing from commercial banks for that purpose on a national basis. 3.31 The project lasted five years, instead of the appraisal forecast of four years, and total costs for the five years were CFAF 2,204 million as compared to CFAF 2,550 million at appraisal. The IDA Credit of US$8.0 million was disbursed at an average of CFAF 227.5 for US$1.0 instead of CFAF 250 on which appraisal estimates were based; for that reason, funds were exhausted by the end of the project span, and financing for a sixth year (to bridge the gap before the Bougouriba II project.could be effective) was obtained through an EEC Special credit of $1.5 million, approved on January 31, 1980. 3.32 The following table shows how the total six-year program was financed: Table 3.6: FINANCING With EEC financing of Sixth Year Original Project (to complete Phase I) US$ Million CFAF Million % US$ Million CFAF Million % IDA 8.0 1,832 84 8.0 1,832 72 EEC - - 1.5 310 1/ 12 Government - 359 16 - 408 2/ 16 2,191 100 2,550 100 1/ Only 1.4 m of EEC funds have been disbursed; the small balance was applied to Phase II. 2/ The auditors, Coopers & Lybrand, have certified the receipt of this contribution. 3.33 As for project costs, apart from the Roads Component, which has been already said was substantially understimated at appraisal, they remained generally in line with original estimates. Auditing of the project accounts was carried out regularly by Coopers and Lybrand. In their final report, dated July 17, 1981, they state that: - 30 - (a) withdrawal requests had been established as per IDA guidelines; (b) goods and services included in these requests originate from countries members of IDA and Switzerland; (c) such goods have been received (or are in transit) and services have acutally been rendered; and (d) all goods and services (including on force account) financed by IDA have been used within the Project framework. - 31 - IV. AGRICULTURAL IMPACT A. Crop Production 4.01 The agricultural program followed a multi-crop approach directed at both cotton and foodcrops. At appraisal it was assumed that farmers' adoption of improved technical packages would result mainly in an increase in cash crop production -- cotton -- of 380%, with foodcrop production, mainly for home consumption, rising by only 5%. Assessing the project achievements in this field has been very difficult, particularly for foodcrops. Apart from cotton statistics for which reasonably reliable information was recorded, both on the project and with the marketing agency SOFITEX, data on cereals is generally sketchy; no data was collected in PY1 as the project was in the process of getting organized, and no comparative figures were available as a baseline starting point. This problem remained throughout the project life, even when material was gathered for the preparation of the follow-up project, specifi- cally since there was no specific project-financed unit in charge of monitor- ing and evaluation. 4.02 By the time the project organized its extension services and started having a good look at areas under production, estimating yields and total output, it became obvious that appraisal estimates were grossly over-estimated, both in its starting figures and in its progress assumptions. Cotton areas were estimated to grow from 5,000 ha in Year 0, to 12,000 ha in Year 4, an increase of 240%; yields were expected to double, from 400 kg/ha in Year 0 to 813 kg/ha in PY4. Total cotton production was estimated to increase fivefold by FY4. As for cereals, the starting yields were estimated at appraisal at 700 kg/ha for sorghum and millet; such yields were never achieved in the project area, Only the maize yields were correct, reflecting the ongoing performance in the project area, but the area under production was over- estimated in the appraisal report at 25,000 ha, while the project's inventory discovered only about 7,700 ha. Progress of maize plantings was estimated at 10% over 4 years (from 25,000 ha in PYO to 27,500 ha in PY4). The project estimates that maize area actually increased by 20% (to 9,237 ha) by PY5. 4.03 Following the 1978 cropping season, project management prepared a production data series for each crop. The matter of greatest importance was the selection of a reference year of yields under average climatic conditions; 1975/76 was retained as the base year by project management because the rainfall that year was 977 mm distributed over 77 days. This is the nearest to the regional yearly average rainfall of 1,050 mm spread over 82 days, as recorded over a 30-year period. 1975/76 was also the project's first year of operations. The first two years (1975/76 and 1976/77) were based on the best estimates of field staff. The 1977/78 and 1978/79 data were developed from best estimates and field survey data. The results thus obtained were examined by the February 1979 supervision mission, commented upon in the supervision report (Annex 7) and discussed with the Project's management. These remarks were taken into consideration when the final project figures were tabulated for the PCR in October 1981. - 32 - 4.04 The final figures are Presented in two tables in Annex 3, Table 1 for cotton and Table 2 for cereals. The first part of each table shows appraisal estimates for areas, yields and production, from which percentages of expected progress (yearly and overall) have been calculated. The bottom part of the tables show the project's best estimate of actual results with the same percentages calculations. Cotton 4.05 Cotton was cultivated in the Bougouriba ORD well before the project started, as records go back to 1963. From 1970 to 1975, the land under cotton average 4,210 ha and yields about 326 kg/ha. The appraisal report decided however to use the 1973/74 area (about 5,000 ha) as baseline, applying to it yields of 400 kg/ha, obtaining a baseline production of 2,000 tons which was seldom reached in the area. Area planted under the project increased sub- stantially, from 4,879 ha, in 1975/76, to 7,036 ha, in 1979/80. With the exception of 1977/78, which experienced a severe dry spell in April and insufficient rains in July, the project cotton area has now levelled at around 7,000 ha per year. Yields and Production have also increased significantly. Table 4.1: COTTON AREA, PRODUCTION AND YIELDS Year Area (ha) Production (tons) Yield (kg/ha) Pre-Project 1970/71 3,940 1:,109 281 1971/72 4,100 1,416 345 1972/73 3,655 1,607 438 1973/74 4,973 1,497 301 1974/75 3,981 1,062 266 Project 1975/76 4,859 1,837 378 1976/77 7,016 2,483 354 1977/78 7,208 1,802 249 1978/79 6,000 3,412 568 1979/80 7,036 4,679 665 1980/81 6,835 4.050 592 4.06 In 1979/80, the Bougouriba ORD produced some 4,700 tons of cotton, or about 6% of the total 75,100 tons purchased by CFDT nation-wide, the highest percentage of the national production ever. Because of adverse climatic conditions in 1977/78, the area under cotton decreased from 7,208 ha to 6,000 ha the following year; farmers probably decreased the amount of land under cotton to switch to foodcrops. From discussions in the field, it is - 33 - believed that the total land under cotton, in the Bougouriba ORD, would remain in the range of 6,000-8,500 ha, for the next few years, depending on climatic conditions. Appraisal forecasts were generally correct regarding the annual area increases for cotton during the first two years of the project, but were over-optimistic beyond 1976/77. As a result, it is expected that there would be a lag of 2/3 years in reaching appraisal targetL for area under cotton; the targets are expected to be achieved under Phase II. 4.07 On examining the yields' column of the table in para 4.05, it is noticed that during first three project year no progress is shown and even the baseline figure of 400 kg/ha was not reached. Several observations have to bc made: (i) as said in para 3.05, the project did not have any impact before PY3, being totally involved in recruiting adequate extension staff, training it, and being otherwised occupied in completing its organization; (ii) 1977/78 suffered adverse climatic conditions, not only affecting yields, but also forcing farmers to reduce their cotton acreage in the following year; and (iii) by that time, on the recommendation of the supervision missions, the project hired an expatriate specialist in cotton-production (from CFDT). Only then could serious extension work (and distribution of inputs) begin; the results are felt in the following years, with yields immediately rising to about 600 kg/ha (665 kg/ha in 1979-80). This is still below the appraisal estimate of 813 kg/ha (-22%), and also below the national average of 837 kg/ha achieved in 1978/79. 4.08 Within the Bougouriba ORD, the area under cotton, number of farmers and size of plots have varied since 1975/76, as shown in this table: Table 4.2: NUMBER OF COTTON FARMERS AND SIZE OF HOLDINGS Area (ha) Average Area per Farmer Number of Cotton Growers Year North South ORD North South ORD North South ORD 1975/76 4,784 75 4,859 0.73 0.15 0.69 6,519 493 7,012 1976/77 6,811 205 7,016 0.92 0.20 0.81 7,347 1,025 8,372 1977/78 7,022 186 7,208 0.99 0.20 0.90 7,061 930 7,99i 1978/79 5,685 315 6,000 1.02 0.20 0.84 5,555 1,565 7,115 1979/80 7,690 445 7,035 1.09 0.25 0.89 6,042 1,780 7,822 The difference bctween the North, the main cotton zone, and South should be noted. The take-off in cotton plantings was slow in the South reaching a total of 445 ha by project's end. On the other hand it increased by 38%, to 6,590 ha, in the North. The number of farmers in the South rose by 72% to 1,780, whereas their number decreased by about 7% in the North; one particular reason for this was the drought in 1977/78. Also of interest, the average size of cotton plots per farm unit amounts to 1.09 ha in the North and 0.25 ha in the South. At appraisal, it was estimated that by 1978, some 16,000 farm units would be involved in growing cotton. In fact, the actual number in 1978/79 was only 7,822 growers. - 34 - 4.09 Total production in PY5 (1979/80) was 4,679 tons, compared to appraisal estimates of 9,756 tons, or 48%; but as the project was very slow in starting having an impact, and the take-off did not really start until PY3, we will compare the project's PY5 achievements to both PY3 and PY5 appraisal estimates: Appraisal Estimates Project's PY5 Percentage of Percentage of Achievements PY5 Achievement PY3 Achievement Number of farmers 7,822 16,000 49 13,200 59 Average plot (ha) 0.89 0.75 119 0.65 137 Area under culti- vation (ha) 7,035 12,000 59 8,750 80 Yields (kg/ha) 665 813 82 720 92 Total production (tons) 4,679 9,756 48 6,300 74 4.10 The main reason why the project's achievements seems low, even if compared to the appraisal's PY3 target, is that the baseline estimates were overestimated in the SAR. One of the assumptions used were that there were 10,000 cotton farmers in the project, which would increase as follows: PYl PY2 PY3 PY4 Number of farmers 10,000 10,500 13,200 16,000 Increase 500 3,200 2,800 Percentage of increase 5 26 21 Overall increase - 32 60 The project seriously questions these figures. Their field surveys lead them to state that there was a core of farmers (about 7,000) who have been cotton growers for years, and will probably stick to that crop and both slightly expand and improve, to whom marginal numbers could be added as a result of the extension action. Even then, the new farmers would remain shy and abandon cotton for foodcrops whenever climatic conditions are adverse and susbsistence becomes the main problem. Even for the core of cotton growers, 60% apply only 105 kg/ha of fertilizer, instead of the recommended 150 kg/ha plus 50 kg of urea. The only way for these farmers to reach the 800 kg/ha yield target is to follow the fertilization recommendation, which, together with improved husbandry already making progress, could even result in yields of between 1,000/1,200 kg/ha; these are the targets of the Second Bougouriba Project. Cereals (Sorghum, Millet and Maize) 4.11 Returning to the Production Table, Annex 3, Table 2 (para 4.04), the comparison of both lines for cereals production is as follows, using PY1 as baseline: - 35 - Table 4.3: COMPARISON BETWEEN APPRAISAL ESTIMATES AND ESTIMATED ACTUAL PRODUCTION OF ALL CEREALS (Sorghum, miliet and maize) PYl PY2 PY3 PY4 PY5 PY6 Appraisal Estimates: Production (metric tons) 116,200 117,800 119,400 122,500 122,500 122,500 Estimated Actual Production: 73,432 69,111 73,028 75,778 78,063 77,115 Appr. Yearly Progress % - 101.4 101.4 102.6 100.0 100.0 Appr. Overal Progress - - 102.8 105.4 105.4 105.4 Estimated Real Achievements: Yearly - 94.0 105.7 103.8 103.0 99.0 Overall - - 99.5 103.2 106.3 105.0 4.12 The percentage of production increase shown by this stream is in line with appraisal estimates, an overall 5% over the 4-year period; the latest production figures indicate that this performance is maintained. Evidently, this percentage applied to a lower baseline shows an incremental production due to the project much lower than appraisal estimates, but then it is obvious that these estimates were inflated. 4.13 Project managemcnt has established its estimates of actual produc- tion (Annex 3, Table 2), essentially by analyzing population figures, average food consumption requirements, the number of farms, and the estimated average farm area under millet and sorghum; gross yields of these crops in 1980/81 in the area were estimated to be in order of 530 kg/ha, for all project farmers. The impact of the project on foodcrops farmers reached only about 10,000 farmers out of the estimated 33,000 in the area. These "progressive" farmers have improved their yields to an estimated 625 kg/ha PY5, thus indicating that the other farmers have remained with yields of under 480 kg/ha. In the appraisal of the Second Bougouriba Agricultural Development Project, four models representing the lowest level farmers were presented, with brackets showing the range of yields within which these farmers were expected to progress. 1/ The appraisal report stated (para 5.02) that the emphasis in the models was on the incremental yield changes possible (on average an increase of 150 kg/ha was forecast for sorghum) 2/, in view of the fact that data on 1/ Sorghum 400-600 kg/ha millet 350-550 kg/ha and maize 700-900 kg/ha. 2/ Which is a reasonable expectation based of Phase I results. - 36 - actual yields was inadequate, and the monitoring service would attempt to improve the statistics of actual yields. Comprehensive survey were carried out in 1979/80 in 1981/82 and the results are currently being analyzed. Consultant services under Bougouriba II will be used to establish a system by which areas, number of farmers and actual yields could be efficiently monitored. Groundnuts and Rice 4.14 Groundnuts. The appraisal data for this crop was that about 20,000 ha were in cultivation before the start of the project, producing about 8,700 tons at yields averaging 435 kg/ha. The project's targets for groundnuts were modest: only 600 ha would shift to improved cultivation (mainly through selected seeds and fertilizer) to reach yields of 800 kg/ha, while the other areas would rise their yields to 450 kg/ha only. Statistics drawn by the project show that groundnuts occupy a much smaller area than estimated at appraisal. Basic data of groundnut cultivation under the project is as follows: Table 4.4: GROUNDNUT PRODUCTION Average size Year Acreage Production Yield of pilot (ha) (tons) kg/ha (ha) 1975/76 5,439 2,311 425 0.17 1976/77 5,820 2,328 400 0.18 1977/78 5,241 2,106 402 0.16 1978/79 5,344 2,538 475 0.17 1979/80 5,456 2,728 500 0.17 4.15 Project management reports that some 100 tons of selected groundnut seed were distributed to farmers between 1976/77 and 1979/80. Assuming a sowing rate of 80 kg/ha of unshelled seed, the 100 tons of selected seed allowed for the sowing of 1,250 ha. Based on numerous FAO demonstration plots (i.e. 182 trials in 1978/79) the average yield for the local varieties chosen was 1,100 kg/ha compared to a potential yield of 1,500 kg/ha. It is estimated that, in the farmers' environment, the obtaintable average yield is about 700 kg/ha, as most cultural practices were not adopted until 1978/79, at the time that the extension services started reaching groundnuts farmers. 4.16 The project also provided farmers with some 65 tons of fertilizers during the last four years. It is estimated that, based on a traditional application of 75 kg/ha, some 860 ha of land received some fertilizer. The combined effect of the selected seed and the use of fertilizers (assuming the same farms are considered) allows a final yield of 1,000 kg/ha over 860 ha and 700 kg/ha over some 410 ha. The ensuing production is evaluated at 1,150 tons. - 37 - It is assumed that the overall production without the project would have been about 570 tons (with an average of 450 kg/ha). Therefore, the induced produc- tion over the past four years is estimated at some 580 tons or 145 tons per year. This represents only 5% of the project's annual average production of 2,500 tons of groundnuts. 4.17 Rice. It was estimated at appraisal that the project area had some 3,000 ha under rice cultivation, producing 2,700 tons with an average yield of 900 kg/ha, in 1972. It was thought that the total area would hardly vary, from 3,000 ha to 3,050 ha, by PY4, but that some 550 ha of both upland and swamps would be improved, of the latter total. The overall production was forecast to increase from 2,700 to 3,017 tons. Yields from the 500 ha of improved upland area would rise from 900 kg/ha to 1,200 kg/ha. Similarly, yields would increase from 1,000 kg/ha to 2,000 kg/ha in the sixth year of development. The following table provides the basic data for rice cul- tivation over the past five years: Table 4.5: RICE PRODUCTION Year Acreage Production Yield Average Size of Plot (ha) (tons) (kg/ha) (ha) 1975/76 3,200 2,240 700 0.10 1976/77 3,200 2,240 700 0.10 1977/78 2,879 2,225 773 0.09 1978/79 3,686 2,949 800 0.11 Project management cautioned that the data provided was based on rough esti- mates, especially for the first two years. Furthermore, although the project sold some 19 tons of selected rice seed that allowed for the sowing of 316 ha (at a sowing rate of 60 kg/ha), no distinction was made between upland and swamps. 4.18 Apart from selected seed, the project has not yet brought in any major improvement in rice cultivation. Water control measures should be improved and the use of fertilizers and pesticides need to become more wide- spread. Based on the quantities of seed sold to farmers and in the absence of other improving cultivation methods, the mission estimates the possible annual production increase at 20% for the limited areas with selected seed. It is estimated that the average annual rice production due to the project is about 20-25 tons, which is minor indeed when compared to the total production of 2,949 tons. B. Livestock Development 4.19 Upper Volta is a transit area for cattle, and the Bougouriba region is on the path of travelling herds, hence the high risk of contagion of cattle diseases. This situation created numerous centers of contagious rinderpest in - 38 - 1974, and pleuropneumonia appeared in 1972 and remained a threat. The main effects of the first phase project have been to improve the health of the various herds in the project area; vaccination of all cattle against rinder- pest was carried during the first years of the project, followed by a sys- tematic vaccination annually for all cattle under one year of age, accompanied by continous vaccination of cattle where contagious bovine pleuropneumonia appeared. The following statistics have been computed: Rinderpest 1974/75 1975/76 1976/77 1977/78 1978/79 Contagious centers 19 - - - - Vaccinations 58,061 40,361 70,764 57,015 65,863 Pleuropneumonia Contagious centers 2 6 - 1 - Vaccination 33,841 38,353 50,042 54,145 21,980 These campaigns have been successful, as no new contagious centers were observed since. Intensive campaigns of deworming have also been carried out throughout the project area. 4.20 The herd population remained constant during those years; the consultant involved in the preparation of the second phase project estimated the herd population to have remained practically unchanged, from 128,485 heads of cattle in PY2 to 130,000 in the final year. The benefits of the health campaign and the ensuing lower mortality rate resulted in a higher output rate which improved by about 18%; this output rate was estimated at appraisal at 10.5%, advancing thus to 12.5%. The value of the incremental meat production has been estimated in the following table prepared by the Consultant (in millions of CFAF, 1981 prices). Table 4.6: MEAT PRODUCTION Meat Production Meat Production Value of Incremental PY with project without project Meat production 1975/76 1 n.a. n.a. - 1976/77 2 421.2 401.0 20.2 1977/78 3 444.6 394.3 50.3 1978/79 4 464.7 410.6 54.1 1979/80 5 469.4 426.2 43.3 1980/81 6 458.4 428.0 30.4 - 39 - V. INSTITUTIONAL PERFORMANCE 5.01 Tb- project was instrumental in establishing at Diebougou a complete agricultuidl development organization. The ORD of Bougouriba, with its offices, equipment, fleet of vehicles, repair workshop and other auxiliary services serves as a support to the main operational divisions: agricultural development, which includes extension services sprcad over seven sectors, seed multiplication services, ox-drawn cultivation training and support, assistance to farmers' groups and a household economics section; a supplies and stores division, a Livestock division, and an anti-erosion works section. The extension services, together with the farmers' group section, supervize the allocation of credit to farmers for cotton inputs (seasonal credit), and for animal traction equipment. The credit officers are now being trained and organized by CNCA (The National Agricultural Credit Organization), which will take over responsibility for agricultural credit, but would continue operating through the ORD's credit offices. 5.02 The agricultural impact of this organization was very slow to develop (mainly due to underestimated difficulties in attracting and keeping qualified personnel, and to the needs to train them); however, its activities in the project area were certainly a determining effort in motivating farmers, in rehabilitating infrastructure (the roads and wells program were certainly a success), in improving input supplies (although practically limited to cotton growers in Phase I, but slowly expanding to the whole farming population), in organizing credit through farmers' groups, in forming seed-production farmers--in general in building itself as the motor of agricultural drive in the project region. 5.03 Project management made the extension service its first staffing priority; recruiting adequate agents in sufficient numbers, training them and trying them in the field, eliminating deadwood and training new recruits -- this process took the project the best of its first two years of existence. All supervision reports from October 1975 to February 1978 described the extension service as "inexperienced". Efficiency of the extension services started improving in PY3 (1977/78). Project management estimates that farmers adopting either part or all of the improved package for cotton started from a low of about 2% in 1976, and had increased to 70% in 1979. Appraisal forecasts were clearly optimistic in the assumption that 90% of the cotton growers would have adopted improved practices by 1977. According to project estimates, the number of cotton farmers reached by the project extension service since 1975/76 amounts to: - 40 - Table 5.1: EXTENSION SERVICE AND COTTON FARMERS REACHED Estimated % Progressive Total number of farmers Ext. farmers per of progressive Number of adopt-new Year Agent ext. agent farmers Cotton growers methods 1975/76 13 10 130 7,012 2 1976/77 129 10 1,290 8,372 15 1977/78 139 20 2,780 7,813 35 1978/79 140 30 4,200 7,115 60 1979/80 134 40 5,360 7,822 70 5.04 It is, however, more difficult to determine the effectiveness of extension agents in relation to foodcrops. Besides the cotton-growing cereals producers, the number of which is well known, there is little information on the other cereals producers likely to have adopted extension agents' advice. Most assumptions are based on the adoption by farmers of better husbandry practices (i.e. early planting and weeding). Farmers having improved their cultural practices are estimated as follows, with the percentages relating to the whole farming population. Table 5.2: FOODCROP AND COTTON FARMERS REACHED Progressive Number of Number of Number of farmers Percentage Year ext. agents farmers foodcrops cotton Total of total 75-76 13 32,000 1,500 130 1,650 5 76-77 129 32,320 2,500 1,290 3,800 12 77-78 139 32,643 3,500 2,780 6,300 20 78-79 140 32,970 4,200 4,200 8,400 25 79-80 134 33,300 4,700 5,360 10,000 30 5.05 The impact of the extension service started to be effective on the cotton farmers after 1978, with the arrival of the expatriate specialist (para 4.08) while its influence on foodcrop farmers has been marginal; there has been no take-off in fertilizer use, and farmers are reluctant to use new varieties for "safety" reasons, as subsistence still remains their underlying goal. The traditional system emphasizes food security and high returns through mixed cropping and considerable variability in time of planting, density, etc. The recommended practices stressing on sole cropping, new varieties and rigid guidelines for time of planting, spacing and other husbandry improvements were accepted very slowly, and even then limited to improved husbandry. Use of selected improved seeds were limited to rice and yam plots. - 41 - 5.06 The project has also been instrumental in developing farmers' groups, a stage preceeding the setting-up of a cooperative; the number of cooperatives in the project area remained the same throughout the project life (six cooperatives), while the number of farmer's groups increased from 24 in PY1 to 63 in PY6. These groups are very important for the development of agricultural credit, particularly for animal traction equipment; the number of members increased from 6,088 in PY1 to about 8,000 by the project's end. - 42 - VI. ECONOMIC EVALUATION 6.01 There are reliable data on cotton production figures compiled by the national cotton marketing agency. Data on production of foodrcrops are less certain and consist of estimates worked out by the project based on various assumptions, the main one being the level of production necessary to keep the area self-sufficient (which it is). Although this approach has its value (and some of the supervision missions who examined the results agreed that these figures were indicative of production levels), these estimates, in the absence of reliable baseline data could not be used in analyzing the economic impact of the project. It was therefore necessary to better evaluate the incremental production of foodcrops as a direct result of the project's activities. 6.02 The supervision mission of December 1979 (report dated February 14, 1980) made an elaborate analysis of higher production due to improved agricul- tural techniques, the use of fertilizer and chemicals, and improved seeds. Incremental production measured in that way was used in the economic reevalua- tion of the project. To the benefits of incremental production have been added the value of labor savings resulting from the availability of the 119 wells built by the project, road users savings resulting from the construction of 660 km of rural roads, and the value of higher meat production resulting from improved animal health achieved by the project. The methodology is explained in a working paper 1/ on file, which must be read in conjunction with the supervision report mentioned above. 6.03 Project costs have been taken from project audit reports. In addition, the value of inputs paid by farmers, Government subsidies on inputs and the estimated value of the additional farm labor necessary to achieve higher production have been taken into account. 6.04 The economic analysis is based on a time span of 20 years, of which five have elapsed. For the future, the project induced incremental production in 1979 (which had about average climatic conditions) reduced by 90 percent to account for periodic droughts, have been retained. On the cost side, the estimated expenditures necessary to consolidate achieved production levels and to maintain roads and wells have been included, together with inputs (at their full economic cost) and the imputed value of incremental farm labor. Project costs and benefits are given in Annex 4, Tables 1-3. 6.05 Under the above assumption, the rate of return of the project is estimated at 16.7%. This is considerably less than the 68% estimated at appraisal. The lower rate is primarly due primarily to lower absolute increases in cotton and cereal production. 1/ Intermediary Evaluation of the First Bougouriba Agricultural Development Project - updated June 1982. - 43 - VII. BANK PERFORMANCE 7.01 ' 3 was one of the first agricultural projects appraised by the Bank in Upper Volta (1973), and it would be useful to try to identify the weaknesses in the appraisal report which led to what appears as below-par implementation. In the first place the sketchy data available at appraisal resulted in setting physical targets which soon proved to be unattainable. Even in cotton production, where relatively reliable information was col- lected, the appraisal report substantially overestimated the number of new farmers that would shift to cotton; this simply did not happen in an area where farmers give priority to foodcrops, and, apparently, venture into a new cultivation only if limatic conditions are particularly favorable, and if they feel sure that their foodcrop production is not threatened by the change. 7.02 The technical assistance offered by the SAR to the Ministry of Rural Development was not a success; the Bank knows now that you cannot expect much from an isolated effort, in one Ministry, with one technical assistant. Providing adequate technical assistance to a Government, at the right level, in sufficient strength and with a clear understanding of purposes and results expected, is one of the most difficult exercises, and we still have a long way to go. The failure of the effort under Bougouriba I should not be a surprise. 7.03 Finally, it is now obvious that more attention should be given to design. In formulating the follow-up to this first project, a number of lessons were drawn, and changes in emphasis which seemed to be desirable in the light of experience introduced. Firstly, although technical packages are now better adapted to the region, emphasis should continue on basic husbandry themes at farmer level, coupled with improved communications in both directions with research. Secondly, whilst the first project has organized in-house training and introduced the training and visit system, greater attention to training, to organization of the extension program, and the separation of commercial functions from extension has proven to be desirable. Thirdly, although much better knowledge of the project area is now avalable, it will be necessary to devote more resources to monitoring project progress. Management attention needs to be focussed on clearly set production objectives, in order to avoid the ever-present risk of attempting to tackle too much in an area with many avenues of development potential. - 44 - VIII. CONCLUSION 8.01 The Bougouriba ADP was the first phase in a long-term program to develop human and agricultural resources of the region; it was also one of the first projects of that nature, centering around small farmers in the South West. The very sketchy, and often unreliable, data on all aspects of agricul- tural production in the area resulted in appraisal estimates (and development targets) which could not be retained as working hypothesis. As a matter of fact the collection of reliable data is still to be carried-out, and is expected to be seriously addressed under the follow-up project, both through better monitoring and the use of consultants. 8.02 The project did not have as large an impact on agricultural produc- tion as originally forecasted; yet it improved productivity on cotton substan- tially and very probably helped keep the area self-sufficient in foodcrops. The influence on foodcrop farmers has been very limited and more vigorous extension activities geared mainly at improving yields are envisaged under Phase II; this improved productivity could release labor for other activities such as expanding cotton acreage. The project has been much more successful in providing the area with essential and badly-needed infrastructure: its wells and roads building programs have achieved their targets. Its institu- tion-building was positive, and the area is endowed now with a complete agricultural development organization, whose potential is expected to be fully exploited under the follow-up project. 8.03 The economic benefits achieved, based solely on incremental produc- tion that could be directly attributed to the project's impact, result in an ERR of about 17% which is reasonable for a first phase project in a backward area. The building-up of agricultural services of the kind existing now in the Bougouriba region would have been justified even if the return were not at that level. It must be stressed that, in the absence of carefully monitored data, most benefits (except for cotton) are best estimates checked and accepted by successive supervision missions. 8.04 The project has also maintained accounts in a most satisfactory manner, as vouched for by the Auditors, Coopers and Lybrand. The need for better cost-accounting and determination of unit prices has arisen, and steps have been taken to remedy that shortcoming under Phase II. Withdrawal applications have been presented in a timely fashion, and together with generally regular Government contributions, enabled the project to run its finances without constraints. Because of the slow start, and the lapse of time before full project development took place, the estimate contained in the SAR for a 4-year project span was sufficient to expand the project life to a fifth-year within the original funds. - 45 - UPPER VOLTA Annex 1 Bougouriba Agricultural Dev. Project Project Completion Report Roads Built and Maintained Roads or Track Opening & Regravelling Maintenance and Repairs Shaping (km) (ko) (km) Djipologo- PA 80 80 15 + 80 + 80 + 80 +8o + o Dano - Dissin 38 38 20 + 38 + 38 + 23 Bouroum - Nako 26 - 1/ 26 Nako-Dapola-Malba-Borml 49 49 49 + 17 + 49 + 49 + 49 Dissin - Zambo 17 17 5 + 17 + 17 + 17 Tiankoura - Diassara 56 56 56 + 56 +-56 Bouroum-Dipio-Lorop6ni 70 70 70 + 70 Lorop6ni - Kampti 20 20 20 + 20 Lonkoura - Mougu6 28 28 28 + 28 + 28 + 28 Bondigui - Wan 5 5 5 Latara - Bati6 66 66 66 Oronkua-Ponlaba-Founzaa 59 59 59 + 59 Oronkua - Indini 34 - 34 Dissin-Binvar 6 6 6 + 6 Gora - Tovor 7 - j/ 7 + 4 Nako-Jonct.Gaoua 31 31 - Dano-Gu6gubri-Nakar 30 - 2/ 17 KoVIoho - Founzan 10 - 2/ 10 Iridiaka - Boussera 14 14 TOTAUX( 646 539 1.492 1/ No need for regravelling. 2/ Spot regravelling only - complete regravelling yet to be done. UPPER VOLTA Bougouriba Agricultural Developnent Project Project Completion Report Comparison between Appraisal Estimates and Total Pro3ect costs (CFAF million) PYI 1975/76 PY2 1976/77 PY3 1977/78 PY4 1978/79 P15 Total Project EEC Financed Appraisal / Actual Appraisal Actual Appraisal Actual Appraisal Actual Actual Appraisal Actual Interim Year TOTAL 4/80-3/81 1st Phase SALARIES & OPER. EXPENSES MRD Support 10.8 1.1 21.2 3.7 23.2 3.3 25.7 2.6 - 80.9 10.7 - 10.7 ORD Headquarters 29.7 34.7 45.6 77.4 49.9 76.1 55.3 79.1 89.7 180.5 357.0 88.5 445.5 Agricultural Development 61.3 41.9 137.1 68.9 156.7 104.3 180.5 143.6 159.0 535.6 517.T 163.2 680.9 Livestock Services 15.8 3.5 24.7 8.4 25.4 16.1 27.8 15.0 18.1 93.7 61.1 14.7 75.8 Training 16.8 0.5 35.3 7.8 4o.o 6.2 44.8 2.7 1.4 136.9 18.6 1.5 20.1 Research & Seed Multiplication 24.7 0.5 48.7 2.3 51.9 7.0 56.8 7.2 10.2 182.1 27.2 8.7 35.9 Wells Program 22.6 6.8 56.1 23.2 59.4 32.5 64.2 32.6 13.3 202.3 108.6 2.8 111.2 Roads Program 6.0 17.2 44.9 134.7 47.3 106.2 51.0 120.9 101.3 149.2 480.3 41.3 521.6 Topographic Unit - 0.9 5.0 4.6 5.4 11.3 5.7 18.9 13.0 16.1 48.7 9.0 57.7 Farm Inputs 12.3 9.3 52.9 28.3 113.8 15.9 151.8 12.6 - 330.8 66.1 - 66.1 Studies 8.4 - 32.8 - 45.1 - 16.2 - - 102.5 - - - Sub-total: 208.4 116.4 04.3 359.3 618.1 378.9 679.8 435.2 406.0 2,010.6 1,695.8 329.7 2,025.5 INVESTMENTS: Buildings 170.0 29.5 67.1 ,6.4 15.4 32.0 15.0 38.1 62.5 267.5 188.5 20.1 208.6 Vehicles 41.5 42.0 17.2 23.3 - 9.3 46.0 14.1 - 104.7 88.7 - 88.7 Road Bldg. Equipment - - 52.2 170.8 4.8 - 5.2 23.3 9.3 62.2 203.4 - 203.4 Other Equipment 29,6 10.2 61.6 3.4 6.7 8.7 7.1 5.3 - 105.0 27.6 - 27.6 Sub-total: 241.1 81.7 198.1 223.9 26.9 50.0 73.3 80.8 71.8 539.4 508.2 20.1 528.3 Yearly Total: 449.5 198.1 702.4 583.2 645.0 428.9 753.1 516.0 477.8 2,550.0 2,204.0 349.8 2,553.8 1/ From Table 1, Annex 15 of SAR (inflated). 47 Annex 2 Table 2 UPPER VOLTA Bougouriba Agricultural Development Project Project Completion Report Breakdown of Expenditure by Categories (Million CFAF) 1975/76 1976/77 1977/78 1978/79 1979/80 1980/81 Salaries Administration 10.8 25.6 29.4 37.2 40.8 40.2 Agricultural Develop. 29.4 53.9 81.3 106.9 132.4 124.1 Livestock 1.0 3.2 7.2 11.3 12.2 10.5 Training 0.5 6.4 4.4 2.7 1.4 1.5 Research & Seed Mult. 0.5 2.0 5.0 5.8 5.9 5.2 42.2 91.1 127.3 163.9 192.7 181.5 Buildings 29.5 26.4 32.0 38.1 62.5 20.1 Vehicles & Equip. 52.2 26.7 18.0 19.4 - - Operating Expenses 40.0 77.4 85.7 86.3 85.7 95.1 Roads Program: Equip. - 170.8 - 23.3 9.3 - Operating Expenses 17.2 134.7 106.2 120.9 101.3 41.3 Sub-total: 17.2 305.5 106.2 144.2 110.6 41.3 Wells Program 6.8 23.2 32.5 32.6 13.3 2.8 Topographical Unit 0.9 4.6 11.3 18.9 13.0 9.0 Farm Inputs 9.3 28.3 15.9 12.6 - - Total: 198.1 583.2 428.9 516.0 477.8 349.8 UPPER VOLTA Bougouriba Agricultural Development Project Project Ccmpletion Report Comparison between Appraisal Estimates and Actual Production Cotton Acreage, Yields and Production 0 197/76 1976/77 1977/78 1978/79 1979/80 1980/81 A. Appraisal Estimates Area:ha 5,000 5,500 6,000 8,750 12,000 12,000 - Production:tons 2,000 2,684 3,798 6,300 9,756 9,756 - Estimated average yields (kg/ha) (4oo) (488) (633) (720) (813) (813) - B. Estimated acutal results Area:ha 4,859 7,016 7,208 6,000 7,036 6,835 Production:tons 1,837 2,483 1,802 3,412 4,679 4,050 Estimated average yields (kg/ha) (378) (354) (249) (568) (665) (592) Estimated yearly increase 134.2% 141.5% 166% 154.9% - - Estimated overall iicrease - 190% 315% 488% - - Actual yearly increase - 135% 72.6% 189.3% 137% 86.6% Actual overall increase - - 98% 185.7% 254.7% 220.5% P(aX F, ~ UPPLR VOLTA Bougouriba Agricultural Development Project Project Completion Report Comparison between Appraisal Estimates and Actual Production PYO PY1 PY2 PY3 PY4 PY5 P6 Acreage, Yields and Production (1974/75) 197/76 1976/77 1977/78 1978/79 1979/80 1980/81 A. Appraisal Estimates Sorghum and Millet Area:ha 141,000 141,000 140,750 140,000 138,500 138,500 138,500 Production:tons 98,700 98,700 99,800 99,900 100,000 -100,000 100,000 Estimated yields (kg/ha) (700) (700) (709) (714) (722) (722) (722) Maize Area:ha 25,000 25,000 25,250 26,000 27,500 .27,500 27,500 Production:tons 17,500 17,500 18,000 19,500 22,500 22,500 22,500 Estimated yields (kg/ha) (700) (700) (713) (750) (818) (918) (818) Total all Cereals Area:ha 166,000 166,000 166,000 166,000 166,000 166,000 166,000 Production:tons 116,200 116,200 117,800 119,400 122,500 122,500 122,500 Estimated average yields (kg/ha) (700) (700) (710) (719) (738) (738) (738) Estimated yearly progress 101.4% 101.4% 102.6% 102.6% 102.6% Estimated overall progress 102.8% 105.4% 105.4% 105.4% B. Estimated actual results Sorgnum and Millet Area:ha 131,474 132,080 133,288 133,269 i32,542 132,126 Production:tons 67,768 63,794 67,443 69,966 71,572 70,160 Estimated Yields (kg/ha) (515) (483) (506) (525) (540) (531) Maize Area:ha 7,696 7,729 7,800 7,802 8,510 9,237 Production:tons 5,664 5,317 5,585 5,812 6,501 6,955 Estimated yields (kg/ha) (736) (688) (716) (745) (764) (753) Total all Cereals Area:ha 139,170 139,809 141,088 141,071 141,052 141,363 Production:tons 73,432 69,111 73,028 75,778 78,063 77,115 Estimated average yields(kg/ha) (527) (494) (517) (537) (553) (545) Estimated yearly progress 94% 105.7% 103.8% 103% 99% Estimated overall progress 99.5% 103.2% 106.3% 105% UPPER VOLTA FIRST BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT INCREMENTAL CROP PRODUCTIONY ECONOMIC PRICES AND VALUES (QUANTITIES IN METRIC TONS) (PRICES IN 1,000 CFA FRANCS) (VALUES IN MILLIONS OF CFA FRANCS) YEAR 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 986-1994 QUANTITIES COTTON 24.0 1043.0 575.0 1444.0 2696.0 2210.0 2210.0 2210.0 2210.0 2210.0 2210.0 2210.0 GROUNDNUTS (UNSHELLED) 0.0 104.0 58.0 145.0 273.0 224.0 22440 224.0 224.0 224.0 224.0 224.0 RICE (PADDY) 0.0 22.0 12.0 31.0 57.0 47.0 47.0 47.0 47.0 47.0 47.0 47.0 CEREALS 0.0 1291.0 717.0 1793.0 3371.0 2764.0 2764.0 2764.0 2764.0 2764.0 2764.0 2764.0 0 ECONOMIC PRICES COTTON 110.5 176.6 154.0 110.5 102,4 118.6 109.6 97.8 109.6 115.9 127.7 127.7 GROUNDNUTS (UNSHELLED) 0.0 67.8 68.3 68.7 69.1 69.4 69.8 69.9 70.1 70.4 71.1 71.1 RICE (PADDY) 0.0 105.1 79.9 79.1 86.9 74.7 84.2 94.5 68.1 70.6 77.1 83.9 CEREALS 0.0 84.1 73.5 70.0 71.0 73.5 74.6 68.1 68.4 71.3 75.1 75.1 VALUE OF INCR. PROD. COTTON 2.7 184.2 88.6 159.6 276.1 262.1 242.2 216.1 242.2 256.1 282.2 282.2 GROUNDNUTS (UNSHELLED) 0.0 7.1 4.0 10.0 18.9 15.5 15.6 15.7 15.7 15.8 15.9 15.9 RICE (PADDY) 0#0 2.3 1.0 2.5 4.9 3.5 4.0 4.4 3.2 3.3 3.6 3.9 CEREALS 0.0 108.6 52.7 125.5 239.3 203.2 206.2 188.2 189.1 197.1 207.6 207.6 TOTAL EC. VALUE 2.7 302.1 146.2 297.5 539.2 484.3 468.0 424.5 450.2 472.3 509.3 509.7 UPPER VOLTA FIRST BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT WELL CONSTRUCTION AND ROAD BENEFITS (MILLIONS OF CFA FRANCS UNLESS OTHERWISE INDICATED) YEAR 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 986-1994 (1) WELL BENEFITS 0.00 17.80 44.40 88.70 132.00 132.00 132.00 132.00 132.00 132.00 132.00 132.00 (2) ROAD BENEFITS 0.00 38.80 77.60 116.40 155.20 194.00 194.00 194.00 194.00 194.00 194.00 194.00 (3) WELL+ROAD BENEFITS 0.00 56.60 122.00 205.10 287.20 326.00 326.00 326.00 326.00 326.00 326.00 326.00 (4) TOTAL BENEFITS 2.65 358.73 300.87 584.09 914.02 880.32 843.20 807.17 832.88 855.00 892.04 892.36 (5) .... (1) AS % OF (4) 0.00 0.05 0.15 0.16 0.16 0.17 0.17 0.16 0.16 0.15 0.15 0.15 (6) .... (2) AS % OF (4) 0.00 0.11 0.26 0.21 0.19 0.25 0.26 0.24 0.23 0.23 0.22 0.22 (7) .... (3) AS % OF (4) 0.00 0.16 0.41 0.36 0.34 0.42 0.43 0.40 0.39 0.38 0.37 0.37 UPPER VOLTA FIRST BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT PROJECT COST AND BENEFITS (MILLIONS OF CFA FRANCS) YEAR 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 986-1994 BENEFITS INCR. COTTON PROD. 2.7 184.2 88.6 159.6 276.1 262.1 242.2 216.1 242.2 256.1 282.2 282.2 INCR. GROUNDNUT PROD. 0.0 7.1 4.0 10.0 18.9 15.5 15.6 15.7 15.7 15.8 15.9 15.9 INCR. RICE PROD. 0.0 2.3 1.0 2.5 4.9 3.5 4.0 4.4 3.2 3.3 3.6 3.9 INCR. CEREALS PROD. 0.0 108.6 52.7 125.5 239.3 203.2 206.2 188.2 189.1 197.1 207.6 207.6 INCR. BEEF PROD. 0.0 0.0 32.7 81.5 87.6 70.0 49.2 56.7 56.7 56.7 56.7 56.7 WELL CONSTR. BENEFITS 0.0 17.8 44.4 88.7 132.0 132+0 132.0 132.0 132.0 132.0 132.0 132'.0 t ROAD USER SAVINGS 0.0 38.8 77.6 116.4 155.2 194.0 194.0 194.0 194.0 194.0 194.0 194.0 TOTAL PROJECT BENEFITS 2.7 358.7 300.9 584,1 914.0 880.3 843.2 807.2 832.9 855.0 892.0 892.4 COST SALARIES 69.1 142.8 175.1 199.6 220.2 184.1 59.0 59.0 59.0 59.0 59.0 59.0 EQUIPMENT 96.6 345.9 69.5 91.7 25,8 2.8 10.0 10.0 10.0 10.0 10.0 10.0 OPERATING 93*6 332.4 264.0 252.4 213.7 138.3 30.0 30.0 30.0 30.0 30.0 30.0 CIVIL WORKS 48.3 41.4 44+0 46.4 71.4 20.4 10.0 10.0 10.0 10.0 10.0 10.0 MISCELLANEOUS EXP. 1.5 7.2 15.5 23.0 14.9 9.1 1.0 1.0 1.0 1.0 1.0 1.0 IMPUTED LABOR COSTS 0.8 102.7 50.7 989 161.9 132.8 135.1 151.9 159.2 166.4 176.8 176.8 INPUT SUBSIDIES 84.2 148.3 97.3 54.0 85.1 75.0 75.0 75.0 75.0 75.0 75.0 75.0 TOTAL PROJECT COST 394.1 1120.7 716.1 766.0 793.0 562.5 320.1 336.9 344.2 351.4 361.8 361.8 m Aw 4 -S3- Annex 5 UPPER VOLTA DOUCOURIBA AGRICULTURAL DEVELOPMENT PROJECT - CREDIT 496-UV PROJECT COMPLETION REPORT Key Indicators for PY5 (1979/80) (Note: For detailed discussion of project results, see Annex 7 and associated tables of Supervision Report of February 14, 1980.) Indicator Unit Appraisal Estimate Percentage Area: Cotton ha 12,000 7,035 1/ 58 Sorghum, Millets ha 138,500) 123,876 75 Maize ha 27,500) - - Groundnuts ha 20,000 5,456 27 Rice ha 3,050 3,686 121 Yams ha 3,000 3,312 110 Yields: Cotton kg/ha 813 665 82 Sorghum kg/ha 722 540 75 Maize kg/ha 818 764 93 Groundnuts kg/ha 462 500 108 Rice kg/ha - 800 - Yams kg/ha 9,000 6,000 66 Production: Cotton tons 9,756 4,679 48 Sorghum, Millets tons 190,000) 71,572 58 Maize tons 22,500) - - Groundnuts. tons 9,247 2,728 29 Rice tons 3,017 2,949 97 Yams tons - 19,872 - Inputs: Cotton fertilizer tons 480 483 100 Urea tons 80 - - Cotton treated ha 10,800 2,588 Improved seed used tons - 58.7 - Infrastructure: Second roads constructed Km 200 ) Feeder roads constructed Km 460 646 98 Wells constructed No. 120 119 100 1/ Project entimates include all cereals. 2/ Treated three times, which is a minimum. 54 Annex 6 UPPER VOLTA BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT - CREDIT 496-UV PROJECT COMPLETION REPORT Project Organization and Objectives Appraisal Actual Project Objectives or Focus Increased Acreage - Cotton 100% 59% - Cereals 100% 96% Increased Yields - Cotton 100% 82% - Cereals 100% 75% Technological Improvement 100% 25% Integrated Rural Development: Rural Roads 100% 98% Village Water Supply 100% 100% Project Organization and Performance Project Unit Just Established Adequate Rural Development Works (HER) Existing Adequate Ministry of Rural Development Existing Adequate Mixed Economy Corporation (SOFITEX Cotton Co.) Existing Good Institution Building Focus of Effort 100% 100% Means - Financial 100% 100% - Qualified Manpower 100% 70% Achievement of Objective 100% 80% IBRD l 728 1 (PCR) UNE 1983 4-UPPER VOLTA J915 HA U TE- VOLTA FIRST BOUGOURIBA AGRICULTURAL DEVELOPMENT PROJECT PREMI/ER PROJET DE DEVE/OPPEMENT AGR/OZE DE / A BOUGOUR/BA FOUNZAN o- K-n Hb 000 Isohyets rn millimeters /soh/yetes en mi//imetres K y'' ORD headquarters ORONKU Chef -/ieu del'ORD ssee Sorkon Sector offices Chef-lieu de secteur O po a Sub-seCtor offices Nakar Chef-/ieu de sous-secYeur O DANO M Supply and credit centers néd.ougou Cenfres approvisionnement et crédit Y ORD boundaries 3PR _, Limite de /'ORD Zanuoua - Sector bounderies DIEBOUGu Limite de secteur r. IVe - DuSSss National roads BAMAKO Routes nationa/es Secondory roads by year5 of project Roufes secondaires en nnée cinq du pro/et \ Rivers zM Riv,/res TIANKOURA International boLndaries IRONIORO Fronfières internationa/es m Diepela. Tn m BOUROUM- °'^OL G H A N A - inoo BOUROUM Komo ä KILOMETERS 0 10 20 30 40 50 MILES 0 0 20 30 Yéri YantanaMALBA T,kera ÅbBatié-Nord Lokosso LOROPEN1 ar Nb GbAngR PERIGBAN Doudou GBOMBOLORA s o o\ Tobo Gbangbaukoram P,,ena KMPT 0 Ingk- ova LEGMOIN- Brifora ?go Dan D igou .. Passene -10° GAGLL MDEBDO0° THIS pCOR MAP.1S BASED ON IBFID 1512719 IMouloupo GLOl NOVEMBER 1980. SECOND BOUGOURIBA ' AGR ICULTURAL DEVELOPMIENT PROJECT - INFORMATION HAS BEEN OMITTED. KpaVTr Kosso- To/ver BENno *~ mo BATIEUSB Douafor. l V O R Y C O A S T M A L l N IG E R C0GERI D V UPPER VOLTA HUEVOITA - ./ .. KPERE ç.. _, ....,.... Fadiq/ BOG URå B E N IN.......... G H -NA OGO (VORY COAST TON 1.GEr A 3 COTE D'/V0/RErli