Mali - Agricultural Projects

Document of The World Bank FOR OFFICL4L USE ONLY Repwt No. 9688 PROJECT PERFORMANCE AUDIT REPORT MALI LIVESTOCK PROJECT (CREDIT 538-MLI) MALI SUD AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 669-MLI) MOPTI RICE II PROJECT (CREDIT 753-MLI) ODIPAC TECHNICAL ASSISTANCE PROJECT (CREDIT 1174-MLI) SECOND MALI SUD RURAL DEVELOPMENT PROJECT (CREDIT 1415-MLI) JUNE 21, 1991 Operations Evaluation Department This ? -ument has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY US$ - FCFA GLOSSARY OF ABBREVIATIONS ADF - African Development Fund AP - Association Pastorale-Pastoral Association AV - Association Villagoise-Village Association BADEA - Banque Arabe pour le Developpement Economique en Afrique BNDA - Banque Nationale pour 16 Developpement Agricole CCCE - Caisse Centrale de Cooperation Economique CBPP - Contagious Bovine Pleuropneumonia CMDT - Compagnie Malienne pour le Developpemert d(; Fibres Textiles ERR - Economic Rate of Return FAC - Fonds d'Aide et de Cooperation FRR - Financial Rate of Return GDP - Gross Domestic Product GNP - Gross National Product IER - Institut d'Economie Rural IFAD - International Fund for Agricultural Development ILCA - International Livestock Center for Africa IRCT - Institut de Rechc-rches du Coton et des Textiles Exotiques IRAT Institut de Recherches d'Agriculture Tropicale ODEM - Operation de Developpement de L'Elevage de la Region de Mopti ODIPAC - Operation de Developpement Integre des Productions Arachiiere et Cerealiere OED - Operations Evaluation Department OPAM - Office des Produits Agricoles du Mali ORM - Operation Riz Mopti OSRP - Office pour la Stabilisation et pour la Regularisation des Prix PCR - Project Completion Report PPAR - Project Performance Audit Report SCAER - Societe de Credit Agrico's et d'Equipement Rural SOMIEX - Societe Malienne d'Import Export ULV - Ultra Low Volume THE WORLD BANK FOR * USE ONLY Washington, D.C. 20433 U.S.A. Office f Dioectwor-Geea June 21, 1991 iBNORANDUM TO THE EXECUTIVE DIRECTORS AND -M PRESIDENT SUBJECT: Project Performance Audit Report on Mali Livestock Project (Credit 538-MLI), Mali Sud Agricultural Development Project (Credit 669-MLI), Mopti Rice II Project (Credit 753-MLI), ODIPAC Technical Assistance Project (Credit 1174-MLI) and Second Mali Sud Rural Development Project (Credit 1415-EMI) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Mali Livestock Project (Credit 538-MLI, Mali Sud Agricultural Development Project (Credit 669-MLI), Hopti Rice II Project (Credit 753-HLI), ODIPAC Technical Assistance Project (Credit 1174-MLI) and Second Mali Sud Rural Development Project (Credit 1415-MLI)" prepared by the Operations Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only In the performance o' . eir official duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PRO-ECT PERFORMANCE AUDIT REPORT LIVESTOCK PROJECT (CREDIT 538-MLI) MALI SUD AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 669-1LI) MOPTI RICE II PROJECT (CREDIT 753-MLI) ODIPAC TECHNICAL ASSISTANCE PROJECT (CREDIT 1174-MLI) SECOND MALI SUD RURAL DEVELOPMENT PROJECT (CREDIT 1415-MLI) TABLE OF CONTENTS Paie No. PREFACE ...................................... . ........... ± BASIC DATA SHEETS .......................................... iii EVALUATION SUMMARY ........................................... xiii PROJECT PERFORMANCE AUDIT REPORT I. BACKGROUND ......... .............................. I II. THE PROJECTS .......................................... 2 A. Livestock Project (Credit 538-MLI) ................. 2 B. Mali Sud Agricultural Development Project (Credit 669-MLI) ................................. 4 C. Mopti Rice II Project (Credit 753-MLI) ............. 7 D. ODIPAC Technical Assistance Project (Credit 1174-MLI) ................................ 9 E. Second Mali Sud Rural Development Project (Credit 1415-MLI) ................................ 11 III. MAIN ISSUES .......................................... 13 A . General ............................................ 13 B. Dealing with the Sector Constraints ................ 15 C. Aspects of Pricing Policies ........................ 22 D. Fostering Self-Help ................................ 27 E. Bank Reporting and Filing .......................... 29 Attachment I - Comments from the Ministry of Planning and International Cooperation.................... 31 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. POJECT PERFORMANCF IUDIT REPORT MALI LIVESTOCK PROJECT (CREDIT 538-MLI) MALI SUD AGRICULTURAL DEVELO"4ENT PROJECT (CREDIT 669-MLI) MOPTI RICE II (CREDIT 753-MLI) ODIPAC TECHNICAL ASSISTANCE PROJECT (CREDIT 1174-MLI) SECOND MALI SUD RURAL DEVELOPMENT PROJECT (CREDIT 1415-MLI) PREFACE This is a Project Performance Audit Report (PPAR) of the Livestock Project (Credit 538-MLI) involving a credit in the total amount of US$13.3 million; the Mali Sud Agricultural Development Project (Credit 669-MLI) with a total credit amount of US$15.5 million; the Mopti Rice II Project (Credit 753-MLI) with a credit amounting to US$15.0 million; the ODIPAC Technical Assistance Project (Credit 1174-MLI) involving a credit of US$6.5 million; and the Second Mali Sud Rural Development Project (Credit 1415- MLI) with a credit of US$26.8 million. Credit 538-MLI was approved in February 1975 and closed after three extensions of the Closing Date, fully disbursed on June 30, 1984. Credit 669-MLI was approved in December 1976 and closed on December 31, 1984 after an extension of the Closing Date. An undisbursed amount of US$ 441,413 was cancelled. Credit 753-MLI was approved in November 1977 and closed on June 30, 1985 after an extension of the Closing Date and an undisbursed amount of US$64,000 was cancelled. Credit 1174-MLI was approved in April 1982 and closed after an extension of the Closing Date, fully disbursed on September 30, 1986. Credit 1415-MLI was approved in October 1983 and will close, fully disbursed after two extensions of the Closing Date, on September 30, 1990. The PPAR is based on five Project Completion Reports (PCR) issued in 1987, 1988, 1987, 1988 and 19901 respectively by the Sahelian Department/Africa Regional Office, the Staff Appraisal Reports, the President's Reports, Credit documents, and a review of project files. Staff involved with the projects were interviewed. An OED mission visited Mali in April/May 1990. The mission had discussions with officials of the Ministries of Agriculture, Livestock, and Planning, and agencies connected with the projects. Field trips to meet participating farmers and herdsmen were undertaken. The kind cooperation and valuable assistance given by all persons met is gratefully acknowledged. 1/ Report numberst 6846 dated June 25, 1987; 7324 dated June 28, 1988; 6799 dated May 29, 1987; 7565 dated December 23, 1988; and 9270 dated December 28, 1990. The PCRs provide a fair account and assessment of project experience, and discuss the performance of the Bank and project executing agencies. This PPAR, prepared several years after completion of some of the projects, re-evaluates project outcome and assesses the Bank's sectoral approach, aspects of the pricing policies, farmers' self-help efforts and Bank administrative procedures. The draft PPAR was sent to the Borrower for comments. The comments received from the Ministry of Planning and International Cooperation are reproduced as Attachment I to the PPAR. - M1 - PROJECT PERFORMANCE AUDIT REPORT LIVESTOCK PROJECT (CREDIT 538-MI) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as X of item estimate Current Estimate Appraisal estimate Total Project Costs (US$ million) 17.3 19.0 110 Credit Amount (US$ million) 13.3 13.3 100 Economic rate of return X 51 n.s. Institutional performance - Satisfactory CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS EY76 FY77 EZ8 E172 E0 EMA 2 EYI FY84 I Appraisal esti-ate (US$ million) 4.7 7.R 10.4 11.8 13.3 Actual (US$ million) 0.2 1.1 3.9 5.9 7.9 9.6 11.5 12.4 12.8 13.S Actual as X of Appraisal (%) 4.0 14.0 37.0 50.0 59.0 74.0 86.0 93.0 96.0 100.0 Date of Final Disbursement: August 20, 1985 PROJECT DATES Original Plan Actual Appraisal 02174 Negotiations Board Approval 02/75 Credit Agreement Date 04175 Effectiveness 07/75 07/75 Date physical components completed n.a. Proportion completed by that date (2) n.a. Closing Date 12/79 06/84 STAF INPUTS (staff veeks) FY74 Y75 FY76 FY77 FY78 FY79 FY80 FY81 ve82 a Y83 a Y4FY85 U.81 FY Total Preappraisal 7.3 7.3 Appraisal 34.7 17.3 51.9 Negotiation 9.1 9.1 Supervision 4.1 25.4 21.4 25.4 24.4 27.8 22.0 e.9 9.7 5.4 1.0 5.6 2.9 184.2 Other .9 18.9 19.9 Total 42.9 49.5 25.4 21.4 25.4 24.4 27.8 22.0 8.9 9.7 5.4 1.0 5.6 2.9 272.4 2M No. of Spectalisations Performance Types of (molyr) RSEAs Reoroented RALtnza Tend Pnobos La Lk La IdentifLoationlPreparation 10170 Appraisal 02/74 6 Supervision 1 04/75 1 A 2 1 H Supervision 2 09/75 1 B 2 1 T Supervision 3 02/76 1 D - - Supervision 4 03/76 1 B - - Supervision 5 06176 2 A,B 2 1 T,M Supervision 6 11/76 2 A,B 2 1 F,T Supervision 7 05/77 3 2A+1C 2 2 T,M Supervision 8 08/77 1 A 2 2 F,T Supervision 9 11/77 2 A,D - - Supervision 10 12/77 1 A,C - - T,M,P SupArvision 11 03/78 2 A,D 3 3 M,F,T Supervision 12 07178 1 D - - Supervision 13 10178 2 A,D 3 3 M,F,T Supervis.on 14 03/79 1 D 3 1 M,T,O Supervision 15 10/79 1 D 3 2 M,T,O Supervision 16 02/80 2 D,D - - Supervision 17 06/80 3 D,D,C 3 2 M,T,P Supervision 18 10/80 2 D,C 2 1 T Supervision 19 01/81 1 D - - Supervision 20 04/81 1 D 2 1 F,T Supervision 21 10181 1 A 2 2 F,T Supervision 22 06/82 2 D,A 2 1 F Supervision 23 04/84 1 A 2 1 F OTHER PROJECT DATA Borr.der: Republic of Mali Executing Agencies: Operation do Developpement de l'Elevage do Ia Region de Mopti (ODEM) La A - Agriculturalisto B * Agricultural Economists C - Financial Analysts D - Technical Specialist. Lk 1 - no or minor problemso 2 - moderate problemst 3 * major problems. La 1 - Lmprovingl 2 - stationary; 3 - deteriorating. L_ F - Financial; M = Managerialt T = Technical; P - Politicals 0 - Other. *V* PROJECT PERFORMANCE AUDIT REPORT MUI NALI SUD AGRICULTURAL DEVELOPMENT PROJECT (CREDIT .669-MLE) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as X of aS *stLmat. Current Estima. Aoraeal estimate Total Project Costs (US$ million) 44.6 42.7 96 Credit Amount (US$ million) 15.5 15.5 100 Economic rate of return X 44 37.5 85 Institutional performance problem free or minor problems CUMULATIVE ESTIMATED AND ACTUAL DISBUR§EMENTS Lp EL77 EY78 ZY79 X80 FY81 EY8 ZX81 EYS4 E1 FY86 Appraisal estimate (US$ million) 0.5 4.0 10.5 13.0 14.5 15.5 Actual (S$ million) - 1.2 1.9 4.3 7.4 - 12.8 - 14.5 15.1 Actual as X of Appraisal (2) - 30 18 33 51 Date of Final Disbursements September 17, 1985 PROJECT DATES Original Plan Actual Appraisal 11/75 Negotiations 10/76 Board Approval 12/76 Credit Agreement Date 12/76 Effectiveness 09/77 Date physical components completed 06/80 10/83 Proportion completed by that date (%) Closing Date 12/81 12/84 STAFF INPUTS (staff weeks) EYZ EIA EZ77 EZ FY79 Y0 8A HY8 I.Y83 FY84 M5 Y86 FY87 EY88 Total Preappraisal 5.1 33.9 - - - - - - - - - - - - 39.0 Appraisal - 91.2 8.0 - - - - - - - - - - - 99.2 Negotiations - - 9.7 - - - - - - - - - 9.7 Supervision - - 1.6 16.6 8.9 8.8 13.0 .8 4.7 1.1 2.5 1.0 6.8 1.9 67.7 Other .8 7.5 9.0 - - - - - - - - - - - 17.3 Total 5.8 13z.6 28.2 16.6 8.9 8.8 13.0 .8 4.7 1.1 2.5 1.0 6.8 1.9 232.8 * Vi7 * ISSiJN DAES VArA No. of Specialisations Performance Types of (molyr) perA Reoresented Rating ITrond Pohlmm Lk Le LA Li Identifteation/Preparation 07175 1 E Appraisal 11/75 6 E,F,A,L Reappraisal 03/76 1 E Supervision 1 11/77 3 ?,A 2 2 Supervision 2 11/78 2 V,A 2 2 Supervision 3 02179 2 F,A 1 1 Supervision 4 02180 1 A 2 2 F Supervision 5 02/81 1 F,A 2 2 Supervision 6 10/82 1 F 2 2 F Supervision 7 I 05/83 1 L 2 2 F Completion 0/87 OTUER PROJECT DATA Borrower: Republic of Mali Executing Agencies Compagnie Malienne pour le Developpement dex Textiles (CMDT) Follow-on Projects Name Mali Sud II Credit Number 1415-MLI Amount (US$ million) 25.9 Approval Date October 11, 1983 LA Data no longer available from Disbursements Section. This (incomplete) data called from supervision reports. Ib E - Economists L - Livestock specialists F w Financial analysti A - Agriculturalist. Lg 1 - no or minor problemss 2 - moderate problems1 3 - major problems. LA 1 - Lmprovings 2 - stationaryo 3 - deteriorating. La F - Financiale M - Managerials T - Technical; P - Political. I SPN 7 covered livestock component only. - Vii - PROJECT PERFORMANCE AUDIT REPORT MOFTI II RICE PROJECT (CREDIT 753-MLI) BASIC DATA SHEET KEE PROJECT DATA Appraisal Actual or Actual as I of U12S *ftimate Current Estimate Aporaisal estimate Total Project Costs (US$ million) 29.9 31.5 105 Credit Amount (US$ million) 15.0 Economic rate of return (M) i 3 Institutional performance Borrower good poor Bank poor CUMULATIVE ESTIMATED AND ACTJAL DISBURSEMENTS EMZ FY8 MX~ Y8 E[M3 EISAfl EMf2 Appraisal estimate (US$ million) 1.2 4.4 8.7 12.7 14.5 15.0 Actual (US$ million) 2.9 7.2 11.0 14.2 14.4 14.3 14.7 Actual as X of Appraisal (Z) 242 164 126 112 100 97 98 Date of Final Disbursements July 30, 1985 PROJECT DATE Original ..Pdln Act-UAL Appraisal 11176 Negotiations 10103-06/77 Board Approval 11/29177 Credit Agreement Date 12/08/77 Effectiveness 05/15/78 Date physical components completed 12.81 08181 Proportion completed by that date (X) 100 100 Closing Date 06/30185 STAF INPUTS (stAff weeks) EL7 Ul76 =1 EZ M M1282 EM8 EMA EMB PY84 MA M I"ss Preappraisal 14.3 3.8 26.9 .7 - - - - - - * * 45.8 Appraisal - - 38.5 1.2 - - - - - - - 39.6 Negotiations - - - 6.6 - - - - - - - - 6.6 Supervision - - - 7.3 6.6 10.3 9.4 25.7 7.0 4.9 6.2 5.4 82.8 Other 1.0 .1 .7 6.5 - - - - - - - - 8.4 Total 15.3 3.9 66.2 22.2 6.6 10.3 9.4 25.7 7.0 4.9 6.2 5.4 183.1 - viii - MISSION DATES RJU No. of Specialisations Performance Types of (melyr) v!EAsgs Represented Ratin . Ta d rblems La La kA LA Identification 11/74 n.a. n.a. - - Preparation 1975 - n.a. - - - Apprasal 11/76 n.a. Supervisiol, 1 11/78 n.a. a,b 1 - - Supervision 2 05/79 n.a. a 1 1 - Supervision 3 11/79 n.a. a,b,c 2 2 F,T Supervision 4 08180 n.a. a i 2 P,T Supervision 5 02/81 n.a. a 2 2 F,T Supervision 6 07/81 n.a. &,a 2 3 F,T Supervision 7 11/81 n.a. o n.a. n.s. n.a. Supervision 8 02/82 n.a. &c,.d 2 1 F,T Supervision 9 09/82 n.a. a,b,b,b,c,c,e 2 2 F,T Supervision 10 03/83 n.a. a 2 1 F,T Super,ision 11 03/84 n.a. o 2 1 F,T OTHER PROJECT DATA Borrowers Republic of MatL Executing Agencys Operation Ris Mopti (ORM) Follow-on Projects Hame Mopti Area Development Credit Number 397-MLI Amount (US$ million) 19.5 Approval Date May 21, 1985 La a - Irrigation engineers b - Economist; a Agronomist; d * Financial analysts a Livestock specialist. Lh 1 - no or minor problemso 2 - moderate problemas 3 - major problems. Lg 1 = improvingo 2 - stationarys 3 - deteriorating. LJ F - financialt T - Technical. - LX * PROJECT PERFORMANCE AUDIT REPORT MALI ODIPAC TECHNICAL ASSISTANCE PROJECT (CREDIT 1174-MLI) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as % of iem estimate Current Estimate Appraisal estimate Total Project Costs (US$ million) 11.4 11.4(est.) Credit Amount (US$ million) 6.5 6.5 Economic rate of return (%) n... n.a. Institutional performance - poor CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS M81 EY83 FY84 FY85 FY86 FY87 Appraisal estimate (US$ million) 1.0 2.5 4.5 6.5 - - Actual (US$ mliton) 1.4 2.0 3.1 4.0 5.4 6.5 Actual as X of Appraisal (%) 104" 80 69 62 83 100 PROJECT DATES Original .Plan. Actul FL-:t Mention in Files 05176 05/76 Negotiations 05/81 05/81 Board Approval 09/81 04/82 Credit Agreement Date 04/82 Effectiveness - 07182 Date physical components completed 03/31/85 02/87 Proportion completed by that date (X) 100 70 Closing Date 09/30/85 09130/86 STAFF INPUTS (staff weeks) EXR EH1 Us EH8. EXA E4Y85 Y86 EY87 Fe Total Preappraisal 4.6 - - - - - - - - 4.6 Appraisal 6.0 17.1 - - - - - - - 23.1 Negotiations 3.2 .8 - - - - - - 3.9 Supervision - - 14.6 15.0 19.1 14.8 15.1 9.5 10.8 98.8 Other 1.1 10.2 2.3 - - - - - - 12.6 Total 11 7~ 15. 66 L0 19A 1 Af 14 8 S1 10.8 143,1 *R- MlSSION DATES Pag No. of Specialisations Performance Types of (molyr) oxons kepresented Rating Trend Proble LA Identification/Preparation - - n.a. n.a. n.a. Appraisal Lq Post-appraisal Supervision 1 02182 3 LA Supervision 2 09182 3 EIF,A Supervision 3 10/83 2 E,A Supervision 4 09184 3 E.A,? Supervision 5 11/85 2 E,A, Supervision 6 02/86 1 A OTBER PROJECT DATA Borrover: Republic of Mali Executing Agency: Office de Developpement Integre des Productions Aracludieres at Cerealieres (ODIPAC) Follow-on Projectt None JA Includes repayment of PPF. Lk Preparation was spread over several years, reflecting changing project objectives. LA Original full scale project March 1978, subsequently converted to TA project for which there is a President's Report only, July 1981. g E - Economists A - Agronomists P - Financial Analyst. * Kl. - PROJECT PERFORMANCE AUDIT REPORT SECOND MALI SUD RURAL DEVELOPMENT PROJECT (CREDIT 1415-MLI) PASIC-AA-JMlSI KEY PROJECT DATA Appraisal Actual or Actual as X of itm estimate Current Estimate Appraisal estimate Total Project Costs (US$ million) 84.0 90.2 Credit Amount (US$ million) 25.9 29.8 Cofinancing-Total (US$ million) 58.1 60.4 IFAD 13.0 15.3 CCCE 11.3 11.3 FAC 4.2 4.2 FRANCE 2.2 2.2 NETHERLANDS 4.1 4.1 BNDA 2.7 2.7 BENEFICIARIES 11.5 11.5 GOVERNMENT 9.1 9.1 Economic rate of return (%) 34 27 Institutional performance - satisfactory CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS CY8 CY85 CY86 067C CY89 2M Appraisal estimate (US$ million) 4.7 10.0 15.0 19.5 24.1 24.3 24.3 Actual (US$ million) 3.0 4.7 15.4 21.8 25.5 27.9 28.2 Actual as 2 of Appraisal (%) 63.8 .0 102.6 112.0 105.8 114.7 115.9 Date of Final Disbursement: June 22, 1990 PROJECT DATES Original Revisions Actual Appraisal 09128/81 09/28/81 Negotiations 06/27/83 06/27/83 Board Approval 10/11/83 10/11/83 Credit Agreement Date 01/23/84 01/!3/84 Effectiveness 08/09/84 08/09/84 Date physical components completed 09/30/88 03/30/90 Proportion completed by that date (%) Closing Date 03/31/89 09130/90 j1 * Kcii STAFF IKUTS (staff weeks) EY77 EY78 EX79 P.8Q EL1 F.1Y. El8i EY84 FY85 EY86 F87 EL.k EI8_9 EX2 EIS Total Preparation 3.2 - - - 3.6 12.4 - - - - - - 8.8 3.8 - 31.8 Appraisal - - - - - 129.0 29.5 - - - - * - - - 158.5 Negotiations - - - - - - 3.1 5.4 - - - - - - - 8.5 Supervision - - - - - - - 18.6 32.7 30.9 16.9 20.2 14.0 11.0 5.0 149.4 Other .6 .3 - - .3 6.8 7.5 4.3 - - .7 .5 - - - 21.0 Total 3. 4.9 148.3 40.0 28.3 32.7 30.9 17.6 20.7 22.9 14.8 5.0 369.2 ISSION DATES bLe No. of Specializations Performance Types of (molyr) RezspA Represented Rata TrEd Problems LLA Identification 01105/81 Preparation 1911/1982 Appraisal 09/28/81 Through appraisal 12/83 1 E 1 Appraisal through Board Approval 12/84 6 E,A 1 Supervision 1 10/85 6 A 2 n.a. F Supervision 2 02/86 1 2 F Supervision 3 11/86 1 E 2 F Supervision 4 03/87 1 E 2 F Supervision 5 09/87 1 E 2 F Supervision 6 11/87 4 E,A 1 Supervision 7 05/88 - - 1 Supervision 8 11/88 8 EIC 1 OTHER PROJECT DATA Borrower: Republic of Mali Executing Agency: Compagnie Malienne pour le Developpement des Textiles (CMDT) Follow-on Project: None Lg Project officially closed on 09/30/90, but no further disbursement is expected. L E - Economistj A - Agronomist; C - Consultant. o I - no or minor problemsi 2 - moderate problemsl 3 - major problems. Id F - Financial. -xiii - PROJECT PERFORMANCE AMIT REPOR HAL1 LIVESTOCK PROJECT (CREDIT 538-MLI) MALI SUD AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 669-MLI) MOPTI RICE II (CREDIT 753-MLI) ODIPAC TECHNICAL ASSISTANCE PROJECT (CREDIT 1174-MLI) SECOND MALI SUD RURAL DEVELOPMENT PROJECT (CREDIT 1415-LI) EVALUATION SUMMARY Introduction and through reorganizing the budgeting, taxation, marketing and 1. The projects supported cost recovery system of the cotton government efforts to improve complex, the "filiere cotton". agricultural production for domestic consumption as well as export, and 4. It was assumed during one project also combined financing appraisal that due to project for a major policy reform program. promoted improvements, incremental Agricultural/livestock production cotton production would reach was supported through project 92,300t, maize 62,300t, investments in research, extension, sorghum/millet 3,300t, rice 46,500t, agricultural credit and irrigation unshelled groundnuts 4,000t, meat development, and through supporting and edible offals 21,240t and milk the establishment of marketinS and 23.8 million liters. processing facilities. Allowances were also made for social development through financing health Implementation Bxperience services, village water supplies and functional literacy programs. 5. Implementation of the project cluster took on average 6.4 years, a time overrun of about 46%. Main Prolect Obiectives reason for the delays were difficulties in getting project 2. The projects' main objectives organizations started and studies were to intensify production of the required for the follow-on projects country's main cash crop, i.e. finished. Overall, however, project cotton, together with the main implementation progressed without staple cereals such as sorghum, major problems. millet, maize and rice. Introduction of draft animals and better use of Mali's vast livestock Proiect Results resources were also objectives. 6. The best results were achieved 3. The Mali Sud II Project had in improving cotton production, specific policy reform objectives. confirming previous OED findings.' It focused on improving the cotton Production expectations were sector's performance through exceeded by about 40,000t or more adjusting input and output prices in than 40% (1989). Also accordance with world market prices sorghum/millet production benefitted - xiv - from the Mali Sud and ODIPAC available, great interest in the projects. Maize and rice production crop and will so most likely in the did not grow as fast as expected. future. A third Mali Sud project Droughts and lack of floods has been appraised and will enable depressed rice yields in the Mopti CMDT to continue its work. polders, maize production suffered from lower than expected yields and 10. The Livestock Project should marketing difficulties due to be considered a pilot project. The processing problems. follow-on project is progressing well and pastoralists show a keen 7. Output of the Livestock interest in continuing with met.sures Project is difficult to quantify in to protect animal health. The the absence of proper monitoring project-financed cattle market has activities. While the 1983/84 been fully accepted by sellers and drought once more contributed to buyers but recent difficulties in substantial depletion of herds, the traditional export markets may not relatively speedy restocking would only have a negative impact on not have been possible without the cattle sales but may also lead to project. Project financed overstocking and as a consequence to vaccination and veterinary campaigns overgrazing of the range. reduced adult cattle and calf Unreliable floods/rainfall are too losses. much of a risk to get farmers interested in investing in 8. The ERRs of the two Mali Sud additional labor/modern inputs for projects, although lower than rice cultivation in the Mopti estimated at appraisal, are polders. Farmers seem to be more nevertheless impressive at 37.5% and interested in fully irrigated rice 27% respectively. The lower rates schemes where annual rice yields of are due to the fall of cotton 10t/ha can be obtained. prices. With inadequate flooding of the Mopti polders prevailing during the eighties, the Mopti Rice Project Lessons Learned has an unsatisfactory ERR of 3% only. No attempts have been made to 11. Increasing the producer price calculate an ERR for the ODIPAC of cotton through reducing Technical Assistance Project. government's share of export Because of insufficient data a earnings caused only a short term meaningful cost:benefit analysis of loss of government revenues. the Livestock project was Farmers reacted to the provided impossible. incentive as expected, and production as well as government revenues grew faster than before. Sustainability 12. Setting producer prices by 9. Cotton is the motor of linking them to world market prices agricultural development in areas of cotton seems to be a simple and benefitting from higher rainfall. workable solution compared with the Despite relatively lower producer formulas tried before under the prices compared with other West projects. African cotton producing countries, farmers always showed, with no other 13. Socio-economic studies could more remunerative crop alternatives be undertaken to determine levels of -xv- incentives required to safeguard the "marginal" low rainfall area. sustained/expanding production on It appears questionable to launch the one hand and adequate revenue extension activities# Bet extension flows to government on the other. densities, etc. without prior determination of available messages 14. Growing competition for lands and the preparation of an annual between farmers and herdemen, and work calendar to avoid erosion problems, deforestation, underemployment of extension etc. call for an early start of workers. Grain surpluses in land-use planning. The Livestock normal/good rainfall years and the Project and the Mopti Rice Projects absence of crop alternatives raise demonstrated conflicting interests: questions about the justification of polders of the rice projects reduced increasing extension costs in the traditional grazing areas. ODIPAC area. 15. Water drilling in Mali's 19. Differences of farmer credit pastoral areas without proper hydro- terms from project to project for geological surveys bears enormous the same kind of investments, i.e. risks. Preparation/studies of the draft oxen and machinery for draft Livestock project were insufficient traction are not justified. and resulted in heavy losses due to Repayment terms should be related to most sunk wells turning out to be the economic life of credit dry. supported investments. 16. Performance of agriculzural 20. On-farm cereal prices are at a research financed under the ;irojects low point during the harvest season. showed mixed results. Cooperation Farmers' obligations to repay short with a French cotton research term credit at that time is institute and detailed research contributing to the fluetuation of objectives specified by the cereal markets. Allowing moratoria appraisal have led to good results. for several weeks could lead to a Research activities of the other strengthening of the producer projects where no specific research prices. tasks were discussed did not produce meaningful results. 21. The Hopti Rice Projects suffered from poor quality seeds. 17. The ODIPAC Technical Seed supplied by the project from Assistance Project demonstrates the seed farm was not clean and led another problem inherent to the to a rapid reinfestation of the Bank's approach to agricultural polders with wild rice. Attention research. The originally planned needs to be given to the project period of three years was establishment of decentralized, completely inadequate to tackle the properly supervised seed Bank's expectations at the research multiplication farms. stations. Adaptive research aiming at new, improved varieties, 22. With grain trade liberalized fertilizer recommendations, etc. and cereal surpluses during the past definitely requires more than three two years, difficulties of selling years to produce results. grains, especially in more remote villages have become commonplace. 18. The same project also These difficulties are in part due supported extension activities in to the lack of market intelligence. - Zvi - Grain dealers need to know about 27. All projects supported saleable quantities at given functional literary programs under locations. which adults were taught to write, read and carry out simple 23. Maize production is suffering arithmetic. The programs have made from sales problems. Although a valuable contribution to the consumers would prefer maize over establishment and self-management of sorghum, they will not buy maize farmers' village/pastoral because it cannot be pounded as associations. These associations easily as sorghum/millet. Maize have taken on tasks which were at mills at village level may be the the charge of government in the solution. past, thus reducing administrative costs substantially and at the same 24. Grain surpluses are leading to time Increasing their members* heavily subsidized exports. High incomes. transport costs by truck or rail render Malian cereals uncompetitive. 28. Mali* with its difficult Improving or extending river landlocked position, harsh climatic navigation might contribute to conditions and fragile soils is now lowering transport costs. increasingly confronted with competition by other developing 25. The government has restricted countries in its traditional export its grain marketing activities to markets. Having lost its groundnut maintaining a 54 days period. exports in the early eighties due to However, the rapid growth of lower priced vegetable oils population needs to be taken into appearing in the markets, and consideration and a modest increase experiencing heavy losses through of the stocks on an annual basis the cotton price collapse some years should be considered. ago, Mali :to now facing the loss of its livestock markets in Ghana and 26. Cereal prices have fallen the Cote dOIvoire due to imports of considerably in the past years low priced beef, amongst others from reflecting the surplus situation. Argentina. Government, through its grain agency OPAM is acquiring 19,500t per year following bidding procedures. Since Endnotes: the lowest bid will win the contract, only producing 1. Impact Evaluation of Cotton areas/dealers with low transport Development Programs in Burkina costs are successful and disposal Paso, Cote d'Ivoire and Togo. OED problems in remote villages remain. Report Number 6878, June 30, 1987. PROJECT PERFORMAJNCE AUDIT REPORT MALI LIVESTOCK PROJECT (CREDIT 538-INLI) MALI SUD AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 669-MLI) HOPTI RICE II (CREDIT 753-MLI) ODIPAC TECHNICAL ASSISTANCE PROJECT (CREDIT 1174-MLI) SECOND MALI SUP RURAL DEVELOPMENT PROJECT (CREDIT 1415-HLI) I. BACKGROUND 1.01 About 85 percent of Mali's total population of about 7 million lives in rural areas. In 1987, GNP per capita was only US$210 making Mali one of the least developed countries, with a subsistence oriented economy, dependence on a few commodity exports, a high population growth rate, and a low level of education. 1.02 Agriculture contributed about 54 percent to the 1987 GDP and 71 percent to Malian exports, with cotton the major foreign exchange earner. Potential for agricultural production is limited by uncertain rainfall, fragile soils prone to exhaustion and low yielding, traditional crop and livestock technologies with little scope for intensification. High transport costs due to Mali's remoteness from the sea and vast internal distances are a further handicap. 1.03 Agriculture depends for tts production on the better Sudano- Sahelian and Guinean zones and on irrigated lands of the inland Niger delta. The vast Saharan zone in the north and east is almost uninhabitated. In the agricultural zones most farmers cultivate 4 to 5 ha. The country's huge livestock resources constitute a major store of wealth and income in rural areas. 1.04 An estimated 2.7 million hectares are cultivated of which about 10 percent is under irrigation. Cotton, the major rainfed export crop accounts for about 282,000 ha. Sorghum and millet, the main staples, are grown on about 1.9 million ha, maize on 143,000 ha and rice on 231,000 ha. Despite losing its traditional groundnut exports in the early eighties, cultivation of this crop still covers about 166,000 ha. The livestock population, subjected to heavy losses during past drought years, is estimated to number 5.5 million cattle and 11 million sheep and goats. - 2- II. THE PROJECTS A. LIVESTOCK PROJECT (CREDIT 538-MLI) Proiect Design 2.01 The project was to permit a more rational use of land and water in the 5th Region of Mali and to help about 100,000 pastoralist families to rebuild and improve their herds, and to protect them against future droughts. An essential feature of the project was to develop effective grazing control which was expected to offer the only means of creating and maintaining the conditions in which efficient and stable livestock production could develop. 2.02 Components of the project included (i) the introduction of extension services and grazing control in three special development areas totalling 7,000 km2; (ii) the provision of improved animal health services throughout the 5th Region; (iii) the construction of 70 wells and 50 ponds; (iv) the construction of an abattoir and hide-drying facilities at Mopti Sevare; (v) the construction of five livestock markets; (vi) the establishment of a 150ha livestock and pasture trial station; (vii) a functional literacy program for pastoralists; and (viii) the preparation of a second phase livestock project. 2.03 The project was to be implemented during a five year period from 1975 to 1979. Total project costs were estimated at appraisal to be US$17.3 million, of which IDA would finance US$13.3 million representing 100% of the foreign exchange (US$9.0 million) and 522 of the local currency cost. The remaining 11% of the total cost were to be at the charge of the government. The economic rate of return (ERR) was estimated at appraisal to be 512. Project Implementation 2.04 The project got off to a good start due to the government speedily fulfilling the conditions for Credit effectiveness. However, organization and establishment cof the Operation de Developpement de 1'Elevage de la Region de Mopti (ODEM), the executing agency proved to be cumbersome and took, with the recruitment of expatriate consultants and senior Malian staff, almost two years. Another delay in project implementation was caused by a mix-up in the Bank's project documents which indicated ODEM responsibility for research and evaluation, while agreement had been reached during credit negotiations that the Malian Instiltut d'Economie Rurale (IER) would be in charge of this component. Further delays were encountered in the selection of the abattoir site and the construction of the Mopti trial station. 2.05 The livestock extension and grazing control components as well as the establishment of watering points, the Mopti trial station, preparation of the follow-on project and the slaughter house construction were subjected to revisions during project implementation. Disputes arose -3- between ODEM and the Bank about the proper development sequence, i.e. establishment of pastoral associations (AP) prior or after successful well drilling. During project implementation it was found that groundwater levels in the project area were deeper than expected, requiring the installation of costly pumps. Consequently the well drilling program was substantially reduced. 2.06 The trial station did not perform any useful function. At the end of 1976 the station's area was reduced from 150ha to 65ha. Due to deficiencies of the protecting embankment only 12ha had full and 13ha partial water control and could be cultivated. Since the station was poorly managed and could not promote any project objectives, the station was closed in October 1975 and is now used as site for rice straw storage. 2.07 The originally included refrigeration plant at the slaughter house was not constructed because butchers preferred to move carcasses immediately after slaughter, thus eliminating the need for cold storage. 2.08 Vaccination campaigns started badly. However, during a later stage of project implementation, following the appointment of a qualified veterinarian, the campaign gained momentum. Extension activities also improved following ODEM's reorganization in late 1979. 2.09 Project preparation for a follow-on project was unsuccessfully attempted by IER but with assistance from the international Livestock Center for Africa (ILCA), the studies were completed. 2.10 Revisions, implementation delays, and the slow disbursement required three postponements of the Closing Date. The last extension was necessary to permit completion of project preparation and smooth transition to the follow-on project. Proiect Results 2.11 The project did not benefit the rural poor as much as expected, but some project components such as the livestock vaccination campaign, animal health care in general, the alphabetization program, and the human health program were more of a success. 2.12 Project benefits have been estimated by ODEM to include (i) an increase in the cattle herd by 20% and the small ruminant population by 30Z; (ii) controlled annual cattle exportt grew from 18,000 heads in 1981 to 23,000 in 1983; the figures for sheep and goats for the same years are given as 66,000 and 76,000 respectively; (iii) an increase of controlled slaughter rose from 6,000 cattle and 36,000 ruminants in 1980 to 11,000 bovines and 42,000 sheep and goats in 1982. According to ODEM, reconstitution of the 5th Region livestock population would not have been possible without the project. The project area compared with the rest of Mali, shows also the highest percentage of cattle protection against Rinderpest and contagious bovine pleuropneumonia (CBPP), a result of the project promoted vaccination campaign. Use of veterinary drugs increased -4- from 200,000 in 1977 to 15 million in 1985 and sales of agricultural feedstuffs during the dry season, increased from 30 tons cottonseed in 1979 to 3,000 in 1985. The nine wells with water for 3,000 cattle each and the 20 ponds constructed under the project permitted the use of new grazing areas. 2.13 The institution building effort did not devalop vell during the Project's initial years. ODEM's substantial reorganisations, changes at managerial and staff level, and finally its integration into the existing veterinary service deflected much of its attention away from the project. However, at present ODBM is recognized by government and livestock producers as a well functioning agency. In particular ODEM now successfully assists in providing (i) infrastructxre for livestock markets; (ii) maintenance of water points; (iii) protection against cattle thefts; (iv) mediation in case of disputes between pastoralists and farmerA: (v) advice to producer groups regarding other developments than re-investit.g in cattle; e.g. veterinary posts, human health centers, schools, tree plantations, wells, etc. 2.14 The pastoralists confidence in ODEM is reflected in APs contributing to (i) construction of houses and offices for livestock agents in 20 villages; (ii) establishment of livestock markets; (iii) changeover of well maintenance; (iv) readiness to make a 50% downpayment for pumps, generators and reservoir construction; and finally (v) the annual clearing of fire breaks. 2.15 Due to the difficulties encountered in keeping records on cattle moves, numbers, private slaughters, etc., calculation of an ERR and a financial rate of return (FRR) has not been possible. Project risks were recognized from the start although the time frame required for such an innovative project was grossly underestimated. The project proved to be a valuable lesson of experience, experience which has been incorporated in the follow-on project. The overall result is satisfactory. B. MALI SUD AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 669-MLI) Project Desigy 2.16 The project aimed at integrating and coordinating efforts of various foreign donors in the Mali Sud area, a region which comprises about 30% of the country's total population and having more favorable rainfall than the rest of the country. Principal objectives of the project were to (i) increase cotton and kenaf production through improving cropping - husbandry practices and through expanding the area under cotton from about 87,000 ha to 135,000 ha and the area under kenat from 2,000 ha to 4,600 ha; also increase maize and rice production mainly through intensification of proven cultivation methods and expanding the improved maize areas from 6,500ha to 15,000ha and rice from 4,OOOha to 12,000ha; (ii) provide project farmers with technical advice, seed multiplication facilities, and applied research; in addition the project was also to provide lines of credit for the purchase of new agricultural equipment and inputs; (iii) improve animal husbandry practices of cotton farmers, including use of draft oxen, stall fattening, and Improved cattle feeding through rational use of natural pastures, controlled burning, cultivation of fodder crops, and feeding of crop by-products; (iv) expand physical infrastructure through rehabilita- tion of a 10,000t ginnery and construction of a 20,000t ginnery, improvement of kenaf rotting ponds, construction of a 3t/hour rice mill, input storage, procurement of 151 vehicles for the Compagnie Malienne pour le Developpement des Textiles (CMDT), and improvement of 500km of tracks annually; (v) undertake studies for the development of areas freed from river blindness (onchoceriasis) including preparation of a resources inventory to constitute a data bank, a broad development plan for the fol3owing 10 years and further project preparation within the development plans (vi) Improve rural health services by equipping and training their staff and financing an initial supply of drugs; (vii) train CMDT staff and the rural population through young farmers' and blacksmiths' training and functional literacy programs. 2.17 The project was expected to take about four and half years. Total project costs were estimated at US$44.6 million of which US$15.5 million was to be IDA's share, and US$15.9 million the participation of the respective cofinanciers (CCCE, PAC, ADF, and BADEA) Project Implementation 2.18 There was a delay of six months in the credit becoming effective due to finalizing financing arrangements of other donors as well as problems in meeting some of the conditions of effectiveness. CMDT was reluctant to start with the project before finalizing arrangements with other donors and this led to only technical assistance and training activities being undertaken during the first year. 2.19 Some reformulation in project design was required also due to late finalization of cofinancing arrangements as well as shortfalls in cotton, kenaf, and rice output- Retting facilities for kenaf and the rice mill were cancelled and so was the additional cotton ginnery. The central vaccination yard planned under the livestock component was abandoned and replaced by smaller local centers constructed by the Associations Villagoise (AVs) at village level. Substantial delays were encountered in the implementation of the health component. During the originally envisaged project period only 502 of appraisal targets were implemented. Studies for the onchoceriasis area were not completed. Savings from project components not implemented were reallocated to investments for infrastructure improvements, especially for agricultural tracks and for CMDT housing and storage facilities. 2.20 Total project costs were about US$1.9 million below appraisal estimate, mainly due to the elimination of the additional ginning facility. The reestimated project costs have to be taken with caution because the government and the farmers' contribution cannot be determined accurately. 2.21 Because of the initial delays in project start-up, the Bank had to agree to the extension of the Closing Date. The third extension was granted to permit finalization of the study on developing the aroas freed of onchoceriasis. Despite this extension aerial photos could not be taken in time to complete the study. The Bank refused a further extention of the Closing Date and an unused Credit balance of US$441,413 was cancelled on December 31, 1984. Proieet Results 2.22 During the first three project years fertilizer and insecticide application expanded rapidly, but declined during the following two years due to substantial increases in input prices. With higher returns on cotton in 1982/83 fertilization and pert control picked up again. By 1984 almost 90% of all cotton planting received fertilizers and 80% of the cotton area was sprayed at least three times with insecticides during the growing season. Technical packages recommended for cotton under the project also showed important spill-over effects, with modern husbandry practices increasingly being applied also to other crops. Distribution of agricultural implements exceeded project targets in almost all project years. 2.23 Cotton area expanded rapidly and reached a record high in 1979/80 helped by well distributed rainfall. After a reduction in the following years due to the shock caused by the reduction of input subsidies, and adverse weather, production and yields have increased continuously up to the present. Total cotton production reached 139,000t in 1984/85, the end of the project. 2.24 Cereals production expanded in the project area beyond expectations. Areas under maize exceeded project targets by 260% in the final project year (1984). Development of sorghum and millet production also increased dramatically. Acreage expanded by 600% up to 1985/86 and total production reached 150,000t. Rice production reached 14,000t in the same year, slightly below the appraisal projection of 15,000t. 2.25 Kenaf )roduction was discontinued after 1982 when increased competition fr-m jute led to unattractive producer prices and finally to the shut-down of the local processing plant. 2.26 Livestock development was hampered by the delay in obtaining BADEA financing. Implementation of this component did not start until 1979/80 and this is t%e main reason for the incomplete implementation of the livestock program. Since no reliable statistics were developed, achievements cannot be quantified. In spite of difficulties encountered, a special livestock mission, visiting Mali in 1982, reported on the success of the project in integrating crop and livestock farming in the Mali Sud area. 2.27 The project's ERR has been recalculated at 37.5% compared with 44% at appraisal. The lower ERR is mainly due to higher opportunity cost of -7- labor. The project also produced important secondary benefits, in particular, better living conditions through improved literacy and health as well as better developed skills at village levels, a consequence of CMDT's work. However, even more important are the changes introduced at village level through the founding and strengthening of the AVs which have become pillars of rural life in Mali Sud and represent a strong, responsive, and efficient rural self-help structure, involved in commercial as well as development activities. C. MOPTI RICE II PROJECT (CREDIT 753-MLI) Proiect Design 2.28 The project was to permit a moderate expansion of the cultivated area in the Niger flood plain through construction of four polders with a total area of 8,800ha and some improvements of polders developed under the first Mopti Rice Project (Cr.277-MLI). In addition the project provided for (i) deep plowing of new polders as well as on about 14,300ha in existing polders; (ii) construction of storage facilities for paddy, farm inputs, and farm equipment, and of workshops, training centers, offices and staff houses; (iii) procurement of farm machinery, civil works maintenance equipment, ferries, vehicles, workshop and office equipment; (iv) an adult literacy program and an audio-visual program to be used by the extension service; (v) applied agricultural research; (vi) credit for fertilizer and agricultural implements; (vii) technical assistance for project management and staff of the Operation Riz Aopti (ORM), the organization responsible for project implementation, and operating costs attributable to project area expansion; and (viii) asaitance to the Study Bureau of the Rural Works Department in Bamako. 2.29 The project was expected to start in mid 1978 and to extend over a 5 year period. Total project costs were estimated at US$31.2 million with a foreign exchange component of L 119.0 million or 61% of total cost. The credit of US$15.0 million was to tiLrance 571 of project costs net of taxes. Additional joint financing (US$2.' million) was obtained from FAC and US$6.0 million from parallel ADF firncing. The ERR estimated at appraisal was 18%. Prolect Implementation 2.30 The project had a timely start with contracts for initial works having been prepared already under the earlier Mopti project. All civil works targets were fully attained within the time envisaged during appraisal. The four new polders were established including dikes, canals with regulating gates and feeder roads. A polder started under Mopti I was completed and rehabilitation works were carried out on nine polders of the first project. Deep plowing was undertaken on 20,400ha and construction of offices, warehouses and accommodation for project staff was executed according to appraisal expectations. -8- 2.31 Land distribution fell about 20% short of appraisal estimates but the number of project participants, who received less land than originally planned, was close to target. The reduction of the overall project acreage was due to doubts if all plots were to receive enough water in normal years. The smaller farm size could not meet the labor requirements of the original holdings. 2.32 The applied research program, financed by FAC, was delayed because of disbursement problems but produced valuable information on possibilities of double cropping, rock phosphate fertilization, and identification of improved rice varieties. 2.33 The adult literacy program, also financed by FAC, progressed satisfactorily. More than 100 villages participated and the program continues under the follow-on project. 2.34 Main problems encountered during project implementation were (i) the poor quality of rice seeds made availabl. to farmers, seeds not properly cleaned, leading to continuous reinfestation of project polders with wild rice; (ii) the inappropriateness of the agricultural credit scheme, requiring farmers to make down-payments amounting to 40-45% of implement costs and repaying their loans within one or two years, leading farmers to shy away from acquiring farm machinery; and (iii) farmers, confronted with inadequate floods for most project years, showed little enthusiasm for adopting improved husbandry techniques. 2.35 OM operating costs were higher than expected at appraisal and its revenues lower. Instead of accumulating an annual surplus of MF 225 million, (for the 1978-82 period), annual losses of MF 270 million were registered. The operating deficit had to be covered by government subsidies and resulted in delayed maintenance works. Due to the difficult financial situation of the project implementing agency, IDA disbursements were delayed and the Closing Date had to be extended to June 30, 1985. Proiect Results 2.36 Project output was severely curtailed by drought throughout tne implementation period. During the years 1982, 83, and 84, and after project completion in 1986, 87, 89, the Niger and Bani rivers remained well below levels that could be considered the statistical average of more than forty years of river flow gauging. Meteorologists point to a 200-400km shift southwards of the Sahelian (100 400mm rainfall) and the Sudanian (400-1,200m rainfall) isohytes based on the ana]-ses of the 1965-1984 rainfall data. 2.37 As a consequence of insufficient or no flooding, rice yields remained continuously below appraisal estimates. Instead of an average of 1.7t/ha for the five year implementation period (1.8t/ha at project completion), yield averages reached only 1.02t/ha, with the lowest yield (0.7t/ha) obtained in the final project year when only 2,500 ha of the sown -9- 24,400ha could be harvested. To sum it up, actual project production reached only 312 of the appraisal projections. 2.38 Project costs increased from US$29.9 million at appraisal to US$31.5 mainly due to higher than expected costs of civil works, a component which accounted for more than half of total project costs. 2.39 Higher costs and lower than expected output has led to a recalculated ERR of 3%. D. ODIPAC TECHNICAL ASSISTANCE PROJECT (CREDIT 1174-MLI) Proiect Design 2.40 The project was expected to support a newly established rural development institution, the Office de Developpement Integre des Productions Arachidieres et Cerealieres (ODIPAC) in defining a new approach to agricultural development in the groundnut belt and in determining the scope for further investments. 2.41 The project aimed at (i) increasing the efficiency of agricultural production by retraining extension staff, improving seed production, distribution, and storage, as well as strengthening the credit and marketing system; (ii) promoting self-managed farmer organizations able to take over village development activities; and (iii) testing diversification programs to reduce the area's dependence on groundnuts as an export crop. 2.42 Project components included the recruitment of six expatriate specialists to strengthen ODIPAC's finance, marketing/credit, seed multiplication, agricultural research, extension/cooperatives, and workshop sections. The project also provided for an adult functional literacy program, replacement of essential vehicles, and purchase of incremental fertilizer and agricultural equipment. A monitoring and evaluation unit was to be established, and applied research and studies as well as audits undertaken. 2.43 The project was expected to be undertaken during a three year period. Total cost was estimated at US$11.4 million with a foreign exchange component of 58Z of total costs. The credit of US$6.5 million was expected to finance 572 of the total cost. A FAC grant of US$2.5 million was equivalent to 22Z of projected costs. The remaining US$2.4 million (212 of total project costs) were to be the government's share. No ERR was calculated. Project Implementation 2.44 Credit signing was delayed until April 30, 1982 due to shortages of IDA funds. The credit became effective two months later. Following the first supervision mission in October 1982, the Association agreed with the government to amend the Credit Agreement taking into account that (i) crop - 10 - marketing functions had been transferred away from ODIPAC; farmer credit terms would be modified so that individual farmers would have no access to seasonal credit, but AVs would provide a line of credit requiring farmers to shoulder 50% of the input costs; (iii) funds for the purchase of building and educational materials, agricultural inputs, as well as equipment for feeder road construction would be increased; (iv) more funds would also be provided for consultant services reflecting the shifting development focus on extension and regional improvement; (v) responsibility for the functional literacy program would be transferred to ODIPAC. 2.45 Supervision missions during project implementation found that the research component had produced some encouraging results but that the extension component remained weak. Further weaknesses were identified in the functional literacy program. ODIPAC requested a relaxation of farmer credit terms to promote wider use of inputs. The Bank agreed in September 1984 and permitted the issuing of loans of up to 1002 of agricultural input and equipment costs for AV members. In the same year the Bank also agreed with the implementation of a livestock component, providing funds for vaccination campaigns, parasite control measures, establishment of veterinary pharmacies, fodder reserves, etc. at a total cost of US$80,000. 2.46 With some funds remaining, the Bank suggested and GOM agreed to an extension of the Closing Date by one year. As a consequence the Credit could be fully disbursed. Proiect Results 2.47 At project inception the Bank was optimistic about groundnuts becoming a viable Malian export commodity. During project implementation it was, however, confirmed that even with new improved varieties, production/transportation costs would remain too high for Mali to successfully compete in the world markets. 2.48 Although about 50% of project costs were aimed at improving crop production through providing farmers with modern inputs, credit, services, etc., the eventual iacremental output could not be quantified. Indications are that the late introductiin of the livestock component was beneficial. Lack of proper definition, organization and implementation of the agricultural research component, and especially the unrealistic three year time frame for research activities could not be conducive to obtaining pertinent results. 2.49 ODIPAC's accomplishments were considerable with establishing extension services, and getting the functional literacy program restarted. Despite official reluctance to support road construction in the project area, unwanted competition with the existing railroad was the main reason, the feeder road component was well executed. 2.50 The objective of using project results as basis for deciding on future investments in the ODIPAC area was not accomplished. Towards 1986 it was found that a follow-on project would not be ready for appraisal - 11 - prior to the Credit's closing in September 1986. Project preparation funds amounting to US$750,000 were allocated to continue ODIPAC's preparation activities. In February 1987 a Bank mission found that the soils of the project area had been over-exploited which was leading to nutrient depletion, soil erosion, and loss of vegetation. The mission concluded that agricultural development efforts should be directed at improving the area's farming systems. Confronted with this information, the Bank informed the government that its own staff constraints and the insufficiency of its project preparation facility would preclude further Bank support for preparing a new project. The Bank suggested the government should address itself to other donors in obtaining assistance for continued project preparation. E. SECOND MALI SUD RURAL DEVELOPMENT PROJECT (CREDIT 1415.-MLI) Project Design 2.51 The main project objectives of the second phase project, following Mali Sud I (Cr. 669-MLI) were to (i) increase cotton production through offering producers higher prices while reducing input subsidies; (ii) improve control and use of the cotton sector's revenues by changing institutional arrangements for marketing cotton and allocating proceeds from it; (iii) protect cotton producer incentives, CMDT's financial viability and government revenues by establishing a fund to buffer temporary revenue fluctuations caused by declines in either price or output; (iv) produce more maize, sorghum millet, cowpeas, rice, and livestock through extension work and input use; (v) strengthen research for cotton and other crops; (vi) promote farmer cooperatives through economic incentives and training; and (vii) meet basic needs by improving village water supplies and health services. 2.52 The projecc was designed to (i) expand cotton production on about 160,000ha and improve farming practices on about 31,000ha maize, 35,000ha millet/sorghum, and 18,000ha cowpeas; (ii) partially rehabilitate and improve rice production on 1,750ha of small irrigation schemes and 1,000ha in the Klela River, and the strengthening of CMDT's capacity in design, operation and maintenance of small irrigation schemes; (iii) supply on medium term credit through the Banque Nationale de Developpement Agricole (BNDA), about 7,500 draft oxen 13,000 plows, 10,000 seed drills, 7,000 cultivators, 2,200 ultra low volume (ULV) sprayers, and 160 low-horsepower tractors; (iv) strengthen BNDA branch offices; (v) substitute imported phosphate fertilizer with about 15,000t locally produced rock phosphates; (vi) deliver drugs and feed supplements and improve animal husbandry practices for cattle and draft animals, and install a small mineral lick factory; (vii) improve and extend agricultural research through four applied research programs on cotton and coarse grains, farming systems, technology transfer to smaller farmers, and fertilizer application; (viii) strengthen CMDT's capacity for the rehabilitation and maintenance of 3,000 km of feeder roads and construction of tracks; (ix) promote 600 additional AVs and transfer all primary marketing of cotton and credit recovery to the - 12 - AVas (x) provide functional literacy training for AV members, train AV leader., blacksmiths, and CMDT staffI (xi) drill 400 wells for improved village water supplies and 50 vell for pastoral usel and (xii) continue the primary health care program and training of village health workers. 2.53 The project was expected to be implemented during a four year period. Total project costs were estimateC at appraisal at US$84.0 million including identifiable taxes and duties (US$4.6 million) and a foreign exchange component of US$59.9 million (71%). The financing plan provided for a US$25.9 million (31% of total project costs) IDA Credit, US$2.2 million (2.62) from the French Ministry of Foreign Affairs, US$4.1 million (4.92) from the Netherlands, US$4.2 million (5.02) from FAC, US$11.3 million from the CCCE (13.5%), from the International Fund for Agricultural Development (hAD) US$13.0 million (15.5%). Of the total external financing US$35.2 million was to be under joint and US$25.5 million under parallel financing. The ERR was calculated at 34%. Project Imnlementation 2.54 Project implementation started without any major problems with a smooth transition from Mali Sud I to Mali Sud II. Serious problems, however, arose as a result of the "cotton crisis" the cotton price collapse of the world markets in 1985/86. The Cotton Guarantee Fund had insufficient means to support the guaranteed producer incomes. Next to the cotton farmers, CMDT suffered dramatic losses. 2.55 To counter CMDT's difficulties significant reallocations of Credit proceeds were introduced with reductions in investments and increases in CMDT's operating budgets. Deficits incurred by the filiere cotton were financed by donors. Most hit by the reallocation was the rural roads component. Project Results 2.56 The project design was innovative because the project did not restrict itself to focus on increasing production and improving the living conditions of small farmers, but also intended to bring about major policy changes in Mali, in particular to (i) phased real increases in cotton producer prices, to be financed through reduced input subsidies and government revenues; (ii) reforms of the price fixing system and institutions through controlled and audited budgeting, and through consolidating most specific taxes on cotton exports into a single distribution of surplus revenue, after all costs had been recovered, as well as reducing the Societe Malienne d'Import-Export's (SOMIEX) role to an export agent; (iii) earmarking gross export receipts to a Cotton Account to be managed by the Office pour la Stabilisation et pour RegularisatiGn des Prix (OSRP); and (iv) the capitalization of a Cotton Guarantee Fund to buffer interannual variations in net cotton revenues. The creation of the "filiere coton" a streamlined cost recovery system was a major institutional reform instrument which made it possible to accomplish the -13 - above mentioned policy changes. The Cotton Guarantee Fund, however, proved not workable in the light of the dramatic fall in world market prices of cotton in 1985/86. A better method which links producer prices directly to world market prices -or cotton has now been adopted. 2.57 The project was also highly successful in reaching and even surpassing its production targets. The area under cotton increased to 142,000ha compared with the appraisal estimate of 126,000ha with average yields reaching 1,314kg/ha in 1987/88 compared with appraisal projections of 1,250kg/ha. Sorghum/millet area and production also exceeded appraisal estimates although average yields stayed below projections. Maize production fell short of expectations with areas planted and average yield. remaining lower than anticipated. Rice and groundnut production did well mainly due to a more rapid expansion of acreage. 2.58 Rehabilitation or improvement of irrigated/rainfed rice production reached lrvs than 30% of appraisal targete with incremental production falling be.ow expectations. The livestock component proved a partial success especially in regard to training/supplying draft animals. The rural water supply component was successful with well construction exceeding appraisal targets by almost 201 and at a lower cost than anticipated. Due to the reallocation of Credit funds (see para. 59 above) only one third of the feeder road rehabilitation planned (3,000 km) was undertaken and construction of 30 km new roads never initiated. Lack of proper collaboration between CMDT and national health services had a negative impact on the implementation of the project's health component. At the end CMDT took over the distribution of medicines and the supervision of the health worker training programs. 2.59 The project's recalculated rate of return is 17%. The lower rate, compared with the appraisal estimate is largely due to the unexpected decline in cotton prices. III. MAIN ISSUES A. General 3.01 The audit of these five agricultural projects permits reaching some conclusions on the impact of Bank support to Mali's agricultural sector during a 15 years period. In addition, the range of projects under review cover the sector's main activities, i.e. livestock, cotton, and to a certain extent also groundnuts, the most important export commodities of the present and past, but also coarse grains and rice, the most important staples of the local population. 3.02 In trying to develop Mali's agricultural sector, the Bank has been, and in many respects still is, confronted with major constraints limiting agricultural/livestock production and development. The main constraint being the low rainfall, with two thirds of the country receiving less than 400mm annual rainfall and only one quarter with more than the - 14 - 550m needed to sustain farming. Even in areas suitable for agricultural production, annual rainfall fluctuates widely leading to the recurrence of famine. The Staff Appraisal Report (SAR) of the Livestock Project (Cr. 538-MLI) points already to the occurrence of frequent droughts (para 2.01) and the vagaries of thj climate which result in substantial yield fluctuations. 3.03 Extended periods of favorable rainfall patterns lead to the accumulation of surplus grain and growing livestock herds result in over- stocking and consequently overgrazing, with degradation of the range, while lack of watering points prevents full use of vast pasture areas. 3.04 Government agricultural policies at the time of project preparation/appraisal were found to be counterproductive to agricultural development due to inTufficient producer incentives (cotton, rice), and to support given to inefficient parastatal supply and marketing agencies (Societe de Credit Agricole et d'Equipement Rural - SCAER; SOMIEX; Office des Produits Agricoles du Mali - OPAM). Although identified in the early seventies, policy constraints remained until the appraisal of the last of the projects subject to this audit. The 1983 SAR on Mali Sud II repeats the findings of the 1976 SAR on Mali Sud I: "three major constraints become apparent. First, the lack of an effective floor price policy for coarse grains which would balance the need to reflect structural price differences through space and time with the modest financial and institutional resources available to carry it out. Second, the exclusion of paddy (in large official schemes) from the market liberalization program, in an effort to safeguard low-priced supplies to urban target groups and protect the turnover of State-owned rice mills. Third, a reluctance to increase producer prices of the remaining administered crops (e.g. cotton) in line with border parities." 3.05 Major sector constraints could also be found in regard to Mali's supporting services and the poor infrastructure. Development planning for the agricultural sector is and was confronted by lack of basic data, be it on statistically significant rainfall/floods data, land use/ownership, groundwater availability, etc. Agricultural research covered cotton, a major export crop, but has little to offer on traditional staples or alternative export crops till now. Consequently extension efforts outside the cotton producing areas had to remain limited. Agricultural credit in general is confronted with high administrative costs, except for the AVs, and suffers from unrealistic credit terms, and split responsibilities (BNDA, CMDT). Input suPlies were the responsibility of SCAER which had difficulties in achieving timely imports and distribution. Marketing, officially monopolized by parastatals showed major shortcomings in both the cotton and cereals sector, while traditional private marketing activities competed successfully (cereals, livestock). Processing of agricultural commodities on a large scale showed mixed results with the cotton sector well covered, while groundnut/kenaf processing suffering from declining world market prices and consequent loss of supplier interest, and finally rice milling battling with system immanent difficulties. Processing facilities (maize mills) at village level together with storage facilities . - 15 - were practically non-existant, traditional rural practices (labor intensive grain pounding) and primitive on farm storage bine prevailed. 3.06 One other major non agricultural constraint hampering production for export and increasing costs of imported inputs is Mali's landlocked position, with distances to foodstuff importing coastal areas exceeding on average 1,000km. In addition, internal roads are widely missing which would link villages to major trunk roads or the railway. 3.07 Furthermore, the human factor is a constraint with the high rate of population growth (2.6% p.a.) combined with health problems, especially among the rural population. The high percentage of rural versus urban population, limits possibilities of marketing excess agricultural output within the country. Finally, there is the high rate of illiteracy which forced government to take over expensive support activities which are handled by farmers themselves in other countries. 3.08 The projects under this audit provide valuable lessons on how the above mentioned constrains were considered during project preparation, appraisal and implementation; whether successfully removed or not tackled at all. Prima faci evidence leads to the conclusion that the Bank was pre-occupied with agricultural economic policy issues, i.e. input subsidies, producer prices, market deregulation, privatisation of marketing, etc., and, as a consequence lost eight of other, overall agricultural sector problems. 3.09 The Bank deserves credit for contributing to removing, or at least partially removing a "human factor constraint" by supporting functional literacy programs in all five projects. The functional literacy programs permitted the early establishment of AVs and APs, farmer/pastoralist groups which permit the takeover of government activities, thus reducing national budget allocations to supporting services and at the same time promoting the often requested but rarely obtained graseroot support for development. 3.10 The projects under review also provide insight regarding the problems encountered by the Sahelian countries where introduction of modern husbandry techniques is leading to marketing problems of surplus, low value crops, especially coarse grains in years of adequate rainfall. Surplus production which can from year to year disappear and be replaced by widescale hunger. B. Dealing with the Sector Constraints 3.11 Overcoming Natural Constraints. To overcome the drought problems, irrigation seems to be the answer. The Bank consequently supported irrigation on a small scale through the two Mali Sud projects and also through the controlled flooding Mopti Rice projects. The decision to support Mopti was mainly based on the assumption that controlled flooding offered a lower cost alternative to any other irrigation undertaking. The unreliability of the Niger and Bani river floods was grossly under estimated and the actual double risk of the project not properly clarified - 16 - in the SAR. This double risk is caused by the dependence on adequate rainfall for the seeding and germination of rice and the reliability of the river floods to provide adequate water during the growing season. Crop failures could thus occur if either rains or floods are inadequate. Farmers are well aware of the high risk and are prepared to follow extensive cultivation practices. Fertilization and weeding are consequently reduced to absolute minima. 3.12 The assumed project yields had been based on relatively high input-labor use with final project production of Mopti II to reach 34,000t with an average yield of 2.0t/ha. Actual production is given in Table 1. Table I Project Production (in metric tons) 1978/79 15,500 1985/86 25,723 1979/80 13,400 1986/87 5,833 1980/81 13,400 1987/88 1,156 1981/82 17,800 1988/89 17,684 1982/83 1,800 1989/90 12,074 1983/84 n.a. 1984/85 n.a. Source: PCR p. 6, ORM 3.13 The relatively low cost of water control structures versus a full scale irrigation scheme was the main reason for deciding in favor of the design adopted in the implemented project. Questions can be asked if the "cheapest" solution was really the best. Netherlands and Arab aid financed two irrigation perimeters totalling 53 he in 1987 at a hectare cost of FCFA 2 million. (US$7,000 equivalent). Farmers are transplanting the rice and are obtaining two harvests per year averaging St per harvest. Investments of US$17.5 million, at current prices would have permitted the establishment of 2,5OOha irrigated rice paddies which would produce 25,000t/year without being jeopardized by either poor rainfall or ina4equate floods. 3.14 Various Bank reports mention the problems of soil erosion, low organic matter content and lack of plant nutrients. None of the projects addressed the problem of soil erosion control. It is, however, interesting to note that several AVs in the Mali Sud projects have started simple soil erosion control activities like contour bunds, waterways, etc. Although not directly intended, draft animal mechanization and some livestock raising/fattening operations has led to farmers recognizing the importance of farmyard manure, which is now properly applied to fields. Fertilizer application has increased over the year. despite the removal of fertilizer subsidies. It is interesting to note the spill-over effect of modern cotton cultivation on cereal production in the Mali Sud project area. On - 17 - the other hand it is rather cynical when the PCR on the ODIPAC Technical Assistance project remarks: "..anarchic exploitation of natural resources was very advanced due to recent soil erosion induced by land reclamation activity in marginal zones." According to supervision mission reports, there had been overcultivation of the traditionally cultivated areas with no replenishment of soil nutrients, leading to soil fati&ue and causing erosion and lose of vegetation. (PCR, p.9, para 3.13). The PCR, reflecting the Bank's response to the problem summarizes the Bank's reply to the Government: "..if government wished to pursue the apparently very elusive goal of project preparation, government should seek grant funding from any agency specialized in preinvestment." (PCR, p.10, para 3.15). The Bank's response to encountered soil nutrient/erosion problems after total investments of US$11.4 million plus US$750,000 from Project Preparation Facilities is surprising. 3.15 Agricultural Development Planning received Bank assistance under three of the five projects, i.e. Livestock, Mali Sud I, and ODIPAC in the form of Bank financed studies. The Mali capacity for development planning/project preparation had proved successful in the work done by the Institut d'Economie Rurale (IER) responsible for the concept for the Mali Sud projects but less successful in the preparation of the livestock follow-on project. The accomplishment of the Mali Sud project preparation can be traced to IER's effort to assure the cooperation of all ministries and agencies concerned with the project at an early planning stage. As a result, agreement was reached also on non-agricultural development priorities of the region. 3.16 The Bank's assistance, 18 studies financed alone under the Livestock project had only a minor impact on the planning/project preparation stage, especially when looked at in the context of identifiable sector constraints. Land use studies were never part of any Bank financing and their absence has at least contributed to conflicts of interest i.e. collision between Bank assisted projects (Livestock and Mopti I/II) over tho use of project lands. Most rice polders financed under the two Mopti projects were traditionally used as waiting areas for herdsmen before entering the Niger delta grazing grounds following the receding floods. 3.17 Landuse studies are also needed in other parts of the country where increasing competition for land can be found between agriculture, livestock and forestry. In addition increasing erosion problems call for restrictive agricultural use of threatened lands. Closely associated with land use studies would be reviews of land ownership. For instance, agricultural production in the Mali Sud area has expanded rapidly at the cost of the traditional fallow periods. The main reason is not only population growth but the traditional tribal land allocation system which leads to the occupation of fallow lands by farmers. Larger farmors consequently avoid fallow periods and continuously cultivate all lands to avoid losing some of their fallow fields. The already poor soil fertility is thus further depleted to avoid "quiet expropriation". - 18 - 3.18 Lack of proper studies has had serious repercussions on the performance of the Livestock project. First, insufficient basic data led to the Seno Mango as the focal area for watering points/grazing development. Malian experts consider the area's potential as less promising due to the relatively low number of livestock owners having traditional grazing rights, compared with other areas on the left bank of the Niger. Immigration of outsiders would be considered as infringements of traditional rights and would lead to frictions. A socio-economic survey of the grazing areas would have led to different development priorities. 3.19 The audit also questions the wisdom of rushing into Aevelopment of watering facilities in the Seno Mango without a proper hydro-geological survey. Under the Livestock Project 60 wells were to be sunk in this area but only 11 could be drilled and of these only 5 produced water flows in volumes as expected at appraisal. New studies undertaken for the follow-on project are better and the rate of successful well development has increased dramatically. 3.20 Agricultural research was suppor--ed by all five projects at a total cost of US$12.3 million. Results obtained from this investment were mixed: successful cotton research versus complete failure of the research under the Livestock project. There were several reasons responsible for the varying research results. On the positive sidet first, cotton research has a long established tradition not only in Mali, but in all francophone countries. Research programs prepared/suggested by the French Institut de Recherches du Coton et des Textiles Exotiques (IRCT) based on economic impact of targeted research work and collaboration and division of labor through a network of stations working under similar natural conditions in other countries of West Africa and Madagascar, are responsible. Second, the appraisal of Mali Sud I and II clearly stated the research objectives: (i) breeding of shorter cycle cotton varieties; (ii) obtaining higher ginning percentages; (iii) breeding oil-rich glandless types with more than 20Z oil content; (iv) test results obtained from farming systems research; (v) update fertilizer recommendations including use of rock phosphate. 3.21 On the negative side the Livestock Project had too wide a scope for inexperienced researchers, i.e. (i) determine methods to protect, regenerate and improve the Niger delta pastures; (ii) establish the economics of different fattening techniques; (iii) develop methods for the control of parasites and insects. While the Bank prides itself for the technical assistance it provides in the context of project implementation, no proper guidance was ever extended on how a workable and feasible research program could be formulated and implemented. The PCR (para 3.28) states laconically that "ODEM's inability to use the trial station to promote project (research) objectives, led the October 1979 supervision mission to recommend its closure". 3.22 Equally general in nature were the recommended research objectives for Mopti Rice II. The SAR mentions "field trials of new varieties, development of optimal cultivation techniques" (para 4.13). The results obtained, however, were better than one could expect but only because the - 19 - work was entrusted to experienced French researchers from the Institut de Recherches d'Agriculture Tropicale (IRAT). Following IRAT's departure, research work faltered. 3.23 Little consideration is generally given by the Bank during project preparation or appraisal to time required for obtaining meaningful results. The ODIPAC project could serve as an example. One of its objectives (President Report, p.14, para 41 ii) was to "test diversification programs adapted to each sub-area's potential and to farmers' identified preferences". And in para 52 of the same report details on planned research activities are given: "Research work would cover responses of traditional cereals and groundnute to economic input packages, varietal trials on maize and other relatively new crops such as cowpea, sesame, etc. and application of Malian rock phosphate. The trials will focus on productivity, pest resistance, cycle length, and optimum dosages of fertilizers and pesticides." 3.24 Results were expected during an implementation period of only three years since the information obtained was expected to lay the basis for the follow-on project. Even under the best of conditions, i.e. no rainfall variation, no pest/disease fluctuation, etc. could a research team come up with new, adapted varieties, fertilizer/plant control recommendations etc, in the timespan envisaged by the appraisal of this project. 3.25 Extension activities, with the exception of CMDT activities, were of little use in three of the five projects (Livestock, Mopti II, ODIPAC) because of limited research messages available. Similar to the agricultural research problems mentioned above, extension activities and goals were not subjected to proper analyses during project preparation or appraisal. 3.26 The success of extension activities under the Mali Sud I and II projects gives an example of the necessary "phased" approach of extension. During the first phase, extension efforts concentrated on proper timing of seeding, seeding in lines, and weeding activities. The second phase concentrated on introduction of draft animal mechanization, land preparation, fertilization, and application of farmyard manure. The third phase familiarized farmers with pest control and introduced them to ultra low volume (ULV) sprayers. 3.27 The experience gained by farmers with improved cotton husbandry has had a significant epill-over effect on the other crops grown on their farms. This is reflected in higher and more consistent cereal yields. As a consequence of the encouraging production developments, CMDT is now considering reducing its extension effort by CMDT staff and use farmers trained under the Young Farmers Training Programs as intermediaries, taking into account that farmers will ask less but on the other hand more sophisticated questions about future production problems. - 20 - 3.28 Extension activities under the Mopti II and ODIPAC projects had little impact. In both projects the risks confronting intensified production were such that farmers were reluctant to invest in additional inputs or labor. In this context it might be interesting to reflect over the role of extension in a country like Mali in general, a country where most regions are suffering from low rainfall which permits the cultivation of low value crops only, low-value crops which cannot support high transport costs and furthermore are confronted with surplus production from more favored areas within Mali. The ODIPAC Technical Assistance project in particular has demonstrated the difficulties of promoting alternatives to groundnute. if adequate rainfall occurs Mali is likely to produce surplus coarse grains in the Mali Sud areas as in the past two years. Extension may have contributed to higher cereal output also in the ODIPAC area. However, ODIPAC management complained that there were only restricted sales from remote villages. In years of drought the same areas will be hit more than Mali Sud and even with the best extension advice, cereal yields will be depressed and famine will return, but Mali will have spent considerable sums on extension activities. 3.29 A special case was also the Livestock extension effort which proved to be a complete failure, in part due to the Bank's lack of appreciation of the herdsmen's priorities. While the livestock owners aimed at rebuilding their herds after the severe droughts, the project's extension effort was meant to focus on controlling over-stocking and consequently prevenzion of over grazing of the range. Cattle fattening was also to become a major extension supported activity, an operation which seems doubtful in light of livestock still not being sold by weight or meat quality. In short, the project appraisal did not properly analyze the scope and feasibility of extension advice likely to be accepted by the pastoralists. 3.30 The InRut SuRR11 situation preoccupied the Bank during appraisal and project implementation. The elimination of an unreliable and inefficient parastatal (SCAER) originally responsible for input supply was a worthwhile accomplishment. Fertilizers and pesticides are now available in time and in sufficient quantities. However, the availability of one basic input, improved and clean seeds, remains unresolved. Depending on the results of agricultural research, as mentioned above, improved varieties of the main staple cereals - millet and sorghum - have not yet been foT eoming. The same applies to improved groundnut and cowpea varieties. 3.31 The seeds issue became a serious problem in the Mopti Rice II (also Mopti 1) project. Due to the project area's heavy infestation with wild rice, the project included funds for deep plowing, a suitable method to rid fields of wild rice. However, rice seeds supplied by the project's seed farm was again infested with wild rice seeds and within a short time the polders were again invaded by this weed, rendering the preceding deep plowing exercise futile. The Bank was aware of the problem at an early stage. A 1981 memo stated .."poor quality of seeds discourages farmers" but to this date no action has been taken. - 21 - 3.32 Atricultural Credit activities developed well and in this context it is necessary to mention the exceptionally fine credit repayment record, which stands at about 992 mainly through the group credit activittes of the AVs. The use of the AVs as credit intermediaries has also contributed to significant coat reductions of credit administration. 3.33 All projects under review with the exception of the Livestock Project provided credit funds for on-farm investments. A striking fact is the variation in credit terms from one project to another. Farmers' shares (downpayments) in financing medium term investments varied from 50% (Mopti II) to 52 (Mali Sud II) and repayment periods for the same kind of investments, i.e. mainly draft oxen and associated equipment from 2 (Mopti TI) to 4 years (Mali Sud II). 3.34 Bearing in mind the Bank's and the government's interest in supporting draft animal mechanization as means for better and more timely land preparation, improved seeding/weeding practices as well as facilitating on-farm transportation it is difficult to understand why Mopti farmers had to face rather restrictive credit terms, high downpayments and repayment time which was less than half the economic life of the financed investment, a condition which cannot be an inducement to draft animal mechanizaticn. 3.35 Aaricultural Marketins shows best the Bank's focus on policy reforms while several sector/hsubsector constraints remain unresolved. While the Bank succeeded in convincing the government to restrict OPAM's previous trading monopoly for cereals to the role of the holder of emergency cereal stocks, inefficiencies with rice marketing and with SOMIEX cotton export marketing remain. 3.36 There are unresolved problems with export as well as domestic marketing. First and above all problems created by the difficult external and internal transportation situation. High transport costs have already contributed to Mali losing its traditional groundnut exports - for years the major foreign exchange earner - and is now, since 1989, in the process of losing also its traditional livestock export markets in the Cote d'Ivoire and Ghana due to the comparatively high transportation costs. 3.37 Some indication of the high cost of moving cattle on the hoof, the traditional way of getting cattle to consumption centers, is explained by ODEM in comparing its cost of moving cattle from Mopti to Bamako. It takes about six weeks to cover the 600km between the two towns, and taxes as well as "parallel taxes" amount to about FCFA 4,500/head in addition to weight and sometimes total losses of cattle. Transport by river barge, however costs FCFA 2,500/head and takes only 4 days to Koulikoro and another two days on-hoof to Bamako. Weight and other losses are minimal. 3.38 Expectations of cereal exports in years of production surpluses are difficult to understand under the prevailing transport cost structure. A 1986 Bank intern study comparing groundnut export cost structures of Mali and Senegal indicates that higher transport costs of Malian groundnuts add - 22 - FCFA 21.7/kg to the Dakar fob price compared with Senegalese groundnute. Only high value crops like cotton are in a position to absorb these high transportation costs but not internationally low priced commodities like coarse grains. Introduction of low cost river transportation to reach the grain importing coastal areas of West Africa would dramatically change the outlook for Mali's agricultural exports and also lower the costs of imported bulk inputs. There is navigation on long stretches of the Niger River, starting from Kankan in Guinea. Looking at the feasibility of improving river navigation may help not only Mali but also soveral other Sahelian countries. 3.39 Malian exports are further hampered by inadequate handling/storage facilities at major export outlets like Abidjan and Dakar. Ways and means have to be found to eliminate these "regional" as opposed to country specific constraints. 3.40 Internal marketing also suffers from an insufficient secondary/feeder road system. With cereal production exceeding domestic consumption during the last two years, farmers in areas of poor or no road access are unable to sell their grain. The ODIPAC area is a case in point. A feeder road project would have had more impact on agricultural development than the Bank's technical assistance project aiming at finding alternatives for groundnute. 3.41 A further weakness of the present marketing system is the lack of market intelligence. Grain dealers complain about insufficient information on quantities of cereals available at different locations. For them it is uneconomic to travel with their trucks to remote villages to find sometimes only 200 or 300kg of saleable grains. 3.42 Finally, lack of improved on-farm storage together with high farmer cash requirements at harvest time have led to "imperfect markets". While some AVe have successfully inititated, albeit still on a small scale, the construction of grain storage facilities at village level, on-farm storage should not be forgotten in the light of reducing storage losses and allowing farmers more amplitude in their sales. Rescheduling credit repayments, i.e. extension of short term loans would also permit a levelling off of sales peaks immediately following the harvest. C. Aspects of Pricing Policies 3.43 Since the beginning of its operations in the agricultural sector in Mali, the Bank has put major emphasis on correcting economic distortions of the sector, i.e. elimination of subsidies and higher producer prices to encourage production, especially for cotton and cereals. While cotton prices were kept low by the Government's need for revenue and also partly due to inefficient and costly marketing organizations, the cereal situation was different. Cereal prices at the "gray markets" offered farmers better returns than what was paid by OPAM at officially announced price levels. - 23 - 3.44 Ove time the Bank succeeded in convincing government to adjust the cotton price structure through the "filiere cotton", albeit after going through trial and error periods. Questions could be asked why cotton producer prices are only now being linked to world market prices. The first price adjustment formula, introduced under the Mali Sud II project was too complicated - taking into account price developments of inputs - and, above all, did not allow for any major price declines in the cotton markets. The 1986/87 decline of cotton prices immediately proved the weakness of the then adopted Malian cotton price structure. 3.45 The international decline in cotton prices and its impact on cotton production teach some interesting lessons. Mali, with its past reluctance to fully embrace the Bank's recommendations about raising producer prices weathered the storm of the "cotton crisis" much better than for instance Cameroon which had followed Bank advice and where farmers received high prices. The following Table 2 illustrates the cotton producer prices in West African countries. Table 2 Cotton Producer Price FCFA/kg Country 1982/83 1983/84 1984/85 1985/86 1986/87 Benin 85 100 100 110 110 Burkina Faso 62 70 90 100 100 Mali 65 75 75 85 85 Cote Ivoire 85 100 115 115 115 Senegal 70 70 70 100 100 Togo 65 75 90 105 105 Cameroon 105 117 145 145 145 RCA 70 80 90 100 100 Chad 70 80 100 100 100 Source: CFDT Resultat Analytique Fibre Compards, 1989 3.46 Despite the lower producer prices in Mali compared with Cameroon, production and average yield developments moved about in parallel between the two countries. Even with the relatively low Malian producer price, cotton offers attractive returns compared with the limited range of other crovi cultivable under Sahelian conditions. The SAR for Mali Sud I (page 65) shows cotton returns per man day domineering with Malian France (MF) 477 over maize with MF 361 and rainfed rice with MF254. Table 4 of the Mali Sud I PCR (page 13) indicates that cotton returns per hectare are 1.4 to 2.5 times higher than those from millet or rainfed rice. 3.47 In the audit's view it would be interesting to study the farmer's response to different producer prices, i.e. what would be the minimum price level required to sustain/expand production of a given crop. Bearing in mind how hamstrung countries like Mali are in their need to generate - 24 - revenuelforeign exchange and eventually repay their credit, such a study could contribute to improved country finances without jeopardizing production. The now applied formula of using a certain percentage of world market prices as appropriate and fair producer share is likely to vary from country to country, especially if cropping alternatives should exist.l/ 3.48 In agreement with the Bank's request, OPAM's marketing functions have been reduced to the agency now maintaining only emergency stocks of 58,000t of coarse grains. With annual purchases of 19,500t OPAM cannot fulfill the function of guaranteeing a floor price, a function assigned to it in the past but inefficiently dealt with. The "free" market has taken over, a free market which in, however, imperfect in several ways. First, as mentioned above, many farmers lacking proper storage and, facing repayment obligations, are compelled to sell their marketable surpluses immediately after harvest; second, private trade shows little interest in acquiring cereals from remote, not easily accessible areas in times of surplue production; third, surplus cereal production is presently not exportable without heavy subsidies. Table 3 shows the development of millet/sorghum/maize production during the past years and the surpluses above domestic requirements. TABLE 3 Coarse Cereal Production and Requirements, Deficit/Surplus (.ccop +100 ----------s T RQ.M based on 187 igWeapit +1000 .7so0 s00 +250 Prouwn 0 -250 4500- 1976 78 80 82 84 86 86 Source: OPAM Wodd Bank4797Sb [/ See OBD Report No. 2034 (1978) Benins Zou-Borgou Cotton Project. - 25 - 3.49 The surplus production has already resulted in declining producer prices. OPAM statistics indicate that the average producer price for millet/sorghum/ maize amounted to FCFA 44/kg during the period November 1988 to April 1989, the period after the harvest when producer prices are usually low. However, due to overproduction of cereals, prices declined further during the usually high price period of May to October, averaging only FCFA 39/kg and took another plunge following the good 1989 harvest with prices reaching only FCFA 33 during the November 1989 and April 1990 period. Due to the absence of proper market intelligence, the audit was unable to obtain year round producer prices at different production areas. However, Table 4 showing price developments at 13 urban centers could give an indication of the magnitude of price differences, differences which are even more pronounced at farm level. TABLE 4 Millet/Sorghum Price Development at 13 Urban Markets FCFA 180 170 - 150 / %MILLET ISO I II 110 so sone*tum 70 50- 01 3 I I I I I I I I 1 I 1 1 1 1 I NOV MAR JUL NOV MAR JL NOV MAR JUL NOV MAR NOV MAR NOV MAR JUL NOV 81 82 83 84 85 88 87 Soue OPAM Word Bark-47973a 27 - 3.50 With the relatively small number of grain dealers, market distortions through dealer agreements on purchasing areas/villages and price rigging is likely. Strengthening the position of the AVe might be a solution to counteracting exploitation by dealers. In thie context it was interesting to note that some AV& are not only active in storing and selling cereals produced by their members, but have also actively entered the cereals trade. An AV visited during the audit mission had about 32t millet/sorghum in stock, purchased from neighboring villages and expected to be sold during the summer months at a minimum profit of FCFA 35/kg, a mark-up of about 80-100. 3.51 With overproduction of cereals on the one hand and the abolition of a floor price policy for cereals on the other, questions should be asked about the fate of farmers, the rural poor outside the cotton belt in Mali's not easily accessible areas - for instance the ODIPAC region - or in areas more distant from consumption centers. The annual purchases of 19,500t of coarse grains by OPAM have no longer any producer price regulating effect. This the more since OPAM's procurement procedures provide for inviting bids from traders and awarding the buying contract to the lowest bidder, i.e. a dealer who buys and sells close to Bamako or another major consumption center thereby saving on transport cost. 3.52 The audit recognizes the high cost of maintaining cereal stocks but questions the rigidity of the system now practiced. Based on the above graph, Table 3, the most serious recent shortfall in local grain production occurred in 1980 when a deficit of 400,000t for the given year materialized. With a daily requirement of 1,090t the OPAM stock would last for 54 days. Taking into account Mali's population growth of 2.6% p.a. the emergency stocks need to be increased by 1,520t annually to cover the 54 days period. 3.53 The reaction of the Malian farmer to continuously declining grain prices cannot yet be foreseen. It is possible that a grain cycle similar to the classical hog cycle may develop, fluctuating maize production seems to point in that direction. On the other hand more farmers may revert back to subsistence production and enlarge the number of the rural poor. 3.54 Exports of surplus production will require heavy government subsidies in the light of the above mentioned transportation problems. With US sorghum prices of FCFA 27/kg fob Gulf port on the one hand and transport costs of FCFA 50/kg to the coast, even the frequently requested devaluation of the FCFA would not ease the export situation. D. Fostering Self-Help 3.55 All projects under review provided assistance for functional literacy programs, a rather unique undertaking for the Bank. The programs' intention was and still is to teach volunteering farmers and also their wives, reading and writing as well as simple arithmetic. But there is a more interesting aspect to this effort. - 28 - 3.56 Past efforts to establish agricultural cooperatives in Africa have been largely met with failure. Africans preferred the establishment of small, easily overseeable units where everybody knew each other. Problems of this set-up usually started when the illiterate members of the cooperative ha' to employ the services of a literate cooperative secretary. Due to the modest turnover of these mini cooperatives, salaries paid to secretaries were low and did not offer acceptable income to the employee. As a consequence embezzlement of cooperative funds frequently occurred resulting in a complete loss of farmers' confidence in cooperative ventures. 3.57 From the start of the functional literacy program in Mali, it was expected that participating farmers would be in a good position - after graduation - to actively contribute to the managemenx ;f the AVs then in various stages of establishment. The Bank can take credit for having recognized the eventual benefits of the alphabetization program and for supporting it through all agricultural projects financed by it. the only question to be asked in connection with this program is why has the Bank not yet promoted similar undertakings in other countries. 3.58 As mentioned above, one possible result expected from the literacy programs was the hoped for impact on self-help, through the AVe. Main purpose of the AV or AP is to function as a kind of pre-cooperative, i.e. take care of certain functions which in the past had been at the charge of government. Distribution of inpute, group credit, collection/grading of agricultural produce for the market, storage of inputs/marketable produce, some processing activities such as maize mills, small scale bottom lands developments, simple soil erosion control works, maintenance of tracks and fire breake, reforestation, village/livestock water supplies, etc. All these activities were previously the responsibility of the government or parastatals at considerable cost. Incentives in the form of receipts for input distribution or cot--.. collection, grading and storage have contributed to strengthen the villagers' self-help spirit. 3.59 Most impressive are the activities of the AVs in the cotton producing Mali Sud area. By the end of 1988 AVs existed in 1,119 villages or about 30% of all the region's hamlets. However, these AVe marketed about 70% of the cotton produced in Kali. Agricultural credit amounting to FCFA 3.5 billion has been channeled through the AVs and 99.9% has been recovered - a more than unique accomplishment. About FCFA 500 million of earned marketing fees were reinvested in 1986 alone. 3.60 Village self-help has been further supported by the Bank-assisted projects through financing the training of blacksmithe who are essential for the maintenance and repair of agricultural machine-.y introduced under the draft animal mechanization scheme. The financed young farmer training programs have also been successful. Graduates will take on simple extension functions by acting as intermediaries between farmers and CNDT staff, thus permitting CMDT to gvadually reduce its extension input. - 29 - 3.61 The Livestock Project emphasized the creation of APs. Little , progress vas made during project implementation due to differing views between project management, insisting that assured water supplies had to be available prior to the establishment of APs to work as an incentive for herdsmen to join the associations and the Bank insisting that only APs having agreed on limiting their herd numbers should benefit from the project financed investments. Due to the unsuccessful drilling program the dispute had merely academic character. 3.62 It was only after project completion and under a different project management, that APs were founded. However, these APs have not yet been put to the crucial test, i.e. limiting their herds to prevent overstocking and consequently overgrazing. The droughts of the seventies and eighties contributed to herd decimations and up to now pastoralists were stocking up their herds to pre-drought levels. 3.63 There are several factors that give reason for concern about future herd development and the APs sticking to the agreements limiting the size of their herds. The Livestock project was rather successful in its animal health component.2/ 3.64 The vaccination campaign against Rinderpest and other diseases has contributed to a substantial reduction of adult cattle losses. In addition, veterinary work, supplementary feeding, etc. have also led to a reduction of calf mortality. As a result, off-take rates which were around 9-11% of the herde annually should be increased to prevent overstocking. In the light of re;ant difficulties confronting cattle exports, it remains to be seen if APs will be in a position to control the herds of their members and prevent overstocking/overgrazing. E. Bank Reporting and Filing 3.65 The Bank's supervision reports have always been important internal documents which permitted Bank staff to communicate to management at various levels the problems encountered in project implementation, procurement problems, difficulties with project staff, etc., and all in a most candid manner because of the reports' restricted circulation. Despite the fact that supervision reports' are of a highly confident nature, at least one report (Livestock I) was freely distributed among Malian officials of various Ministries and State Agencies. This led to difficulties with the Minister concerned and strained Bank-Borrower relations. In addition it turned out that the distribution was completely counterproductive to steps taken and action recommended by the Supervision Mission. 3.66 The format of supervision reports in the more recent project (Mali Sud II) does not conform with normal Bank standards. Supervision reports 2/ J.C. Deveze, Un Example de Developpement Rural d'une Region a Partir d'une Filiere Porteuse: Mali Sud et l'Action de la CMDT. - 30 - of this specific project consist only of covering standard form 590 which is prepared in English and which is attached to an aide memoire prepared in the field and therefore by necessity entirely in French. No "Key Indicators", Disbursement Schedules, Paragraph for the President's Report, or Lists of people met and places visited are given. 3.67 The use of a French aide memoire as the main body of the supervision report raises several questions. First, it will be impossible to include confidential information/considerations in a document which will be signed and immediately released to officials of the Borrower. Second, is thus the Supervision Mission fulfilling its obligation to completely inform managerial bank staff on project progress/problems if it is evident from the distribution list that several recipients of the supervision reports have no command of French. 3.68 Third, does managerial staff fulfill its supervision function when it is not requesting reports in English when not understanding French. 3.69 The Mali Sud II project gives rise to another concern. Despite being an exceptionally quick disbursing - no problem project, implemented during a planned six year period Bank files on the project fill 23 folders. The presently practiced filing system - a euphemism - is inadequate and costly in terms of manpower and materials input. There is no guarantee/control possibility that all incoming mail is filed since no logging is being done. The same unsatisfactory situation exists with outgoing correspondence. 3.70 In the audit view time savings could be realized if separate operational, procurement, and disbursement folders/files were to be established. Alone for the preparation of the PCR as well as the Audit Report many hours were required to wade through the volumes of not always pertinent correspondence, for instance copies of cables confirming procurement decisions, requests for fielding supervision missions, usually copies to several Ministries, agencies, officials of project entities, etc. 3.71 Files could be further slimmed down if filing clerks were to be better trained. At present, one finds frequently several copies of the same memo/letter, due to some recipients of copies pass them on to files after reading them where they are dutifully added to the folders without checking if they are not already filed. - 31 - Attachment I Comments from the Ministry of Planning Page 1 and International Cooneration REPUBLIC OF MALI MINISTRY OF PLANNING AND INTERNATIONAL COOPERATION Office of the Minister Bamako Mr. Graham Donaldson May 15, 1991 Division Chief (Agriculture, Infrastructure and Human Resources) Operations Evaluation Department World Bank Washington, DC Ref.: P[?]00318 MPCI/CAB RE: Livestock Project (538-MLI), Mali-Sud Projects (669-MLI and 1415-MLI). Second Mopti Rice Project (753-MLI). and ODIPAC Technical Assistance Project (1174-MLI) In response to your letter of March 29, 1991, I would like to make the following observations and comments on the above Projects: The Livestock Project: For various reasons, this Project did not achieve all its objectives. Nevertheless, the results of the livestock vaccination campaign and the functional literacy program were very satisfactory. The Project also increased ODEM's capacity for providing support for pastoralists. The following lessons can be learned: - The Bank should ensure that local specialists are involved to a greater degree in project preparation. On the Malian side, the Project suffered from the many personnel changes and the delays in the establishment of a management structure. As regards the Bank's performance, there was a lack of the sort of monitoring that would have certainly made it possible to adopt the necessary corrective measures. - The diagnostic study performed as part of project preparation should have taken account of all aspects of the Project and established priorities for the various components. As a result, insufficient attention was given to certain important items (such as marketing), and others (such as the trial station) were inadequately planned. In the latter case, it might have been better to select a number of pastoralists for a pilot project, - 32 - Attachment I Page 2 providing them with sufficient support to enable them to develop successful model farms. Institutional issues are very important in project execution, and in future the Bank and Mali should give particular attention to these. Overall, the activities implemented under this Project should be consolidated, and extended to other regions of Mali. The Mali-Sud Agricultural Project and the Second Mali-Sud Rural Development Project: These Projects were successful, and the results were very satisfactory. Seed selection, applied research, and the agricultural credit system should be improved. There appears to be considerable justification for the proposal that prices be indexed (Phase TI of the Project). - Good organization always provides a very sound basis for project execution. - The World Bank and the international community should give greater thought to the problem of stabilizing commodity prices, since price instability is a serious threat to development activities in developing countries. On several occasions, falling prices have thwarted the efforts made by those countries and by the World Bank. In the case of this Project, it should be noted that the Guarantee Fund for cotton is severely limited. The Second Mopti Rice Project: The outcome of this Project was very unsatisfactory, and it suffered from cost increases. In our opinion, these shortcomings were the result of overoptimistic assumptions regarding flood water levels, and problems relating to research, extension and the credit system. The ODIPAC Project: Likewise, this Project was not a great success. It suffered from a number of shortcomings that some observers have traced to the original planning stage. Assumptions regarding groundnut prices were overoptimistic, and some research and extension activities fell short of expectations. Nevertheless, the World Bank and the Government of Mali should continue to seek an appropriate strategy for providing greater assistance for the communities in the project area. Sincerely, /s/ Denis Traore On behalf of the Minister of Planning and International Cooperation, and by order of the Director of the Office of the Minister