Malawi - Shire Valley Agricultural Consolidation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 6171 PROJECT PERFORMANCE AUDIT REPORT MALAWI SHIRE VALLEY AGRICULTURAL CONSOLIDATION PROJECT (CREDIT 823-MAI) May 5, 1986 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only In the performance of their official duties. It! contents may not otherwise be disclosed without World Bank authorization. ABBREVIATIONS ADD Agricultural Development Division ADMARC Agricultural Development and Marketing Corporation CADO Chief Agricultural Development Officer CCDP Chikwawa Cotton Development Project CTO Chief Technical Officer DA Development Assistant DTC Day Training Centre EAPSA Eastern and Southern Africa Projects Department EARSA Southern Agriculture Section of RMESA EPA Extension Planning Area ERR Economic Rate of Return FA Field Assistant FMD Foot and Mouth Disease GOM Government of Malawi IBRD International Bank for Reconstruction and Development IDA International Development Association IRR Internal Rate of Return MANR Ministry of Agriculture and Natural Resources MK Malawi Kwacha MG Malawi Government NADD Ngabu Agricultural Development Division NRDP National Rural Development Program OED Operatione Evaluation Department PCR Project Completion Report PM Program Manager PO Professional Officer RMESA Regional Mission in Eastern and Southern Africa RTC Residential Training Centre SAR Staff Appraisal Report SVACP Shire Valley Agricultural Consolidation Project TA Technical Assistant T&V Training and Visit System of Extension Currency Exchange Rates Appraisal Year Average (1977) US$1.00 = MK 0.91 Intervening Years Average (1978-81) US$1.00 - MK 0.93 Completion Year Average (1982) US$1.00 = MK 1.07 Fiscal Year of Borrower April 1 - March 31 THE WORLD BANK FOR OFFICIAL USE ONLY Washington, D.C. 20433 U.S.A. Ofke of Director-Cmneaal Operations Evaluation May 5, 1986 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report - Malawi Shire Valley Agricultural Consolidation Project (Credit 823-MAI) Attached, for information, is a copy of a report entitled "Project Performance Audit Report - Malavi Shire Valley Agricultural Consolidation Project (Credit 823-MAI)" prepared by the Operltions Evaluation Department. Attachment 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. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT MALAWI SHIRE VALY AGRICULTURAL CONSOLIDATION PROJECT (CREDIT 823--MAI) TABLE OF CONTENTS Page No. Preface *e***.*******ooo*** ***** o****o **o**** **** 1 Basic Data Sheet ooooo#ooo#ooo*ooooo*oooooo It Evaluation Summary .................................................. I PROJECT PERFORMANCE AUDIT MEMORANDUM I. PROJECT SUmARY ............................................ 1 Project Area ....o........................................ I Project Description *.*************************** 1 Project Cost 2........................................... Production Services ..................................... 3 Natural Resources o********e************************ 4 Hydrology and Water Supply ............................... 5 Health Services *......................................... 5 Technical Services ...................................... 5 Project Administration o......*.......*....*..**........ 6 Economic Rate of Return .................................. 6 II. MAIN ISSUES .........*.........o........................... 7 A. Achievement of Objectives .............................. 7 B. Appropriateness of Project Design ...................... 7 C. Replicability .................0**00.000*0......... 8 D. Sustainability o...o................................... 9 Table 1: Rural Development Projects in Malavi ..................... 11 PCR OVERVIEW I. Introduction ...........................*..*..* oeo**o . 15 11. Project Objectives ......................................... 15 III. Physical Implementation ..............**...............*.... 15 IV. Project Management ......................................... 16 V. Services Supported under the Project ....................... 16 VI. Project Costs, Financing and Disbursement .................. 19 VII. Project Benefits *....................*.................... 22 VIII. Government Performance ............*........................ 23 IX. IDA Performance ............................................ 24 Xs Lessons Learned ........................................... 25 This document hes a rtricted distribution and may be used by recipiena only in the peformmnce of their official duties. Its contents may not otherwise be disclosed without World Bank authoriution. TABLE OF CONTENTS (cont'd) Page No. GOVERNMENT PCR I. Introduction and Summary ................................. 29 HIe Background ................................................. 30 III. Project Formulation ........................................ 31 IV. Implementation ............... ........................ 35 V. Financial Performance .................................... 48 VI. Impact ..................................................... 54 VII. Institutional Performance ................................. 58 VIII. Economic Benefits .......................................... 60 IX. D:scussion and Conclusion .................................. 63 Appendix 1: Revised Economic Analysis Appendix 2: Statistical Tables (1-10) Chart: SVACP Organizational Chart !!L IBRD No. 13348 PROJECT PERFORMANCE AUDIT REPORT MALAWI SHIRE VALLEY AGRICULTURAL CONSOLIDATION PROJECT (CREDIT 823-MAI) PREFACE This is a Project Performance Audit Report (PPAR) on the Shire Valley Agricultural Consolidation Project for which a credit of US$10.7 mil- lion was approved in June 1978. The credit Closing Date of December 31, 1982 did not need to be extended and final disbursements were made on January 7, 1983. The project under review forms part of a 15-year development effort by the Government of Malawi in the Lower Shire Valley which was supported by three consecutive IDA credits (Credits 114, 363 and 823-MAI). An Impact Evaluation Report on the three-phase program was issued by OED on Decem- ber 23, 1983 (Report No. 4850). The audit report consists of a Project Performance Audit Memorandum (PPAM) prepared by the Operations Evaluation Department (OED), a Project Com- pletion Report (PCR) and a Revised Economic Analysis, both prepared by the Borrower's Ministry of Agriculture, and an Overview dated December 2, 1985, prepared by the Bank's Regional Mission in Eastern and Southern Africa (RMESA), which summarizes and cor.ments on the Government's PCR. The audit memoraadum is based on a review of the Appraisal Report (1945-MAI) dated May 25, 1978, the President's Report (P-2302-MAI) dated May 24, 1978 and the Credit Agreement of July 7, 1978. Internal Bank memo- randa on project issues and correspondence with the Borrower, as contrined in relevant Bank files, have been studied, and Bank staff associated with the project have been interviewed. An ORD mission visited Malawi in July 1985. Discussions were held in Lilongwe with Government officials and a field visit made to the Shire Valley to inspect project works and discuss project achievements and problems with field staff of participating agencies, local officials and farmers. The PPAM supports the principal conclusions of the PCR and Revised Economic Analysis, and the PCR Overview. It summarizes the objectives and results of the project and discusses issues concerning the project design, factors contributing to implementation delays, and experience with develop- ment ad operations. Following customary OED procedures, the draft report was sent to the Borrower for comments on January 30, 1986. No comments were received. The audit mission acknowledges gratefully the assistance provided by officials of the Ministry of Agriculture, cooperating agencies, project field staff and farmers. - it - PROJECT PERFORMANCB MIT BASIC DATA SMTST xAAis tiniR VALLEY AGRICULTURAL CONSOLIDATION PROJECT (CRBDIT 823-MAI) EST PROJECT DATA Appraisal Actual or Actual as 2 of Estimate lsttatee Actual Appraisal Estimate Project Costs (1S$ million) 12.6 11.6 92 Credit Amount 10.7 10.7 100 Date Board Approval 06/06/78 - Date Credit Agreement 07/07/78 Date ffectiveness - 9/78 - Date Physical Components Completed 3/82 12/82 116 Houses (No.) 169 80 47 EPA Headquarters (No.) 5 5 100 wealth Facilities (No.) 6 6 100 Boreholes and shallow tubewells (No.) 40 32 85 Roads (ga) 139 111 80 Training Centers (No.) 6 4 67 Smcking Kila (No.) 60 60 100 Fish Landing Sites (No.) 4 4 100 Village Centers (No.) 50 0 0 Dip Takso (No.) 2 2 100 Irrig. Seed Fare (Ha) 150 0 0 Closing Date 12/31/82 12/31/82 - Economic late of Return (2) 20 8 40 Livestock 18 n.e. Fisheries 44 a. Institutional Performancez Poor management; high staff t-trnover throughout the project; procurement delays In the first year of Implementation Agronomic Performaneat Variable due to climtic conditions and price policies. Farmers' profit unsatisfactory for cottta and groundanuts. Farmers reacted favorably to price increases for aefte. Ouar beans Introduced with success. Insufficient development of new extension messages. Financial Performsa t Budget allocations generally satisfactory, but with some excep.ilons. Direct Senaficiariest About 7,000 farmers benefited from group seasonal credit provided under the project. About 35,000 farmers representing 502 of the farming community had direct contact with extension. Others benefited from roads, water supply and health facilities constructed under the project. CIMULATIVE DISBURSEMENTS TY79 FY80 F81 FY82 FY83 Appraisal estimate (US$ million) 4.5 7.0 9.1 10.7 10.7 Actual (US$ million) 2.3 5.7 8.1 9.9 10.7 Actual as I of estimate 51.5 81.4 89.0 92.5 100 Date of Final Disbursemeatt January 7, 1983 STAFF INPUTS /a FY77 FT78 FT79 FY80 7(81 FY82 FT83 FY84 PY85 Total Identification/Preparation (Staffeeks) 15.7 2.8 18.5 Appraisal (Staffveeks) 111.1 111.1 Negotiations (Statfueeks) 4.6 4.6 Supervision (Staffweeks) 0.3 8.6 13.4 21.7 4.8 3.7 2.7 2.0 57.2 Total 15.7 118.8 8.6 13.4 21.7 ;.8 3.7 2.7 2.0 191.4 MISSION DATA Month/ No. of Staffdsys Specialisations Performance Types of ission Year Persons in Field Represented R /ting b TreAd /c Problems /d Superviston I 12/78 2 10 Ag.1con./Agr. 2 - T Supervision II 07/79 1 2.5 Ag.8con. 2 2 MF Supervision III 05/80 2 5 Ag.lcou./Ftn.An. 2 2 PM Supervision IV 01/81 2 20 Agr./Fin.An. 2 3 PHT Supervision V 06/81 2 20 Agr./Ag.Scon. 2 2 T Supervision VI 02/82 2 15 Agr./Fia.An. 2 1 T Supervision VII A8/82 1 10 Agronomist 2 2 TN TOTAL 82.5 OTHER PROJECT DATA Borrower Government of Salawt Executing Agency Ministry of Agriculture Follow-on Project None a Source: Planning and Budgeting Department. 1 * problem-free or minor problems; 2 - moderate problems; and 3 * major problem. 1 - Improving; 2 - stationary; and 3 - deteriorating. T - technical; F - financial; and M - managerial. - III - PROJECT PERFORMANCE AUDIT REPORT MALAWI SHIRE VALLEY AGRICULTURAL CONSOLIDATION PROJECT (CREDIT 823-MAI) EVALUATION SUMMARY Introduction and Objectives This project was the third puase of a 15-year development effort by the Government of Malawi in the Lover Shire Valley supported by IDA. The first project (Credit 114-MAI for US$3.7 million) focused mainly on increas- ing cotton production by smallholders. It was considered successful, and during its implementation the Bank appraised a second project (Credit 363-MAI for US$10.5 million) in 1973. The second project retained cotton production as its central component but widened its scope to include promotion of food- crops and livestock development, and enlarged its coverage to include the entire valley. The third credit (823-MAI for US$10.7 million), the subject of this audit, was aimed at consolidating the earlier Shire Valley invest- ments through continued support for development of rainfed agriculture, livestock, fisheries, land husbandry, and forestry, and for drinking water, health and road services. The last disbursement was made in January 1983 and the amended Project Completion Report and Overview was received in OED in December 1985. Implementation Experience The project reached appraisal estimates in the establishment of administrative blocks for extension and the construction of health facili- ties, markets, laboratories, and fish landing sites. Construction of staff houses, roads and boreholes fell short of the appraisal targets due mainly to local funding problems, cost escalation and breakdown of equipment. The extension and research services under this, as well as the two previous Shire Valley development projects, had only limited success in gene- rating new technologies and farming systems which could be adopted by the majority of the farmers and which could have assisted in counteracting the adverse climatic conditions in the valley. During the implementation period covered by the PCR, the Govern- ment-set prices for cotton and groundnuts provided inadequate incentives to farmers. However, for the past two years (1983/84 and 1984/85) better price incentives have been provided for cotton and this, combined with more favor- able rainfall and lifting of the legal constraint to estate production of cotton, has resulted in a considerable revival of cotton production in the Valley (PPAM para. 28). - iv - Results The Government reestimate of the economic rate of return for the project is 8%, well below the appraisal estimate of 20% (PCR Appendix 1). However, the uncertainty over future price prospects and the un-reliability of available data make any calculation only indicative (PPAM paras. 26-28). Sustainability At the time of audit, two and one-half years after the final disbursement under Credit 823-MAI was made, the activities initiated under the project were being largely maintained but without new capital investment. Seasonal credit is expanding at a moderate pace, now that group responsibility procedures have been introduced. Input supply and crop purchase is under the responsibility of ADMARC and technical *upervision is provided by the Agricultural Development Division. Annual recovery rates have averaged above 88% over the past seven years (P7AM para. 12 and PCR para. 63). The Bank has no immediate plans for a direct follow-up investment in the Lower Shire Valley. The irrigation study fin"nced under Credit 823-MAI has been completed, but no decision has been taken as to whether an irrigation investment in the Shire Valley is merited at this time. Nevertheless, it appears that the experience gained in the Lower Shire Valley Projects has contributed to at least four changes in Government policy and in Bank support: (i) a less intensve but nationwide structure of Agricultural Development Divisions (ADDs) including one for the Lower Shire Valley supported by the Bank's National Rural Developent Program Projects as well as other donors; (ii) a nationwide project to strengthen agricultural research (Credit 1549-MAI for US$23.8 million approved February 19, 1985); (iii) a nationwide project to support agricultural extension (Credit 1626-MAI for US$11.6 million approved September 19, 1985); and (iv) a national health project (Credit 1351-MAI for US$6.8 million approved April 26, 1983) which now provides some of the suppor* for health centers that was provided earlier under the Shire Valley projects (PPAM para, 38). Other Findings Important lessons and findings include: (a) The project had an important impact in providing increased social services--wells for drinking water, health clinics, schools, roads and markets--of which the wells were especially appreciated by the villagers. Maintenance of these wells remains a problem, however, and consideration should be given to orcanizing beneficiary groups and providing training for this purpose (PPAM Paras. 19-21; PCR Overview paras. 18-19; and PCR paras. 58, 71). -v - (b) The expected increases in agricultural production did not material- ize, partly because of drought, but more substantially because of the lack of improved technology and inadequate price incentives (PPAM para. 28; PCR Overview paras. 24-30; and PCR paras. 89-91). (c) For cotton production, the main cash crop, price incentives appear to have been the major factor influencing both the area planted and the use of purchased inputs such as pesticides and fertilizers (PPAM para. 34 and PCR Overview paras. 28-29). (d) Extension efforts had limited success, due in part to the few adequately proven and profitable technological innovations avail- able for the variable soil and rainfall conditions of the Lower Shire Valley (PCR Overview para. 7 and PCR paras. 33-37). (e) Obtaining useful research results under rainfed conditions was a tedious and time-consuming process, and especially frustrating because of the extreme variability in rainfall. Such research also required considerable expertise and was difficult to supervise adequately in a multi-component project in which research was only a mir. tategory (PPAM para. 10; PCR Overview para. 10; and PCR paras j-47). (f) The lack of a full-scale program of technical and leadership train- ing in the earlier phasqs was a limitation for this consolidation phase. Little progress had been made in gradually replacing expat- riate staff with experienced local staff which also resulted in a considerable cost overrun for international staff (PPAM para. 9; PCR Overview paras. 7-8; and PCR paras. 30-32). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM MALAWI SHIRE VALLEY AGRICULTURAL CONSOLIDATION PROJECT (CREDIT 823-MAI) I. PROJECT SUMMARY Project Area 1. The Lower Shire Valley has some 300,000 people, the majority small farmers without mechanization who produce subsistence crops and cotton by hand and in some cases with the help of ox-drawn equipment. They grow mainly maize, cassava, millet and groundnuts, primarily for home consumption, and cotton as a cash crop. At appraisal, average annual income per smallholder family was estimated at US$75. 2. The Valley covers 672,000 ha at the southernmost tip of Malawi. The climate is harsh in agricultural terms. The Valley is only 60 meters above sea level and temperatures can reach 48* C. Rains occur primarily between November and April with heavy downpours likely from December to February which sometimes cause water logging on flat land near the marshes. There is large variation in rainfall between micro regions, between years, and in the distribution within each year. The yearly rainfall over the past 25 years has varied from drought conditions of 538 mm up to 1199 mm. Thus, farmers contend with weather risk each year. In spite of this, conditions are generally adequate to obtain some harvest of rainfed cotton even in the poorest years. The Valley's extensive marshes and riverside flats are good grazing areas and support more than 80,000 head of cattle and a large goat population. 3. Although a paved highway now connects the Valley to the rest of the country, Malawi does not have its own access to the sea and the route through Mozambique has been essentially closed since 1982. The high transport costs via alternative routes have a severe impact on availability of inputs and on production and exports. Project Description 4. The Shire Valley Agricultural Consolidation Project was the third phase of a 15-year development effort by the Government of Malawi in the Lower Shire Valley.j The first project (Credit 114-MAI for US$3.7 million) 1/ See also: Impact Evaluation Report, Malawi Lower Shire Valley Agricul- tural Development Project (Credits 114, 363, and 823-MAI), OED Report No. 4850 dated December 23, 1983. -2- focused mainly on increasing cotton production by smallholders, It was considered successful,2/ and during its implementation the Bank appraised a second project (Credit 363-MAI for US$10.5 million) which was approved in early 1973. The second project retained cotton production as its central component but widened its scope to include promotion of fooderops and livestock development, and enlarged its coverage -to include the entire Valley. / The third credit (823-MAI for US$10.7 million), the subject of this audit, was designed to assist the further development of Malawi's Shire Valley region and to consolidate the agricultural and infrastructural investments made in the region since 1968. It was intended to intensify the Provision of agricultural extension services; establish an irrigated seed multiplication farm and an afforestation program; provide production credit; improve livestock extension services; expand the fisheries development program; improve village health facilities and potable water supply; and extead farm to market road networks. Project Cost 5. At appraisal, the project was estimated to cost US$12.6 million and to be implemented in four years. The IDA Credit of US$10.7 million would finance 85% of project costs including all foreign exchange (US$5.8 million) and 72% (US$4.9 million) of local costs. In fact, the Credit was fully disbursed early in the fifth year. However, cost overruns in certain categories required reallocation of funds. Although the unit costs of houses, roads and boreholes were higher than the appraisal estimates, the civil works category was kept within the original allocation. As a result, not all of the houses, roads and boreholes were built (for details see Basic Data Sheet). At appraisal it was anticipated that most of the expatriate staff required to initiate the project would be gradually replaced by local staff. This did not happen,/ probably reflecting both too optimistic an assumption and too little effective training. In any case, the result was a considerable cost overrun in the international staff/disbursement category. Funding for operating costs also exceeded the appraisal estimate. Project Impact 6. The project had an important impact in providing increased social services: wells, health clinica and roads of which especially the wells were greatly appreciated by villagers. However, the expected production increase 2/ PPAR, Malawi Shire Valley Agricultural Development Project Phase I (Credit 114-MAI), OED Report No. 895 dated October 22, 1975. 3/ PPAR, Malawi Shire Valley Agricultural Development Project Phase II (Credit 363-MAI), OED Report No. 2593 dated June 29, 1979. 4/ An earlier PPAR also commented on this issue: "The major shortfall of this program is that despite the progress made, it d.Ld not go far enough in lessening Malawi's dependence on outside technical assistance and in contributing to the creation in Malawi of an internally generatel, self-sustained development process." Highlights, PPAR Malawi Lilongwe Land Development Program Phase III (Credit 550-MAI) OED Report No. 3414 dated April 7, 1981. - 3 - did not materialize. The implementation of various services in support of crop production was only partially successful and generated only minor impact. Production Services 7. In order to increase crop production the project provided support for extension services, crop research, training facilities ane equipment, farmer and staff training, credit administration, seed multiplication and animal husbandry. At appraisal these production services were to have accounted for 54% of project cost. In practice they accounted for less than 30%. 8. Extension expenditures were only slightly below appraisal esti- mates. However, extension efforts had limited success, mainly because there were few adequately proven and financially attractive technological innova- tions to recommend with respect to drought-resistant crop varieties and soil management practices./ 9. Training used only 70% of the amount projected at appraisal, large- ly because of a Government decision not to construct 50 Village Training Units. Farmers and their wives were trained instead at Residential Training Centers and no information is available on numbers trained. It appears that training of extension workers was sporadie as was technical support from subject matter specialists. 10. Research funding was about half of the expected and the program suffered from a lack of qualified staff and high staff turnover. Some variety trials were carried out on cotton, sorghum, millet and food legumes in the context of the National Trials Program. From these a new cotton variety and a drought-resistent sorghum variety were selected, recommended for the Valley toward the end of the project, and appeared to be acceptable to the farmers. The crop/weather plots, established to determine best adapted crops for different ecological areas of the Valley, were abandoned because of financial constraints. Trials with vegetable crops were not carried out. The fundamental weakness of the research program was the under- lying concept that quick solutions could be found on which an improved tech- nological package could be designed. Experience has shown that obtaining 5/ Similar findings have been included in other audits. See, for example, PPAR, Togo Maritime Region Rural Development Project (Credit 638-TO), OED Report No. 5137 dated June 18, 1984. The poor results of that proj- ect with regard to crop and farm development were inter alia due to rainfall extremes, failure of smallholders to widely adopt the technical packages proposed, absence of basic data, and lack of any demonstration effect. The PPAR of the Senegal Sine Saloum Agricultural Development Project (Credit 649-SE/Loan 1113-SE), OED Report No. 5171 dated June 29, 1984, also noted that technical packages were not adapted to fit far- mers' resources of land, labor and capital. results and carrying out adequate field testing for rainfed agriculture is a tedious and time-consuming process, especially frustrating if there is extreme variability in rainfall. Such research also required considerable expertise and was difficult to supervise adequately in a multicomponent proj- ect in which research was only a minor category. / 11. The seed multiplication component was abandoned in 1980 after eight staff houses, an office block and a workshop had been built. There- after, seed supply was entrusted to the National Seed Company. 12. Credit uptake by farmers was significantly below appraisal esti- mates, partly reflecting the low Government-set cotton prices. The number of beneficiaries per year gradually declined from 1978-79 through 1982-83, partly as a result of tightening-up of lending policies whereby defaulting individuals or groups became ineligible for new credit. Credit use increased again in the 1983-85 period in response to better cotton prices. During this time there has been a shift from individual to group credit with some improvement in the recovery rate which has averaged above 88% over the past seven years. 13. Livestock services utilized two-thirds of the funds allocated for this category. The animal husbandry and livestock extension section remained understaffed during the project period. Staff transport was insufficient to properly supervise activities. Although the cattle population increased by 26%, the annual off-take averaged only 7%, well below the appraisal estimate of 12% per annum. The restriction of cattle transfer out of the Valley as a measure to control foot and mouth disease (FMD) is a major reason for the population increase. Overgrazing has become a problem, and with cattle popu- lation in the Valley now well over 80,000, this requires attention. The appraisal indicated that under the project an FMD vaccination program would be initiated by the Veterinary Department with technical guidance from the Animal Virus Research Institute of Pirbright, UK. The project was to finance purchase of vehicles, field equipment, FMD vaccine and vehicle operating costs. The PCR indicates that this was only partially successful with 79% of cattle vaccinated for FMD in 1980 and 731 in 1981 (PCR Overview paras. 48-50). Natural Resources 14. The natural resources component consisted of a group of activities in land husbandry, fisheries, forestry and wildlife, which together accounted for about 6% of project costs, somewhat less than the 101 estimated at appraisal. 6/ On the topic of relatively small research components, see also PPAR, Ethiopia Coffee Processing Project (Credit 290-ET), OED Report No. 3976 dated June 21, 1982. -5- 15. Under land husbandry, land use maps were produced for each Exten- sion Planning Area (EPA) on a scale of 1:50,000. About 40 farm plans were also prepared each year with particular emphasis on contour ridging. 16. Only 23% of the estimated funds for forestry were used and very little was accomplished on reforestation. 17. The wildlife conservation activities for the Mwabvi and Lengwe game reserves, initiated under the second phase project, and the planned invest- ments in demarcation and construction of fences and ditches as well as provi- sion of housing and vehicles, were completed as planned. Funding and opera- tion of the Lengwe National Park were taken over by the parent department in 1981. 18. The fisheries work initiated in Phase II of the Shire program was only partially completed in Phase III. Landing sites and improved smoking kilns were constructed and training in new techniques was provided for fishermen. Credit facilities also enabled fishermen to buy improved gear and plank canoes. However, the planned investigation to determine unexploited fish stock and the most effective methods of fishing under present conditions of relatively stable water flow were not conducted partly because staffing of the program was not completed. Hydrology and Water Supply 19. An experienced hydrologist was recruited as planned to conduct hydrology studies, supervise the construction of boreholes and to assist in designing a master plan for water development. About 80% of the planned boreholes and shallow wells were drilled and pumps installed. However, the contract of the hydrologist expired before a master plan had been completed. The National and Shire Valley Irrigation Study was contracted and carried out by an international firm and the report has been presented to the Government 20. Maintenance of the boreholes and shallow wells remains a problem and measures should be taken to train and structure beneficiary participation for this task. Health Services 21. The small public health component was staffed and managed by the Ministry of Health under supervision of its Health Officer at Ngabu. The project assisted with construction of two health subcenters and four health posts as well as renovation of a public health office. It was also involved in bilharzia control, a school vaccination program and provision of furniture and transport to the Shire Valley Tuberculosis Project. Technical Services 22. Of the 139 km of roads planned, 111 km were either constructed or upgraded under the project. An increase in construction costs, breakdown of equipment and lack of spare parts hindered completion of this component. The - 6 - costs of mechanical maintenance were more than triple the appraisal estimates# Project Administration 23. At appraisal, administration costs were expected to account for 16% of project costs. In practice, these costs accounted for more than 30%. Although costs of the evaluation unit were only 73% of appraisal estimates, the costs of the finance division quadrupled and those of project management nearly doubled. In spite of this, intermittent vacaIcies of senior posts occurred throughout Phase III, and at the end of the project only 40% of the senior staff positions were filled. This contrasts with overstaffing of some junior grades. 24. Difficulties in reaching agricultural production targets and unat- tractiveness of staff assignments to the Shire Valley, along with uncertain promotion prospects for technical and supervisory staff, contributed to high turnover and low morale, and to management and operational problems. 25. The existing evaluation unit from Phase II was continued under the project. Several surveys were carried out. However, there were weaknesses in survey design, and the data when analyzed were of limited assistance to management in decision-making and forward planning. As a result the PCR suf- fered from lack of reliable and consistent agricultural production data. Economic Rate of keturn 26. At appraisal, it was estimated that the economic rate of return (ERR) for the project would be 20%. After completion, Government submitted a Project Completion Report (PCR) indicating a re-estimated ERR of 15%. How- ever, this was not adequately documented and the audit requested clarifica- tion from EAPSA on several points related to methodology and assumptions used by GOM in the ERR calculations. 27. EAPSA, in turn, pursued this with GOM Ministry of Agriculture and found that the assumptions underlying the production cost, financial and economic data could not be reconstructed because the officer involved and the back-up papers were no longer available. The Ministry agreed to redo the calculations, and this was sent to OED by EAPSA just prior to the audit field mission (see PCR Appendix 1). 28. This recalculation spells out more clearly the methodology and assumptions used for analysis and results in an ERR of 8%. Overall, these assumptions appear reasonable. However, the definition of benefits remains somewhat conjectural; the empirical data cover only four years and there is some concern about the statistical sample and validity of the data. For example, in the cotton production data, a key item in the ERR, the PCR shows "with project" production of 16,451 tons in 1981-82. This was a year with very low rainfall in which ADMARC, the sole buying firm, purchased only 7,456 tons according to data obtained at Ngabu by the audit mission. The following year, 1982-83, was slightly better with ADMARC purchases at about 8,500 - 7 - tone according to data obtained at Ngabu by the audit mission. The following year, 1982-83, was slightly better with ADMARC purchases at about 8,500 tons. This was followed by a substantial increase in ADMARC purchases in 1981-84 and 1984-85 at 15,500 tons and 15,200 tons, respectively. Thus, these indicate that two additional years passed before attaining production in the range of 15-16,000 tons. This delay in benefits would further decrease the ERR. The audit also notes the point made by the Overview that an unquantifiable part of the project benefits assumed in the analysis are attributable to investments made under the previous Phase II project, and that if one quarter of the Phase II investment cost was added to this proj- ect, the ERR would decline to zero (PCR Overview para. 31). Nevertheless, the notable improvement in cotton production over the past two years provides some grounds for projecting further improvements. II. MAIN ISSUES A. Achievement of Objectives 29. The Issues Paper and Appraisal Report for this project recognized that Phase II had failed by a wide margin to meet its appraisal estimates of total crop production but had been successful in meeting most of its infras- tructure targets. The present project (Phase III) was justified by the fact that there would be a sharp reduction in infrastructure development which would permit full attention by project staff to agricultural production. 30. The project's major objectives as defined at appraisal were: (1) to increase the number of farmers applying improved crop practices by about 15,000 so that at least 55% of the Valley's smallholders (40,000) would have adopted at least some of the practices by full development in 1982/83; (ii) to provide technical assistance to these farmers and increase land under improved practices (12% of cropped area at appraisal) to about 50,000 ha (47% of cropped area) by 1982/83; and (iii) to provide increased social and infrastructural services directly or indirectly affecting just about all 70,000 families in the project areas. It was estimated that the overall economic rate of return (ERR) of the project would be 20%. 31. In practice, as documented in the PCR Overview, the results of Phase III have been similar to Phase II in that the project has been rela- tively more successful in meeting infrastructure and staffing targets than in increasing agricultural production. Consequently, the ERR is now reestimated at 8% and possibly less. B. Appropriateness of Project Design 32. The major limitations to increasing agricultural production in the Shire Valley that were identified in Phase II have been reaffirmed in Phase III, namely: (i) the serious impact of rainfall variability between years and within years; and (ii) the need for adequate price incentives, especially for crops grown primarily for market, such as the main cash crop,.cotton. - 8 - 33. To contend with the rainfall problem, project identification of Phase III took account of the drought risk in the Valley and Government pro- posed that the project should have a substantial irrigation component. At appraisal, this drought risk was noted, but the decision was taken that additional studies would be needed before an irrigation investment could be considered. The audit believes that this was a correct decision, in view of the lack of technical feasibility studies as well as the economic and social complexity of a new small farmer irrigation scheme. However, the project as brought forward did not fully address the weather risk. The project assumed that a combination of moisturc-conserving cultivation practices, selection of drought-resistant cereal varieties, and a change of extension system could be introduced and would result in a substantial increase in crop production. 34. With respect to maret prices, project documentation shows that concerns were expressed by Phase II supervision missions and by the Bank staff during the review process for Phase III that the levels of cotton prices established by ADMARC were not keeping pace with increases in general price levels and thus were not providing sufficient incentive to farmers. A price convenant was proposed, but at the Decision Meeting it was agreed that the need for such a covenant would be reviewed on return of the appraisal mission for the National Rural Development Project (NRDP). In the end, it was decided not to include a price covenant in the Credit Agreement. Also related to this, the Decision Memo notes "that there would probably be little direct cost recovery for the Government, although the marketing agency, ADMARC, would make substantial profits which would be reinvested in agricul- tural and industrial enterprises which were part of the Government's develop- ment program" and noted that the meeting felt this should be acceptable. This approach has since been questioned on the grounds that ADMARC has not been effective in using the resources generated from the smallholder sector for productive purposes. In any case, the end result of these decisions was the design of a project which built infrastructure and strengthened govern- ment services but had little success in increasing cotton production during the disbursement period, apparently due in large part to tihe negative incen- tive of low farmgate prices. Only during the past two years, since disburse- ment was completed, has considerable price policy reform been undertaken in the framework of the Structural Adjustment process, and production has sub- stantially recovered. C. Replicability 35. When this consolidation project was approved, it was already under- stood that the intensive approach of Phase II, and of similar projects in the Lilongwe and Karonga areas, corld not be directly replicated nationwide. A National Rural Development Program (NRDP) was being formulated at that time (Credit 857-MAI, approved December 21, 1978) which would essentially expand the area development approach nationwide by dividing the country into eight Agricultural Development Divisions (ADDs) that would be integrated into the federal civil service. The Lower Shire Valley is now one of these ADDs. Thus, to some extent the general area approach of the project is being repli- cated nationally, although less intensively in terms of staffing and invest- ment. -9- D. Sustainability 36. The sustainability of the project achievements appears still very tenuous. Gains in crop production appear to correlate more closely with Government price incentives and with weather than with the additional techni- cal assistance provided. In part this reflects the lack of available new technology. It also reflects rather clearly the tendency of the small farmer to retreat into subsistence production as soon as he decides that price incentives do not justify the financial risk of using additional purchased inputs such as fertilizer and pesticides. Thus, maintenance of any improve- ments depends heavily on Government pricing policies and continuing flow of funds from national budget resources or international lending. 37. Social infrastructure, such as village wells and health centers, are now the responsibility of the Ministry of Health and maintonance has been somewhat irregular. Villagers have not been trained and equipped to maintain their wells for drinking water and there has been very little emphasis on farmer organization. Thus, sustainability at this point still depends large- ly on central Government decisions and the availability of funds for imple- mentation. 38. Nevertheless, it appears that the experience of the Lower Shire Valley projects has contributed to at least four changes in Government policy and in Bank support: (i) a less intensive but nationwide structure of Agri- cultural Development Divisions (ADDs) including one for the Lower Shire Valley supported by the Bank's National Rural Development Program Projects as well as other donors; (ii) a nationwide project to strengthen agricultural research (Credit 1549-MAI for US$23.8 million approved February 19, 1985); (iii) a nationwide project to support agricultural extension (Credit 1626-MAI for US$11.6 million approved September 19, 1985); and (iv) a national health project (Credit 1351-MAI for US$6.8 million approved April 26, 1983) which now provides some of the support for health centers that was provided earlier under the Shire Valley projects. 一r必一 才辭二妒→？g《汗冷 Table I Rural Development Projects In Malawi Loan/Credit Approval Audit Pr2jeet Name Number Amount Date Report (USUI Lilongwe I Cr. 113 7.17 02/68 751 (05/23/75) Lilongwe II Cr. 244 7.41 05/71 1597 (05/17/77) Lilongwe III Cr. 550 8.50 03/75 3414 (04/07/81) Shire Valley I Cr, 114 4.30 02/68 895 (10/22/75) Shire Valley 11 Cr. 363 10.50 03/73 2593 (06/29/79) )/a Shire Valley III Cr. 823 10.70 06/78 /b Karonga I Cr. 282 6.61 01/72 2576 (06/29/79) Karonga II Ln, 1286(T) 9,20 06/76 5340 (11/28/84) Nat, Rur. Dev. I Cr. 857 22.00 11/78 Mat. Rur. Dev. II Cr, 992 13.80 03/80 Nat, Rur, Dev. III Cr, 1183 7.30 10/81 Nat. Rur. Dev. IV Cr. 1343 10.60 04/83 118.09 NNOWOMMMOM Other Agricultural Projects Smallholder Fertilizer Cr. 1352 5.00 04/83 Nat, Agr, Research Cr. 1549 23.80 02/85 Agr. Ext, & Plan, Sup, Cr, 1626 11.60 09/85A /a All three projects have been reviewed in an impact evaluation, OED Report No. 4850 dated December 23, 1983. A Audit report being prepared (for distribution June 1986). /c Not yet effective (as of 04/11/86). W~~ ~~ s -z> & - 13 - SBINB FALLEY AGRICULTURAL CSOLIDATION PROJECT CREDIT 823-MAI PROJECT CO08ZTION 83P01R AN OVERVIEW December 2, 1985 Government of Malavi Ministry of Agriculture and Eastern and Southern Africa Proj(.cts Department Regional Mission in Eastern and Southern Africa cP ~ 9 - 15 - MALAWI SHIRE VALLEY CONSOLIDATION PROJECT (CREDIT 823-MAI) PROJECT COMPLETION REPORT AN OVERVIEW I. Introduction 1. The Planning Division of the Ministry of Agriculture and Ngabu ADD personnel prepared the PCR for the above Project, although this was not a condition of the Credit Agreement. A draft report was extensively com- mented upon by both Headquarters and RMESA staff. Many of the comments made by IDA staff on the draft have been incorporated in the final version. The Planning Division also prepared a revised economic analysis which provides a recalculation of the economic rate of return and clarifies the methodology and the financial and economic assumptions underlying the analysis. II. Project Objectives 2. The Shire Valley Agricultural Consolidation Project supported by IDA Credit 823-MAI was appraised in August 1977 and became effective in June 1978. It was prepared by the Government with the assistance of RMESA staff. The project was a direct continuation of the second Shire Valley project and was expected to consolidate previous development activities taking particular cognizance of the production weaknesses identified during the prior phase (SAR para. 2.15). Over a four year period (1978/79 - 1981/ 82) the Project was to: (i) continue to support development of rainfed agriculture production services, reorganize and strengthen the extension service, provide improved seed varieties and credit; (ii) improve the quan- tity and quality of meat and fish production; (iii) continue to support the land husbandry program with an added forestry unit; and (iv) extend water, health, and road services. III. Physical Implementation 3. The civil works construction program started at a slower pace than anticipated. The construction of administrative blocks for extension, health facilities, markets, dip tanks, laboratories, and fish landing sites met 100% of the appraisal targets. Because of cost increases and Govern- ment's decision not to increase the allocation for civil works, only 47% of the planned staff houses and 80% of the planned boreholes were built. Of the 139 kms of roads planned, 111 km (872) were either constructed or up- graded under the Project. The increase in road construction costs, break- down of equipment and lack of spare parts hindered the completion of this component. None of the 50 planned village training centers were built because residential training was given preference. - 16 - IV. Project Management 4, The organizational structure of the Phase II Project continued in Phase III with slight modifications. Lack of staff continuity developed into a serious problem and the Project had three managers and three financial controllers during the implementation period. Intermittent vacancies of senior posts occurred throughout Phase III and seriously hampered Project implementation. At the end of the Project, only 40% of the senior staff positions were filled. Unsuitably qualified support staff, difficulties encountered in reaching agric,itural production targets and the related unattractiveness of staff assignments to the Shire Valley, lack of job security for the staff employed from the development budget (directly paid by Project funds), and uncertain promotion prospects for technical and supervisory staff contributed to high turnover and low morale of staff, and to management and operational problems. 5. At the beginning of Phase III, 71% of the staff occupied positions not yet officially "established". Although the Government had agreed to absorb these staff gradually into permanent positions, by the end of the Project 55% of the staff remained in positions not yet permanently estab- lished. Furthermore, the local funds made available so far have not been adequate to retain all staff and to maintain all buildings, vehicles and equipment procured under the Project. 6. A large number of surveys were carried out by the monitoring and evaluation unit established under the Project (paras. 65 and 66 and Table 12 of PCR). The data was not always adequately evaluated, and was of limited assistance to management in decision-making and forward planning. This also affected the preparation of the PCR, especially with regard to determining reasonably reliable and consistent agricultural production data for assessing the Project's impact.1/ V. Services Supported under the Project 7. In support of the crop production goals, the Project provided for crop research, extension services, training facilities and equipment, farmer and staff training, and credit. Efforts to reorganize the extension system along the lines of the Training and Visit System and to provide farmers with more up-to-date advice on farming methods and practices did not succeed to the extent hoped for, mainly because: (a) senior staff at the Ministry of Agriculture and at the Project level were reluctant to move quickly with the system, with which they were not yet very familiar and of whose advantages they were not yet convinced; 1/ See also Impact Evaluation Report Malawi : Lower Shire Valley Agricultural Development Projects (Credits 114, 363, and 823 MAL). Report No. 4850 of December 23, 1983 Annex II. - 17 - (b) there was a shortage of well-trained personnel available to the extension service due to the low output of students from the Colby College; and (c) there were very few adequately proven and financially attractive technological innovations to recommend on drought resistant crop varieties or land husbandry practices. 8. Due to the Government decision not to construct the 50 Village Training Units, farmers and farmers' wives were trained instead at the Residential Training Centers. The PCR notes (para. 32) that no information is available on the number of farmers and farmers' wives trained during the Project period. Nonetheless, given the weakness of the technical packages to be transmitted to farmers, the impact of the farmers' training program on improving agricultural technology should probably be considered to have been less than hoped for. Training of the extension workers was sporadic, as was the technical support to them from subject matter specialists. 9. The surveys conducted by the Project Monitoring and Evaluation Unit did not adequately distinguish between Project participants and non-participants, and it has proven difficult to assess the impact of the Project on production and yields of those farmers directly reached by extension. In the absence of adequate direct data, the PCR used production and yield data from the National Sample Survey of Agriculture for the Project areas and compared them with the appraisal estimates (PCR - Tables 6, 7 and 21), It appears (data is very poor) that most crops did not reach the planned hectarages or yields. Appraisal assumptions regarding the possible impact of an improved extension effort, the availability of suitable cereal varieties, the adoption of moisture conservation practices and the existence of adequate producer prices all proved overly optimistic. 10. Crop Research. Adequate progress in crop research is one of the keys to improved agricultural production in the Shire Valley. (An appropriate pricing policy has proved to be another such key.) However, the crop research program suffered significantly from shortages of qualified staff, inadequate local funding and high staff turnover. Maize cultivar trials were conducted throughout the Valley using mainly exotic hybrid varieties. These trials led to the limited release of three imported hybrid varieties suitable for the Valley. Distribution of seed of these varieties was, however, discontinued on acrount of its low quality, and a new series of trials was initiated to screon the suitability of locally bred hybrids. Some research work was done on cotton, legumes, sorghum and millets in the context of the National Trials Program. As a result of these trials, a new cotton variety (Makoka 78) and a drought resistant sorghum variety (PN3) were identified and recommended for the Valley toward the end of the project, and appeared to be acceptable to the farmers. The crop/weather plots, established to provide information on yields of major crops in different ecological regions of the Valley, were abandoned because of financial constraints. Trials with phaseolus beans, sweet potatoes and vegetables on drainage areas were not carried out because suitable areas could not be made available to the Project. As a result of the shortcomings of the research effort, extension was left with little to offer. - 18 - 11. Seed Multiplication. Mainly because of financial constraints, the 150 ha irrigated seed multiplication farm for the selection and multiplication of maize, sorghum, millet and groundnut seeds was abandoned in 1980. Eight staff houses, an office block, and a workshop had been constructed. Seed supply was entrusted to the National Seed Company. The Government is currently studying the future use of the project built facility, possibly for its national agricultural research program. 12. Livestock Production. During the Project period, the cattle population increased by 26% (from 61,550 to 77,470 head). The annual off-take averaged about 7%, significantly below the appraisal estimate of 12% per annum. The special permit required for the slaughter of female animals, and the restriction of cattle transfer out of the Valley as a measure to control foot and mouth disease, contributed to the increase in the cattle population. Overgrazing has become a problem. Some 16 dip tanks were constructed or rehabilitated. However, only about half of the cattle were dipped, mostly at infrequent intervals. The animal husbandry and livestock extension section remained understaffed during the Project period, and insufficient transport was available to staff to properly supervise the activities. The tsetse study was completed and vaccination campaigns on trypanosomiasis were conducted in the Project area. The diagnosed incidence of trypanosomiasis declined by about 10% following the campaigns. 13. Fisheries. The development of the fishing industry received emphasis in both Phase II and Phase III of the Shire Program. Landing sites and beaches were constructed, improved smoking kilns erected and fishermen trained in new techniques. Credit facilities were provided to enable fishermen to buy improved gear and plank canoes. Planned investiga- tions to determine unexploited fish stock, and the most suitable and effective methods of fishing under the prevailing water regime were not conducted. This was partly because the planned post of Gear Development Officer was not filled. Fish catches were estimated to be higher than expected at appraisal. 14. Credit. There was a shift from individual to group credit. About 700 credit groups with some 7,000 members benefited from seasonal credit. The average loan per farmer was MK 25 per season. Credit recovery was 80% for short-term and some 63% for term lending. Credit uptaki by farmers was significantly below appraisal estimates. This partly reflected the low (Government-set) cotton prices. The number of beneficiaries per year gradually declined. Most sub-loans were for short-term credit and term lending remained insignificant. The risk of using fertilizers to increase yields proved unacceptably high given the incidence of drought years. Partly as a result, a more general risk aversion inhibited the adoption of practices requiring cash outlays or credit. The tightening-up of credit policies, whereby defaulting groups or individuals became ineligible for new credit, also contributed to the lower use of credit. 15. Land Husbandry. Land use maps were produced for all EPAs on a scale of 1:50,000. About 40 farm plans were prepared per year as a joint effort between Land Husbandry and other sections within the Ngabu Agricul- tural Development Division (ADD). Particular emphasis was given to contour ridging. Although there is no precise data available on farmers' adoption - 19 - of the above practices, supervision missions reported progress in this regard. The cultivation of crops such as cashew nuts, macadamia nuts, pineapple, banana, and citrus was adopted gradually by some farmers in the hill districts. 16. Forestry. Shortage of senior staff during most of the Project period resulted in implementation delays and in shortfalls in performance against most of the original targets of the forestry component. Three adaptability investigations were conducted and several tree species were recommended for the Project area. The proposed re-afforestation program was, however, not successful. Farmers proved very reluctant to "give up" land for woodlots and shelter belts. 17. Wildlife. All facilities were completed as planned. Funding and operation of the Lengue National Park were fully taken over by the parent department in 1981. 18. Hydrology and Water Supply. The Project financed the National and Shire Valley Irrigation Study, which addressed issues related to: national irrigation policies; settlement procedures; water charges; scheme management and maintenance; cost recovery; and the preparation of a water resource plan for the Valley. The Government has since engaged a consultancy group to update the study of the economics of possible irrigation and - if found justified - conduct feasibility studies for the rehabilitation and expansion of irrigation schemes in Malawi. 19. About 80% of the planned boreholes and shallow wells were constructed during the Project period. IDA also agreed that Project funds could be used to supplement the financing of rehabilitation of about 100 boreholes constructed under previous projects. However, maintenance of the boreholes and shallow wells remains a problem not yet properly dealt with by the Government. The rotary drilling rig with compressor unit which was procured with Credit funds proved to be inappropriate. Supervision missions suggested the transfer of the drilling rig to the Department of Lands, Valuation and Water. However, that Department had neither the funds nor an operator to use the rather complicated rig. The proper use for the rig is again being studied by the Government. The originally envisaged groundwater survey program was not fully completed because of the expiry of contract of the expatriate driller, lack of spare parts, and, on occasions, lack of petrol. VI. Project Costs, Financing and Disbursements 20. Total Project costs were estimated at appraisal at US$ 12.6 million (MK 11.4 million equivalent). Due to a reduction in the project's scope and the appreciation of the Malawi Kwacha during the early years of the Project implementation period, actual Project costs in local currency were 9% below the appraisal estimate. As indicated in the table below, project expenditures were significantly below appraisal estimates for the agricultural services component, the natural resource component, health services and the evaluation unit (M 6 E). The actual costs for Project management and financial control as well as for mechanical maintenance exceeded appraisal estimates by a wide margin. - 20 - Appraisal Actual as Project Cost (MK '000) Estimate Actual % % of SAR Agricultural Services Extension 993 921 93 Training 374 264 71 Research 714 389 54 Seed Multiplication 774 54 7 Credit Administration 331 n.a. n.a. Animal Husbandry 651 438 67 Sub-total 3s837 2s066 54 Natural Resources Land Husbandry 232 188 81 Fisheries 208 152 73 Forestry 332 76 23 Wildlife 123 132 107 Sub-total 895 548 Hydrology and Water Supply 897 921 103' Health Services 363 148 41 Technical Services Roads 1,025 864 84 Building Maintenance 599 n.a. n4a. Mechanical Maintenance 127 468 369 Sub-total 1752 1,332 76 Administration Project Managerent Office 941 1,712 182 Evaluation Unit 274 202 74 Finance Division 304 1 276 420 Sub-total 1,519 3 190 210 Contingencies Physical 463 Price 1,718 Total Project cost 11,443 8,205 (10,234) 1 1/ Excluding MK346,000 disbursed in 1983 on Project commitments and unidentifiable expenditure on central credit administration and building maintenance. Actual expenditure of IDA and Government funds totalled 110.234 m for the 4 years of the project and 110.580 a when that disbursed in 1983 is included. - 21 - 21. IDA provided US$ 10.7 million or 85% of the cost as projected at appraisal. Financing Sources A praisal Estimate Actual (MK al.)(K aM .) IDA 9.7 85 8.7 85 Government 1.7 15 1.5 15 11.4 100 10.2 100 22. Overall Credit disbursements remained within appraisal estimates. However, cost overruns in certain categories required . reallocation of funds. The civil works category was kept within the original allocation, although the unit costs of houses, roads and boreholes were higher than the appraisal estimates. As a result, not all of the houses, roads and boreholes were built. At appraisal it was anticipated that most expatriate staff required to initiate the Project would be gradually replaced by local staff. This did not happen, probably reflecting both too optimistic an assumption and too little effective training. In any case, there resulted a considerable cost over-run in the category for international staff. Funding for operating costs also exceeded the appraisal estimate. Allocation Actual 1/ Credit Category at Appraisal Disbursements % 1 Vehicles, equipment 1,410,000 1,182:403 83 2 Civil works 1,660,000 1,660,000 100 3 Salaries and allowances international staff 280,000 578,258 206 4 Operating costs 4,910$000 6,473,488 131 5 Consultants 400,000 805,851 201 Unallocated 2,040,000 - - Total 10,70,000 10,700,000 100 1/ IBRD Loan Department - Disbursement I: Status of Funds as at December 31, 1983. - 22 - 23. As indicated in the PCR (Table 15), disbursements of credit pro- ceeds were below appraisal expectations in the first and third years, exceeded appraisal estimates in years two and fOour and extended 10 months beyond the original schedule. Actual accumulated disbursements compare to the appraisal estimates as follows (in US$ million): FY 1979 1980 1981 1982 1983 Actual 2,318 5,690 8,070 9,991 10,700 APR 4,500 7,000 9,100 10,700 - Actual as % of Appraisal Estimate 52 81 89 93 100 VII. Project Benefits 24. Most Project benefits were expected from increased production of cotton, maize, sorghum, millet, rice, groundnuts and beef. Improved research, extension, training, and land husbandry services were to provide the basis for the improvements of the traditional technology for an increase in yieldt of major crops. Projected crop yields and total production were based upon expected adoption by smallholders of a basic package of practices designed to reduce the risks of periodic dry spells which affected Valley production. 25. Crop Areas and Yields. Data provided in the PCR regarding estimated developments in crop areas and yields in the Project area are very difficult to interpret and do not appear to be entirely consistent. The PCR data should be considered with caution. 26. Generally, it is believed that cotton areas and yields have been well below appraisal estimates both because of the inadequate adjustment of cotton prices and the risky nature of the technical packages which were recommended. Maize areas increased substantially, and those for drought tolerant crops such as sorghum and millet increased marginally. The combined crop area, however, is thought to have remained below appraisal estimates. Reliable yield data are not available, but recent estimates for most crops in the Project area have been below the appraisal targets. 27. Crop Production. Project production varied considerably from year to year, mostly depending on seasonal conditions (rainfall) and producer prices. Only guar bean and sorghum production exceeded appraisal estimates in the last two years of the Project. Cotton and groundnut production actually declined during the Project period and maize production increased. (Cotton production has very recently - reflecting price adjustments - begun to recover.) Although millet and rice production increased during the Project period, it did not reach appraisal estimates. - 23 - 28. Official Crop Purchases. The average annual cotton purchase by ADMARC during the Project period was about 9,000 tons which is considerably lower than purchases in the mid 1970s. Purchase of maize in 1981/82 was 2,833 tons and compares with an average of 437 tons in the five previous years. Official groundnut purchases declined from 259 tons in 1977/78 to 7 tons in 1980/81. ADMARC purchases of guar beans, a newly introduced crop during the previous Project phase, reached an average of 3,300 tone per year. 29. Causes for Production Changes. It is apparent from the informa- tion available that producer prices and rainfall conditions were critical factors in determining production. Because of inadequate producer prices, cotton and groundnut production did not develop as expected. The introduc- tion of guar beans by extension has provided some farmers with an addi- tional source of income. Although extension could only offer a limited crop package, some advice on ridging and contouring was followed by the farmers. Similarly, marsh areas have benefited from the activities of the fisheries component. 30. The population of the Valley has benefited from improved water supplies, health facilities and roads. The Project-supported effort to explore the potential for future irrigation development has given useful guidance to Government. 31. Economic Rate of Return and Fiscal Viability. The Government's revised analysis estimates an economic rate of return to the directly pro- ductive investments under the Project of 8. It assume for Project Years 5 to 8 a 5% annual increase in the incremental output of cotton, a 3% annual increase in the incremental output of maize, rice, groundnuts, sorghum and millet and no increase for beans, over the level of Project Year 4 (1981/ 82). These assumptions, as well as the methods used for determining cost and economic values, appear reasonable. However, it should be noted that an unquantifiable part of the Project benefit assumed in the analysis is attributable to investments made under the previous Phase II project. If one quarter of the Phase II investment cost were added to this project, the economic rate of return would decline to zero. 32. Available data do not allow ready or clear quantification of the fiscal impact and viability of the Project. However, it is probable that the Government budget has benefited little from increased taxes and duties, while public expenditures on account of capital investments and operating and maintenance costs of the Project have been substantial. VIII. Government Performance 33. Government commitment to the Project was uneven. It introduced annual work plans and budgets, aimed at assisting Project management. Throughout the Project, the Ministry of Agriculture provided assistance to Project management and closely supervised implementation. However, senior staff turnover was a major problem during implementation. There were three project managers, three financial controllers and four evaluation officers during the four-year Project period. About 40% of the senior staff posi- tions foreseen at appraisal remained vacant, and Government only trans- formed a small part of the staff posts to permanent positions. - 24 - 34. Government-set prices for cotton and groundnuts were inadequate and this contributed to the decline of production of these crops. Agricul- tural research did not meet appraisal expectations, leaving extension with limitee technical advice to offer to farmers. 35. Government did maintain a group credit approach which facilitated credit administration. The focus was, however, mostly on farmers who had already been beneficiaries of the credit programs in Phase I and II. 36. Although price escalation for civil works was higher than expected, and baseline civil works costs had been underestimated by the appraisal mission, overall Project expenditure remained within initially budgeted amounts. This can be partly attributed to careful financial monitoring. During tie first half of the Project period, the Government did experience difficulties in providing the required funds, which slowed down implementation and/or led to reduction in the scope of some project components. The preparation of accounts was generally timely but audits were delayed throughout the Project period by about one year. With the exception of the staffing, counterpart fund and audit covenants, the Credit covenants were complied with in a timely manner. 37. Progress reporting by Government was comprehensive and timely, but a less than satisfactory commitment to and execution of the monitoring and evaluation component, combined with financial and staff constraints, resulted in a paucity of really hard and reliable information required for project management and project impact evaluation. 1X. IDA Performance 38. The appraisal team reduced the originally proposed scope and cost of the Project by deleting a smallholder irrigation scheme and eliminating several other proposed components. Infrastructure development was mostly limited to the completion of ongoing works, in order to permit concentra- tion of available resources on agricultural production, livestock, and other productive components. This restructuring or "consolidation" was positive inasmuch as it reduced the burden on Government's limited finan- cial and manpower resources. It did, however, too little to focus Project intervention more clearly on the crucial constraints for production devel- opment in the Valley, i.e., research and output pricing. The capability of the Project's agricultural research program to develop new technological packages during the life of the project was seriously over-estimated. To the same extent the extension component was overdesigned. Based on the disappointing results of Phase II, the appraisal team had projected what it thought were very modest levels of improvements in crop production and yields for cotton and other crops, which however proved to be still optimistic in the circumstances. While the cotton producer price was in fact increased somewhat at the time of appraisal, it proved not to be enough to provide a sufficient incentive to farmers in the Valley to continue with cotton production at earlier levels or expand it. The Government's PCR suggests rightly that the drought risk was not sufficiently dealt with at appraisal. - 25 - 39. Government's capacity to recruit the required number of well qualified staff and its capability of absorbing 261 staff members into permanent positions was over-estimated. The Project contributed to the establishment of a network of services and infrastructure, which so far has made a much smaller than expected impact on production, but which will be costly for the Government to maintain. 40. Seven supervision missions visited the Project at an average interval of seven months for about 2.4 staff weeks per mission. They provided considerable technical assistance, particularly on extension organisation, project management, and agricultural production and also tried to assist in resolving staffing and financial problems. Although the missions for this as well as the previous project pointed out the negative impact of low producer prices on cotton production, the Bank was for some time ineffective in getting action on price policies. At no time did supervision missions or IDA propose to substantially alter the project design to bring its scope and component structure better into line with the borrower's apparent manpower and financial constraints and with the more immediate basic research and technical needs for the agricultural development of the Valley. X. Lessons Learned 41. The limited success of this broad based effort to increase the income and quality of life of the Valley's population suggests that: (a) a more rigorous analysis of historical production data and more careful and conservative judgements in assessing production prospects is called for - especially for areas with risky rainfall conditions; (b) project size and the structure of investment proposed need to be reasonably in line with the Borrower's capacity in terms of manpower and financial constraints to implement the project and sustain project investments; (c) in situations where agricultural production development depends so centrally on the generation of new technologies through research, one should not expect substantial research results to be extended to the farmers within the same period; and (d) the success of production projects depends heavily on a supportive policy environment, especially in the area of producer prices. \�'' � ! � � 1 `,1 � .и, � - 27 mm~ GOV PROJ= CONPLVrm R~RT SEM VALLU åGRICULTURåL ComæLIDMM PROJ= (r^ R~RT) Dec~er 2, 1985 ージアアー オく」ノ勢介6ガ溶そ 29 MALAWI SHIRE VALLEY AGRICULTURAL CONSOLIDATION PROJECT (.CREDIT 823-MAI) PROJECT COMPLETION REPORT I. INTRODUCTION AND SUMMARY 1. This is the third Project Completion Report to be prepared by Malawi Government, the first being that of Lilongwe Land Development Programme Phase III prepared in 1979 and that for Karonga-Chitipa Rural Development Project Phase II prepared in February, 1983. Shire Valley Consolidation project was a follow-on project to Phase II which was terminated in early 1978. SVACP began in 1978 and was completed in 1982. 2. The purpose of SVACP was to increase crop production, livestock off take and fish catches. The Project was expected to consolidate all previous development activities such as infrastructure and institutional development for increased agricultural production. To increase crop production, Phase III planning took particular cognizance of drought risk in the valley a-ad proposed solutions to minimize this risk. 3. In order to achieve the overall goal of increased agricultural production, there were Inputs to be delivered and employed and consequently outputs to be realized. 4. This report attempts to assess to what extent the Project has been successful in as far as input delivery, output realization and overall impact is concerned. The report is presented following guidelines detailed in the World Bank's Operational hanual Statement (IBRD, February, 1979). Findings and Conclusions 5. The report finds that the Project commenced on schedule in April, 1978, with a retrospective financing. However, this phase started at a slow pace because incremental staff and equipment were not yet in place. As a result of this, a number of Project activities such as building of staff houses and construction of roads took time to complete. Inevitably, the slow start and the consequent delay in the construction program resulted in cost penalties, Despite the slow start, the report finds that input delivery during the phase was satisfactory. Incremental staff, equipment and funding were delivered and used satisfactorily. Project outputs such as the building and road construction programme, water and development and expansion of the Natural Resources Division were consider- ably realized. Envisaged project impact, however, has not been realized. Crop production, cattle off take and fish catches did not increase during the phase as anticipated at appraisal. In crop production, this is because the proposed solutions such as reorganized extension system, moisture con- serving cultivation practices and the release of drought resistant cereal varieties from research were either not implemented or were not achieved sufficiently enough to create the necessary impact on crop production. In case of increased cattle off take, this is a result of lack of congruency between the Project objective and Government policy. As for low fish catches, no conclusive reasons have been noted. It is, however, understood that the failure to recruit the Gear Development Officer was the main reason, - 30 - 6. Actual Internal Economic Return of 15% has been calculated based on benefits from crop production only. The actual IRR is lower than what had been anticipated at appraisal. This is not surprising considering that incremental crop production throughout the phase has been lower than anticipated. II, BACKGROUND 7. The Project area comprises the administrative districts of Chikwawa and Nsanje. The area is located in the Southernmost tip of Malawi between latitudes 150 45' south and 17* 8' south. The total area is 6,700 square km. The valley is the southern-most extremity of the Great African Rift Valley and is bounded to the East by the Shire Highlands and on the West by the Mtundwe and Marengwe Hills which form the divide between the Shire and Zambezi River catchments. The valley floor is an elongated alluvial plain about 44 km long by 16-32 km broad whose height above sea level changes gradually from 150 m at the foot of the valley sides to 60 m through which the Shire River passes on its way to join the Zambezi. 8. Annual average rainfall for the valley floor is 64 cm rising to 89 cm on the surrounding hills. Eighty-seven percent of this rain falls in the months of October to March. The movement of the inter-tropical convergence zone gives rise to a large variation in the timing of the onset of the rains and consequently affects the planting dates. 9. By 1977, population in the valley was 303,030 with an estimated annual population growth rate of 2.3%. Actual rural population in 1977 was 274,500 with a household size of 4.6 persons giving 62,700 farm households. Within the valley, high population densities are in Naanje district, especially along the edge of Ndinde marsh. Population is also dense on the East Bank where alluvial soils from numerous streams and marsh edges provide opportunities for Intensive cropping. 10. Major dry land crops of the Lover Shire Valley are cotton, maize, bulrush millet, sorghum and groundnuts. Pigeon peas and cassava are important in hill areas. Rice is widely grown along the Shire flood plain. The people of the valley also keep cattle, goats, pigs and chickens. 11. The valley has always been considered one of Malawi's disadvantaged regions in terms of climate, incidence of disease, and low agricultural productivity. Consequently, after independence, detailed proposals were prepared for the Chikwawa Cotton Development Project (CCDP) which tmmenced in 1968 with a foreign financial support from IDA Credit (114-MAI), totalling MK4.07 m. The Project was terminated on schedule in 1973. The completion report and the performance audit considered the Project to be well conceived, effectively implemented and generally successful. This success was based not only on the achievements of physical targets, but also on overall impact of the Project on the attitudes of those living in the valley. The Project was noteworthy in that it was very successful in involving people at all levels in the rural development effort in an area whose inhabitants had been considered resistant to change. The Project emphasized increased cotton production. - 31 - 12. The CCDP was immediately followed in the same year by a second phase of Agricultural Development also financed by an IDA Loan of MK11.5 m, Credit 363-MAI. Development activities were expanded to cover both adminiatrative districts of Chikwawa and Neanje (6,734 sq km) with emphasis on cotton, maize, groundnuts, livestock and fish production. 13. The results of Phase II were measured by two different criteria. On one hand, excellent progress vas achieved in completing the Project's physical infrastructure; on the other hand, total crop production failed by a wide margin to meet appraisal estimates. In fact total cotton and maize production declined because of frequent drought which devastated the area, low farm gate prices for cotton, and the unsuitability of the varieties of maize and sorghum recommended for the valley. 14. Both the Phase II Completion Report and the Crop Review of 1976 prepared jointly by project management and an IDA team noted these problems. As a short term solution, the reports recommended a careful choice of crops/varieties for the valley. In the long run, the Crop Review Report recommended a gradual shift to supplementary Irrigation. 15. Apart from the smallholder projects, there have been other activities going on in the valley such as the 11,000 he irrigated sugar estate; few Chinese Rice Schemes; three Game Reserves (Mwabvi, Majete and Lengwe) and the Kasinthula Agricultural Research Station carrying out both trials under irrigation, and settlers growing irrigated rice. III. PROJECT FORMULATION 16. The thinking of SVACP began in 1976 at about the same time that the Crop Review Report was being prepared. Project identification of Phase III took particular cognizance of drought risk in the valley and its effects on crop production. Consequently, Phase III was expected to mini- mize the effects of drought risk on crop production. The Project proposals were made along the lines of an intensive development phase where a rapid move towards irrigated smallholder agriculture would take priority over rainfed agriculture. These proposals were prepared by a joint team from Planning Division of the Ministry of Agriculture and SVADP management. The proposals were also made in full consultation with other Ministries of Works and Supplies (as regards construction programme). 17. The Malawi Government proposals for Phase III were for a four- year project commencing April 1, 1978 and amounting to a base cost of MK15.259 m giving an estimated cost of MK19.491 m. Detailed ccsts break- down by components were presented and an IRR of 29% was estimated over a 20-year period. Primary benefits of the project were to be increased crop, livestock and production, in order to improve farm incomes. Sensitivity analysis indicated that with an increased cost of 10% coupled with a reduced level in benefits of 10%, the proposals would result in a return of 21%. 18. Major objectives of Phase III were as follows: (I) to increase the number of farmers applying improved crop practices by about 15,000 so that at least 55% of the valley's smallholders (40,000) would have adopted at least some of the practices by full development in 1982/83; - 32 - (ii) to provide technical assistrnce to these farmers to increase land under improved practices (12% of cropped area) to about 50,000 ha (47% of cropped area by 1982/83); (iii) to provide increased social and infrastructural services directly or indirectly affecting about 20,000 families in the Project area; (iv) improve quantity and quality of meat and fish production; (v) extend water, health and road services; and (vi) continue support of the land husbandry programme with an added forestry unit. 19. An IBRD appraisal mission of six team members visited Malawi in August 1977, and the appraisal report followed in 1978. The appraisal mission changed the direction of Phase III from that with a major input in smallholder irrigation to that based exclusively on rainfed agriculture. Although the drought problem in the valley was noted, it was believed that with moisture conserving cultivation practices, availability of drought resistant cereal varieties and a change of extension system, increased crop production would be achieved. In view of this, the project was therefore specifically to provide for: (a) establishing headquarters for 5 additional BPAs with required staff housing, training and administrative facilities, (see Figure 2 for location of BPAs); (b) establishment of a 150 he irrigated seed selection and multiplication farm directed by an expanded research team; (c) continuation of seasonal and medium credit for farm and fisheries inputs; (d) construction of additional livestock markets and dips, expansion of FMD and Tsetse Control, and increasing the ox- training programme; (e) expansion of the Natural Resources Division to include a Forestry Development Unit, and continued Improvement of wildlife facilities; (f) ground water survey and development including construction of 40 additional boreholes and 40 shallow wells, providing consultants' service to review experiences gained under existing irrigation schemes and to undertake an izrigation feasibility study of the valley; (g) continuation and improvement of project area health facilities; (h) completion of the access road network, and maintenance of project roads, buildings and equipment; and - 33 - (1) continuation and expansion of the project Evaluation Unit, and two-year financing for the position of Chief Projects Officer in the MANR. 20. There were considerable differences between the final appraisal document and the original Malawi Government proposals. Main points of difference were as follows: (a) the appraisal mission reduced the overall project cost by MK8.05 m (40%) according to a reduced project design (see Table 1): Table 1: Financing Arrangements, compared (MK '000) MG Project Appraisal Source Proposals Prov. (1945-MAI) World Bank loan 16.567 (85%) 5.234 (46%) Malawi Government 2.924 (15%) 6.209 (54%) Total 19,491 (100%) 11.443 (100%) (b) The components of irrigation and rural development were dropped out at appraisal and the leucaena development component became part of the credit section. (c) A complete change of extension approach and management was agreed at appraisal; the Training and Visit (T&V) system of extension was to be adopted. Farmer to technical assistant (extension worker) ratio was reduced from 1:255 to 1:600 at appraisal. (d) Appraisal document emphasized that the land husbandry work programme should give emphasis to adoption of moisture conservation cultivation practices. (e) For research component, the appraisal document placed emphasis on a programme of project specific research and, together with seed multiplication component, the development of drought resistant crop varieties for the Shire Valley. 21. A detailed breakdown of project costs by component is shown in Table 2 below: - 34 - Summarized comparison of component costs are as follows: Table 2: Component Costs Compared (MK '000) a/ Appraisal Project Document Proposals (1945-MAI) Component (4 years) (4 years) Agricultural Services Extension 1,722 993 Training 452 374 Research 795 714 Seed Multiplication 476 774 Credit Administration 338 331 Animal Husbandry 827 651 Leucaena development 195 - Irrigation 2,735 - Sub-Tota. 7,540 3,837 Natural Resources Land Husbandry 279 232 Fisheries 268 208 Forestry 393 332 Wildlife 141 123 Sub-Total 1,081 895 Hydro-(Geology) 642 897 Health Services 455 363 Technical Services Roads const. and maintenance 1,797 1,025 Buildings 628 599 Mechanical maintenance 280 127 Sub-Total 2,705 1,751 Administration Project Manager's office 1,584 941 Evaluation Unit 328 274 Finance Division 608 304 Rural Development 315 - Sub-Total 2,835 1,519 Base Cost 15,259 9,262 Physical Contingencies 762 463 Price Contingencies 3,470 1,718 Total Project Cost 19,491 11,443 a/ These are costs on Development Account only. - 35 - From this table, it is evident that at appraisal, funding of some components was either reduced, increased or eliminated completely. IV. IMPLEMENTATION Effectiveness and Start-up 22. Negotiations for Phase III funds (Credit 823-MAI) were held in Washington D.C., U.S.A. during the period May 1-6, 1978. Signing of the Development Credit Agreement was d)ne in July 1978, and funds were finally released on September 2, 1978. Prior to the Development Credit Agreement becoming effective, it was agreed that project expenditure could be incur- red on a retrospective basis from the beginning of the financial year. By the end of September 1978, the total retrospective financing amounted to MK800,000. 23. The policy or retrospective financing allowed the Project to start on time and also enabled the start of tender procedures to be set in motion early during Phase III, and sectional internal budgets redrafted on the basis of appraisal data that had now become available. International tender for vehicles, plant and equipment was advertised o July 30, 1978 and vehicles and equipment ordered started arriving early 1979. 24. Revisions: There were no major revisions during the implementa- tion to the basic design of the Project or its principal objectives. Seed multiplication component was, however, abondoned during implementation after a considerable amount of infrastructure was developed at the seed farm at Makanga. Mnagement Organization 25. The programme management structure is summarized in the organizational chart at Annex 2. Phase III was a direct continuation of Phase II, and therefore no major changes in staffing or organizational structure were necessary. The overall implementation of the Project was the responsibility of the then Ministry of Agriculture and Natural Resources. The Programme Manager (PM) was responsible for day-to-day running of the Project and he was responsible to the Principal Secretary through the Chief Agricultural Development Officer (CADO). The PM's office was responsible for four main operational divisions (Agriculture, Natural Resources, Finance and Technical) and the Ancillary Services Division. 26. Monthly management meetings were organized under the chairmanship of the PM to discuss implementation problems. These management meetings were attended by the two District Development Officers, Assistant Programme Manager and subject matter specialists. Instead of discussing overall broad implementation problems affecting different components of the Project, it is understood the main topic of discussion was agricultural progress reports. Staffing 27. The staff included expatriate staff as well as Malawi Government staff employed on permanent and temporary basis except industrial also employees such as drivers, carpenters, mechanics and general labourers. - 36 - 28. Appendix Table 1 show how SVACP staffing by project component has changed during Phase III. By the beginning of the Phase or end of Phase II, 80% of required posts for Phase III were filled suggesting that these were the already existing staff. By the end of Phase III, 99% of all posts were filled. 29. Appendix Table 2 presents staff changes by grade during Phase III. It is apparent from the TAble that from STA/SCO grade and below the Project generally suffered overstaffing. From TO/PO and above, the Project however generally experienced understaffing. Relative scarcity of qualified staff between these two broad categories of staff explain this difference. Appendix TAble 3 presents staff situation broken down to show number of posts filled by established personnel and those posts filled by non-established staff. By the beginning of Phase III, 71% of all filled posts were occupied by non-establshed staff. By the end of the Phase, the percentage had gone down to 55%. Training 30. In Phase III, there were two types of training activities namely, staff training and farmer training. Staff training took the form of refresher courses or seminars intended to introduce or refresh grass root staff to some kin of skill. Courses for staff were either organised at Residential Training Centers (RTC) or Day Training Centers (DTC). At appraisal it was envisaged that staff training for grass root staff, especially village extension workers, would become systematic in order to complement with the new extension method (T&V System). Unfortunately, the T&V extension system was never adopted by Malawi Government as noted in paragraphs 36 and 37 and as a result staff training programmes did not follow the system agreed at appraisal. 31. Farmer training in Phase III was in two parts: farmers' courses and farmers' wives courses. Both types of courses were organized at farmer group or club level, DTCs and RTC. The intention of Phase III planning was to reserve the Residential Training Center for specialized courses for selected farmers or farmers' wives and to move all general courses which were originally held at the RTC to DTCs and to farmers groups. It was believed with this new arrangement many farmers would be within reach of some kind of training center and be trained. Despite this arrangement, general farmer courses were run at RTC especially in the first two years of the project because of the delay in forming training groups and the failure to construct the 50 Village Unit Training Centers. Since the RTC was always engaged with general farmer courses or staff courses, there was insufficient room for all required specialized courses for farmers as agreed at appraisal. 32. No information is available on the number of farmers or farmers' wives trained during Phase III. It is however understood that a considerable number of farmers of farmers' wives have been trained during the Phase and furthermore evaluation section reports that 60% of farmers' wives courses were agricultural based rather than being exlusively home economics as was the case in the previous phases. - 37 - Extension 33. Until 1977, agricultural extension services in Malawi had been on dualistic structure. Each of the major integrated rural development Project areas had its own extension service managed from the projects headquarters. However, farmers outside the Project areas were served by embroynic national extension service. 34. With the launching of NRDP in 1978, the Ministry instituted a geograhpical unit approach to rural development based on EPAs. As a result of this, the SVACP established five new EPA headquarters at Mikalango, Dolo, Kalambo, Livunzu and Nyachilenda. It is now felt the EPA network is adequate for the valley. 35. Prior to 1978, the extension technique in the Project area, emphasized on individual visits and to maximize farmers contact, a considerable number of DAs were employed as noted in Table 3 below: Table 3: Extension Staff-Farmer Ratio End of Phase III (1982) Description 1978 Projected Actual Field Assistance (FAs) 49 120 80 Development Asats.(DAs) 202 - 153 Ratio of TAs* to farm families 1:255 1:600 1:297 *The term TA covers both FAs and DAs. 36. During Phase III, the T&V system of extension was proposed for the Valley. This system has been employed in India and Thailand with considerable success. The main features of this extension system is the group approach and its complete union with agricultural research and training components. With the adoption of the T&V system it was envisaged that the farmer staff ratio would also be reduced tc 1:600 as noted in Table 3. The projected number of FAs, was never achieved because of inadequate supply from Colby College of Agriculture. With the Natural Resources College implemented, the supply of FAs in future will improve. 37. The extension service during Phase III was never reorganized to follow the T and V system as was envisaged at appraisal because of the following reasons: (a) By the beginning of the Phase, very few senior staff members either in the Project or the Ministry were familiar with the T&V system. (b) There was no proven technology available from research to be imparted to farmers (see section on agricultural research). - 38 - The Government is still aware of the inadequacy of the extension technique in the Project area, and it is considering introducing a modified T&V system in the area as has been done in other ADDs in the country. Crop Production 38. The main thrust of SVACP was on smallholder crop production. At the beginning of the Phase, there were 64,000 farming households with an average holding size of 1.88 ha and growing such crops as cotton, maize, sorghum, guar beans, groundnuts and bulrush millet. By the end of the Phase (1982) there were 69,200 farming households - an increase of 8%. This increase of farming households meant the reduction of holding size from 1.88 ha to 1.30 ha per family by the end of Phase III. Reduction of holding sizes in the valley meant reduction of proportion of cultivation land devoted to cotton production as shown in Table 5, 6 and 8. Table 4: Percentage of Land Devoted to Cotton by Holding Size Holding Size Cotton 0-0.25 0.26-0.5 0.6-1.00 1.01-1.5 1.6-2.50 0.07 13.49 13.50 18.79 23.95 Source: SV.? Evaluation Surveys 1978/79. 39. Table 5 presents cropping pattern in 1975/76 and 1981/82 seasons. It is apparent from this Table that percentage of cultivated land devoted to cotton between 1975/76 and 1981/82 seasons decreased due to reduced holding sizes and shift of land mainly to food crops. From the same Table, proportion of cultivated land devoted to sorghum increased during the same period possibly to meet farmers' subsistence needs as individual holdings became smaller and fragmented. Table 5: Cropping Pattern (%) Crop 1975/76 1980/81 1981/82 Cotton 32 23 20 Maize 31 28 29 Sorghum 9 17 18 Guar n.a. 6 9 Groundnuts 6 n.a. 1 Millets 11 n.a. 6 Other 1/ 11 26 17 1/ Includes vegetables and pulses. Source: Shire Valley Land Use Survey 1975/76 IBRD, 1980/81 NSSA Preliminary Report and Land Husbandry Section, ASA for 1981/82. - 39 - 40. Table 6 compares planned and actual hectarages by year and by different crops during Phase III. In general, the planned hectarages were never met in most crops except for sorghum because the assumption made during planning such as improved extension effort, availability of suitable cereal varieties, moisture conserving cultivation practices and favourable producer prices never lived up to the expectations of the appraisal team. 41. The projected and actual yield levels are shown in Table 7. In 1981/82, the cotton yields exceeded the projected ones by 13% whilst the maize and sorghum yields were below the projected levels. It can be observed, however, that the trends of maize and sorghum yields from the initial year have been improving. For instance, the maize yield improved from 933 kg/ha in 1978/79 to 1,350 in 1981/82 and the sorghum yield improved from 391 to 710 kg/ha. The major factor for not achieving the targeted yield was drought which was not properly addressed during the Phase. The projected yields assumed the use of drought resistant seeds which were unavailable. Table 6: Crop Development (ha) 1978/79 1979/80 1980/81 1981/82 Crop Planned 1/ Achieved Planned 1/ Achieved Planned I/ Achieved Planned I/ Achieved Cotton 28,000 17,000 29,000 18,990 31,000 18,800 32,000 19,354 Maize 2/ 39,000 27,360 40,000 27,250 40,000 28,580 41,000 28,490 Sorghum 11,000 18,680 11,000 24,000 12,000 28,310 12,000 17,907 Millet n.a. 5,000 n.a. 5,400 n.a. 5,488 n.a. 5,575 Rice n.a. 1,500 n.a. 1,400 n.a. 1,264 n.ao 1,127 G.nust n.a. 500 n.a. 400 n.a. 827 n.a. 1,254 Guar beans 800 3,990 1,000 5,850 1,200 6,780 1,400 9,197 1/ Combines unimproved, Stage I and Stage II. Combines maize grown on Dimba gradens. Table 7: Yield Levels by Crop and by Year (kg/ha) Actual 1981/82 as % of Crop Expected 1/ 1978/79 1979/80 1980/81 1981/82 expected yield Cotton 750 625 635 570 850 113 Maize 1,500 933 619 1,310 1,350 90 Sorghum 1,200 391 544 640 710 59 Millet 700 256 156 n.a. 206 - G.nuts 700 579 49 n.a. 314 - Guar n.a. 110 294 407 435 - Rice 1,600 2,460 428 n.a. 1,534 - Note: Dimba maize is excluded. 1/ Based on Stage 11. - 40 - 42. Table 8 presents a comparison of actual crop production estimate and the projected. The Table has been derived by multiplying yield and hectarages of different crops as given in Tables 6 and 7. It is apparent from the Table that cotton production targets were never achieved. Maize production also lagged behind projected estimates, but actual production during the Phase shows an increasing trend - thus showing increasing productivity. Sorghum production in SVACP showed an increasing trend up to 1980/81 to that of maize production because of large proportion of cultivated land devoted to the crop then it dropped considerably. 43. Production targets of guar beans were achieved during the Phase; production of other crops such as bulrush millets, groundnuts and rice however never met expectations of the appraisal report. Seed Multiplication 44. The 150 ha seed multiplication farm at Makhanga was intended to complement the Agricultural Research effort to produce and release suitable cereal varieties for the Shire Valley. Work on the farm started early in the phase with clearing, fencing and laying of irrigation canals. Eight staff houses, office block and a workshop were constructed. The seed multiplication project was, however, abandoned in 1980. The future of this facility cannot be stated here as this will depend upon the Department of Agricultural Research which is currently under review. Agricultural Research 45. Considerable progress has been achieved under the objective of developing drought resistant varieties of maize. Maize cultivar trials were conducted throughout :he Valley since 1978/79 season using mainly exotic hybrid varieties. These trials resulted into limited release of three hybrids for the Valley as follows: PNR 353 for the East Bank and Western Escarpment. PNR 95 for West Bank north of Bangula. R201 for the Valley South of Bangula. Table 8: SVACP - Projected and Actual Production 1978/79 1979/80 1980/81 1981/82 Caop Projected Actual Projected Actual Projected Actual Projected Actual Cotton 21,000 10,625 21,750 12,059 23,250 10,716 24,000 16,451 Maize 58,500 25,380 60,000 16,250 60,000 37,440 61,500 38,462 Rainfed rice n.a. 3,960 n.a. 599 n.a. na. n.a. 1,919 Sorghln 1/ 13,200 7,304 13,200 13,056 14,440 18,118 14,400 12,714 Millet n.a. 1,278 n.a. 835 n.a. n.ea n.. 1,148 G.nuts (un- shelled) n.ea. na.. 16 n.a. a.a. n.a. 394 Guar beam 485 3,289 485 5,163 1,620 n.a. 1,620 4,000 1/ White and red sorgiun Source: Author's estimate based on Tables 6 and 7. - 41 - In 1980/81 season, however, these hybrids were discontinued because it was observed that seed viability of these hybrids was low. New series of cultivar trials have since started screening locally bred hybrids. It is hoped a promising hybrid from these new trials will replace the PNR and R201 hybrids. 46. Other research work was done on cotton, legumes (groundnuts, guar beans and cowpeas), sorghum and millets in conjuction with national trials programmes. As a result of these trials, a new cotton variety (Makoka 78) was released during the Phase and a new sorghum variety PN3 was also recommended for the Valley. 47. During the Phase, maize trials were also carried out on dimba land along the Mwanza and Shire rivers using residual moisture. The objective of these trials was to select high yielding and stable maize cultivars. Results from these trials are not conclusive. Trials on phaseolus beans, sweet potatoes and vegetables were not carried out on Dimba land as originlly intended because of shortage of suitable dimba land. The section was also involved in joint activities with Land Husbandry Section at Zunde farm. Livestock Development and Disease Control 48. In Phase III, Project objectives as regards livestock development centered primarily on cattle. The proposed two cattle markets at Chimombo and Mthumba were constructed. The five ox-training units at the new EPA centers were also built. A number of dipping tanks were renovated and two new dipping tanks were built at Kaombe holding ground and Mthumba. 49. The Tsetse survey was completed in the first two years of the Project and a report is available. Trypanosomiasis vaccination programme and a FMD vaccination programme have been carried out during the Phase. Under Trypanosomiasis vaccination programme, 39 and 44% of all valley animals were vaccinated in 1981 and 1982 respecitvely. In case of FMD, 79 and 78% of all cattle were vaccinated in 1980 and 1981. 50. The objective at appraisal was to increase the rate of cattle off take with a view to stabilizing the total herd at 60,000-65,000 which is the carrying capacity of the valley. Table 9, however, reveals that the total herd increased and off take did not increase as envisaged at appraisal. The following explains the failure to stabilize cattle population: (a) policy prohibiting selling or slaughtering of heifers and productive cows; and (b) restriction of cattle transfer out of the valley as control measure of Foot and Mouth Disease. Natural Resources 51. Introduction. Because of duplication of activities between land husbandry and agricultural research, limited planning and coordination of relevant research between agriculture/land husbandry and crop production - 42 - and consequently, piece-meal approach of land husbandry in previous phases, it was decided to leave all project related research under the direction and supervision of the Senior Agricultural Research Officer and then create a natural resources division. The Natural Resources Division embraced Land Husbandry, Forestry, Fisheries, Hydrology and Wildlife. The division was to protect and utilize all available resources following appropriate land use. 52. Land Husbandry. The Land Husbandry Section was expected to continue with its overall concern for the entire Valley's ecological progress and the development of programmes related to long term preservation of the valley's natural resources. This general objective was broken down as follows: - Natural Resources Mapping - Biological control of soil erosion. - Mapping of suitable irrigation areas. - Planning of physical infrastructure. (i) Natural Resources Mapping - Very little was achieved on this programme because the second Professional Officer was not recruited in the first two years of the Project, and when he was recruited, he was untrained and had no experience and therefore could not take up the detailed programme of this kind.. The first Land Husbandry Officer who had some experience was expected to deal with the forestry component because of the unavailability of a Forestry Officer and as a result, not much could be done to assist the inexperienced Land husbandry Officer. Some of the mapping however was done. Maps showing present services and organization by EPA were drawn. The work of A.R. Stobbs, Natural Regions and Areas of Malawi Environmental Conditions and Agriculture was also drawn onto Maps of 1:50,000 scale for each EPA providing a useful reference for extension staff. Some work was done on soil maps by EPA and current land use on few using 1977 aerial photos. (ii) Biological Control of Soil Erosion - In addition to investigating appropriate farming systems, Zunde farm continued to demonstrate the effect of early cultivation, early planting, proper spacing and rotations on crop yields and soil structure as a method of moisture conserving cultivation practices. Farmers and extension staff visited the farm regularly. The effect of this demonstration has been noticed in CK5 and CK6 EPAs. Extension staff are now running after own demonstrations in EPA centers basing on Zunde guidance. (iii) Mapping of Suitable Irrigation Areas - This was adequately covered by the National and Shire Irrigation Study by Huntings Technical Services and a report is available. The Land Husbandry Section assisted the study team, and the recommendations of the report are still being studied by the Government. (iv) Planning of Physical Infrastructure - Major roads could not follow ridged crests because the Project area is valley with drainage lines cutting across it from East to West whilst main roads linking major centers run in a north-south direction. There were some liaison between Roads Section and the Land Husbandry Section during implementation and the section planned 65 km of roads for Majete Game Reserve. - 43 - Table 9: Cattle Numbers, Growth Rate and Offtake SVACP Description 1976 1977 1978 1979 1980 1981 Total herd 63,109 61,550 74,970 1/ 76,110 1/ 78,900 80,600 Slaughter 2,923 652 1,520 1,820 2,390 1,990 Market sales 2,124 1,940 4,150 4,963 4,660 5,000 Total offtake 2,923 2,592 5,670 6,783 7,050 6,990 % of offtake 2/ 4.6 4.2 7.6 8.9 8.9 8.7 Source: Project Management and Ministry of Headquarter files. 1/ Includes cattle on private estates and Government stations. i/ Sum of growth rate (%) and % of offtake. 53. Some work was done at Chididi, Lulwe and Gaga hills on the introduction of perennial crop growing. Cashew trees, bananas and citrus were recommended. The uptake of these tree crops by farmers was slow because of organizational problems on the part of the Project, and insufficient enthusiasm on the part of the farmers. Overall, there has been a considerable achievement of Land Husbandry objectives. Forestry 54. The rapidly expanding population in previous phase with inherent demand for land, building poles and fuel led to large areas of land being open up to cultivation and settlement without any form of integrated long term planning for afforestation. It was with this background that a forestry component was established in Phase III. The primary goal of this forestry component was to involve local population in the planting and care of forestry plantations. There were also research and natural woodland management activities as part of this component. 55. Because of late appointment of forestry staff, as a result of normal delays of the Government's appointment procedures, forestry programmes started late. Reafforestation on escarpments was not done in Phase III. A total of 12 ha of woodlot, however, were established at fishing landing sites using tree species suitable for smoking fish. A total of 6 ha demonstration wood lots were also established at EPA centers. Sixteen tree nurseries were established during Phase III with a total of 222,627 seedlings. Only 16% of these were sold to farmers; 70% were issued to schools during National Tree Planting and 14% were left in the nurseries. 56. Planting of more trees in woodlots, during National Tree Planting days or as part of an agro-forestry system of production is partial approach to afferestation problem. To overcome the defforestation problem in the valley, there is need to encourage the establishment of gazetted forestry reserves either run by the Project through the forestry component or by the District Councils. - 44 - Wildlife 57. The.appraisal targets for improvement of Langwe and Majere roads, housing for staff and chalets for visitors were achieved. Water supply was also provided at Lengwe; fencing and game ditches along the t)ark boundaries were also provided at Lengwe. Bydrology 58* The appraisal targets for the hydrogeology section were achieved in various levels as follows: the preparation of a master plan for water development by the then Hydrologist has not be prepared, and it is, therefore, difficult to make a detailed assessment of this component. Finances for the National and Shire Irrigation Study were provided (see Table 17) and the study was carried out by Huntings Technical Services Ltd., and the Government is still studying recommendations of the report. As noted in Appendix Table 5, 80% of the planned boreholes were drilled and pumps fitted. Seventy percent of the shallow wells were completed. Some of the planned activities such as the purchase of Gamma ray, an electric lagger and a Neutron Probe were not bought. The proposed eight sediment sampling stations were also not established. Reasons for the failure of some of the proposed activities are not known because there was no proper handover notes prepared by the then Hydrologist. Fisheries 59. Objectives for fisheries component during Phase II were in three broad categories as follows: (a) Infrastructural development; (b) Improving level of fish catches in the valley; and (c) Carrying out various research. 60. As noted in Table 5 in the Appendix, the infrastructural develop- ment was achieved. Construction of roads to these new landing sites was also completed. Fish catches statistics are presented in Table 10. Looking at the column for actual catches, the figures are lower than before Phase III, but when the actual figures are compared with appraisal figures an Impressive achievement of appraisal targets is revealed. 61. Various research or investigations were to be carried out to determine unexploited fish stocks, pesticide pollution of the Shire River, and the most suitable and effective methods of fishing in the valley. Determination of unexploited fish stocks and the study of fishing methods were never done because of the failure to recruit the Gear Development Officer. As regards to post pollution, the Project has sent water and soil samples to Tropical Products Institute for testing. Results for these tests are not yet available. Table 10: Fish Catches SVACP Appraisal Actual estimates catches Actual as a Year (M. tons) (N. tons) % of appraisal 1976 - 3,128 1977 - 4440 1978 - 3,933 - 1979 675 3,305 490 1980 1 350 2 654 197 1981 2,025 3,249 160 Source: Project Management. - 45 - Credit 62. In Phase III, group credit policy was proposed for seasonal credit rather than the individual credit system which had been employed during the previous two phases. The issue of medium credit, however, remained on an individual basis and this was mainly for fishing boats for fishermen and ploughas ridgers, ox-carts and other form of a farm implements for farmers. Unlike Phases I and II, Phase III issued credit for leucaena development in addition to the normal crop inputs. 63. Table 11 presents number of farmers receiving seasonal credit by year, financial aspects of credit are outlined in Chapter 5. In the first year of Phase III, 9,635 borrowers or 92% of all short term credit borrowers received seasonal credit as individual borrowers. By 1981/82 season the percentage of seasonal credit borrowers receiving loan individually had dropped to 4%. The lifting of the 10% deposit requirement in 1980/81 for all groups; lower service charge (10% for groups and 15% for individual borrowers) and, the intensification of group credit campaign were responsible for this rapid shift from individual to group credit. Table 11: Number of Farmers Receiving Seasonal Credit Description 1978/79 1979/80 1980/81 1981/82 Number of group members 821 104 6,932 6,959 Number of individuals 9,635 8,813 1,792 253 Total number of farmers receiving credit 10,456 8,917 8,724 7,212 Source: NADD Credit Section. 64. Although there was a rapid shift from individual to group credit as noted above, the total number of seasonal credit borrowers during the Phase decreased with time because of declining crop production. Credit recovery during the Phase, however, improved as compared to Phase II. Recovery has been at or above 82% by September 30, of each year (official closing date of seasonal credit repayment). The intensification of a no- credit-to-defaulters policy in phase and the effect of group credit during the Phase ezplain this improved credit recovery. 65. As noted in Appendix Table 6, medium-term credit was extended to fishermen in the first two years of the Phase although the plan was to extend this facility to fishermen throughout the Phase. Reasons for the discontinuation are not known and recovery figures for this credit item ar not available. It is, however, understood that repayment was satisfactory. - 46 - Table 6 also shows disbursement of both medium- and seasonal credit in physical quantities. In general, appraisal targets for both short- and medium-term credit were not achieved because of the declining number of borrowers. Evaluation and Monitoring Section 66. Some success in the implementation of the section's programmes during Phase III was achieved in as far as collection of statistics and developing common computer programme are concerned. Several surveys were conducted (see Text Table 12) and cover was quite good ensuring more representative data base. As noted from the Table, there were delays in processing survey information because of lack of experienced staff. However, in 1980/81 season with the coming of DDA staff, the section managed to pool the three past survey data, analysed it and produced a report. This report revealed a number of socio-economic factors (such as land pressure and farmers' priorities) that influence farmer behavior in the Shire Valley. 67. A number of reports/working papers were also produced by the section. some of these papers were useful in gross margin calculation; crop estimates and Annual Workplan exercises carried out by the Project management. Information from Evaluation Section was also used in compiling this report. Table 12: Ngabu Evaluation Section Surveys by November, 1982 Type of Survey conducted Processed Written Up Remarks 1978/79 H/Hold Comp. / Drafted Garden / / Yield / H/Hold (Muona Rice Garden (Kasinthula Rice ( Winter / / 1979/80 H/Hold Comp. / / Garden / Yield / Under Five Clinics Cattle Census Dipping Tank around Ngabu Area Requested by Animal Husbandry Officer - 47 - Type of Survey conducted Prcessed Written Up Remarks 1980/81 H/Hold / / Garden 1 and 2 / / Yield / / Resources / Wood Energy / Income and Expenditures with NSO Livestock Nutrition Extension Crop Storage 1981/82 B/Hold no Garden no Yield no Cotton growers follow-up survey being processed 1/ Processed and written up by NSO as part of NSSA. 2/ Processed and written up by Wood Energy Unit. Source: Evaluation Section, Ngabu A.D.D. Infrastructural Development 68. A summary of infrastructure achievement compared to appraisal target is given in Appendix Table 5. the contruction of staff houses during the Phase was carried out by the Project's Construction Unit and the Ministry of Works and Supplies. The construction programme was behind schedule during the first two years of the Project but the situation Improved somehow in later years and by the end of the Phase only the road programme remained incomplete. The slow start of the Project in general and the construction unit in particular resulted in cost penalties as noted in Chapter 5. Figures on the number of houses built during the Phase are incomplete, the information there suggests that 47% of the proposed houses were constructed during the Phase. The inavailability of Chief Works Supervisor in the last tow years of the Phase as noted in Figure 2 affected the construction programme. 69. Of the 139 km of roads planned 111 km were either constructed or upgraded. Five km of fisheries roads were upgraded during the Phase. It is understood the section was, however, unable to construct roads at Lengwe Game Park as agreed at appraisal, and to complete construction of East Bank road in time because of frequent breakdown of construction equipment. - 48 - 70. All the proposed Public Health construction were completed; the five EPA centers were built but none of the proposed Village Training Centers were built because of other technical reasons. In general, construction programme for other infrastructure except housing has been satisfactory. Scheduling of construction programme between years was satisfactory. Public Health 71. Under Phase III, two health sub-centers (one at Dolo and the other at Kakoma) and four health posts (Kavalo, Kubalalika, Mkango and Mchacha) were constructed. A number of houses were constructed at various places, and an administration block at Chikwawa Hospital was also renovated to a public health office. Apart from the infrastructural work the public health component was also involved in: bilharzia control at Kasinthula and Mlolo schemes; mobile eye unit programme, and school health programme involving other vaccination against communicable diseases and health education. The section also provided furniture and transport to the Shire Valley Tuberculosis Control Project. Phase III involvement in health programme in the Valley was largely on infrastructural development and this has been achieved. V. FINANCIAL PERFORMANCE 72. This chapter reviews the Projects' financial affairs with particular emphasis on: (a) performance in terms of comparing actual and planned expenditure, and drawing out key issues; and (b) agricultural credit, in terms of its financial management and resources. The previous chapter provided a physical description of the Project. The various components are now set in perspective by detailing the financial provision and expenditure. Contribution to sources of finance as compared with those planned at appraisal are given in Table 13 below. Table 13: Contribution to Project Funding (MK million) Source Appraisal Plan Actual Funding Amount % Amount % IDA 9.7 85 8.7 85 MG 1.7 15 1.5 15 Total 11.4 100 10.2 100 Source: Appraisal Report and Project Management. Note: The difference between total planned and actual expenditure is a result of changes of exchange rates between the time of planning and implementation. - 49 - Actual contribution to Project financing by IDA was 90% of appraisal plan because of changes in exchange rates between the U.S. Dollar and the Malawi Kwacha. At appraisal, the exchange rate was MK.91 equivalent to one U.S. dollar, and this declined to MK.8 during implementation, a change of 10%. As a result of this change, local contribution correspondingly declined to MK1.5 m. 73. Table 14 presents Malawi Government contribution by year. By the end of the Phase, local contribution was 94% of the appraisal estimate. Table 14: Malawi Government Contribution by Year (MK '000) Appraisal Actual as a % of Year Estimate Actual Appraisal 1978/79 226 380 168 1979/80 467 465 99 1980/81 471 320 68 1981/82 462 267 58 Total 1,626 1,535 94 Source: Appraisal Report and Financial Coordinator, Ministry of Agriculture. 74. From this Table, it is clear that actual local contribution as a percentage of the appraisal plan has been decreasing with time. The decrease has resulted the Project activities and salaries of local staff that were to be paid from local contribution to be paid on foreign contribution. Consequently, this situation resulted in overstretching of various financial categories especially that for operating costs, and thus leading to overclaiming on that category. Draw Down of the Credit 823-MAI 75. The draw down of loan is an important aspect of the financial management of any project. Ideally, an advance of loan will be drawn down, from which relevant project expenditures can be financed. If this is not possible, it is desirable to draw down the funds as rapidly as possible after the expenditure is incurred in order to minimize the bridging finance required which would be met by Malawi Government. 76. Table 15 provides a comparison of the anticipated draw down at appraisal with actual. A major feature of this Table is that the first draw down was 50% of the planned, indicating a slow start of the Project. This is not surprising considering that the planned draw down in the first year of the Project assumed that all required incremental staff and equipment would be in place within the first year of the Project. The slow start of the Project resulted in the rescheduling of a number of project activities particularly the construction of roads and staff houses. - 50 - Table 15: Planned and Actual Expenditure (MK '000) 1978/79 1979/80 1980/81 1981/82 Total Planned IDA (including contingency) 4,128 2,294 1,927 1,468 9,817 MG (including contingency 226 467 471 462 1,626 Sub-Total 4,354 2,761 2,398 1,930 11,443 Actual IDA 2,155 2,843 1,600 1,857 8,699 1/ MG 380 465 320 267 1,535 1/ Sub-Total 2,535 3,308 1,920 2,124 10,234 1/ I/ Includes balance to be spent in 1983; breakdown of the balance is as follows: IDA 243,000; MG 103,000; and total 346,000. Source: Appraisal Report pages 25 and 45. Financial Coordinator, Ministry of Agriculture. 77. Delay in the construction programme led to cost penalties as noted in Table 16 below. Table 16: Average Cost by Year DL 3 EL 2 Description Houses Houses Borehole Health Post 1978 n.a. n.a. 4,200 n.a. 1979 n.a. 4,500 4,500 n.a. 1980 n.a. 9,000 5,400 n.a. 1981 30,000 10,000 9,700 15,600 1982 30,500 10,300 n.a. n.a. Source: Project Management. The cost of a borehole or an EL2 house doubled during the Phase. This situation partly explains the failure to build all houses agreed at appraisal. 78. Table 17 presents total expenditure by section excluding contingencies. Overall, actual expenditure during the Phase was within the appraisal provision. There was, howevsr, overexpenditure between sections because of internal reallocation of project finance. - 51 - Table 17: Total Expenditure by Section (MK '000) Actual as a % of Component Appraisal Actual Appraisal PMs Office 941 1,712 182 Finance 304 1,276 420 Evaluation 274 202 66 Extension 993 921 93 Training 374 264 Research 714 389 54 Seed Multiplication 774 54 7 Credit Administration 331 U.a. Animal Husbandry 651 438 67 Land Husbandry 232 188 81 Fisheries 208 152 73 Forestry 332 76 23 Wildlife 123 132 107 Hydrology 465 266 57 Irrigation Studies 432 655 152 Health 363 148 41 Building Maintenance 599 n.a. Roads 1,025 864 84 Mechanical Maintenance 127 468 369 Total 9,262 9,888 1/ 89 1/ This total is not the summation of the column because of the missing data on credit and maintenance. Source: Appraisal Report and Project Management. 79. Table 18 shows withdrawal of the loan by category. Inter- category comparison of expenditure shows differences and this must have been the result of internal reallocations of funds. Table 18: Withdrawal of the Loan by Category (MK '000) Amount of the Final Category Credit Allocated Draw down 1. Vehicles & Equipment 1,507 1,265 2. Civil Works 1,771 1,485 3. Technical Assistance Staff 300 519 4. Operating Costs 5,251 6,310 5. Consultants 432 655 6. Unallocated 2,181 - Total 11,442 10,234 Source: SVACP Credit Agreement document, and Project Management. - 52 - satisfactory. Financial information was kept in a systematic manner and was easily retrieved, and financial audits have been carried out annually. The following are the amounts that have been submitted to Treasury for onward transmission to Donors: Development Fund - Accounts at 30/9/1979 at 30/9/1980 at 30/9/1981 at 30/9/1982 Credit Fund - Accounts at 30/9/1979 " " at 30/9/1980 9 " at 30/9/1981 at 30/9/1982 Credit 80. The provision of credit facilities to farmers has always been a central feature of project activities, although the scale and nature of credit has changed over the years since Phase I. This section is concerned primarily with financial aspects of credit, while para 61 deals with physical aspects and broader issues. 810 The background to credit provision to smallholders in Malawi has been set out in GITEC, 1980, a study commissioned by the Ministry of Agriculture. It is proposed here, therefor, to deal only briefly with general background. 82. Credit may be either seasonal, granted to individuals or groups, or medium-term, granted to individuals. National policy emphasizes the desirability of farmer clubs for seasonal credit. The main benefits of the group/club are considered to be reduced administrative costs and Improved recovery rates. 83. Currently, an interest fee of 15% on seasonal credit issues is payable by individuals and non-self accounting groups, and 102 by self accounting groups. A ten percent annum interest is payable on medium-term loans. 84. Loan. The value of seasonal loans issues within SVACP since 1968/69 is shown in Table 19. Details of medium-term loans issues are not available. The Table also indicates the number of borrowers and credit recovery. Table 19: Value of Seasonal Credit Number of Amount Loaned Recovery Year Borrowers (MK '000) Rate (Z) Ii0771 17 132 96 1971/72 5,4 218 87 1972/73 6,760 474 63 1973/74 10,396 294 80 1974/75 15,320 335 64 1975/76 24,031 46 4 1976/ 7 n.a. n.a. n.a. 19777 8 n.a. n.a. nba. 1 97 9 10,456 208 88 1979/80 8,917 228 82 1980/81 8,724 285 82 1981/82 7,212 178 92 Source: SVACP Appraisal Report Working Papers. - 53 - One main feature emerging from the Table is: the steady growth in seasonal loans advanced until 1975/76. In 1978/79, seasonal loans shows a fall back to 1971/72 season reflecting a decline in cotton production as noted in para 39. The Credit Fund 85. Separate financial accounts have been maintained in respect of credit finance from the beginning of the Phase. Table 20 presents Credit Fund Trading Profit and Loss Account from 1978 to 1982. Main features emerging from this Table are as follows: (a) Credit sales have been declining with time reflecting a general decline of credit uptake; (b) The operational costs have been increasing with time; (c) A decrease of net profits from MK160,623 in 1978 to HK20,308 in 1981 and subsequently to a net loss of MK4,907 in 1982. Table 20: Trading Profit and Loss Account as at 30/9 of Each Year Description 1978 1979 1980 1981 1982 Sales 886,691 222,580 257,546 316,038 22?,638 Purchases 1/ 65,132 1?,012 226,311 277,494 183,240 Gross Profit 231,806 42,525 57,233 59,785 46,212 Other Incomes 2/ 48,286 80,434 70,116 60,082 65,780 Salaries and Wages 35,233 70,256 59,025 58,140 75,288 Plant and Vehicles 7,885 22,697 24,422 17,794 22,102 Other Costs 12,292 27,177 22,545 23,625 19,519 Net Profit/loss 160,623 2,829 21,357 20,308 -4,907 1/ Purchase of all pesticides, fertilizer, seeds, sprayers and other farm equipment. 2/ Includes interest on daily Bank balance and interest on overdue accounts. Source: Extracted from Credit Fund Accounts, 1978-1982, Ngabu. Loan Transaction Costs 86. According to GITEC, 1980, at face value, SVACP incurred the second highest loan transaction costs of the four main agricultural projects. Such comparisons are of limited value, however, in view of the changing nature and content of credit over the years* However, a brief remark on credit administration should be made as follows: The growth in group borrowing does not appear to have brought about tangible reduction in credit administration costs, as relatively few groups, if any, achieve self accounting status. In view of the above remark, considering that credit - 54 - uptake is decreasing because of declining crop production, and since a considerable amount of staff input in respect of the credit programme is undertaken not only by credit staff, but also by extension staff, it is suggested that a comprehensive review of staffing for credit may be appropriate, together with projections of credit fund to ascertain if credit fund can genuinely be self sustaining. VI. IMPACT Introduction 87. Project impact is mainly achieved through increased agricultural production as a result of project activities. SVACP aimed at increasing offtake of livestock, production of crops such as cotton, maize, sorghum, groundnuts, millet, rice and guar beans, and also increasing fish catches. According to the Appraisal Report, incremental crop production would be based upon the adoption by smallholders of a basic package of practices designed both to reduce drought risks and increase production (see Appraisal Report pages 28-31). 88. The Project's Monitoring and Evaluation Section, however, did not identify the "adopting farmers" as defined at appraisal and monitor their progress in crop production. This exercise could not be done because as stated already, no new suitable technology ever became available for farmers to take up. Phase III continued to use the previous non-drought resistant varieties. Consequently, it is not easy to determine to what extent the first three project objectives mentioned in para 18 were achieved. As a result of this shortcoming, the crop production figures uced in this report have been derived using a method different from the one employed at appraisal. This report merely looks at crop production increments from the base year without having to distinguish between adopting and non-adopting farmers before at such figures. Incremental Crop Production 89. Table 21 compares actual and projected incremental agricultural production during Phase III. Table 21: SVACP Incremental Production (mt) Description 1977/78 1978/79 1979/80 1980/81 1981/82 Cotton Rectarages (ha) 25,000 17,000 18,990 18,880 19,354 With project production 15,000 10,625 12,059 10,716 16,451 Without project production 10,000 6,800 7,596 7,520 7,742 Incremental production 5,000 3,825 4,463 3,196 8,709 Phase III incremental - (1,175) (537) (1,804) 3,709 Appraisal estimate - 80 520 1,080 1,200 - 55 - Description 1977/78 1978/79 1979/80 1980/81 1981/82 Maize Hectarage 36,000 27,360 27,250 28,580 28,490 With project production 34,634 25,380 16,250 37,440 38,462 Without project prod. 21,600 16,416 16,350 17,148 17,094 Incremental production 13,034 8,964 100 20,292 21,368 Phase III incremental - (4,070) (12,934) .,258 8,334 Appraisal estimate - 216 1,100 2,675 5,310 Rice (rainfed) Rectarage 2,000 1,500 1,400 1,264 1,125 With project production 2,673 3,960 599 n.a. 1,919 Without project prod. 2,000 1,500 1,400 1,264 1,125 Incremental production 673 2,460 (801) n.a. 792 Phase III incremental - 1,787 (128) n.a. 119 Appraisal estimate - 270 575 1,210 1,860 Sorghum Hectarage 6,000 18,680 24,000 28,310 17,90? With project produdction 7,500 7,304 13,056 18,118 12,714 Without project prod. 3,500 11,208 14,400 16,986 10,744 Incremental production 3,900 (3,904) (1,344) 1,132 1,970 Phase III incremental prod. - (4) (5,244) (2,768) (1,930) Appraisal estimate - 863 665 803 1,11? Millets Hectarage 34,970 5,000 5,400 5,488 5,57? With production 6,994 1,278 835 n.a. 1,14? Without project prod. 5,455 3,750 4,050 4,116 4,18? Incremental production 1,539 (2,472) (3,215) n.a. (3,03?) Phase III Incremental prod. - (4,011) (4,754) n.a. (4,57?) Appraisal estimate - 55 250 585 n.a. Groundnuts Rectarage 1,455 500 400 827 1,254 With project production. 582 69 16 n.a. 394 Without project prod. 290 100 80 165 251 Incremental production 292 (31) (64) - 143 Phase III incremental prod. - (323) (356) - (149) Appraisal estimate - - 54 81 117 Guar Beans Hectarage 5,000 3,990 5,850 6,780 9,197 With project production 3,136 3,289 5,163 4,150 4,000 Without project prod. 500 399 583 678 920 Incremental production 2,636 2,890 4,578 3,473 3,080 Phase III incremental prod. - 254 1,942 837 444 Appraisal estimate - 485 425 1,620 1,620 - 56 - Actual Phase III incremental production for all crops was lower than what was anticipated at appraisal. In fact, Phase III incremental crop production for cotton, sorghum, millet and groundnuts has been negative throughout the Phase. Reasons for the crop production failure are subdivided as follows: those affecting cropped area and those affecting yields. These problems have been discussed in paras. 38 to 43. 90. Factors such as low producer prices, low plant population densities and poor crop husbandry practices are common throughout the country and even in more successful projects, such as Lilongwe Rural Development Project. These factors, therefore, do not serve to explain the real problem of crop production in the valley. The availability of off-farm employment at SUCOMA as an explanation of low crop production is difficult to substantiate because SUCOMA has been there even before Phase I. The principal factor affecting crop production in the valley is unreliable rainfall and its poor distribution. Within a four year period, for instance, Ngabu Station has had an annual minimum rainfall of 550 mm and a maximum of 1,174 mm supporting the unreliability of rainfall in the valley. 91. The problem of drought was noted in the 1976 Crop Review Report; Phase III Completion Report and Phase III Project proposals. In the Phase III proposals the drought risk as it affects crop production was given priority. As a starting point to minimize this drought risk, a smallholder irrigation component was proposed at an estimated cost of MK2.7 m. It is noted with concern that this proposal was omitted at appraisal. Moisture conserving cultivation practices were instead recommended and an immediate solution to the drought problem. The moisture conserving cultivation was based on box ridging. As a result, it was difficult to implement in an area where only 5% of farmers ridge their land. In short, Phase III the Project was aware of the drought risk in the area, but did not address itself directly to the problem. Livestock Development 92. Cattle population in SVACP was 80,600 by the end of the Phase. The objective of stabilizing cattle population at about 65,000 was not achieved. Annual offtake did not increase as envisaged at appraisal because as already noted in para 49, only males and non-productive could be sold or slaughtered, and there was a restriction of cattle transfers out of the valley as a measure to control FMD spread. 93. As regards disease control, vaccinations have been carried out for the FMD and Trypanosomiasis programme. Cattle dipping continued during the Phase. The impact of disease control programme is noted from the high annual cattle population increase in an area where cattle diseases are common. Agricultural Credit 94. Overall, credit disbursement in Phase III was generally lower than in Phase II and when compared with appraisal targets, the achievements were generally low. The number of borrowers in Phase III were also lower than in Phase II because of toughening of credit policy against defaulters. - 57 - Group credit almost replaced individual credit during the Phase as a result of intensified credit campagins and the lifting of the 10% deposit requirements in 1980/81 for all groups. In Phase Ill, credit recovery improved considerably as compared to Phase II recovery rates. Impact of credit component during the Phase has been on the reorganization of credit activities and therefore, setting a new basis for future agricultural credit in the Shire Valley. Direct impact of credit activities to either crop production or fisheries has not been noted. Extension and Training 95. By the beginning of the Phase most of the grass root staff were non-established and were Development Assistants by grade (staff with no formal training in agriculture). The situation did not change very much during the Phase. Individual visits and meetings with farmers were principal methods of reaching farmers. Training component agumented extension activities by organizing both staff and farmer courses at RTC and DTCs. Training programmes for the RTC ere drawn up annually by the training officer through consultations with extension staff and other subject matter specialists Extension and training components did not have proven technology to impart to farmers, and as a result, the impact of these two sections has been minimal. Agricultural Research 96. Efforts to identify cereal varieties suitable for the valley started in Phase III. The PNR maize varieties, which looked promising early in the Phase, were discontinued because it was noted that seed viability was low. Drought resistant sorghum variety still looks promising and furhter trials are being made. To date, no suitable cereal variety has been recommended to extension staff. Consequently, the extension staff do not have proven technology (that would minimize drought risk) to impart to farmers. Economic and Social Impact 97. The weakness of data in many areas makes it impossible to appraise the overall economic and social impact of SVACP. Because of data limitation, no attempt has been made to use such indicators as business licences and resource ownership. Evidence on income from sales to ADMARC, cattle offtake and fish catches show a marked downturn throughout the Phase. 98. Actual Economic Rate of Return (IRR) of 15%, however, has been estimated based on the actual incremental crop production as the project benefits and using project costs that are believed to have had a bearing on crop production. Details of the IRR calculation are summarized in Chapter 5. 99. The population in some parts of the project area has grown significantly faster than elsewhere from 1977 as people have moved into new areas (3% average growth rate). Population growth has been rapid in Mwanza Hills, Mwanza Valley and East Bank (EPA CK/4 especially) as these were areas of relatively low proportion of arable land under cultivation before Phase III. - 58 - 100. The development of water supplies and health services has substantially raised the level of social service provision at the rural level. Project staff assisted at "Under-Five Clinics" and taught health subjects to women groups and primary schools. The Project also assisted in Tuberculosis Control Project and Vaccination campagins of communicable diseases. The Project was also involved in Bilharzia control programme at Kasinthula and Mlolo agricultural schemes. All this contributed to the improvements in health in the Project area. 101. By financing an irrigation study and preparing a water resource plan in the valley, Phase III set a basis for future irrigation development. The financing of Tsetse study and the initiation of MFD and Trypanosomiasis vaccination programme also set a strong basis for the livestock industry in the valley. 102. Phase III has continued to provide employment opportunities to the people. The Project has also contributed to local institutional development through the promotion of group credit. VII. INSTITUTIONAL PERFORMANCE 103. Institutional performance is reviewed in three broad categories: performance of the Project itself; Ministry of Agriculture and the donor agency. The performance of institutions within and related to the project have been touched upon in varying degrees of depth in earlier chapters. Each issue discussed has, in a sense, been a comment upon institutional performance. This chapter, however, brings all relevant points form those chapters together in an institutional context, discusses them by reference to individual institutions and adds some commentary. Project Performance 104. Management organization. The organizational structure effected in Phase II continued in Phase III with slight modifications. The overall implementation of the Project was the responsibility of the then Ministry of Agriculture and Natural Resources. The Programme Manager was responsible to the Principal Secretary through the Chief Agricultural Development Officer. As noted in para 25, the PM1's office was responsible for four operational divisions and Ancillary Service Division. Each division was headed by a division chief who was fairly senior to command executive power except Natural Resources Division, which was headed by a Professional Officer (PO). Project components were headed by section heads at PO/CTO grade. There was high turnover of staff in key posts as noted in Figure 2. There have been three Programme Mangers in four years; fou: Evaluation Officers and three Financial Controllers. Management meetings were organized monthly under the chairmanship of the Programme Manager and these meeting assisted in the running of the Project. Finance Section 105. As noted in Chapter 5, financial control in Phase III was adequate; expenditure kept pace with what wats anticipated at appraisal. Actual draw down of the loan was constantly lower than anticipated draw down especially in the first two years as a result of a slow start of the - 59 - Project, but this picked up well in later years of the Phase. Malawi Government contribution fell short of appraisal estimates especially in the last two years of the Phase. This resulted in over-stretching of the development funds and consequently over-claiming in some of the categories followed. The high turnover of Financial Controllers during the Phase did not affect the activities in the finance section because the accounting system was already in place. Evaluation and Reporting 106. A number of surveys were conducted during the Phase but not all this data was written up, let alone processed. The few papers that were produced have been useful in gross margin exercises, crop estimates and annual work plan for the Project. The post of Senior Economist for the section as agreed t appraisal was never created. This is noted with concern because such experienced person would have enough executive power and experience to coordinate activities of the section with other divisions and sections and also be able to design monitoring procedures for all activities in the Project. 107. Reporting was done monthly and quarterly following formats design by Ministry of Agriculture headquarters. Many quarterly reports were not produced in time. Malawi Government Performance 1.08 Performance of the Government during Phase III was satisfactory. Sixteeen percent of the non-established staff were taken on the established warrant during Phase III. Government introduced Annual Workplans and Budgets in the Project in 1979 with a view to assisting project management and implementation. Local costs contribution to the Project funding, though lower than expected, was a considerable achievement if seen in th' context of the international economic situation of the time. In fact, when cost is imputed on all assets developed by Government in the Project, the local contribution to Project funding is higher than 67%. Donor Agency Participation 109. The involvement of the donor agency is reviewed from two points of view: (a) The ability of the appraisal document to differentiate between project activities and objectives; (b) Supervision Missions during project implementation. 110. There is clear differentiation between project activities and overall project objectives in the appraisal report. As a result of this, it has been easy to determine what inputs and outputs were required during Phase III and consequently, what impact was anticipated. As regards to supervision missions, there were four supervision missions during the Phase. Implementation weaknesses or delays were always pointed out in round-up meetings. The supervision missions were flexible and prepared to accommodate changes in emphasis or implementation schedule as evidenced in the discontinuation of the seed multiplication component. - 60 - VIII. ECONOMIC BENEFITS1/ 111. In 1982/83 season, annual incremental crop production attributable to the Project was valued at MK2.75 m based on the ectual farm gate prices which are based on export parity prices. Sorghum, millet and leucaena have been excluded in the benefits for economic analysis because of lack of actual information from which to calculate farm gate values. 112. In addition to the quantifiable benefits, the population of the valley has benefited form improved potable watar supplies, health facilities, roads and future firewood supply. The Projqct has also promoted future irrigation development by financing an irrigation study (report available). 113. The overall economic rate of return (IRR) of the Project over 25 years is 14.92%. The economic analysis, like at appraisal, project and incremental farmers costs (crop input costs) were included to the extent they contribute to development of quantifiable benefits. Costs of extension, training, credit administration, seed multiplication and land husbandry have been included. Some 80% of research costs have been included as these are attributable to activities with a direct impact on present crop production while at least 20% represents an investment for valley development. Two thirds of forestry costs have been included, as these are estimated to have had an impact on agricultural production by reducing erosion and preventing excessive siltation; the balance of the forestry costs are excluded as they are directly related to fuelwood production the benefits of which have not been quantified. Half the cost of hydrology and water supply component have been included as they are related to crop production, since the village boreholes are used to supply water for crop spraying and livestock; the other half of the costs are directed towards potable water supply for human consumption. The costs of irrigation studies, fisheries, livestock, wildlife and Foot and Mouth Disease have been excluded. Some 80% of the road construction and mainte- nance costs have been included in the economic analysis because they directly support the development of agriculture, livestock and fisheries; the remainder have been attributed to roads for public use, the benefits of which have not been quantified. Costs of building and mechanical maintenance and administrative services (management, evaluation and finance components) have been included at 90% of their financial costs, since the remaining 10% covers overhead costs which are used in administering the non-quantifiable aspects of the Project. 114. The crop production benefits during Phase III have adversely been affected by: the inavailability of drought resistant varieties, drought and the failure to adopt the T&V system of extension. As a result of this, the actual IRR is lower than what was anticipated at appraisal. 1/ The economic analysis was revised in 1985 by Malawi Government staff and the revised analysis is at Appendix 1. - 61 - Table 22: IRR for SVACP Net Year Costs Benefits Benefit DSCPL 1 1,736 -409 -2,145 -1,866.5 2 1,619 -1,072 -2,691 -2,037.581 3 2,058 -152 -2,210 -1,456.11 4 2,324 2,622 298 170.85145 5 1,926 2,753 827 412.58034 6 1,926 5,661 3,735 1621.4152 7 1,926 3,073 1,147 433.279 8 1,926 3,097 1,171 384.91224 9 1,926 3,097 1,171 334.93638 10 1,926 3,097 1,171 291.44924 11 1,926 3,097 1,171 253.60834 12 1,926 3,097 1,171 220.68059 13 1,926 3,097 1,171 192.02808 14 1,926 3,097 1,171 167.09573 15 1,926 3,097 1,171 145.40052 16 1,926 3,097 1,171 126.52215 17 1,926 3,097 1,171 110.09489 18 1,926 3,097 1,171 95.800496 19 1,926 3,097 1,171 83.362046 20 1,926 3,097 1,171 72.538566 21 1,926 3,097 1,171 63.120375 22 1,926 3,097 1,171 54.925014 23 1,926 3,097 1,171 47.793714 24 1,926 3,097 1,171 41.58832 25 1,926 3,097 1,171 36.18861? NPV - -0.019108 DSF = 15% Table 23: Economic Costs (MK '000) Component 1979 1980 1981 1982 1983 Management 156 126 156 741 741 Administration 151 129 107 80 80 Evaluation 43 45 54 35 35 Finance 94 80 192 154 154 Forestry 0.6 11 23 16 16 Health 22 52 42 - - Water Resources 41 25 58 84 - Extension 215 212 195 242 242 Land Husbandry 44 48 55 49 49 Animal Husbandry 86 77 87 88 88 Training 74 77 60 40 40 Research 91 86 92 36 36 Construction 232 218 277 314 - Building Maintenance 224 155 201 171 171 Mechanical Maintenance 74 52 182 91 91 Crop Inputs 189 226 277 183 183 Total 1,736 1,619 2,058 2,324 1,926 - 62 - Table 24: Value of Incremental Production (MK '000) Year Cotton Maize Rice Groundnuts Guar Beans Total 1 (447) (273) 354 (79) 36 (409) 2 (339) (905) (28) (80) 280 (1,072) 3 (823) 537 - - 134 (152) 4 1,873 683 31 (33) 68 2,622 5 2,003 683 31 (32) 68 2,753 6 2,133 708 31 (32) 68 5,661 7 2,281 725 31 (32) 68 3,073 8 2,281 750 31 (33) 68 3,097 9 2,281 750 31 (33) 68 3,097 10 2,281 750 31 (33) 68 3,097 11 2,281 750 31 (33) 68 3,097 12 2,281 750 31 (33) 68 3,097 13 2,281 750 31 (33) 68 3,097 14 2,281 750 31 (33) 68 3,097 15 2,281 750 31 (33) 68 3,097 16 2,281 750 31 (33) 68 3,097 17 2,281 750 31 (33) 68 3,097 18 2,281 750 31 (33) 68 3,097 19 2,281 750 31 (33) 68 3,097 20 2,281 750 31 (33) 68 3,097 21 2,281 750 31 (33) 68 3,097 22 2,281 750 31 (33) 68 3,097 23 2,281 750 31 (33) 68 3,097 24 2,281 750 31 (33) 68 3,097 25 2,281 750 31 (33) 68 3,097 Table 25: Economic Far, Gate Values Crop 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988---2000 Cotton 380 445 456 505 540 575 615 615 615 615 615 Maize 67 70 74 82 82 85 87 90 90 90 90 Rice 198 220 240 260 264 264 264 264 264 264 264 G.nuts 244 224 220 220 215 215 215 215 215 215 215 Guar 140 144 160 152 1/ 152 1/152 1/152 1/152 1/152 1/152 1/ 1521/ I/ Average of years 1980 and 1981. - 63 - Table 26: SVACP Incremental Production (metric tons) Year Cotton Maize Rice Sorghum Millet Groundnuts Guar 1 (1,175) (4,070) 1,787 (4) (4,754) (356) 254 2 (537) (12,934) (128) (5,244) (4,754) (356) 1,942 3 (1,804) 7,258 - (2,768) - - 837 4 3,709 8,334 119 (1,930) (4,572) (149) 444 5 3,709 8,334 119 (1,930) (4,572) (149) 444 6 3,709 8,334 119 (1,930) (4,572) (149) 444 7 3,709 8,334 119 (1,930) (4,572) (149) 444 8 3,709 8,334 119 (1,930) (4,572) (149) 444 9 3,709 8,334 119 (1,930) (4,572) (149) 444 10 3,709 8,334 119 (1,930) (4,572) (149) 444 11 3,709 8,334 119 (1,930) (4,572) (149) 444 12 3,709 8,334 119 (1,930) (4,572) (149) 444 13 3,709 8,334 119 (1,930) (4,572) (149) 444 14 3,709 8,334 119 (1,930) (4,572) (149) 444 15 3,709 8,334 119 (1,930) (4,572) (149) 444 16 3,709 . 334 119 (1,930) (4,572) (149) 444 17 3,709 8,334 119 (1,930) (4,572) (149) 444 18 3,709 8,334 119 (1,930) (4,572) (149) 444 19 3,709 8,334 119 (1,930) (4,572) (149) 444 20 3,709 8,334 119 (1,930) (4,572) (149) 444 21 3,709 8,334 119 (1,930) (4,572) (149) 444 22 3,709 8,334 119 (1,930) (4,572) (149) 444 23 3,709 8,334 119 (1,930) (4,572) (149) 444 24 3,709 8,334 119 (1,930) (4,572) (149) 444 25 3,709 8,334 119 (1,930) (4,572) (149) 444 IX. DISCUSSION AND CONCLUSION 115. This Chapter brings together the previous key issues from Chapters 4 to 7 to provide an overall discussion of SVACP. Discussion in this Chapter is presented for the following components: (a) Project activities and overall implementation. (b) Financial performance. (c) Project impact. Repetition of previous comments is inevitable but serves to emphasize the points already made. Project Activities and Implementation 116. Phase III started on sche!ule in April 1978, with a retrospective financing. However, the Phase started at a slow pace because incremental equipment and staff were not yet in place. As a matter of fact, by the beginning of the Phase, tender procedures for Phase III vehicles and equipment had not yet been done, and posts for incremental staff were not - 64 - yet created. As a result of this, a number of Project activities such as: construction of roads and staff houses took time to complete. The slow start resulted in cost penalties due to inflation. It is not known, precisely to what extent the slow start has affected the overall Project achievement; it is, however, considered that the effect is considerable. In view of this, it is recommended for future phases to have year zero phases to allow procurement and recruitment of staff. 117. Generally, however, specific activities which were expected to be provided in Phase III as noted in para 19 have been achieved. The five new EPA centers and construction of additional cattle markets and dipping tanks was done. The Natural Resources Division was expanded as planned and the ground water survey and development has generally been implemented. Project management has generally implemented the activities which were agreed at appraisal. 118. The T&V system never took off the ground because of the following reasons: (a) By the beginning of Phase III, few senior staff at the Project as well as at Ministry Headquarters, if any, were familiar with T&V system, and there was no plan in the Project plan at least to allow Project staff to acquire that knowledge. (b) Over 60% of the grass root staff of extension section were non-established with very little knowledge of agriculture nd it would be difficult to implement this new method through such staff; and (c) As already noted in Chapter 4, there was no proven technology both from research in form of drought resistant varieties, and from land husbandry in form of moisture conserving cultivation practices to impart to farmers. Financial Performance 119. Financial performance during Phase III was satisfactory. The high turnover of staff in the section appear not to have affected the activities of the section because proper accounting system was established early in the Phase. Credit draw down was generally within appraisal plan. Annual audit reports were always prepared but not as timely as the Bank requires because of various reasons. Project Impact 120. Phase III as a consolidation phase was aimed primarily on increased agricultural production. The Project activities proposed and carried out in Phase III did not address sufficiently well to the drought problem in the valley. The moisture conserving cultivation technique as a solution involves, among other things, deep ploughing and box ridging. Deep ploughing in the Shire Valley where the soils are dry and of clay type is difficult especially when a hand hoe is used. Box ridging in an area where only 5% of farmers ridge their land is impossible to implement. Even - 65 - if deep ploughing was easy and ridging was common, the technique would not be taken up quickly enough to give the anticipated impact because it would take sometime to teach the extension staff. 121. The agricultural reearch component devoted its effort and resources to developing drought resistant varieties suitable for the valley. A number of foreign varieties were screened in the valley with a view of identifying suitable varieties. By the end of the Phase, there was no suitable varieties recommended and as a result, the extension staff did not have suitable package to recommend to farmers. Under research, releasing a crop variety and even screening a number of varieties in order to pick out the appropriate ce takes a number of years. It is surprising that Phase III planning propo-ad that the development of the variety, recommendation to farmers and the impact to be realized would happen simultaneously. 122. The cattle disease control programme made considerable impact as evidenced from the 5% annual cattle population increase. Cattle offtake, however, has not increased as expected because of lack of congruency between what was planned and Government policy as regards cattle marketing. This was an oversight in the planning. 123. The new extension system recommended in the appraisal report for the valley was developed in Asia where the culture of the people and agriculture system are different from those in the Shire Valley. The method has not yet been tried out or modified to suit local conditions, therefore, even if the method was implemented, it would not be effective. Conclusions 124. Conclusions for Phase III are as follows: 1. Project inputs such as incremental staff, equipment and required funding were successfully delivered; 2. The Project outputs such as water development, increased public health facilities, and establishment of new EPA centers, have been achieved considerably; 3. The anticipated project impact in crop production and livestock development has not been achieved because the Project outputs/activities were insufficient to deal with the drought problems in the valley. -. И" � � � � � г�, � �� 1 � � � - 67 - Appendix I Page 1 of 6 REVISED ECQfiOMIC ANALYSIS Introduction 1. This paper presents a revised economic analysis of the Shire Valley Agricultural Consolidation Project which was financed by the World Bank between 1978/79 and 1981/82 as presented in the Project Completion Report produced by the Planning Division, MOA, in October 1983. A number of questions have been raised by the World Bank regarding the assumptions underlying the first analysis and since no documentation existed It has been found necessary to re-do the analysis and incorporate the underlying assumptions. No effort has been made in this paper to include details of project implementation and social impact as these are adequately covered by the Project Completion Report# The Analytical Approach 2. In order to determine the effects of project investment, an attempt has been made to identify costs and benefits between "with project" and "without project" situations. The difference between these two situations represents net incremental benefits arising out of project activities* The economic rate of return (ERR) has been calculated using the net incremental benefits for 20 years, the expected life of the project. The costs and benefits involved in the calculation of ERR are described below in turn. 3. Costs. The cost figures for the first four years of the project are derived om actual expenditures, taking cognizance of the modifica- tions on page 36 of the Staff Appraisal Report. These cost figures have been projected to Year 8 at at 5% annual increase, From Year 8 to Year 20 al costs are assumed to remain constant. Other assumptions made are as follows: (a) Vehicles will be replaced every 5 years; (b) Equipment will be replaced every 10 years; and (c) Fifteen percent of the total expenditure has been assumed to represent the "without project" expenditure. This level of expenditure, which was required to maintain a minimum of services to the farmers, has been deducted to come up with incremental expenditure* 4. All capital and operating costs presented in Table I are economic costs since duties and taxes have been removed. (The Government purchases capital items duty free; vehicle maintenance and running costs may have taxed elements, e.g., fuel and certain spare parts. An adjustment has been made for this by deducting 20% from the operating costs)* - 68 - Appendix 1 Page 2 of 6 Table 1: Shire Valle Costs (MK '000) Capital Operating Total 85% Year Costs Costs Costs (1) (2) 1 1,736 1,476 2 1,619 1,376 3 2,058 1,749 4 2,324 1,975 5 278 1,459 1,737 1,476 6 159 1,532 1,691 1,437 7 51 1,609 1,660 1,411 8 19 1,689 1,708 1,452 9-20 1,708 1,452 Notes: (1) It has not been possible to separate actual total costs into actual capital and operating costs for the first four years. (2) Eighty-five percent represents project incremental costs, after deducting 152 need to maintain a minimum of services for a "without project" situation. 5. Benefits. Project benefits included in the analysis have been derived from incremental production of crops in the consolidation phase. The crops involved are cotton, maise, rice, groundnuts, sorghum, millet and guar beans. Actual incremental production figures for the first four years of the project have been used. hrese have arrived at by deducting crop production "without project" from observed production figures "with project". To obtain the "without project" figures 1976/77 production figures have been taken to represent a base situation. Crop production has then been assumed to increase at 32 per annum (i.e., at the same rate as that of population growth) in the "without project" situation.The incre- mental production figures for the first four years of the project life are shown in Table 2. - 69 - 696 &Pedix1I Table 2: Shire Valley Incremental Production (replaces Table 21) 1976/77 1977/78 1978/79 1979/80 1980/81 1981/82 Cotton: Incremental Production Area (ha) 17,000 18,990 18,800 19,354 Production (with project) 10,625 12,059 10,716 16,451 Production (without project) 9,500 9,785 10,079 10,380 10,692 11,013 Incremental production 546 1,679 24 5,438 Appraisal estimate (incremental) 80 520 1,080 1,200 Maize: Incremental Production Area (ha) 27,360 27,250 28,580 28,490 Production (with project) 25,380 16,250 37,440 38,462 Production (without project) 28,400 29,252 30,130 31,033 31,964 32,923 Incremental production (4,750) (14,783) 5,476 5,539 Appraisal estimate (incremental) 216 1,100 2,675 5,310 Rice: Incremental Production Area (ha) 1,500 1,400 1,264 1,127 Production (with project) 3,960 599 1,688 1,919 Production (without project) 700 721 743 765 788 811 Incremental production 3,217 (166) 900 1,108 Appraisal estimate (incremental) 270 575 1,200 1,860 Groundnuts: Incremental Production Area (ha) 500 400 827 1,254 Production (with project) 69 16 44 394 Production (without project) 880 906 933 961 990 1,020 Incremental production (864) (945) (946) (626) Appraisal estimate (incremental) 54 81 117 Sorghum: Incremental Production Area (ha) 18,680 24,000 28,310 17,907 Production (with project) 7,304 13,056 18,118 12,714 Production (without project) 10,200 10,506 10,821 11,145 11,480 11,825 Incremental production (3,517) 1,911 6,638 889 Appraisal estimate (incremental) 863 665 803 1,116 Mllet: Incremental Production Area (ha) 5,000 5,400 5,488 5,575 Production (with project) 1,278 835 1,375 1,148 Production (without project) 1,200 1,236 1,273 1,311 1,350 1,391 Incremental production (3) (476) 25 (243) Appraisal estimate (incremental) 55 250 585 n.a. Ouar Beans: Incremental Production Area (ha) 3,990 5,850 6,780 9,197 Production (with project) 3,289 5,163 4,150 4,000 Production (without project) 3,040 3,131 3,225 3,321 3,421 3,524 Incremental production 64 1,842 729 476 Appraisal estimate (incremental) 435 425 1,620 1,620 - 70 - Appendix 1 Page 4 of 6 6. Projections for incremental production havebeen made for Year 5 through Year 8. The following growth rates of incremental production have been assumed; cotton - 5% per annum; maize, rice, groundnuts, sorghum and millet - 3% per annum; and guar beans - zero percent per annum. From Year 9 through Year 20 incremental productions have been assumed to be constant at the Year 8 levels. Table 3 gives a summary of incremental productions of various crops. Table 3: Incremental Production Snamary (Mt) Year Cotton Maize Rice Groundnuts Sorghum Millet Guar leans 1 546 (4,750) 3,217 (864) (3,517) (3) ' 64 2 1,679 (14,783) (166) (945) 1,911 (476) 1,842 3 24 5,476 900 (946) 6,638 25 729 4 5,438 5,539 1,108 (626) 889 (243) 476 5 5,710 5,705 1,141 (607) 915 (235) 476 6 5,995 5,876 1,175 (589) 943 (228) 476 7 6,295 6,052 1,210 (571) 971 (221) 476 8 6,610 6,234 1,247 (554) 1,000 (215) 476 9-20 6,610 6,234 1,247 (554) 1,000 (215) 476 7. The incremental production figures have been multiplied by the farmgate export parity prices in order to arrive at the incremental benefit in monetary terms. Table 4 gives the economic farmgate values of crops and Table 5 gives the benefits in monetary terms. - 71 - Appendix 1 Page 5 of 6 Table 4: Economic FaMate Values of Crops - (11K/mt) Year Cotton Maize Rice Groundnuts Sorghum Millet Guar Beans 1979 442 57 168 682 110 60 189 1980 432 71 213 784 119 65 233 1981 386 86 224 725 128 70 234 1982 436 99 343 645 136 74 236 1983 449 102 353 664 138 76 243 1984 463 105 364 684 140 78 250 1985 476 108 375 704 142 80 258 1986 491 111 386 726 144 82 266 1987 491 111 386 726 144 82 266 Note: Prices for the first four years have been determined through observed World market prices. Projections (3% per annum) have been made for the next four years (1.5% for sorghum). Thereafter prices have been assumed to remain constant. Table 5: Incremental Benefits in Value Termw (MKO QJar Year Cotton Maie Rice Groundmits SorgL= Millet Beans Total 1 241,322 (270,750) 54,040 (589,248) (386,870) (180) 12,096 (453,164) 2 725,328 (1,049,593) (35,358) (740,880) 227,409 (30,940) 429,186 474,848 3 9,264 470,936 201,600 (685,850) 849,664 (1,750) 170,586 1,017,950 4 2,370,968 548,361 380,044 (403,770) 120,904 (17,982) 112,336 3,110,861 5 2,563,790 581,910- 402,773 (403,048) 126,270 (17,860) 115,660 3,369,503 6 2,775,685 616,980 427,700 (402,876) 132,020 (17,781) 119,000 3,650,725 7 2,939,769 653,616 453,750 (401,984) 137,882 (17,680) 122,550 3,887,899 8 3,245,510 691,974 481,342 (402,204) 144,000 (17,630) 126,616 4,269,608 9-20 3,245,510 691,974 481,342 (402,204) 144,000 (17,630) 126,616 4,269,608 Economic Rate of Return 8. In Table 5 the net benefits are presented and the resulting ERR for a period of 20 years has been shown to be 7.85%. This is much below the 20% ERR anticipated during the appraisal. This is not surprising since crop production in the Valley was adversely by drought. - 72 - 4ppendix 1 Page 6 of 6 Table 6: Summary of Costs and Benefits (MK '000) Net Year Costa Benefits Benefits 1 1,476 (453) (1,929) 2 1,376 (475) (1,851) 3 1,749 1,018 (731) 4 1,975 3,111 1,136 5 1,476 3,370 1,894 6 1,437 3,651 2,214 7 1,411 3,888 2,477 8 1,452 4,270 3,118 9-20 1,452 4,270 3,118 ERR - 7.85% - 73 - ApRendix 2 Table 1 Page 1 of 11 Staffing Situation Phase by Section Filled Approx. March March March March March Filled as a % Section Estimate 1978 1979 1980 1981 1982 ( 1982 only ) Management 26 24 14 18 15 17 65 Finance 34 25 19 26 26 25 74 Extension 257 267 274 289 244 266 104 Fisheries 7 - 5 5 5 6 86 Credit 16 2 14 3 15 15 94 Hydrology 9 4 2 9 10 5 56 Seed multipli. 4 - - 1 - - - Forestry 6 - - - - - - Land husbandry 11 9 9 7 9 9 82 Training 28 9 11 28 20 31 111 Evaluation 34 1 6 7 7 33 94 Building constr. 12 7 13 14 8 8 67 Agr. research 35 11 15 31 36 33 94 Roads 10 3 55 4 4 4 40 Mechancial 12 5 4 5 6 5 42 Total 459 367 391 447 405 456 99 Source: Quarterly staff returns, Ngabu, ADD. - 74 - Appendix 2 Table 2 Page 2 of 11 Staffing Situation Phase III by Grade Appraisal March March March Grade Plan 1978 1980 1982 P6 1 1 1 1 P7 - - - - P8/S8 5 4 3 3 CTO/CEO 10 5 7 4 PO 13 7 7 8 SEO/STO 8 2 5 4 O/To 42 17 28 22 SCO/STA 25 15 27 32 CO/TA 190 113 163 224 D5/6/7 4 1 4 5 DA 160 202 202 153 Total 459 367 447 456 Source: Quarterly staff returns, Ngabu ADD. - 75 - Appendix 2 Table 3 Page 3 of 11 Established and Non-Established Staff Project 1978 1982 Component Established Non-Established Established Non-Established Management 10 14 9 8 Finance 1 24 10 15 Extension 58 209 107 159 Fisheries - - 4 2 Credit 1 1 13 2 Hydrology 2 2 4 1 Seed multiplication - - - - Forestry - - - - Land Husbandry 9 - 9 - Training 9 - 29 - Evaluation 1 - 1 32 Building Construction 1 6 - 8 Research 11 - 17 16 Mechanical Maintenance 2 3 1 4 Roads 1 3 - 4 Total 106 261 204 252 Appendix 2 Table 4 Page 4 of 11 Expenditure by Section and by Year (Fiscal Years) 1/ (1K) (Current Prices) Description 1978/79 1979/80 1980/81 1981/82 Total Managmnt and Administration 213,625.95 381,392.03 204,360.00 912,986.24 1,712,364 Finance 02,772.58 84,549.43 84,536.63 253,200.98 1,275,959 Evaluation 42,196.57 54,803.65 50,204.82 54,690.09 201,896 Extension 207,727.66 237,145.04 200,522.67 275,893.97 921,290 Training 62,489.85 76,587.90 69,861.02 55,204.63 264,143 Land Husbandry 34,115.29 51,151.97 44,313.60 58,571.61 188,153 Livestock 107,930.47 149,031.30 91,974.39 88,674.91 437,610 Building and Maintenance 208,491.75 213,247.58 156,480.22 286,047.19 864,267 Mechanical Maintenance 57,391.41 72,379.76 89,274.02 249,173.75 468,220 Wildlife 29,137.43 70,151.57 32,911.94 43.24 132,244 Credit 88,090 76,688 49,427 116,899 331,104 Visheries 29,211.97 36,933.81 35,931.09 49,890.16 151,966 Forestry .00 4,646.52 26,344.80 45,100.17 76,092 Health 27,515.22 42,871.83 41,229.75 36,365.83 147,982 Seed Multiplication 31,461.01 14,208.32 8,682.23 .00 54s351 Research 88,703.38 125,137.17 88,634.55 86,900.76 389,375 Water Resources 48,240.33 47,728.17 98,114.14 71,867.44 265,949 Crop Production 189,012 226,311 277,494 183,240 876,057 Total 1,529,511 1,964,966 1,650,295 2,824,750 7,969,522 1/ Excludes contingencies and direct debts by IDA. Source: Project Management. - 77 - Appendix 2 Table 5 Page 5 of 11 Infrastructural Development a Phase III Achievement as a Description Appraisal Achievement % of Appraisal Dip Tanks - - - Housing: C 2 2 100 D 11 10 90 E 90 49 54 P 60 12 20 Low Cost 8 7 88 Sub-Total 169 80 47 EPA Headquarters 5 5 100 Training Centers DTC's 6 4 67 Village Centers 34 - 0 Laboratories 1 1 100 Fish Landing Sites 4 4 100 Health Posts 4 4 100 Roads (construction and upgrading) kas 139 111 1/ 20 Boreholes 40 32 80 Shallow wells 40 28 70 Smoking kilns 60 60 100 1/ "cludes fisheries roads. Source: Appraisal Report and Project Management. - 78 - Appendix 2 Table 6 Page 6 of 11 Loan Requirements (Physical Quantities) 1978/79 1979/80 1980/81 1981/82 Description Target Achieved Target Achieved Target Achieved Target Achiev Seasonal Loans Sulphate of Amonia (a tons) 248 119 523 131 833 174 1,238 77 20:20:0 (m tons) - 6 11 2 23 42 68 16 Pesticides (thousand sprays x ha) 37 n.a. 66 n.a. 94 n.ea. 114 n.a. Medium Term Loans Knapsack sprayers 1,500 158 200 195 2,500 77 3,000 11 ULU sprayers 300 3 400 4 400 15 600 118 Farm carts 80 14 100 5 120 22 120 13 Ploughs 30 4 40 3 60 3 80 - Ridgere 30 1 40 2 60 1 80 - Ox yoke 100 - 150 - 200 - 250 - Trained oxen (pairs) 60 21 80 6 90 17 100 13 Berhed wire (rolls) 100 - 150 - 200 - 250 - Fishing boats 50 59 70 7 100 - 120 - Source: NAD Credit Section and Staff Appraisal Report. - 79 - Appendix 2 Table 7 Page 7 of 11 Population in the Shire Valley 1977 Growth 1982 Family 1981/82 Cultivat E.P.A. Population Rate % Population Size Families Holding Size Area CK1 (Gaga) 16,163 3 18,737 5 3,747 1.50 5,621 CK1 (Mwanza) 10,256 3 11,889 4.3 2,765 2.72 7,521 CK2 37,728 3 43,737 4.7 9,306 1.00 9,306 CK3 25,931 3 430,061 4.7 6,396 1.00 6,396 CK4 40,219 3 46,625 4.1 11,372 0.92 10,462 CK5 38,187 1 40,135 5.2 7,718 1.92 14,819 CK6 26,149 1 27,483 5.2 5,285 1.92 10,147 NS1 (Biton) 8,319 1.5 8,962 5 1,792 0.82 1,470 NS1 (Valley floor) 24,740 3 28,680 4.1 6,995 0.92 6,435 NS2 18,834 2 20,794 4.5 4,621 2.08 9,611 NS3 12,627 2 13,941 4.5 3,098 2.08 6,444 NS4 (Chididi) 8,379 1.5 9,026 5 1,805 0.82 1,480 NS4 (Valley floor) 14,743 1 15,495 4 3,874 0.84 3,254 NS5 (Lulwe) 3,547 1.5 3,821 5 764 0.82 626 NS5 (Valley floor) 17,208 1 18,086 4 4,521 0.84 3,798 Total 303,030 74,059 97,390 Source: Land Husbandry Section, Ngabu ADD. - 80 - Appendix 2 Table 8 Page 8 of 11 Plant and Vehicles Including Motorcycles Available at Beginning of Bought as Z of Type Phase Planned Bought Planned Land Rovers ( 4 x wd) 8 31 14 Flat Lorries 8 4 3 75 Personnel Carrier Lorry - 1 1 100 Lorry Tipper 5 3 3 100 Station Wagon Car - 1 1 100 Peugot Pick-ups 8 7 16 229 Motorcycles - 79 75 94 Tractors 6 7 6 86 Graders 2 1 1 100 Roller - 1 1 100 Dumpers 2 1 1 100 Tractor Mounted Digger - 1 1 100 Water Bowser 4 4 5 125 Tractor Trailers - 4 4 100 Caravans 4 3 3 100 Motor Boats - 1 1 100 Outboard Engines - 4 2 50 Drilling Rig - 1 1 100 Compressor - 1 1 100 Caterpiller D.6 1 - - - - 81 - Appendix 2 Table 9 Page 9 of 11 ADMARC Crop Purchases (metric tons) Season Cotton Maize Rice Groundanuts Guar 1967/68 5,100 - 279 174 - 1968/69 10,900 150 530 273 - 1969/70 14,600 85 704 279 - 1970/71 14,500 25 1,193 409 - 1971/72 15,200 465 844 389 - 1972/73 10,200 106 1,407 274 - 1973/74 15,200 476 1,496 95 - 1974/75 9,700 161 1,066 127 - 1975/76 7,700 161 946 180 22 1976/77 9,700 252 1,327 63 705 1977/78 9,700 731 1,633 259 3,138 1978/79 9,300 613 1,548 76 3,486 1979/80 9,300 353 2,778 52 3,189 1980/81 9,561 196 1,160 7 3,769 1981/82 7,419 2,833 n.a. n.a. 3,112 Source: NADD Evaluation Unit. 必 必 .’〕〕’〕’〕〕！〕〕’〕〕〕〕〕〕”〕！:!!/- 6 cn cn # # @ # a t l e t Ch i i i Oh 9 00 i I &n t 9 SM4 i g s I I i v, 4 i 1 $3 M P;f I t i i i t i i i e on IN 4n i e I i t tn 2 in -o PIK. Ity; - 84 - Cat 8.V.A.CJ. ORGANIZATIONAL CHART P8MW MVLMR- EA&'O I I nSIaTL" A D AMR. DIVI INff. lam vm IEVSI nvm =2 ,=EISE 'mIMcx SmmImS ANIm" mmI umm mmmT w~- - 'IRTAITNIMS ..C LV)WIUQm M Souw: Staff Appsalssl Report Note: B.M.C. - btil1dg, MWntensnc and Constructlon. N.M. - MechwicalNaner - IBRD 13.s48(PPA) S30'E TANZANIA MALAWI SHIRE VALLEY AGRICULTURAL DEVELOPMENT PROJECT ZAMBIA METEOROLOGY 'IMOZAMBIQUE ZIMBABWE OS ANA 100035-00- 10000f\ 16-00 '600 \ N æ,*oweg,I KKdl -~~~6 1'0p •30'- o \-' Project Soundary> (~~ SMeteorological Stations & 1*-. ~~' ~ k Roinguoge - Agriculture Rainguage - Other 1 .---- Isohyet in mm 760- Conjectural Isohyet in mM - Main Roads - Minor Roods Railroad C N - Rivers 1 00 M 17'00, Non -Customary Land - International Boundcry 0 5 10 15 20 25 MILES 5 10 15 20 25 30 35 40 KIILOMETERS 34 3 35* 0* FEBRUARY 1986