Malawi - Fourth National Rural Development Program (NRDP Dedza Hills and Lilongwe Northeast Development Areas) Project

Document of The World Bank , FOR OFFICIAL USE ONLY Report No. 4087-MAI STAFF APPRAISAL REPORT MALAWI NATIONAL RURAL DEVELOPMENT PROGRAM (NRDP) PHASE IV PROJECT (Dedza Hills and Lilongwe Northeast Development Areas) March 17, 1983 Eastern Africa Projects Department Southern Agriculture Division This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit = Malawi Kwacha (MK) SDR 1.00 = MK 0.9487 (pre-devaluation, 26 April 1982) SDR 1.00 = MK 0.8249 (15 June 1982) US$ 1.00 = MK 1.07 MK 1.00 = US$ 0.93 WEIGHTS AND MfEASURES 1 hectare = 2.47 acres 1 kilometer = 0.62 miles 1 square kilometer = 0.39 sq. miles = 100 ha 1 kilogram = 2.2 pounds 1 metric ton = 1000 kg = 2204 pounds 1 liter = 1.05 US quarts ABBREVIATIONS ADB - African Development Bank ADD - Agricultural Development Division ADMARC - Agricultural Development and Marketing Corporation CIDA - Canadian International Development Agency DA - Development Area DAR - Department of Agricultural Research DHI - Dedza Hills DLVW - Department of Lands, Valuation and Water DRIMP - District Roads Improvement and Maintenance Program DTC - Day Training Center EDF - European Development Fund EPA - Extension Planning Area FHA - Farm Home Assistant FRG - Federal Republic of Germany GOM - Government of Malawi GRAMIL - Grain and Milling Company ISNAR - International Service for National Agricultural Research LADD - Lilongwe Agricultural Development Area LLDP - Lilongwe Land Development Project LNE - Lilongwe Northeast MOA - Ministry of Agriculture MOF - Ministry of Finance MOWS - Ministry of Works and Supplies MU - Management Unit (in each ADD) NRDP - National Rural Development Program RTC - Residential Training Center RTO - Research and Trials Officer STA - Senior Technical Assistant STO - Senior Technical Officer TA Technical Assistant TO - Technical Officer UK United Kingdom USAID - United States Agency for International Development Government Fiscal Year April 1 - March 31  FOR OFFICIAL USE ONLY MALAWI NATIONAL RURAL DEVELOPMIENT PROGRAM (NRDP) PHASE IV PROJECT (Dedza Halls and Lilongwe Northeast Development Areas) Table of Contents Page No. I. BACKGROUND A. General ............ I1 B. The Agricultural Sector ...............................,.1 C. Agricultural Development Strategy ............................. 3 D. Government Agricultural Services. 5 E. Commercial Services. 7 F. Bank Group Assistance in Rural Development in Malawi. 8 II. THE PROJECT AREA 9 III. THE PROJECT A. General Description ....................... 13 B. Detailed Features .15 C. Project Costs ................ 21 D. Financing ................................. 23 IV. PROJECT IMPLEMENTATION A. Organization and Management .24 B. Implementation Schedule .25 C. Management and Implementation of Specific Components .25 Extension ....... 25 Training .............. 26 Research ..26 Horticulture ..27 Land Husbandry... 27 Livestock ..27 Credit.. 27 Marketing ....... 28 Water Supplies ..29 D. Procurement ................................................... .29 E. Disbursements .30 F. Accounts and Audits .31 G. Project Monitoring, Evaluation and Reporting .31 H. Policy Issues .32 This report is based on the findings of a appraisal mission comprising Mlessrs. A. Nyberg, A. Spurling, R. Wallace, Y. Doka and Ms. L. Effron (IDA). The mission visited Malawi in April/May 1982. 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.  - ii - Page No. V. PRODUCTION AND MARKETING A. Foodcrops ................ ...... 33 B. Tobacco .............................. 34 C. Horticulture Crops ..................... 36 D. Livestock .......*s..*........ ..................... 36 VI. BENEFITS AND JUSTIFICATION A. Farmers Benefits ..................... 38 B. Government Cash Flow and Recurrent Cost Implications ......... 40 C. Economic Analysis ...... ................ ...... so .... 40 VII. ASSURANCES AND RECOM1MENDATIONS . ............... ....... 44  - iii - Annexes, Tables, Charts, Maps I. Supporting Charts and Tables Chart 1 - Organization Chart Chart 2 - Implementation Chart - Dedza Hills Chart 3 - Implementation Chart - Lilongwe Northeast Table 1 - Key Indicators - Dedza Hills Table 2 - Key Indicators - Lilongwe Northeast Table 3 - Disbursement Schedule II. Cost Tables Table 1 - Project Component by Time D.H. Table 2 - Project Cost Summary D.H. Table 3 - Project Component by Time LNE Table 4 - Project Cost Summary LNE III. Crop and Farm Budgets Table I - Hectare Budget, Maize D.H. Table 2 - Hectare Budget, Maize/Beans D.H. Table 3 - Hectare Budget, Groundnuts D.H. Table 4 - Hectare Budget, Potatoes D.H. Table 5 - Hectare Budget, Wheat D.H. Table 6 - Hectare Budget, Maize LNE Table 7 - Hectare Budget, Groundnuts LNE Table 8 - Hectare Budget, Tobacco LNE Table 9 - Farm Budget, Maize/Pulse D.H. Table 10 - Farm Budget, Maize/Groundnuts D.H. Table 11 - Farm Budget, Maize/Potato/Wheat D.H. Table 12 - Farm Budget, Maize LNE Table 13 - Farm Budget, Tobacco LNE Table 14 - Farm Budget, Dairy & Tobacco LNE IV. Economic Analysis Table 1 - Farm Input Costs D.H. Table 2 - Crop Benefits D.H. Table 3 - Costs for Crop Component D.H. Table 4 - Benefits from Livestock D.H. Table 5 - Costs of Livestock/Veterinary D.H. Table 6 - Farm Input Costs LNE Table 7 - Crop Benefits LNE Table 8 - Costs for Crop Component LNE Table 9 - Benefits from Livestock LNE Table 10 - Costs of Livestock/Veterinary  - iv - Maps IBRD Map 16547 - General Map of NRDP Areas in Malawi IBRD Map 16548 - Lilongwe Northeast Sub-Project Area IBRD Map 16549 - Dedza Hills Sub-Project Area Selected Documents Available in the Project File Project Preparation Report - Lilongwe Northeast Project Preparation Report - Dedza Hills.  MALAWI National Rural Development Program (NRDP) Phase IV Project I. BACKGROUND A. General 1.01 The proposed Project was prepared by the Planning Division in the Ministry of Agriculture (MOA) and submitted to the World Bank in February 1982, at which time it was reviewed by a mission from the Regional Mission in East Africa (RMEA). The preparation mission proposed some alterations and requested additional data be included in the preparation documents. These data were incorporated into a set of revised documents and a World Bank mission comprised of Messrs. A. Nyberg, A. Spurling, Y. Doka, and R. Wallace and Ms. L. Effron (IDA) appraised the Project in Malawi during April/May 1982. The Project would be part of the National Rural Development Program (NRDP) with the overall objectives of raising smallholder production to meet domestic needs and provide exports; of assuring smallholder access to inputs and services; of preserving Malawi's natural resources; and of providing some social service infrastructure. 1.02 Malawi is a land locked country with a total area of approximately 118,900 km2 of which 23,900 km2 is lake surface. It is bordered on the north and northeast by Tanzania, to the east, south and southwest by Mozambique, and to the west by Zambia. The land area consists of three distinct topographic regions: the northern mountainous region, the central plateau and the southern lowlands. The 1977 census indicated the population of Malawi was 5.6 million with an annual growth rate of 2.9%; it also revealed that Malawi has one of the highest population densities in Africa, 60 inhabitants per km2 and 170 inhabitants per km2 of arable land. However, the population is unevenly distributed with 24 inhabitants per km2 in the Northern Region, 59 per km2 in the Central Region and 88 per km2 in the Southern Region. B. The Agricultural Sector 1.03 Agriculture is the most important sector of the Malawi economy, engaging about 85% of the employed population. The sector accounted for 45% of wage employment, 40% of GDP and 89% of export earnings in 1981. Tobacco provides the largest share of agricultural export earnings (48% in 1981), followed by sugar (26%), tea (12%) and groundnuts (4%). The relative importance of various crops have altered in recent years with sugar assuming considerable importance after the establishment of a second sugar estate. Maize has been imported and exported during years of deficit and surplus, but is not expected to be a component of external trade under the recently established policy of maize self-sufficience and strategic reserves. Malawi's soils are relatively fertile and fairly reliable rainfall permits a variety of food and cash crops to be grown. About 3.3 million hectares, 36% of the total land area, is considered suitable for arable agriculture. Of this area, some 2.9 million hectares are estimated to be under cultivation at present. Forest reserves occupy 9% of the total land area and urban, infrastructure and National Parks areas each account for 11% of the total. Of the remaining 33% of total land area presently unused, an estimated 60% contains natural, woody vegetation. Over 80% of the total land area is held under customary tenure, about 2% under freehold/leasehold and 16% is public land.  - 2 - 1.04 The average farm size is estimated at 1.7 ha and the average farming family has five members. At this level of cultivated farm holding, the 3.3 million ha of arable land could probably sustain a population of about 9 million. At present population growth rates this would be reached in the quinquennium 1995-2000, when all the remaining arable land would be utilized. This also assumes that farm production has first claim to arable land. Over the past decade, increased food requirements for the expanding population have been met for the most part through expansion in cultivated area. This has meant that in many areas land more susceptible to erosion has been cultivated and that fallow periods have been reduced. Consequently, land use planning, land conservation and maintenance of soil fertility through the introduction of improved techniques, inputs and afforestation has become increasingly important. Crop Production 1.05 Malawi's agricultural production derives from two subsectors: smallholder agriculture and estate agriculture. In 1981 the smallholder subsector accounted for 81% of all agricultural production. In most years smallholder production meets the country's demand for food staples (maize, beans, groundnuts, cassava, sweet potatoes and rice), provides agricultural raw materials for industry (cotton and fire-cured tobacco) and some export surplus (cotton, tobacco, tea and groundnuts). However, the export surplus from this subsector comprises only about 20% of all agricultural exports. Smallholder output is estimated to have been expanding at about 3% p.a. in real terms (equal to the rate of population growth) in recent years. Better performance by the smallholder subsector has been constrained by (a) poor incentives provided by low producer prices to smallholders; (b) the transfer of some 150,000 persons and 470,000 ha of cultivable land to the estate subsector; and (c) limitation of smallholder participation in the production of high-value, high cost agricultural export crops such as burley tobacco, tea, and guar beans by Government regulation. On the other hand, recent performance of the estate subsector has been spectacular, averaging a growth of 10% p.a. in real terms since 1973. In 1981, the estates contributed 19% of the total agricultural production in the country, but accounted for 80% of all agricultural exports. Estate production is primarily flue-cured tobacco, burley tobacco, sugar and tea. Livestock 1.06 The livestock population is estimated at about 877,000 cattle, 800,000 sheep and goats, 200,000 pigs and several million poultry. The cattle herd is estimated to be growing at 4.4% p.a. and contains approximately 2,000 grade dairy animals. Grazing areas are being steadily reduced by the expansion of crop enterprises but there is minimal challenge from tsetse fly infestation on the grazable land. Most cattle are found in the less densely populated central and northern regions, although some are in the Shire Valley. They are typically raised by traditional methods and they are considered by many farmers more as a source of wealth than as commodities for sale. However, overstocking is becoming appreciable in some areas and will probably worsen unless the current offtake rate of 10% is substantially improved in selected problem areas. Offtake rates have improved in the Central and Southern regions in recent years. Goats are an important source of meat in Malawi as are pigs and poultry which are found  in many households. There are no official marketing systems for smallstock and most of the organized pig and poultry production is, however, supplied to the urban markets from large commercial producers. Malawi still imports much of its dairy requirements, though recent progress in local production has been impressive. Two dairy projects with external support (from Canada and Denmark) have contributed to this rapid production increase. Milk throughput at the commercial dairies doubled between 1975 and 1981, to approximately 5.8 million liters, which is estimated to fulfill 45% of current demand. C. Agricultural Development Strategy 1.07 The Government's aim has been to maintain self-sufficiency in food staples, expand agricultural exports significantly and improve rural incomes. Given the limitation on high quality land available for new cultivation, the increasingly high cost of reclamation on marginal lands or lands exhausted by continuous cultivation, and the high cost of introducing/expanding irrigation schemes, agricultural policy is focussed on sustaining productivity increases. It has attempted to provide a growth climate for both smallholder and estate production. 1.08 Two approaches have been pursued in attempting to improve smallholder productivity. One has focussed on the gradual improvement of extension, land husbandry and farmer training services throughout the country; these effects have been supported by small, low-cost rural development projects and by special activities and programs such as ox-training, dairy improvement, and tea development (usually funded by the United Kingdom (U.K.) or by the United Nations Development Program). These activities have been relatively successful but have impacted upon relatively few farmers. The other approach which the Government introduced in 1968/69 concentrated on relatively expensive and management-intensive integrated development programs in several areas. Four reasonably large geographic areas have been the recipients of continuing projects: Lilongwe with Lilongwe Land Development Projects I, II and III (Cr. 113 MAI, Cr. 244 MAI and Cr. 550 MAI, respectively) followed by the National Rural Development Program, Phase I (Cr. 857 MAI); Shire Valley with Shire Valley Agricultural Development Projects I and II (Cr. 114 MAI and Cr. 363 MAI) followed by Shire Valley Agricultural Consolidation Project (Cr. 823 MAI); Karonga with Karonga Rural Development Projects I and II (Cr. 282 MAI and Ln. 1286 MAI) followed by the National Rural Development Program Phase III (Cr. 1183 MAI), all funded by the World Bank Group; and Lakeshore Rural Development Project (LRDP) formerly funded by the Federal Republic of Germany (FRG) and now assisted by the European Development Fund (EDF). All of these programs provide infrastructure (roads, markets, water, health facilities, etc.) land improvement and conservation measures, improved extension, credit facilities and other services. These four comprehensive programs - which are concentrated in the more agriculturally productive and densely populated areas - affect about one million persons, or 20% of Malawi's population. Several other smaller geographic area projects are in their first phase of development and include Phalombe, Namwera, Kawinga, Ntchisi and Dowa West. 1.09 Malawi's experience in carrying out these rural development projects has so far been generally favorable. A number of completion reports have been prepared by the World Bank on IDA/IBRD financed projects  - 4 - that have come to the end of their disbursement periods. The first two reports on LLDP, Phases I and II, found that infrastructural development and introduction of Government services had been excellent but that incremental output had proven difficult to measure and probably did not reach appraisal targets. The report on Phase III of the LLDP, however, using a number of assumptions, found a 24% rate of return for the three phases combined. For Shire Valley Agricultural Development Project, Phase I was considered a success, where farmers' cotton yields incresed significantly, appraisal production targets were reached and the Project was credited with successfully mobilizing people at all levels in an area considered resistant to change. The rate of return was estimated at 15%. Phase II was less successful in encouraging production and was affected by several years of severe drought, although again infrastructural development and provision of Government services were good. The rate of return for Phase II was estimated at 6-12%, depending on the assumptions regarding the introduction of guar beans, in which the Project staff were instrumental. Finally, the completion report for Karonga Rural Development Phase I concluded that despite misfortunes of some components of the Project such as an irrigated rice component washed away by flood, a lake barge that cracked and a jetty that was under water shortly after completion, the Project was generally successful in achieving its physical targets as well as incremental output targets and had an estimated rate of return, without the irrigated component, of 33%. The National Rural Development Program 1.10 In the mid-1970s, it was realized that such an intensive level of investment could not be replicated in the rest of the country within a reasonable time. A strategy redesign was undertaken resulting in the formulation of the National Rural Development Program (NRDP) to provide a more extensive level of services to a large population with a concentration of resources on the more immediately productive aspects of agricultural investment. NRDP is designed to increase the level of smallholder production through the provision of agricultural inputs and farm services and to increase the efficiency and scope of extension, input supply, marketing and credit services. Emphasis is also placed on soil conservation, watershed management, and on afforestation. 1.11 The MOA is responsible for the implementation of the program. The central services and management capability of MOA have been strengthened; instead of three regional agricultural areas with four major projects, the country has now been divided into eight Agricultural Development Divisions (ADDs).1/ On average, each ADD is divided into five Development Areas (DAs), containing an average of about 25,000 farm families. 1.12 The Government envisages a 15 to 18 year development period for each area divided into four phases. The activities would start with a two or three year preparatory phase. This would include land and agro-economic surveys, followed by detailed physical and economic planning; the establishment of agronomic trials where necessary; construction of 1/ Karonga, Mzuzu Kasungu, Lilongwe, Salima, Liwonde, Blantyre, and Ngabu (Shire Valley3 - see map.  - 5 - essential additional staff housing; allocation of new extension staff; and construction of additional water sources, marketing facilities, rural roads and health posts. The subsequent extensive phase would be implemented over a period of about five years and would include investments in production-related items such as improved extension and training, marketing, supply of inputs and provision of credit. Research would also be accented as a few years of research are needed to generate the 'new' technologies for the next phase. During the following five-year intensive phase, more attention would be given to increasing research and to the introduction of new crops and processing technologies; intensification of cropping programs; opening up of new areas where available for settlement; land conservation and consolidation; animal husbandry; and the construction of new roads. Finally, the areas would enter a consolidation phase, which would involve a continuation of more intensive development, further improvements to social infrastructure, especially education and health and may also include the development of rural industries. In practice, there will not necessarily be a clear dividing line between activities in each phase. The initial NRDP project (Cr. 857 MAI) began in 1978 for several DAs and is being co-financed by IDA, EDF, UK, FRG, Canadian International Development Agency (CIDA) and the African Development Bank (ADB). In addition the United States Agency for International Development (USAID) has a large program to strengthen the country's agricultural research program (para. 1.15). D. Government Agricultural Services Agricultural Extension 1.13 Crop and animal health extension are important services provided to farmers by the Government. The past performance of the smallholder subsector tends to indicate, however, that extension efforts provided only marginal stimulus to agricultural development. This led MOA to restructure its extension services by instituting geographical units, called Extension Planning Areas (EPAs), based on ecological principles consistent with administrative boundaries. These EPAs are as uniform in terms of physical resources as possible, and number 180, covering the entire country. They are, in turn, grouped together for administrative purposes into Development Areas of which there will be about 40 when NRDP is fully implemented (para. 1.11). Initially, the extension workers made individual farm visits but his farmer contacts are evolving to working with groups of farmers, which utilizes his time more efficiently. The extension worker is supported by district supervisory staff and specialist staff at the ADD level. The extension service has expanded substantially in recent years and now contains a number of young and relatively inexperienced staff; however, they are well disciplined, well motivated and should mature into an effective service. Farming Training 1.14 Training facilities for farmers (both men and women) have become increasingly available in recent years. Three levels of training are provided at farm institutes, residential centers and day centers. There are three farm institutes at Mbelwa (Mzuzu), Chitala (Salima) and Thuchila (Blantyre), with a total capacity of 260 student places for courses lasting two or three weeks. There are 22 residential centers conducting courses of one to two weeks' duration for staff and about 17,000 farmers (male and  female) in a year. All centers have demonstration gardens, plots or farms. Day centers have the use of a room at an EPA field office. The number of these centers varies annually as some are upgraded and transferred to specific projects or as new ones are constructed. The courses are centered on specific topics of current interest or importance. Only limited evidence exists on the impact training programs have on farm operation, production and productivity. Evaluation by questionnaire indicates that RTC participants have adopted, to some extent, some practices learned in the course(s). However, this has not been translated into increased production or productivity. Agricultural Research 1.15 Agricultural research is carried out by the Department of Agricultural Research (DAR) of MOA, the Faculty of Agriculture of the University of Malawi, the Tea Research Foundation and the Tobacco Research Authority. The University carries out its own research on behalf of MOA at Bunda College of Agriculture. Its main contribution to date is in the field of bean and livestock research. The DAR is a full-time research department with national responsibilties covering all crops and livestock as well as seed multiplication and certification. Administratively, agricultural research is organized on the basis of research stations which, in turn, have substations and district sites. Scientifically, research is organized under research projects based on crops or disciplines. The DAR has conducted some very relevant and effective research, although the Department has established too many research stations to operate them effectively within their staffing and financial constraints. However, the research structure and organization is currently under a joint review by the MOA and the International Service for National Agricultural Research (ISNAR). In the past the link between research and extension has been weak but the incorporation of research activities in NRDP projects has improved coordination. Coordination is expected to improve further as responsibilities are delineated and the current review is completed. USAID has a large program to provide graduate training to about 50 researchers and to provide technical assistance. Agricultural Marketing 1.16 The Agricultural Development and Marketing Corporation (ADMARC) is a statutory company established to purchase smallholder crops (foodcrops and non-food crops), to sell food crops and to sell and deliver crop production inputs (seeds, fertilizers, chemicals, and sometimes farm implements). ADMARC operates a national network of marketing sheds and storage facilities for these purposes. In addition, ADMARC invests in agricultural and agro-industrial enterprises and pays/charges prices as agreed and authorized by the Price Commission of the Malawi Government. Until recently, the authorized prices permitted ADMARC to realize profits on a number of its crops, but recently higher transport costs and improper crop handling combined with higher costs of borrowing for large investments have resulted in substantial losses. Nevertheless, it is one of the more effective crop marketing parastatals in Eastern Africa.  -7- Credit Services 1.17 No central agricultural credit institution exists in Malawi. Smallholders have access to seasonal and medium-term agricultural credit through the MOA's rural development projects, which have their own credit funds and credit staff, settlement schemes under the MOA (which are small comprising 100-200 farmers at most), or through the Loan Board under the Ministry of Trade and Industry (negligible amounts in recent years). Under NRDP, a study has recently been conducted on agricultural credit in Malawi recommending the establishment of a national agricultural credit institution, but a decision has not been made on the institutional form that should be established. 1.18 Most seasonal credit to smallholders is through self-accounting groups at 10%.2/ Repayment rates have been high, approximately 98-99%. Repayment is either at the time of commodity sales to ADMARC or refunded by the secretary/treasurer of the self-accounting groups. In addition, the credit staff provides strict supervision; seasonal credit is given only to groups (even where they may not be self-accounting) and no credit is given to a group where any member has defaulted. The intense supervision by credit and extension staff contributes to the high level of credit recovery - a supervision intensity which probably would not be maintained under an institutionalized credit system. Medium-term credit is available at 10% per year, repayable over 3-5 years, and repayment is less satisfactory, at 60-80%, because there is no systematic method of collection as there is for seasonal credit. Annual inflation rates for the past four years have averaged about 11% and are projected to be about 12% for the near future. Afforestation 1.19 The Forestry Department in the Ministry of Forestry and Natural Resources has a nationwide program underway, financed by IDA and called NRDP II (Wood Energy Project, Cr. 992 MAI), to provide fuelwood, charcoal and poles to both the urban and rural populations. The program includes the establishment of District Plantations and Forestry Department nurseries, the latter for production of seedlings for sale to farmers (smallholder or estate) who wish to establish woodlots. The Project is in its early stages but promises to be, if successful, an innovative means of reafforesting the country and providing an accessible source of energy to the population. E. Commercial Services Banking 1.20 The estate farming sector obtains credit from the two commercial banks - National Bank and Commercial Bank. Credit is primarily seasonal financing but medium term investment financing is also provided. The volume of outstanding credit is about MK92 million, most of which is seasonal overdrafts; at interest rates of 14-15%. However, poor recovery 2/ Annualized rates are higher, however, because farmers pay loans back after only 7-9 months.  - 8 - has caused the banks to establish an Agricultural Technical and Management Service for protection of their estate investments and to screen and reduce their volume of new investment financing. The banks also provide extension to tobacco estates as the Ministry of Agriculture's extension staff services only the smallholder. Input Supplies 1.21 Fertilizers for the estate sub-sector are imported under monopoly rights granted to Optichem and other agro-chemicals are provided by commercial firms. The National Seed Company of Malawi stocks a wide range of seed available for any purchaser. However, most direct sales are to estates and ADMARC, smallholders normally purchase their seed requirements through ADMARC. Commodity Marketing 1.22 Estate produce is marketed almost entirely through the private sector. Grain and Milling Company (GRAMIL) is the major maize purchaser and the tobacco auctions at Limbe and Lilongwe provide the vehicle for tobacco sales. Tea and coffee are exported through international trading companies. F. Bank Group Assistance in Rural Development in Malawi 1.23 The Bank Group has been involved in financing rural development in Malawi since 1968 and the proposed NRDP IV would be the eleventh such project. The total cost of these projects were about MK116.1 million (US$152.9 million) with a Bank Group contribution of US$105.5 million. The Bank Group has assisted in three geographical areas as noted above (paras. 1.08-1.09). In addition to the Project Completion Reports and Project Performance Audit Reports written on a number of these projects, a national rural development program (NRDP) review has been produced on the overall structure and content of the rural development approach. 1.24 The NRDP Review concluded that, in general, investment in rural development was an appropriate and economic means of improving smallholder production, provided the incentives to produce and input delivery were adequate. The Review did make a number of recommendations to improve the effectiveness of these projects, however, which are relevant to the proposed Project. First, in view of the constrained availability of financial resources the Review recommended that the complexity of the rural development projects should be reduced and efforts focussed mainly on production related agricultural activities: extension, land husbandry, training, research, animal husbandry, and marketing. Investments in other sectors - rural roads, rural health, forestry, etc. should be included wherever practical in ongoing or forthcoming national investment programs. In this way, proper maintenance and staffing of non-agricultural activities and infrastructure would be better implemented and operated than if they were administered under the MOA. Agricultural staff could focus their attention on production and conservation. However, in some sectors national priorities have not been identified and the capacity to develop national plans does not exist. Where this prevails, the appropriate sectoral component should be included if that sector's participation is perceived to be a limiting agricultural development  - 9 - constraint. The proposed NRDP IV project follows these recommendations. Second, to ensure better coordination of veterinary, research and agricultural staff within an Agricultural Development Division, some reorganization both at headquarters and at the field level was recommended (see para. 4.01). 1.25 Third, to reduce the cost of civil works, which have become an increased proportion of Project costs, the Review identified housing as an area of potential investment cost savings by using a consulting architect, greater use of private contractors and more efficient housing designs. To reduce the recurrent costs, a less intensive extension staff to farmer ratio was recommended. The proposed Lilongwe Northeast and Dedza Hills Project has followed these guidelines (paras. 3.03 and 4.01) and has also recommended, to ease the incremental burden to Government, that new posts for this Project be transferred from existing projects elsewhere in the country, thus reducing the intensity of ongoing projects (para. 4.07). The financing arranagements proposed for this Project are also designed to facilitate handover to Government's own accounts from donor-financing by having the Government cover an increasing portion of recurrent costs during the five years of the Project (para. 3.28). 1.26 Finally, a number of specific recommendations were made in the NRDP Review on the activities directly related to agricultural production to rationalize the use of the resources and to improve their effectiveness. These included, among others, developing a national livestock program that would set priorities for disease control and marketing and reorienting both extension and research towards existing farming systems and constraints faced by farmers. The NRDP IV Project incorporates these recommendations by including, for example, financing for a livestock review aimed at establishing national priorities and by focussing both extension and research efforts on local problems. 1.27 The Bank Group has supported a Structural Adjustment Loan (SAL) (Loan 2026 MAI) and a Technical Assistance Loan (Loan 2027 MAI), both of which addressed some rural development issues. The recurrent budget of the MOA had been inadequate to maintain the rural development projects even at their reduced staffing intensity, the SAL prescribed and provided funds for an immediate increase in the recurrent budget for agriculture and subsequent MOF budget projections for agricultural development have incorporated future requirements. A pricing methodology to be used in establishing the prices of crops produced by smallholders was agreed upon, under the auspices of the SAL. Similarly, two commodity studies, a tobacco industry study and a livestock industry study, were proposed and/or financed under these two loans. II. THE PROJECT AREA General 2.01 The proposed project area would cover Dedza Hills (DH) and Lilongwe Northeast (LNE), two of the five development areas (DA) in the Lilongwe ADD (Map No. 16547). The other three DAs, Lilongwe, Ntcheu and Thiwi/Lifidzi currently have ongoing rural development projects, all financed by IDA. Lilongwe Northeast and Dedza Hills cover areas of 1,581 km2 and 1,847 km2 respectively although a forest reserve in the latter  - 10 - occupies 622 km2. LNE is divided into 5 EPA centers and 2 EPA sub-centers and Dedza Hills is divided into 4 EPA centers (Map Nos. 16549, 16548). The area is bounded, to the extent possible, by natural barriers - the escarpment and rivers - and to a limited extent by a national frontier and roads. The major North-South road traverses and bounds the area making communication with Lilongwe and Blantyre relatively easy. Several regional and district roads also traverse the area but road transport to many locations within the project area is difficult, particularly during the rainy season. Population 2.02 Population was estimated to total about 288,000 people in the 1977 census, composed of 164,000 in LNE and 124,000 in Dedza Hills. Thus, the project encompasses about 5% of the population and about 3.6% of the land area. Population densities are quite variable within the areas, but the overall averages are about 100 per km2 for each of the two sub-project areas (excluding the forest reserves). The population growth rate for the areas are reported to be 2.6% for LNE and 2.2% for Dedza Hills (see Table 1).  - li - Table 1: Socio-Economic Data and Indicators Lilongwe _/ Dedza b/ National a/ Northeast Hills A. Demographic 1. Total population ('000 1989 est.) 5,960 178 124 2. Population density (pop/km arable land, 1980 est.) 180 112 101 3. Average family size (rural 1977) 5.0 5.0 4.6 4. Annual population growth rate (%, 1969-1978) 2.9 2.6 2.2 B. Economic i. Gross Domestic Product (US$ million, 1981) 1,271 Na Na 2. Growth rate of GDP (%, 1980-1981) 0.9 Na Na 3. Per Capita -DP (1980) 200 Na Na 4. Rural Relative Poverty Income Level (1978 US$) 52 52 52 5. Rural Family Cash Income (1977 MK) 24 17 6. Total Land Area ('000 ha) 9,500 c/ 158 184 7. Total Cultivable land ('000 ha) 3,300 c/ 107 123 8. Total Land Cultivated ('000 ha) 2,900 c/ 56 34 9. Average farm size (cultivated ha) 1.7 c/ 1.3 1.4 10. Total livestock herd size ('000 animal units, 1981*) 1,900 Na Na 11. Total number of cattle ('000 head, 1981*) 877 59 23 12. Farmers per Agricultural Extension Agent (1981) 1,040 1200 1900 13. Classified Roads (kin, 1981) 10,772 Na Na C. Social Water 1. Rural Population with safe water (%, 1981) 15 19 Health 1. Life Expectancy at Birth (years, 1981) 41 d/ Na Na 2. Crude Birth Rate (per thousand, 1970s) 56.0 d/ Na Na 3. Crude Death Rate (per thousand, 1970s) 27.0 d/ Na Na 4. Infant Mortality (per thousand, 1970s) 190 d7 Na Na 5. Incidence of Malaria (% adult population, 1979) 26 d/ Na Na Education 1. Primary school enrollment ('000, 1980/81) 810 Na Na 2. Secondary school enrollment ('000, 1980/81) 18 Na Na 3. Adult literacy rate (%) 25 Na Na 4. Pupil-teacher ratio (primary, 1980/81) 64 Na Na - provisional a/ Source (unless otherwise indicated) Malawi Government, Economic Report 1982. b/ Source (unless otherwise indicated) Preparation Reports, April 1982. c/ World Bank; Malawi, The Development of the Agricultural Sector, May 8, 1981. d/ World Bank; Malawi, Health Sector Review, November 17, 1981.  - 12 - Physical Environment 2.03 The Project area encompass a diverse set of topographic and climatic conditions. About 70% of the LNE area is part of the Lilongwe plain, lying about 1,150-1,300 meters above sea level. The Dowa hills in the northwestern portion of LNE range from 650 to 1600 meters in elevation and the Dedza hills section in the southeastern portion range from 1300 to 1600 meters. Rainfall is roughly 800-900 mm annually with over 95% of the rainfall occurring between November and April. Soils in the area are predominately ferruginous soils of various depths except in the hills of the southeastern portion which are primarily lithosols. 2.04 Topography in the Dedza Hills sub-project is quite undulating with steep slopes in much of the area. Most of the area is 1000-1600 meters in elevation and annual rainfall is about 900-1000 mm. Soils are primarily stony lithosols with some poor ferralitic and fertile ferruginous soils occurring. Erosion, both potential and actual, is considerable due to the steep slopes which have been deforested and cultivated. Much of the area should be in perennial crops and/or terraced. 2.05 Rivers and streams are numerous in the area with two major rivers, Linthipe and Lilongwe, and several of their tributaries have an all-year flow. Several other streams are seasonal and cease flowing during the latter part of the dry season. However, access to clean potable water is relatively limited, as only 125 boreholes and 160 protected wells service about 51,000 people. Thus, less than 20% of the rural population has access to an improved water supply. 2.06 The nation's major north-south bituminized road traverses LNE and borders Dedza Hills. In addition to this main road (designated M), the main road connecting Salima and Lilongwe traverses LNE. About 40 km of secondary roads (designated S) exist in LNE and about 50 km exist in Dedza Hills. A network of several hundred km of district roads (designated D) exist in the project area of which about 310 km will be improved under the IDA supported DRIMP program. Many of the district roads and most of the undesignated feeder roads are impassable for motor vehicles immediately after rains. However, when the DRIMP program is completed (scheduled completion in 1983), transport will be considerably enhanced and should be adequate for NRDP IV implementation. Land Tenure 2.07 Although estates are on leasehold or freehold land, this is applicable to a minor portion of the project area. Most of the project area falls under customary or communal land tenure where only cultivation rights rather than ownership is granted by the village headman. Within most of the project area matrilineal inheritance of cultivation rights prevail, however, in some locales in Dedza Hills patrilineal inheritance rights exist. In Lilongwe DA a pilot land registration program exists but there are currently no plans to extend it to other areas. Average size of cultivated holdings (in 1977/78) were 1.3 ha and 1.4 ha for LNE and Dedza Hills respectively. About 25%-30% of total holdings are cultivated. Holdings are quite fragmented as farmers, on average, have rights to four non-contiguous plots.  - 13 - Economy 2.08 Although agriculture is the principal employment activity in the area, cash income from agriculture is disproportionately low. In 1977/78 cash income from agricultural sources was 65% and 43% of total cash income for LNE and Dedza Hills respectively, the balance being derived from non-farm sources. Livestock contributes about 10% of the agricultural cash income in LNE and only 4% in Dedza Hills. Cattle ownership is relatively low in the area and the cattle owners have relatively few animals. However, over 50% of the farmers own goats. 2.09 Lilongwe Northeast has an arable area of 107,000 hectares or two-thirds of the total land area. The cropping pattern is dominated by pure stands of local maize occupying almost two-thirds of the cultivated land (1981). Improved groundnuts are the next most widely grown crop. Mixed cropping (intercropping) is not common in the area. Tobacco, which is grown under quota, occupies less than one percent of the cultivated area but provides the largest proportion of agricultural cash income. The area is a surplus production area and in 1980/81 ADMARC purchased the following quantities of agricultural commodities: maize-8,700 mt (net maize purchases were 7,400 mt as ADMARC sold 1,300 mt of maize during the year), groundnuts-1,600 mt, tobacco-430 mt and beans-60 mt. Marketed maize has increased in recent years but marketings of other commodities have declined - particularly tobacco which has declined to less than one-half of the marketed tonnage of the late 1970's. 2.10 Dedza Hills has a relatively small arable area as slopes are severe and the erosion hazard is high. Cultivated area is approximately 34,000 hectares (1977) with a maize-pulse crop mixture being the overwhelming production pattern occupying over two-thirds of the area. Pure stands of maize occupy another 12% of the cultivated land. Other crops produced are groundnuts, pulses, wheat and root crops (sweet potatoes and Irish potatoes) and some vegetables. The area basically produces food crops although the food crops can obviously be sold as cash crops. ADMARC commodity purchases (1981) in the area included maize-3,700 mt, beans-150 mt and groundnuts-90 mt. Maize purchases, although relatively small, have increased rapidly due to the introduction of hybrid maize. However Dedza Hills is still a net deficit maize area. Dedza Hills is one of the poorest agricultural areas in the country although the potential for improvement exists through increased productivity of maize, groundnuts, wheat and horticultural crops. III. THE PROJECT A. General Description 3.01 Project Objectives. The proposed activities would be the first phase of a five-year rural development project for LNE and Dedza Hills. These are two of the five project areas in Lilongwe ADD and are the only areas in the ADD that have not previously been the focus of an NRDP project. LNE has good agricultural potential and the Project would attempt to increase yields on the principal crops of maize, groundnuts, and fire cured tobacco and support adaptive research on these and on new crops. The Project would also improve veterinary health services and increase milk and meat production. Living standards would be improved through the provision  - 14 - of potable water supplies. The agricultural potential of Dedza Hills is more limited as much of the land is quite steep. However, the Project would attempt to increase maize yields, re-establish groundnuts as a major part of the cropping system, promote wheat production as a dry season crop and support research on horticultural crops. Meat production would be promoted through improved veterinary services. Potable water supplies would be provided to improve living standards. 3.02 The Project would consist of the following: (a) Agricultural Extension would be provided with additional resources including staff to reduce the farmer:extension worker ratio to 780-800:1 from over 1000:1. Additional transport and other infrastructure (houses and office space) would also be provided. Financial support would be provided to enable the extension service to conduct crop demonstration trials. A women's program would be included to address both agricultural production and homecraft activities, primarily in female headed households. (b) Training facilities would be expanded and additional staff would be recruited to permit increased numbers of farmers and staff to be trained. (c) Research funds would be provided to re-introduce a field trials program and expand the research staff. One of the incremental staff would concentrate on horticultural crops. (d) Horticulture would be supported through the provision of additional staff and attendant infrastructure. The program will include a fruit tree distribution activity and will involve trials at EPA centers and on farm gardens. (e) Land Husbandry would be supported by recruiting staff to be posted to EPA centers and employing senior supervisory staff to assist in farm planning and design. An agro-forestry component would also be incorporated into the unit. (f) Livestock and Veterinary Services would be improved through the renovation of existing and construction of additional dipping facilities and the provision of improved dipping fluids. (g) Credit would be made available for the purchase of incremental inputs for crops, farm implements, dairy cows and facilities and stall-feeders. Funding for incremental credit assistants would also be provided. (h) Marketing Sheds for input supply would be constructed together with housing and incremental staff would not be required. (i) Potable Water Supplies would be expanded through the construction of new water points and the renovation of existing boreholes and wells. (j) Separate Management Units for the two sub-projects would be established at Mkuza (for LNE) and Dedza town (for Dedza Hills). The units would have appropriate staffing, housing, transport and  - 15 - office space. A monitoring and evaluation component would form a component of the management unit. (k) Lilongwe ADD Headquarters would expand their technical support staff and their land husbandry technical capacity. B. Detailed Features Agricultural Extension (Annex II, Tables 1 and 16) 3.03 Adequate technical packages exist to improve yields of the major crops grown in LNE. Yields of both local and hybrid maize can be increased substantially through better husbandry (particularly improved plant densities) and correct fertilization. Tobacco yields can be improved through the increased use of fertilizer and yields of groundnuts can be improved through the use of properly dressed improved seed. In Dedza Hills there is some scope for increasing maize and groundnut yields and there is some potential for expanding fruit and horticultural crop production, but a major focus of the Extension Service will be on conservation - in cooperation with the Land Husbandry staff. The transmission of technology information is a fundamental function of the extension service which, along with credit for these technical packages, is the key to increased productivity. The Project would provide funds to support an additional 16 and 22 extension field assistants in LNE and Dedza Hills respectively, as well as housing, transport and supervisory and support staff. These additional field assistants would reduce the farmer-extension worker ratio to 780-800:1 from the current ratio of 1000-2000:1 (para. 1.25). An Extension Planning Area (EPA) center, consisting of a small office block, would be established at each of the five EPAs in LNE and the four EPAs in Dedza Hills. In addition single offices would be established at the two EPA sub-centers in LNE. Funds would also be included for agricultural inputs to conduct farm demonstrations and trials as an extension complement but planned in cooperation with the research component. 3.04 In both of the sub-project areas significant numbers of households are headed by females and in all households a high proportion of the agricultural work is carried out by women. The extension staff would make an effort to involve women in regular extension activity and two Senior Farm Home Assistants (SFHA) will supervise the women's programs. A women's program will be developed at each EPA Center as an adjunct to the extension service. Farm Home Assistants (FHAs) will provide training and advice to women in both agricultural production and homecraft activities. The Project would provide funds for an FHA, with supporting infrastructure, at each EPA center but the placement of the FHAs would be phased over the first three years of the project to permit a sufficient intensity of supervision in order to ensure that the program is as effective as possible. Training (Annex II, Tables 2 and 17) 3.05 Most of the needed training facilities exist in the sub-project areas, although some renovations and improvements will be undertaken and funds will be provided for farm equipment and materials to be used in demonstrations and trials at the Residential Training Centers (RTCs). Additional training and support staff will be provided with supporting infrastructure (housing, transport, etc.). Day Training Centers (DTCs)  - 16 - will be established at each of the EPAs together with demonstration plots. Training component activities will focus on farmer training courses devoted primarily to general agriculture and crop husbandry topics of broad interest. In addition, special courses on such topics as dairy husbandry will be conducted periodically. 3.06 Staff training emphasis would be on extension techniques, communication and management in addition to the technical courses on crop production, livestock management, horticulture, land husbandry, crop storage, etc. General management training would include work planning and budgeting, monitoring and evaluation and NRDP concepts. Training activities would be conducted by the Training Center staff and by other selected staff from the Extension Service and the Research Department. In addition, funds are provided to employ the services of external consultants for special subjects, including communication and management. Research (Annex II, Tables 3 and 18) 3.07 Most foodcrop research is performed at Chitedze Research Station, which is the central research station. Although located outside the project area it is representative of much of LNE and some of the Dedza Hills areas. Limited foodcrop trials work has been conducted at Bembeke Experiment Station in the Dedza Hills RDP. However, applied field trials have not been carried out in either of the sub-project areas for several years. The Project will provide funds to reactivate a comprehensive field trials program covering the major ecological zones of the Project area and include a greater variety of crops than was previously encompassed. As noted above (para. 1.15), the research structure is under review by the MOA and ISNAR. The composition of the research component in this Project conforms with the preliminary proposals for reorganization. 3.08 A collaborative extension/research diagnostic survey would be conducted to ensure that the trials programs are designed to provide solutions to local farming problems. These will be repeated annually to evaluate the effectiveness and appropriateness of extension and adaptive trials programs. Additional research staff, with appropriate support infrastructure, would be employed consisting of a Senior Technical Officer (STO) who would supervise the research program in the project and a TO (or STA) and 2 TAs in each sub-project area who would conduct the research activities. One of the research staff in Dedza Hills would specialize in horticultural crops. Horticulture (Annex II, Table 4) 3.09 A wide range of fruit and vegetable varieties are under test at Bvumbwe Research Station near Blantyre and the Project will provide funding to enable this program to be extended to the Dedza Hills RDP. Funds will be provided for an STA post in horticulture with the support infrastructure. The horticultural program will involve trials with different varieties at EPA centers and on farmers' plots, in cooperation with research and extension staff. The extension component will focus on improving quality and handling of existing production.  - 17 - Land Husbandry (Annex II, Tables 5 and 19) 3.10 Much of the land within the project area is steeply sloping and highly erodible and technically is not suitable for crop cultivation. However, the high population density will result in much of this land being planted to annual crops, particularly maize. In addition, if the trend into a monoculture of maize is not reversed, long-term fertility problems are probable. Specialist land husbandry staff, with supporting infrastructure will be included at the EPA centers. To assist in planning and supervision a Chief Technical Officer and a Senior Technical Officer will also be funded. Land husbandry staff, in cooperation with research staff, will work on perennial crop diversification which includes fruit trees (agro-forestry), forage trees and grasses. Veterinary Services and Livestock Production (Annex II, Tables 6 and 20) 3.11 Tick-borne diseases are major constraints in livestock production in the project area. East Coast Fever and Anaplasmosis are the most common diseases. Previous inadequate budgets have resulted in the deterioration of existing dip tanks in the area, thus requiring renovation and in several instances improvements (installing roofs, adding calibration tanks, etc.) are needed and improved acaricides should replace the arsenicals. Additional animal health drugs and vaccines would be obtained and made available to farmers at cost recovery prices. Local animal slaughter, under hygienic conditions would be facilitated by constructing slaughter facilities. Veterinary services would encompass the following elements. (i) Dip tanks. The 18 existing dip tanks in LNE and 9 of the 10 existing tanks in Dedza Hills would be renovated. In the absence of a national program with priorities for livestock investment three new dip tanks will be constructed in LNE. Renovation would include roofing, calibration tanks and drainage so that organophosphate acaricides could be used. (See map Nos. 16548, 16549 for dip locations.) (ii) Animal health drugs and vaccines. Livestock drugs, including drenching materials and vaccines would be financed through a revolving fund managed by the Veterinary Department and sold with complete cost recovery by veterinary assistants at dip tanks. (iii) Seven slaughter facilities would be built at the following population centers: Chimutu, Nkhomo, Nathenje, Nanjiri and Kamphata in LNE and Mayani and Linthipe in Dedza Hills. This would support the Animal Slaughter Act and the design of the facilities would conform to the Central Veterinary Services standard which would provide a concrete slaughter area, a structure for hanging and cutting carcasses and a covered meat-selling area. Animals would be slaughtered under veterinary supervision and inspection. Administration and maintenance of the facilities would be the responsibility of the District Councils, who would levy a user charge. (iv) Incremental staffing would be provided for the new dip tanks, including veterinary assistants, dip attendants and tank laborers. Housing would also be provided for the veterinary assistants and dip attendants.  - 18 - 3.12 Animal production would be promoted through the encouragement of stallfeeding beef animals and smallholder dairy units. Credit would be available for financing these activities. The on-going program of distributing improved poultry breeding stock will be expanded by constructing additional poultry distribution houses which will be operated by the Veterinary Service. The specific activities would include: (i) Approximately 1500 beef stallfeeders in LNE and 750 in Dedza Hills would be fattened by encouraging farmer-to-farmer animal transfers without LADD being used as a physical intermediary. LADD would facilitate transporting the animals at cost, to Cold Storage Company (CSC) where quality price differentials are paid. (ii) Smallholder dairy would be promoted in LNE. A total of 175 new farmers would receive, on credit, 2 dairy animals each during the project period. The animals would be obtained as they became available, from the Dairy Multiplication Farms. New farmers would be selected from existing dairy farming areas currently served with a milk collection service and A. I. facilities. (iii) Three poultry distribution houses would be constructed at high population density locations in each of the project areas. The LNE locations would be Chimutu, Nkhomo and Nathenje. The Dedza Hills locations would be at Mayani, Linthipe and Bembeke. (iv) Six incremental staff would be funded for LNE, including two Artificial Insemination (A.I.) staff, and one additional staff in Dedza Hills. These staff would support both the smallholder dairies and beef stallfeeder units. 3.13 A Livestock Policy and Program Review would be financed under -e Project. Developing the Livestock Industry through individual projects ia the Rural Development Program is difficult as the NRDP focus is crop agriculture. Therefore, consistent with the NRDP Review recommendations a study would be conducted focussing on livestock production constraints. This study would complement and build upon the Livestock Marketing Study currently underway which is focusing on marketing, pricing, demand and supply relationships. It would then form the basis for developing a National Livestock Program. With the incorporation of the Veterinary Department into the NRDP organizational structure (as agreed in the NRDP Review) a National Livestock Program could be implemented in parallel with the NRDP. Credit (Annex II, Tables 7, 8, 21 and 22) 3.14 Credit supervisory staff would be provided for the two sub-project areas and incremental credit, both seasonal and medium term, would be provided. Seasonal credit would be available for crop production inputs to farmers who adopted technical packages and improved husbandry and to farmers who engaged in stallfeeder beef fattening. Medium term credit would be available to new dairy farmers (in LNE) for animal and utensil purchase and to farmers for the purchase of farm implements.  - 19 - 3.15 In LNE it is expected that the number of farm families adopting improved maize packages will increase from the current 20% to 28% by the end of the project. It is assumed that the incremental rate of adoption will diminish over the 5 year project period and that approximately 5,450 farmers will adopt improved maize packages during the project. New adopters of improved groundnut and tobacco packages are assumed to total 3,750 and 1,000 farmers, respectively. In Dedza Hills the number of farm families adopting improved practices are assumed to be 5,000 for maize, 550 for groundnuts, and 250 for wheat and potatoes3/. A total amount of MK1.1 million would be required during the project to provide seasonal crop credit. Crop budgets for various types and sizes of farms are contained in Annex III, Tables 9-14, illustrating potential profitability of the technological packages. Stallfeeder steers have relatively low gross margins but require only small labor inputs and, therefore, constitute a relatively attractive farm enterprise. It is assumed that 100 and 50 incremental animals per year would be stalled in LNE and Dedza Hills respectively. The incremental credit requirements for stallfeeder operations is estimated to be MK151,600during the project. A stallfeeder steer budget is illustrated in Annex III, Table 23. Credit for crop production inputs would be available only to farmers belonging to farmers' clubs or groups at 10% (higher on an annualized rate). Individual farmer credit would be available for stallfeeding at a 10% annual rate. 3.16 Medium term loans, repayable over five years would be available for the purchase of dairy animals. Dairy animals will be provided in pairs and it is assumed that 35 farmers will obtain animals each year. Credit funds totalling MK64,300 will be available for dairy units, all in LNE. A dairy unit budget is illustrated in Annex III, Table 24. Credit for the purchase of farm implements such as ploughs and ox-carts has not been widely used in the past, but based upon recent experience elsewhere in LADD MK65,800 will be available for this purpose. All medium term credit will be at 10% annual rate. 3.17 To administer the credit a credit supervisor with office space, housing, equipment and a motorcycle would be posted at each of the sub-project headquarters. A credit assistant with a bicycle would be posted to each EPA center. Markets (Annex II, Tables 9 and 23) 3.18 Permanent input stores of 350 m.t. capacity each would be constructed at four locations where either ADMARC currently operates a seasonal market or where an EPA center would be located and where the expected incremental input requirements would merit a permanent facility. The stores could also be used for temporary commodity storage. The stores will be located at Chimbalanga and Mpenu in LNE and Bembeke and Kanyama in Dedza Hills. (See map Nos. 16548, 16549). In addition to the storage facility, appropriate housing and equipment would also be provided under the project (para. 4.20). 3/ Farmer numbers are based on a farmer adopting a full credit package, however, if two (or more) farmers wish to share a package they may do so.  - 20 - Water Supplies (Annex II, Tables 10 and 24) 3.19 In the absence of a national water plan, funds would be provided for a potable water component. The Department of Lands, Valuation and Water (DLVW) would, in cooperation with project staff, determine the priority areas for providing water points. The DLVW would make site determination of whether to provide a borehole or a dug well, depending upon the hydrogeologic conditions. An implementation plan would be jointly agreed upon between DLVW, the sub-project management and local authorities which would include the construction program, training of village attendants and proposals for maintenance cost recovery. Approximately one-half of the water points deemed necessary would be provided under the Project.4/ Based upon standard costings, funds would be provided to construct 880 water points in LNE and 600 in Dedza Hills. In addition, the project would rehabilitate the boreholes and shallow wells already in place. It is assumed there are 87 boreholes and 160 shallow wells in the area. 3.20 Maintenance stores for spare parts, etc would be constructed and drilling equipment would be provided. The equipment would consist of: a drilling rig, tractor with trailer and a land rover. With provision of the drilling/construction equipment, the cost of constructing water points excludes capital depreciation. A proposal for maintenance and cost recovery would be submitted as a condition of component disbursement (para. 7.08). 3.21 Water point construction would be phased with construction in LNE during project years 2 and 3 and construction in Dedza Hills during project years 4 and 5. Sub-project management will be responsible for coordinating the activities of this component. Other commitments prevent DLVW from initiating construction at the commencement of the project. Project Administration (Annex II, Tables 11,12,13,25 and 26) 3.22 A Monitoring and Evaluation structure will be established similar to those found in other project areas. The Monitoring and Evaluation Unit (MEU) would be responsible to project management but would be coordinated by the LADD Evaluation Officer and the Central Evaluation Unit in the Ministry of Agriculture to ensure standard and consistent sets of data are collected. Three teams of enumerators-would be established (two teams in LNE and one in Dedza Hills). Each team would consist of a supervisor and 5 enumerators. Appropriate transport and equipment would also be provided. Preliminary data analyses would be undertaken at the sub-project headquarters, although most of the analytical work would be undertaken at LADD headquarters. Eventually much of the data collection role of the MEU and many of the staff are expected to be absorbed by the National Statistical Office (NSO) as part of its nationwide statistical data collection system, although no timetable currently exists for this transition. 4/ A water point is defined as a facility with a daily discharge of approximately 3,375 liters. A dug well is considered a single water point, a borehole is considered two water points and they will service 125 and 250 people respectively.  - 21 - 3.23 Project Administration. This Project will support NRDP activities in two Development Areas (DA), in NRDP nomenclature is also termed project areas, and which are managed by project officers. In this report the DAs are identified as sub-project areas as each constitutes approximately one half of the Project. However, to maintain consistency with NRDP nomenclature the top managers of the sub-projects will be identified as project officers, not sub-project officers. The Project activities will be routinely managed by a team of technical supervisors and administrative support staff. Each management team would consist of a project officer, an assistant project officer and technical officers for the components included in the sub-project. The two sub-project offices will be located at Mkuza (LNE) and Dedza town (Dedza Hills). Headquarters facilities in both locations have been partially constructed under pre-investment funding. The balance of the office space, transport, housing and staff will be provided under the project. 3.24 Lilongwe ADD Headquarters would provide project supervision on a general basis. Technical support, financial control, monitoring and evaluation assistance and policy direction would be provided by LADD. With the increased activity in LADD, four incremental management staff would be provided to the LADD management unit. These staff include a Planning and Evaluation Officer, a Chief Technical Officer in Horticulture and a Chief Executive Officer for the Credit Accounts. A chief technical officer in Land Husbandry would be employed as headquarters staff but would be located in Dedza to supervise work in both Dedza Hills and Ntcheu. Also included in the LADD headquarters component are technical assistance funds for the Livestock Sector Review and external training staff. C. Project Costs 3.25 Total project costs during the five-year life of the project are estimated to total MK13.4 million (US$12.5 million) of which foreign exchange costs would account for MK6.8 million (US$6.4 million), or 51% of the total. The phasing of detailed costs for the respective areas, by Project component are detailed in Annex 2, Tables 1 to 28 and are summarized in the following table.  - 22 - MALAWI National Rural Development Program IV Project Costs 'MK I '000) (USs '000) % Of -------------------------- -------------------------- Foreisn Z of Total Local Foreign Total Local Foreign Total Exchange Base Costs Lilongwe Northeast EXTENSION 596,65 559.68 1,156,33 555.99 521.55 1h077.54 48.4 22.6 TRAININf 125,20 177.30 302.50 116.67 165.22 281.89 58.6 5.9 RESEARCH 74.57 110.74 185.31 69.49 103.19 172.68 59.8 3.6 LAND HUSBANDARY 124.00 134.51 258.51 115.55 125.34 240.89 52.0 5.1 LIVESTOCK AND VETERiNARY SERVICES 234.46 308.56 543.01 218.48 287.53 506,02 56.8 [0.6 CREDIT SUPERVISION 56.42 39.40 95.82 52.57 36.72 89.29 41.1 1.9 CREDIT FUNDS 210,12 279.63 489.75 195.81 260.58 456.38 57,1 9,6 MARKETING 79,50 103.33 182.83 74.08 96.29 170.37 56t5 3,6 WATER SUPPLIES 519,91 546.02 1,065.93 484.48 508.82 9?3.30 51,2 20.9 MONITORING AND EVALUATION 86.19 40.85 127i04 80.32 38.06 118,38 32.2 2.5 PROJECT HEADQUARTERS 294.34 405.54 699.88 274.28 377.91 652.19 57.9 13,7 Total BASELINE COSTS 20401.36 21705.55 5F106.91 29237.73 2f521.20 40758.94 53.0 100.0 Physical Contingencies 127.16 162.09 289.26 118.50 151.05 269,55 56.0 5.7 Price Contingencies 806.20 471,08 1,277.28 751.27 438.98 1190.25 36.9 25.0 Total PROJECT COSTS 39334.73 3f338.72 6!673.45 3f107,50 3,11123 6Y218.74 50.0 1:30.7 ======== =:====== ======== =L====== ======== ======== ======== ====== === Dedza Hills EXTENSION SERUICEP. 6.10 445.22 17061,32 574.13 414.89 989,03 41.9 22.1 TRAINING 112O,5 75.53 288.43 105.20 163.58 268,78 60.9 6.0 RESEARCH 100.83 113,.66 214.49 93,96 105,92 199,88 53.0 4 5 HORTICULTURE 30.45 2-6.96 57i42 28.38 25.13 53.51 47.0 1.2 LAND HUSBANDRY 133.58 134,01 267,59 124.48 124,88 249.36 50.1 5.6 LIVESTOCK AND iEhNVETERINAR!Y 45,94 ........... l Q102.86 148,79 42,81 95.35 138.66 69.1 3-1 CREDIT SUPERVISION 57.01 3639 93.340 53.13 33.91 87.04 39.0 1.9 CREDIT FLIND 143.81 323,04 466.84 134.01 301.03 435.04 69.2 9,7 MARKETING 79.74 103.05 182.79 74.31 96.03 170A34 56.4 3.5 WATER SUPPLIES 336,72 316.77 653.49 313.78 295.19 608.97 48,5 13.6 MONITORING AND EVALUATION 58.09 31.34 89,43 54.13 29.21 53.34 35.0 1.9 PROJECT HEAiDQUARTERS 340,93 416.93 757.86 317.71 388.53 706.23 55.0 15.8 LADD HEADQUARTERS 133.99 392i51 526.49 124,86 365.77 490,63 74,6 10.9 Total BASELINE COSTS 2)190.08 2,618.26 4,808.34 2,040.89 2,439.90 40480,79 54,5 100l 0 Ph2sical Cmntingencies 114.88 139.05 253.94 107.06 129.58 236.64 54.8 5.3 Price Continsencies 937,11 707.44 1!644,56 873.28 659.25 1532.53 43.0 34.2 Total PROJECT COSTS 3i242,08 30464,76 6b706.83 37021.22 3,228,74 6,249.96 51,7 139.5 Februtar- 9. 1983 ?20!38  - 23 - 3.26 Project costs have been estimated at mid-1983 prices. Duties are levied on vehicle fuels but Government bodies are subsequently reimbursed by the amount of the duty. Income taxes are levied on the salary components but are negligible. Physical contingencies have been included at 10% on all civil works, vehicles and equipment and are 5% of the base cost. Price contingencies have been included: (i) on the international component of civil works, vehicles, equipment and supplies at 8.0% in 1983, 7.5% in 1984, 7.0% in 1985 and 6.0% thereafter; (ii) on the local cost component of vehicles, equipment and supplies and on operating costs at 12% annually; and (iii) on the local cost component of civil works at 18% annually, reflecting the continuing inflationary pressure on the construction industry. Price contingencies total 29% of base costs. 3.27 Project costs provide for a limited amount (3 months) of technical assistance consultancy to assist in the training activities. These funds would be used in conjunction with similar funds in other projects whereby a consultant would assist with training activities in several project areas during the consultancy. D. Financing 3.28 Eighty-five percent of the total project costs would be financed by IDA. The proposed IDA credit would be on standard terms to GOM and would cover all the foreign exchange costs of the Project, except those associated with the marketing component, (US$6.2 million) and US$4.4 million or 72% of local costs. The financing of project costs would be as follows: Proposed Financing Plan US$ Million Percent IDA 10.6 85 GOM 1.9 15 Total 12.5 100 To reduce the budgetary impact of transferring large numbers of staff from development to revenue account at the termination of the project, assurances were obtained during negotiations that an orderly transfer of staff would be made and all staff would be on revenue account by the end of the project period. A list of staff positions to be transferred would be submitted to IDA by September 30, 1983. Consequently, much of Government's contribution to the project would be through the revenue account. 3.29 Part of the Government's contribution would be the construction of input sheds, staff housing and market complexes to be operated by ADMARC at an estimated cost of MK365,000 (US$340,000). The MOF and ADMARC have agreed to a financing and lease-back arrangement. Government's contribution would also include all staff salaries and wages (excluding consultants).  - 24 - IV. PROJECT IMPLEMENTATION A. Organization and Management 4.01 The organizational structure of this Project will be similar to that of previous NRDP projects although slight modifications will be made to conform to the NRDP Review recommendations. An organogram indicating the organizational links to parent ministries and overall NRDP management are illustrated in Annex 1, Chart 1. The MOA has overall responsibility for NRDP projects. Under the reorganization, NRDP would be under the direction of a Deputy Secretary (DS) as the NRDP coordinator and the Program Managers of the ADDs would report directly to him. Assurances were obtained at negotiations that this position would be maintained at a DS level (para. 7.01(f)). The Chief Agricultural Development Officer (CADO) would retain the technical lines of responsibility in the agricultural extension effort. Similarly, the Chief Veterinary Officer (CVO) and Chief Agricultural Research Officer (CARO) would maintain technical links with their respective staff but administrative linkage would be the DPS. The Project would also have an administrative linkage to other ministries through the NRDP coordinator but the working relationships would be between the ADD Program Manager and other technical ministries. Thus, the Water Supply Component would be implemented by DLVW but coordinated with the Program Manager. Similarly, the complementary roads development which is funded independently, would be implemented by MOWS but coordinated with the Program Manager. Considerable civil works will be required in the Project and while coordination with MOWS is necessary it is recommended that LADD management retain the services of a consulting architect, to coordinate and manage the construction of houses. The architect would develop site plans, issue and evaluate tenders, develop contracts, supervise construction and generally exercise cost control in the construction of houses through design alteration, alternative materials, etc. (para. 1.25). As a cost control measure civil works procurement would generally be through local bidding rather than the more common use of force account (para. 4.22(c)). The Ministry of Works would, of course, be eligible for bidding. A condition of disbursement against civil works would be the preparation by MOA-LADD of a civil works program, acceptable to IDA, which specifies designs and construction methods (paras. 4.23 and 7.07). 4.02 The NRDP IV project includes two sub-project areas, or two Development areas, and as such the MOA has appointed a Project Officer for each of these two areas. These project officers would report to the Program Manager of Lilongwe ADD and would have day-to-day operational responsibility for the Project. The Program Manager of LADD would have overall management responsibility and would report to the NRDP coordinator. The technical officers at the project level would have two lines of responsibility: one to the Project Officer for daily activities and coordinating work programs, and one to specialist staff in the technical departments or ministries. 4.03 Project management and staff would actively collaborate with local groups and farmers clubs. With all of the credit and increasing amounts of the extension work done with farmers clubs it is imperative that collaboration be close. Staff will also collaborate with local groups on infrastructural development such as livestock dips, slaughter slabs and water points. Local participation in the Project will be encouraged.  - 25 - B. Implementation Schedule 4.04 Project investments and financing for operating costs would be provided over a five-year period. It is expected that the procurement of vehicles and equipment, staff recruitment and most of the civil works construction (except for water points) would be achieved during the first year of the project. The implementation rates of the various components are shown in Annex I, Charts 2 and 3. Details for each component are shown in the cost tables (Annex II, Tables 1 to 28). Key indicators are summarized in Annex I, Table 1 and 2. C. Management and Implementation of Specific Components Extension 4.05 The extension service, in consultation with other units, would develop annual work plans outlining objectives, plans for achieving those objectives and the substantive composition of the extension component. Integrated planning with the training, research, horticulture, land husbandry and veterinary services will ensure consistent objectives are pursued and mutually reenforcing work plans are designed. The Project Officers located at Dedza Hills and Lilongwe Northeast Headquarters would have overall responsibility for the coordination and development of the work plans. To ensure the program is oriented along implementable lines, participation of District and Village Committees will be sought. The annual work plans will be submitted to IDA for comment (para. 7.01(c)). 4.06 To improve the efficiency of the extension workers they will work primarily with farmer groups/clubs composed of 40-50 farmers. The extension worker would visit each farmer group about once each two weeks. Part of the extension workers' time would be devoted to the formation of additional groups/clubs and this would be specified in their individual work plans. Each extension worker would be trained to look at farming problems from a holistic farming systems perspective where the complete farm enterprise would be considered and advice/assistance would be directed toward the farm enterprise with due recognition of the competing demands on labor, land, etc. Interaction and support would be provided by research, horticulture, land husbandry, livestock and veterinary services staff in addition to extension support staff. The extension workers would also be responsible, with assistance from research staff, for setting up demonstration plots on farmers' fields. A component of the cooperation with and assistance from Research would be the incorporation of high analysis fertilizer trials into these demonstration plots. 4.07 The extension workers would be organized by EPA. Each EPA center would have a Development Officer and an assistant who would coordinate staff and farmer training and the general work program. The ratio of farmers to field assistants would be improved to about 750:1 (see para. 3.03), in the project area. Assurances were obtained that new staff would, to the extent possible be obtained by transferring in, from other project areas, staff made redundant by changing the farmer to field assistant ratio to recommended levels. This would result in the acquisition of experienced staff and minimize the incremental Governmental cost by restricting the recruitment of new staff. Assurances will be sought at negotiation that staff transfers from other projects would be undertaken to the extent possible (para. 7.01(b)).  - 26 - Government will specify the numbers of staff to be transferred and the projects from which they will be obtained. 4.08 The Farm-Home Assistants will complement the work of the general extension workers by identifying and focusing on the specific farming problems faced by women. To facilitate supervision and evaluation of the FHA program the incremental number of FHA staff will be phased in over the first three years of the project. Where horticulture is important (Dedza Hills) the FHAs would be specially trained in horticulture. Training 4.09 Both farmers and project staff would be the recipient of training. The training structure for farmers would be through day training sessions, both at Day Training Center (DTC) demonstration plots and on farmers' field and with farmers' animals organized by the Training Center staff. Emphasis would be placed on increasing the effectiveness of farmer training rather than increasing the number of trainees, although the numbers trained are expected to increase slightly. Additional DTCs would be established at the EPA centers where none currently exist. Training sessions at Residential Training Centers (RTCs) would be limited to specialized crop or livestock training and would normally be a one-week course. Such courses would be practical production and problem oriented. Crop production on the RTC farms would be intensified to improve the demonstration value of these farms and to supplement the supply of food to the RTC kitchens. A special effort would be made to involve more women in the training program. 4.10 Extension staff would receive refresher courses on crop and livestock husbandry and emphasis would be placed on the understanding and interpreting of research results and implementing technical recommendations. Periodic training sessions by subject matter specialists would be introduced to complement the regular visit schedule and meetings with research staff. They would receive technical courses on horticulture, silviculture, land husbandry and crop storage and general courses in extension methods, communication and management. Short term consultant trainers would be engaged where appropriate and available, particularly in the generalized methodological courses. Research 4.11 The Project would support and expand food crops (including horticulture) research at Chitedze and Bembeke Research Stations. The Research and Trials Officer (RTO) would be ADD based but will exchange some of his time for other specialized staff time to participate in the adaptive research work. A comprehensive field trials program covering the major ecological areas and encompassing a more diversified set of crops would be initiated. The field trials would consist of simple single question replicated trials and farming systems research. The horticultural research program will consist principally of varietal trials sited at EPA centers. Research station trials using high analysis nitrogenous fertilizers will be extended into field trials. 4.12 Annual work plans would be developed well in advance of implementation, (for budgetary purposes), with close liaison with the  - 27 - Department of Agricultural Research and in cooperation with other Project staff. The Research and Trials Officer would have overall responsibility for developing and coordinating the work plans. Detailed research activities would be discussed and agreed upon by the end of September. Assurances were obtained at negotiations that the annual work plans would be submitted to IDA for comment by June 30 (para. 7.01(c)). After harvest the RTO would analyze and interpret the trials results, maintaining close liaison with the LADD Monitoring and Evaluation Unit to facilitate management feedback. Research staff would also participate in the training program for extension staff. Horticulture 4.13 Although considerable quantities of vegetables and fruits are currently produced in the Dedza Hills area, production could be expanded. However, the local market for additional production is very limited and is a production constraint. Therefore, only carefully planned production programs would be supported. The fruit and vegetable varietal testing at Bvumbwe would be extended to Dedza Hills, primarily at EPA centers but also on farmers' gardens. The objective would be to extend the present marketing season through the use of a wider varietal selection combined with improved crop production and protection techniques. Demonstration storage facilities for potatoes and onions would be constructed. Land Husbandry 4.14 The land husbandry staff posted to the EPA centers would work with the extension staff in land use planning and conservation. Conservation emphasis would be on biological, rather than physical, methods such as establishing perennial crops on steep slopes. Balanced rotations would be demonstrated, where feasible, including restorative cropping breaks of legumes and grasses in conjunction with the livestock program. Livestock 4.15 The livestock and veterinary staff would liaise closely with the extension staff and farmers to ensure proper adoption of stallfeeder and dairy credit packages. Farmers would be encouraged to obtain stallfeeders directly from other farmers, requiring close supervision by the veterinary staff to verify farmers met the criteria for credit (para. 4.17). Veterinary supervisory staff would be responsible for visiting dipping centers, ascertaining acaricide and drug requirements and identifying sites for livestock dips, poultry distribution houses and slaughter facilities. Veterinary assistants would be available for meat inspection on slaughter days at the slaughter facilities. Credit 4.16 Seasonal crop production credit for seeds and fertilizers would be available for tobacco, groundnuts, maize and wheat, with the credit packages being provided for a minimum of one acre (except for tobacco where the minimum would be 0.5 acre), although farmer sharing of credit packages is permissible. Credit would be available primarily in a package and improved seed would not be provided unless the fertilizer component could also be provided (inapplicable to groundnuts which do not require  - 28 - fertilizer); however, partial packages involving only the fertilizer element will be permitted. Credit would be provided only through farmers clubs, which if self-accounting would be charged 10% interest and if not self-accounting would be charged 15% interest. (Since seasonal loans are repaid in 6-9 months the annualized rate is considerably higher.) Repayment of the loans would be through the clubs whereby the treasurer would collect the funds and turn them over to the credit assistant during periodic visits. The credit packages are indicated in Annex III, Tables 27 and 28. 4.17 Stallfeeder credit would be provided upon approval by the veterinary staff, whereby the recipient received the animal and the seller received the financial resources. The project would collect the finished animals, after 5-6 months, and transport them to the Cold Storage Company (CSC) at which time the original loan would be recoverd as well as the 10% interest and other costs incurred. To participate in the program, a farmer would need access to a minimum of 2 ha of utilizable crop residue and the ability to provide an animal pen and feed storage. The number of stallfeeders estimated under the program start at 150 in the first and subsequent years of the project. 4.18 Dairy loans would be provided "in kind" to eligible farmers in a package consisting of two pregnant heifers, spraying equipment and supplies, utensils and pasture establishment materials. The Livestock Officers would supervise the dairy credit package adoption and the implementation of the dairy unit. Criteria for participating in the dairy program include the availability of a minimum of 1.2 ha for improved pastures and be located on an existing milk collection route. Loan recovery would be through milk sales to the Capital Dairy which would withhold 50% of the value of sales until the entire amount due was collected. The annual interest rate would be 10% and it is estimated that recovery would require about five years. The profitability of milk production is low compared with alternative land use. However, it is an attractive enterprise as it provides a steady source of funds distributed over the year. 4.19 General farming loans would be provided for the purchase of oxen and ox-drawn farm equipment. The loan would be provided "in kind" and would be repaid over five years at a 10% interest rate. The credit section has established criteria for selecting recipients of this type of credit and loan repayment would be directly to the credit assistants. Marketing 4.20 ADMARC is responsible for distributing improved technical packages, fertilizer and improved seed, to smallholder farmers. Funds for constructing additional input sheds would be provided but the responsibility for construction of the input stores and support infrastructure, staffing and operating the stores would rest with ADMARC. ADMARC would be responsible for staffing and operating the stores.  - 29 - Water Supplies 4.21 The Groundwater Section of the Department of Lands, Valuation and Water (DLVW) would be responsible for implementing this component. A hydrogeologist would be appointed to the Project and would have supervision responsibility. Together with Project management, Development Area and local officials he would develop a plan for siting water pc4nts with due consideration given to water deficit areas. Determination of the type of water point, borehole or dug well, as well as their precise location would be determined by the hydrogeologist supervisor. A scheme for training village attendants, pump mechanics, village maintenance and repair, and maintenance cost recovery would be developed and submitted to IDA and would be a condition of disbursement for this component (para. 7.08). D. Procurement 4.22 Procurement under the Credit would be in accordance with Bank/IDA guidelines. Such an assurance would be sought at negotiations (para. 7.01(d)). Specifically: (a) orders for vehicles, spare parts, machinery and equipment (US$830,000)5/ would be bulked as far as possible, and orders of US$100,000 and above would be subject to international competitive bidding (ICB) procedures. The bidding documents would specify that suppliers will maintain or agree to maintain an adequate after-sales service and inventory of spare parts. Procurement bulking for ICB, as well as items for local procurement would be approved by IDA prior to tendering/purchasing; (b) orders for vehicles, spare parts, machinery and equipment that cannot be bulked in packages of US$100,000 and above, up to US$250,000 in aggregate, would be procured in accordance with existing local procedures which are acceptable to IDA; (c) civil works construction, including housing, stores, offices, wells and boreholes, and rehabilitation and construction of livestock dips (US$4.25 million), would be too small and scattered to attract international interest. The largest single item of construction that could be bulked would be construction of housing at an estimated cost of US$2.27 million, but at sites in the two districts which are scattered and somewhat difficult to reach. Procurement would, therefore, be by contracts awarded following locally advertised bidding (although embassies would be notified and foreign firms would be entitled to compete); if approved in advance by IDA, procurement could be by force account of the Ministry of Works or of MOA. (d) consultants' services for the livestock review, (US$200,000) would be procured according to the August 1981 Bank guidelines; the consultants would be employed with prior IDA approval. 5/ All figures in this paragraph exclude contingencies.  - 30 - E. Disbursement 4.23 Disbursements of funds from the Credit would be on the following basis: (a) 100% of foreign exchange and 90% of local expenditures for vehicles, spare parts, equipment and other materials excluding marketing and the items to be financed for rural water supplies (US$825,000); (b) 100% of total expenditures on civil works, excluding marketing and rural water supplies if by force account and 90% if by private contractors (US$2.92 million); (c) 100% of total expenditures for incremental credit (US$1.23 million); (d) 100% of foreign exchange and 90% of local expenditures for agricultural inputs for research, extension, training, land husbandry and horticulture activities and veterinary chemicals and drugs (US$500,000); (e) 100% of total expenditures for incremental operating costs excluding operating costs for marketing and rural water supplies (US$1.6 million); (f) 100% of total expenditures for consultants' services (US$235,000 million); and (g) for the rural water supply component: (i) 100% of total expenditures on civil works (US$2.08 million); (ii) 100% of foreign exchange and 90% of local expenditures on equipment, vehicles, spare parts and materials (US$82,000); and (iii) 100% of operating costs (US$130,000). An amount of US$1.0 million would be unallocated. A condition of disbursements against category (b) would be that MOA had submitted a program for civil works acceptable to IDA, specifying the designs and method of construction (private contract or force account) to be used (para. 7.07). A condition of disbursement against category (g) would be submission by DLVW to IDA of a proposal for maintenance and cost recovery (para. 7.08). Disbursements against categories (a), (b(i)), (f) and (g(ii)) would be fully documented. Disbursements against the other categories would be made against Statements of Expenditures (SOEs) signed by the project officer in charge of the area and certified by the program manager and the financial controller of LADD and signed by the Principal Accountant of MOA. The documents for expenditures covered by SOEs would be retained by the borrower and made available for IDA inspection during normal supervision missions.  - 31 - F. Accounts and Audits 4.24 All Project funds would be channelled from the Ministry of Finance to the Ministry of Agriculture and be budgeted and administered by the Lilongwe ADD, except for the livestock review, which would be administered by the Planning Division of MOA. Project financial records would be maintained at LADD headquarters, except for the account for the livestock review, which would be retained by the Principal Accountant of MOA. Credit fund operations would have separate accounts, and although all credit funds from the various projects are to be pooled into one general account, a separate record would be maintained at LADD headquarters for credit operations for Dedza Hills and Lilongwe Northeast. In previous projects, the accounts have been audited by the Auditor General and this arrangement has been satisfactory to IDA. Assurances would be obtained at negotiations that all Project accounts, including statements of expenditures and the account for the credit operations, would be audited by the Auditor General or by independent auditors whose experience and qualifications are acceptable to IDA and that the audited accounts together with the auditor's report would be submitted to IDA not later than six months following the end of the fiscal year (para. 7.01(e)). G. Project Monitoring, Evaluaton and Reporting 4.25 The Monitoring and Evaluation unit would conduct annual surveys on farmers' plantings, yields and output. To permit the inclusion of more ad hoc studies required by Project management the household composition survey would be conducted biennially. Three survey teams, two in LNE and one in Dedza Hills, would be established. To the extent possible these staff would be obtained by re-deploying staff from other project areas which are overstaffed. Each team would consist of a supervisor and five enumerators. The supervisors (SCO) would supervise the units' activities and perform preliminary analyses, would assist project management to identify data requirements and would ensure proper processing. The supervisors would liaise closely with technical staff to ensure adequate information feedback and monitoring and be responsible to the Senior Evaluation Officer in LADD. 4.26 The Senior Evaluation Officer in LADD headquarters would provide professional support and direction to the Monitoring and Evaluation Units and would, on request, assist in designing ad hoc studies. He would also monitor the use of survey information and ensure the surveys meet project and program needs and would assist the planning officer use the survey data in developing annual work plans. The Central Evaluation Unit in the Planning Division of the MOA has standardized survey forms and computer programs to analyze and tabulate the annual surveys. This Unit would coordinate evaluation activities and provide technical support. 4.27 The Monitoring and Evaluation Unit would be responsible for compiling Quarterly Reports. Project and LADD management would revise the reporting procedure and modify the format of Quarterly Reports such that comprehensive reports would be prepared semi-annually and only brief reports highlighting major issues important for management decisions would be prepared for the other two quarters. The semi-annual reports would record adequate information for the eventually required Completion Report but comprehensive detail would not be required quarterly. Consistent with  - 32 - the NRDP Review recommendations the reporting timetable would be as follows: 1st quarter (A-J): Brief report highlighting major problems - staffing, procurement, infrastructure, etc. 2nd quarter (J-S): Comprehensive semi-annual report. 3rd quarter (O-D): Brief report highlighting major issues - crop and livestock situation, credit recovery, input problems, etc. 4th quarter (J-M): Comprehensive annual report. This would free some of the Evaluation Officers' time for assisting management with interpreting surveys and monitoring reports, particularly farmer adoption of recommended practices. A Completion Report would be prepared no later than six months after the credit closing date (para. 7.01(i)). H, Policy Issues 4.28 The lack of economic incentives has been one of the constraints to increased production by the smallholder sector. Although commodity prices are presently adequate to make crop production profitable, in the late 1970's the profitability eroded and labor income from farm activities was often below the opportunity cost of labor. Further, producer prices have often been far below an export parity price. Smallholder Crop prices established for the 1982/83 season reflected a movement toward parity pricing and assurances were obtained at negotiation that prices for the years of the Project would be set at levels acceptable to IDA (para. 7.01(g)). 4.29 The bases for increased production in this Project, as in other NRDP projects, are improved husbandry and improved technical packages. The technical packages are highly dependent upon imported fertilizer which must be applied in a timely fashion to provide the incremental production. As the quantity of fertilizers required has increased in recent years, the procurement and delivery mechanism has become strained and less effective in the timely provision of these inputs. Without these inputs the anticipated benefits from the Project cannot be achieved. A joint IFAD/IDA cofinanced project has been proposed which will address this issue and if processed as scheduled such a procedure should be in-place for the 1983/84 crop season. Assurances were obtained at negotiation that fertilizer procurement would have a high priority in the allocation of foreign exchange (para. 7e01(h)).  - 33 - V. PRODUCTION AND MARKETING A. Foodcrops Foodcrop production 5.01 Agriculture in the two project areas is dominated bv smallholder production, apart from 34 percent of the Dedza Hills area which is forest reserve. Both areas are characterized by relatively large areas of technically non-arable land but on which considerable numbers of families have settled. Maize, groundnuts and sweet potatoes are the major foodcrops grown in the LNE area, while in Dedza maize and various types of beans predominate. Farm sizes are generally small in both project areas and all the arable land is already intensively cultivated and it is improbable that much more arable land will be brought under cultivation. There is likely, however, to be some expansion of cultivation onto the marginal and steeply sloping areas for reasons of population pressure. In Lilongwe NE where tobacco and groundnut acreage have been considerably reduced to the benefit of the maize acreage due to relative price changes, it is possible that there will be some shifting back to the more traditional cropping pattern as price relativities are adjusted. It is assumed, however, that the benefits of the project would be mainly from improved yields resulting from better husbandry practices, increased use of fertilizer and higher yielding varieties. Yields in both project areas are currently low and substantial increases should be possible through application of a more intense and better focussed extension service; the increased availability of seasonal credit for fertilizer and seed of improved maize varieties should enable yields to be increased still further. (See Annex III tables for yield assumptions.) The incremental production expected by year 7 is as follows: Incremental Production at Full Development (1990) Crop Tonnage Lilongwe North East Maize -local 628 -improved 5,665 Groundnuts 126 Tobacco 207 Dedza Hills Maize -improved 6,441 Groundnuts 15 Beans 199 Wheat 80 Potatoes 515 With production in both project areas now being devoted largely to subsistence requirements, estimates of foodcrop production are approximate, but these incremental amounts represent an increase of roughly 8%, 3%, and 50%6/ of without-project production for maize, groundnuts, and tobacco respectively in LNE; and for Dedza Hills the increases are 6/ Tobacco purchases by ADMARC from LNE project area have fluctuated greatly over the past several years. The 50% increase statistic refers to a 1979/80 - 1980/81 base.  - 34 - estimated at 12% and 7% and 4% for maize, groundnuts and beans, wheat would be essentially a new crop for the area since only a few acres were grown in 1981 and there are no reliable estimates of potato production. Incremental production of groundnuts are likely to exceed these estimates as some shift of cropping back to groundnuts is expected to follow from improved crop price relativities. Consumption 5.02 At present, maize consumption requirements are met in Lilongwe Northeast but not in Dedza Hills. The incremental production in Lilongwe North East would be sold, on both the informal market and to ADMARC. In Dedza Hills it is expected that most of the increased production of composite maize will be consumed within the area while the hybrid maize will be sold to either ADMARC or to the informal market. 5.03 It has been assumed that the incremental groundnuts produced in both areas would be sold to ADMARC for eventual export. Marketing 5.04 Although ADMARC is responsible for buying surplus smallholder produce, it is a residual buyer of foodcrops. Local markets provide the major foodcrop marketing channels in both project areas. In 1980/81 ADMARC purchased 8700 tons of maize in Lilongwe North East, or about 11% of estimated production. For groundnuts in Lilongwe Northeast the corresponding figures were 1600 m.t. and 35%. Although Dedza Hills is a net maize deficit areas, ADMARC purchased 3700 m.t. of maize or approximately 7 percent of production. Groundnuts purchased in 1980/81 were 90 m.t. approximately 42 percent of production. Prices 5.05 Prices are panterritorial and fixed, they are agreed by a Price Commission and announced prior to the planting season. Price levels reflect the need to provide production incentives to farmers as well as permitting ADMARC to earn profits. The Government has introduced a pricinEg methodology which considers economic export and import parities, ADMARC's profits, and gross returns to farmers. This was used for the 1982/83 rouncl of producer prices and should have a beneficial effect on stimulating both foodcrop and cash crop production in both Lilongwe Northeast and Dedza Hills. Better relativities between crop prices should assist the extensiort service in their efforts to introduce a more balanced cropping program within both areas and so reduce the risks associated with the drift to monocropping of maize. B. Tobacco Production and Exports 5.06 Tobacco production is critical to Malawi's economy. Although lower in 1981, tobacco typically provides about 55% of export crop earnings; it provides the major source of cash income to approximately 20 percent of Malawi's farming families either as smallholders on their own plots or as labor on estates. The composition of production has changed in  - 35 - recent years. Typically estates have produced 55-60% of the tobacco (fire-cured and burley) on 40-45% of the tobacco area whereas smallholders produced 40-45% of the tobacco (fire-cured, sun or air cured and oriental) on 55-60% of the land. However, estimates for 1981/82 production indicates production was 36% flue-cured, 46% burley, 15% fire-cured and 2% sun or air cured. (Oriental tobacco production while important in some locales is unimportant nationally.) A small percentage of the flue-cured tobacco is grown by about 900 smallholders under the Kasungu Flue-Cured Tobacco Authority. Burley is still restricted almost entirely to estates although small pilot schemes have now been introduced under the IDA-financed NRDP III project in Karonga and Chitipa and under the IFAD-funded project in Dowa West. 5.07 Under the Project, fire-cured tobacco production in Lilongwe Northeast would increase by about 200 m.t. at full development (year 6 of the project), from both increased yields and from some substitution for unimproved maize. This would represent an increase of about 50% of total production over the 1980/81 crop base from Lilongwe Northeast but the 1980/81 crop was only 50% of that produced in the late 1970s. Although operating under a quota system (para. 2.09), production has declined markedly over the past few years and this incremental production could therefore be accommodated within the system. Tobacco is not.grown by smallholders in Dedza Hills. 5.08 Exports of tobacco from Malawi increased rapidly in the 1970s, but there has been a decline in the 1980s in both flue-cured and dark-fired tobacco production and exports. The major importers are the United Kingdom, Netherlands and Germany. Marketing and Prices 5.09 ADMARC is responsible for marketing the dark-fired, sun and air cured tobacco from smallholders, which it then sells on open auctions at Limbe and Lilongwe. ADMARC pays producers on a two-tier system; the first payment is at a price announced prior to the planting season and which is paid on receipt of the leaf at the primary market, the second payment is dependent on the amount realized on the auction floor and is paid usually in September when the inputs for the new crop have to be purchased. Second payments were not made in 1979 and 1980 due to the low prices received at auction but were re-instituted in 1981. Second payments range from 15%-60% of the first payment. Producer prices for tobacco have been less than one-fifth of the auction price, and ADMARC has realized substantial profits on its tobacco accounts in the past. This relatively low producer price has been used in conjunction with a quota system to control the production of fire, sun and air cured tobaccos. Malawi's share of the world market for these types is high, about 20%, thus overproduction could have a disastrous effect on prices, as it has in the past. 5.10 World market prices for tobacco are projected by the World Bank for flue-cured only and these are expected to remain at approximately the same level as in the last few years with some narrowing in regional differentials as developing countries upgrade the quality of their crops. Prices projected for tobacco in 1990 and 1995 are about $2000 per ton in 1981 constant US dollars. The price trend is thus thought likely to remain largely stationary over the medium term. ADMARC projects that the prices  - 36 - of sun, fire and air cured tobaccos will increase slightly in real terms in the 1980s. GOM is conducting a study of the tobacco sector including: world market prospects, implications for Malawi's tobacco production and Malawi's potential for responding to future market prospects. C. Horticulture Production 5.11 Fruits, vegetables and potatoes are grown in the higher altitude locations of the project area, particularly in Dedza Hills. There have been no quantified estimates of the amount of land devoted to horticultural crops or the production of fruits and vegetables. However, surveys indicate that one-third of the households in Dedza Hills derive income from the sale of horticultural produce which provides about 12% of the cash income of the rural community. Potatoes are the most important of these crops, but cabbage, onions, tomatoes, apples, and peaches are also important in varying degrees. Benefits from the Project would be mainly from improved yields of potatoes resulting from better husbandry practices and improved seed stock. But improved storage facilities would reduce wastage and extend the marketing season. The planting of fruit trees on the steeper slopes would also provide a conservation-erosion control benefit. The only quantified benefits of the project is the estimated incremental production of about 500 tons of potatoes. Marketing and Prices 5.12 No organized marketing system exists for fruits and vegetables (including potatoes) and producers make their own arrangements with middle men or sell directly to customers on the roadside. Major market areas, such as Lilongwe and Blantyre, are characterized by gluts and shortages with consequent significant fluctuations in price. Some institutions such as hotels, schools, military barracks and supermarkets invite annual tenders for the supply of perishable produce and groups of farmers cooperate to tender for those orders. The potential for domestic market expansion of fresh fruits and vegetables is relatively small, during the main harvest seasons. However, the production and marketing season can be extended by using early/late maturing varieties and improving the storage facilities, Small quantities of deciduous fruit are imported each year and both fruits and vegetables have been exported (by air) to Seychelles and Europe. However, exporting has been abandoned due to the infrequency of intercontinental flights and the relatively low profitability, The development of the new International airport near Lilongwe may reopen the possibility of export, but no export benefits are assumed under the project. The ADMARC Canning Company in Malawi processes locally grown fruits and vegetables but their requirements are limited and would not likely impact upon Dedza Hills. D, Livestock Production 5,13 Livestock production would increase moderately due to Project activity. Rehabilitated and new dipping facilities would result in improved disease control, which would lower mortality rates and improve  - 37 - productivity (weight gain). Cattle density is relatively high in LNE, at about 36/km2 but only 18/km2 in Dedza Hills. A population plateau is expected to be reached soon, with or without the Project. With current grazing practices (i.e., without large-scale adoption of stallfeeding), a stable herd size will soon be achieved and calving rates will decline; under the Project, the stable herd size would be achieved sooner, and the animals would be larger. The proposed expansion of dairying under the Project would probably occupy land currently utilized for crop production. Cattle Marketing and Prices 5.14 Two separate livestock markets operate in parallel although some overlap does occur. The Cold Storage Company (CSC) buys animals at a fixed price per kilogram, depending upon grade. Until April 1980, the fixed price had been below local market prices and CSC was a residual purchaser, but with current prices more animals are being sold to CSC. Several auction markets exist where animals are purchased by CSC, local butchers, other farmers and rural development programs for stallfeeders. Most purchases, however, are for slaughter: in recent years about 80% of the animals sold were bought by CSC or local butchers. 5.15 Aniamls sold at auction on a liveweight basis are classified in one of five categories, but only Grades A and B are suitable for feeders. Recent average prices in local markets, compared with official fixed prices are as follows: Local Market Prices LWT (K/kg) Standard .44 Commercial .33 Inferior .26 Grade A .44 Grade B .39 Official Prices (1981) CDW (K/kg) Choice 1.10 Prime 0.99 Standard 0.88 Commercial 0.77 Inferior 0.44 The price established in 1980 was the first change in official cattle prices since 1975 and the price increases of about 60% (for most grades) have stimulated offtake. Local auction sales have increased by 50% between 1979 and 1981. However, it is possible the auctions are replacing former sales between individuals. Under a livestock study, financed by the World Bank Technical Assistance Loan, the issues of offtake and meat supply, livestock pricing and meat demand will be studied by the Malawi Government. Milk Collection and Prices 5.16 Milk is already collected from the Project areas by Capital Dairy for processing and sale in Lilongwe. Producer prices vary with the amount of bulking, but the present average for smallholders in Lilongwe Northeast is 11 tambala (KO.11) per liter.  - 38 - Milk prices (Malawi Milk Marketing Board) (i) for non-registered farmers = 8.58 t/kg for deliveries up to 12290 kg. (ii) for registered farmers Delivery (kg) t/kg 0 to 850 9.02 851 to 1700 9.68 1701 to 3400 10.12 3401 to 6801 10G56 6802 to 13603 11.00 13604 to 15360 11.44 15361 negotiable The milk price schedule has remained unchanged for 7 years and dairying does not now favorably compete with alternative cropping enterprises. Although farmers are quite interested in dairying as demonstrated by the queue of farmers awaiting dairy animals, this interest is probably based on some non-quantifiable benefits (see para. 6.07 for discussion). This issue will also receive attention under the planned Livestock Review. VI. BENEFITS AND JUSTIFICATION A. Farmers Benefits 6.01 The objectives of the Project are to increase farmers' food supply and cash incomes, and to increase the standard of living through improved nutrition and improved access to rural water supplies. The Project would also introduce conservation measures to maintain the productivity of the soils and yields of traditional crops and would introduce new varieties and crops and improved husbandry practices to increase crop yields and livestock productivity. About 10,500 farming families altogether are expected to benefit from Project activities on crops, representing about 20% of the total population (by 1987/88) in both areas. In Lilongwe Northeast, about 5,500 farming families are expected to adopt new packages and husbandry practices over the Project period and in Dedza Hills about 5,000 farming families are expected to participate in the Project's crop activities. The difference in the two areas is based on the fact that Dedza Hills is an area of lower potential for improving yields than is Lilongwe Northeast. The additional water supplies would provide about 50% of the area's population with easy access to safe water, 6.02 Details of farm budgets are in Annex III, Tables 9-14; it is assumed that farm sizes and cropped areas would remain the same under the Project, so all benefits would be generated by maintaining/improving yields and by farmers' shifting to more profitable crops (with improvements in yields of subsistence crops to meet food requirements, more land would be available for cash crops). The budgets are illustrative only and should be taken as orders of magnitude. 6.03 In Lilongwe Northeast virtually all farmers have most of their land in local maize and the remainder in hybrid maize, and, to a lesser  - 39 - extent, in tobacco as cash crops. In Dedza Hills as well, almost all farmers have most of their land in local maize, intercropped with pulses, and other subsistence crops like cassava, with a small portion of their land used for hybrid maize as a cash crop. With the recent dramatic increase in the producer price for maize (an increase of over 60% was effected for the 1981/82 season), hybrid maize has become the most attractive cash crop in terms of returns to labor (except for potatoes in Dedza Hills, for farmers with assured market access). 6.04 Illustrative budgets have been worked out for various types of farms in each area. In Dedza Hills, for a "typical" maize and groundnut farmer, assuming adoption of composite and hybrid maize and improvements in groundnuts, as well as in the intercropped pulses, estimated gross incomes would go up from MK109 (US$106) to MK216 (US$201) and net cash income would rise from MK6 (US$5) to MK62 (US$58). A maize/potato/wheat farmer would be able to increase gross income by the largest amount, from MK139 (US$130) to about MK374 (US$349), with most of the incremental income coming from potatoes followed by a wheat crop. Net cash income, after deducting for the value of subsistence, would go up from MK35 (US$33) to MK270 (US$252). These farmers are limited in number and would not create marketing problems for the disposal of potatoes. Only those farmers currently growing potatoes would be encouraged. 6.05 In Lilongwe Northeast budgets were developed for a "typical" tobacco farmer, assuming improvements in maize and tobacco and for certain farmers, the addition of dairy cattle, and a "typical" maize farmer, assuming improvements in maize and groundnuts. The tobacco farmer, with a farm size of 2.5 ha, would increase his gross income by about 30%, from 1mK291 (US$271) to MK381 (US$355); after deducting for subsistence and paid labor, his net cash income is estimated to increase by 63%, from MK138 (US$129) to MK224 (US$209). The increased income would come equally from improving tobacco and adopting composite and hybrid maize. The maize farmer, assuming a switch to composite and hybrid maize varieties and improved groundnut seeds and husbandry, would increase his gross income by 70%, from MK150 (US$140) to MK256 (US$239); his net cash income from agriculture after deducting for subsistence, would increase from MK9 (US$8) to MK106 (US$99). 6.06 The returns per manday to crops under the Project vary considerably, from MKO.52 for tobacco to MK1.64 for hybrid maize in Lilongwe Northeast, and in Dedza from MK0.93 for groundnuts to MK1.51 for potatoes. This compares to an estimated opportunity cost of labor for the whole Project area of about MK0,50 per day. 6.07 For farmers in Lilongwe Northeast who received credit for dairy cattle, the benefits at current milk prices would consist mainly of the steady flow of cash income throughout the year from the sale of milk. In addition, the non-quantifiable benefits make the dairy enterprise attractive: the crop rotations with Napier and Rhodes grasses would maintain or improve the soil fertility and the animals' manure could replace some purchased fertilizer. Although the return per manday and per hectare to a two-dairy cow enterprise are less than for the other crops, at MK0.33 per manday compared to an estimated MK0,52 for tobacco, the demand for dairy cattle in Lilongwe Northeast exceeds the supply. The opportunity cost for labor devoted to the dairy operation is relatively low, much of the labor is performed by children and the adult labor is concentrated in  - 40 - periods of the day when alternative labor use is limited. Sensitivity analysis shows that a small increase (of 10%) in. the milk price would more than double the net benefits and the Government is now considering an increase in milk prices, but according to information gathered at appraisal, such an increase would not be necessary to stimulate farmers' interest in dairy cows in the Lilongwe Northeast area. 6.08 For livestock owners, rehabilitated and improved dipping facilities and operations would mean some weight gain and reduced mortality for their cattle from better health. About 7,200 farmers in LNE (or 20% of the farm families) own cattle, and because coverage is very good, it is assumed that all of the cattle would be affected by improved dipping facilities and better access to drugs. Benefits were not calculated for the "average" livestock owner as ownership is quite skewed, but for an individual herd of 10 animals about MK2O (US$19) could be realized after year 8 of the Project from an estimated 8% increase in cattle weight. B. Government Cash Flow and Recurrent Cost Implications 6.09 The Project will be the first phase of a long-term development input into the area, Thus, at the end of the project implementation period it would continue to require development resources and a succeeding phase is anticipated. Project's impact on Government's budget would be negative and the contribution to Government revenues and foreign exchange would depend upon ADMARC profits, farmgate and export prices, By the end of the project annual incremental operating costs, in constant terms, would be about MK811,00 (US$756,000), including vehicle replacement costs, which would amount to about MK525,000 (US$489,000) in the two years immediately following the close of the Project. Assuming a five-year life, the annualized costs of vehicle replacement are MK105,000 (US$98,000). The annual incremental net profit likely to accrue to ADMARC by the end of the project period is about MKG50,000 from the marketing of incremental groundnuts, tobacco and wheat. Annual incremental foreign exchange earnings are estimated to be US$530,000 by the end of the Project. While the economic benefits to the country would exceed costs (para. 6.18), the cash flow impact for Government would be negative. C. Economic Analysis 6,10 The main objectives of the rural development projects for Lilongwe Northeast and Dedza Hills are to improve the productive capacity as well as to conserve the existing production potential of the areas. Extension and research efforts focused on adaptive trials to improve crops currently grown, diversifying into new crops, introducing agroforestry, and improved veterinary services combined with access to credit are expected to generate economic benefits. The benefits from increased crop production and livestock production are quantifiable and are calculated below. 6.11 Economic analyses have been performed on the incremental costs and benefits, a separate analysis for crops and livestock in the separate sub-project areas of Lilongwe Northeast and Dedza Hills. One rate of return was also calculated for the Project as a whole. The assumptions on cost allocations and incremental benefits are discussed below.  - 41 - Quantifiable Benefits 6.12 Increased crop production would result from expanded and improved extension, land husbandry, training, credit and marketing activities under the Project. Most incremental production would be attributable to improved yields from better varieties, purer seed, proper fertilizer application and better husbandry practices. Although the total area cropped is not expected to increase under the Project, individual crops may be expanded and the increases in production would, in these cases, come from increased acreage as well. The assumptions on yield increases and areas are shown in Annex III, Tables 1-8, and amounts of incremental production were calculated by assuming varying adoption rates by year and by area, found in Annex III, Tables 25 and 26. 6.13 The increased livestock products, meat and milk, would come from several sources. Improved disease control would reduce cattle mortality and for several years would result in a larger herd size, and therefore larger offtake, than in a without-project situation. This would exist until a plateau for the herd size is reached, created by either increased offtake or reduced fertility (or both) thereby maintaining a stable herd size. In addition, a gradual increase in weight is expected from improved disease control. The third source of incremental meat under the Project would be the farmer-to-farmer stallfeeder program. A small amount of milk (about 600,000 liters at full development) would be produced by the dairy program. 6.14 For Lilongwe Northeast, incremental maize production would be the main benefit, with smaller amounts of groundnuts and tobacco. Incremental meat and milk would also be produced. For Dedza Hills, incremental maize would also dominate the benefits, with potatoes, beans, wheat, groundnuts, and meat produced in small incremental amounts. Where local maize and maize/pulse mixtures have been replaced by other crops, the opportunity cost of this foregone production was incorporated into the analysis. Economic Prices 6.15 For incremental maize an average import-export parity price was used to calculate the economic farmgate price, representing foreign exchange savings (or earnings). This is because Malawi has imported maize in bad years, been self-sufficient or exported in better years, and although Government has a policy of self-sufficiency and a large strategic reserve capacity, it is difficult to predict whether incremental production from a certain Project area will serve as import-substitute or additional export. For tobacco and groundnuts export parity prices were used to calculate the economic farmgate price, representing increased foreign exchange earnings, with a premium for the groundnut confectionary nuts. The incremental wheat and milk expected from the Project were valued at import parity prices as both would be import substitutes. For milk, a calculation was made for imported reconstituted powdered milk as the closest substitute, For incremental beef, potatoes, and beans produced by the Project, local prices were used as these products are neither imported nor exported to/from Malawi. For meat, the official prices offered by Cold Storage Company, adjusted to 1983 prices, were used, and are close to prices prevailing at auctions.  - 42 - Costs 6.16 The incremental costs included in the economic analyses are the portions of each activity directly relevant to crops and livestock. The allocations for the two areas (LNE and Dedza Hills) are marginally different and are detailed in Annex IV. Fifty to 60% of extension and training costs were allocated to the costs of the crops component and 5% to livestock, reflecting the proportion of extension and training effort devoted to each. For credit, 90% of the costs of administration have been attributed to crops and 10% to livestock. All of the marketing costs were allocated to crops and all of the costs of veterinary services were allocated to the costs of the livestock component. For land husbandry and horticulture activities, 70% and 25%, respectively, were allocated to the crops component, with the remainder serving mainly non-quantifiable benefits such as research and demonstration. The remaining costs of extension and training were allocated to activities external to the Project and non-quantifiable activities such as working with farmers clubs, conservation work and training in nutrition, hygiene, or homecraft skills or staff training in technical and administrative matters. The costs of monitoring and evaluaton and project headquarters for each area, together with 50% of the costs of strengthening LADD headquarters for each area7/ were added together to arrive at costs of administration, and 50-60% of these costs were allocated to the crops component, 5% to livestock and the remainder being for Project activities with non-quantifiable benefits. (See para. 6,18 for proportion of total costs included in the analyses.) General Assumptions 6.17 The incremental costs and benefits of the Project components are in Annex IV and incorporate the following general assumptions: (a) all prices are in constant mid-year 1983 prices except for those of civil works, which are increased in real terms over the project period to reflect their rising real costs; (b) all costs include physical contingencies for capital items and are net of identifiable taxes; (c) the shadow price of foreign exchange is equal to the nominal price as the official exchange rate is considered to reflect its appropriate value; and (d) the conversion factor for unskilled and agricultural labor wage rates is 1.0 because although alternative employment opportunities and the value of alternative output varies throughout the area, on average the estimated opportunity cost is about MKO.50/day, which is the official minimum rural wage. 6.18 The estimated rates of return are as follows: 71/ With five project areas, the costs of LADD headquarters attributable to any one project area should be 20% of the total.  - 43 - Dedza Hills Lilongwe Northeast Total Project ERR Total Costs ERR Total Costs ERR Total Costs (percent) Crop 12 49 21 50 17 50 Livestock 13 6 21 18 17 12 Total 12 55 21 68 17 62 6.19 Sensitivity Analysis. The switching values for the entire Project show that benefits would have to drop by 17% or costs increase by 20% in order for the rate of return to drop below 10% or for the net present value to drop below zero at a 10% discount rate. This might happen if all crop yields or crop values were to be 31% below expectations or if the uptake rate by farmers were to be 12% of the population over the Project period rather than the estimated 20%; or if the real cost of construction were to increase drastically. These values are presented in the table below: Switching Values of Critical Variables Variable Switching Values 1/ Maize yield per hectare -24 Yield per hectare - all crops -31 Farmers' uptake rate -40 Fertilizer costs +52 Construction costs +161 1/ The percentage change which would reduce the net present value of the entire project to zero at a 10% opportunity cost of capital. If no benefits were generated by the conservation efforts of the Land Husbandry component or if all benefits were delayed by one year the overall rate of return would decrease to 15% or 12% respectively. Risks 6.20 Because the project areas are already densely populated and farmed, the risk of overcrowding exists, particularly for Dedza Hills. Unless land husbandry is successful in introducing conservation measures into Dedza Hills crop yields may not be realized and, in fact, deterioriation of the environment would probably cause a gradual decrease in crop yields. However, the estimates for incremental yield increases are conservative and considerably less than those obtained in nearby Lilongwe RDP. Further, maize yields in the hill areas were estimated to be 10% lower than maize yields on the plain. For these reasons, the Project has put emphasis on land husbandry, on adaptive research including agro-forestry, and on focussing extension efforts on farm systems and the importance of intercropping. In addition, unless relative prices change, the tendency to monocrop with hybrid maize (particularly in LNE) may be difficult to overcome. The Government has indicated its intention to  - 44 - increase both groundnut and tobacco prices relative to maize and its willingness to follow a pricing methodology that should further improve relative prices in future years (para. 3231). Finally, some risk exists for the marketing of potatoes, but because the number of farmers to receive improved seed for potatoes is limited, the incremental amounts are modest. VII. ASSURANCES AND RECOMMENDATIONS 7.01 During negotiations assurances were obtained that: (a) Project staff employed under the Development Budget would be transferred in an orderly fashion to the Revenue Budget over the duration of the Project (para. 3.28); (b) incremental staff for the Project would be obtained, to the extent possible, by transferring surplus staff from other project areas and Government would provide IDA with a staff redeployment schedule by September 30, 1983 (para. 4.07); (c) annual work plans would be submitted to IDA for comment after clearance by MOA (paras. 4.05 and 4.12); (d) procurement procedures as outlined in paragraph 4.22 would be followed; (e) Project accounts would be audited by the Auditor General, or by independent auditors whose experience and qualifications are acceptable to IDA, and the audited accounts together with the auditor's report would be submitted to IDA not later than six months following the end of the fiscal year (para. 4.24); (f) Government would maintain the position of NRDP coordinator at an administrative level of at least Deputy Secretary; (g) smallholder commodity prices for the years of the Project would be established at a level consistent with an agreed methodology and acceptable to IDA (para. 4.28); (h) that GOM would ensure the allocation of foreign exchange for the acquisition of adequate imported farm inputs (para. 4.29); and (i) reporting procedures as outlined in paragraph 4.27 would be followed; no later than six months after the closing date, Government would prepare and submit to IDA a Project Completion Report; 7.02 A condition of disbursement against civil works would be the preparation of a program acceptable to IDA, specifying the designs and methods of construction (paras. 4,01 and 4.23). 7.03 A condition of disbursement against expenditures for water facilities would be the submission to IDA of a proposal for maintenance and maintenance cost recovery (paras. 3.20, 4,21 and 4.23). 7.04 The proposed Project constitutes a suitable basis for an IDA credit of SDR million (US$10.6 million equivalent) on standard terms to the Government of Malawi.  MALAWI NRDP IV MINISTRY OF AGRICULTURE ORGANIZATION CHART (UNDER REVISION) CARL AR CCPrincipal Secretaryi Pri R DStation - Land Husbany - Di r D - aktin DvisionAResg)rt l MgmtFa M eUnitsmer l vision Ag. Ir Ag~~- Irigation e rnut (CAROCADOD Womn S ervice Asst Ass.t AssF,t Asst Health & Vet. CARO CARO CADO CADO Production Research HQ Specialist HQ Specialists Spe cSiael tisetis - National Research - Extension - Veterinary Stations - Land Husbandry - Dairy - Research Coordinators - Crops - Animal Husbandryr - Soil Survey - Credit - Livestock - Training - Marketing - Farm Management - Irrigation R Women's Services - Construction Dv. Aea -Specialist Staff, Specialist Staff, Vt ervisors Field Services l'evelopment Services Extension - Marketing . _ ~~~~~- Veterinary - '31nputs Specialists . _- Research - Forestry - Credit E A- Animal Husbandry - Fisheries - Extension [Supervisors - Credit - Roads r - Veterinary - Land Husbandry - Community Development - etc. - Buildings  MALAWI NRDP IV - Dedza Hills Implementation Schedule Pre- Py 1 PY 2 PY 3 PY 4 PY 5 investment 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 Civil Works Construction Housing Iflhf*Dilhhhh1 - Office Space *ua m mma* Livestock Dips mia lS lli DI_ Wells/Boreholes iil Market Sheds 111111 iI _ Vehicle Procurement al1 l Training P- M - Demonstrations and Trials aSg Ii li 'i I 31135 urn. usa I 51 _1 I ___ _ _____________________ ___ __ =_ _ 7z_ _ __ _ _____________________ i*i**IEEI Planning, Tendering, Contracting _ _N ~ Construction, Implementation rt - World Ban k-241t18 H  MALAWI NRDP IV - Lilongwe Northeast Implementation Schedule Pre- PYI PY2 PY3 PY4 PY5 investment 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 Civil Works Construction Housing 3113133UK Office Space 11311h131313 Livestock Dips I 1 111 1 335 _ 1 Wells/Boreholes - - _ll _11 Market Sheds IEhI3hI***hS3I 3Fl3 m3 m Vehicle Procurement 1311 1111 111M1 3m mm mm 3m3 woo Training I I go I I so I I I m - m - Demonstrations and Trials if let if l 3 l1_1 13 I loll a1t ' _1 o I I iISSEIIUPlanning, Tendering, Contracting _C Construction, Implementation World Bank-24119 L H  - 48 - ANNEX I Table 1 MALAWI Dedza Hills RDP Key Indicators Units PYI PY2 PY3 PY4 PY5 I. Credit Seasonal items Maize seed mt 10 25 55 80 100 Groundnut seed mt 2 4 5 7 9 Wheat seed mt 0.6 1.2 1.8 2.4 :3.0 Bean seed mt 4 10 22 32 40 Calcium ammonium nitrate fertilizer mt 48 121 266 387 484. 20:20:0 fertilizer mt 77 194 425 618 771 Dipterex insecticide packets 7,760 19,760 43,760 63,760 79,760 Feeder cattle No. 50 100 150 200 250 Value of seasonal credit (crops and livestock) MK 1,000 52 130 298 464 623 Value of medium-term credit MK 1,000 9 10 11 12 13 II. Incremental Credit Area Improved maize ha 310 790 1,750 2,550 3,190 Improved maize intercropped with beans ha 78 198 438 638 798 Groundnuts ha 22 44 66 88 11( Potatoes ha 20 40 60 80 10(0l Wheat ha 20 40 60 80 10( III. Civil Works constructed Staff houses No. 55 52 15 2 2 Offices sq. ft. 1,292 2,492 0 0 (0 Training centers (included with offices) No. ADMARC input shed with office No. 1 1 0 0 ( Livestock dips - new No. 0 0 0 0 0 - renovated No. 9 0 0 0 ( Slaughter slabs No. 0 0 2 0 0 Poultry distribution house No. 0 0 3 0 0 Water points No. 0 0 0 298 298 Boreholes rehabilitation No. 0 0 0 23 24 Wells rehabilitation No. 0 0 0 50 50 Water supply, maintenance stores sq. ft. 0 0 0 300 300 August 18, 1982  - 49 - ANNEX I Table 2 MALAWI Lilongwe North-East RDP Key Indicators Units Pyl PY2 PY3 PY4 PY5 I. Credit Seasonal items Maize seed mt 6 20 40 53 65 Groundnut seed mt 12 24 36 48 60 Calcium ammonium nitrate fertilizer mt 47 133 261 347 426 20:20:0 fertilizer mt 73 199 382 508 624 Dipterex insecticide packets 5,280 15,840 32,160 42,720 52,320 Daconil fungicide ha 11 22 33 44 55 Tobacco nursery inputs ha 80 160 240 320 400 Feeder cattle No. 100 200 300 400 500 Dairy cows 2-cow units 35 70 105 140 175 Value of seasonal credit (crop and livestock) MK 1,000 60 150 296 424 552 Value of medium-term credit MK 1,000 28 30 34 38 42 II. Incremental credit area Improved maize ha 264 792 1,608 2,136 2,616 Groundnuts ha 150 300 450 600 750 Tobacco ha 80 160 240 320 400 III. Civil Works constructed Staff houses No. 65 47 8 0 0 Offices sq. ft. 1,863 1,863 0 0 0 Training centers (included with offices) No. - ADMARC input shed with office No. 1 1 0 0 0 Livestock dips - new No. 0 3 0 0 0 - renovated No. 18 0 0 0 0 Slaughter slabs No. 0 0 5 0 0 Poultry distribution house No. 0 0 3 0 0 Water points No. 0 441 441 0 0 Boreholes rehabilitation No. 0 20 20 0 0 Wells rehabilitation No. 0 30 30 0 0 Water supply, maintenance stores sq. ft. 0 450 300 0 0 July 12, 1982  ANNEX I - 50 - Table 3 MALAWI DOWA WEST RURAL DEVELOPMENT PROJECT ESTIMATED SCHEDULE OF DISBURSEMENTS IDA Quarter Quarterly Cumulative Fiscal Year Ending Disbursements Disbursements (US$ 000) 1982/83 June 30, 1983 200 200 1983/84 September 30, 1983 200 400 December 31, 1983 400 800 March 31, 1984 600 1,400 June 30, 1984 700 2,100 1984/85 September 30, 1984 700 2,800 December 31, 1984 700 3,500 March 31, 1985 700 4,200 June 30, 1985 600 4,800 1985/86 September 30, 1985 600 5,400 December 31, 1985 600 6,000 March 31, 1986 500 6,500 June 30, 1986 500 7,000 1986/87 September 30, 1986 500 7,500 December 31, 1986 500 8,000 March 31, 1987 400 8,400 June 30, 1987 400 8,800 1987/88 September 30, 1987 400 9,200 December 31, 1987 300 9,500 March 31, 1988 300 9,800 June 30, 1988 300 10,100 1988/89 September 30, 1988 200 10,300 December 31, 1988 200 10,500 March 31, 1989 100 10,600  -51 - ANNEX II TaDle 1 MALAWI NRDP IV - DEDZA HILLS RURAL DEVELOPMENT PROJECT PROJECT COST SUMMARY (MK '000) (US$ '000) Z of -------------------------- --------------------------Foreign % of Total Local Foreign Total Local Foreign Total Exchange Base Costs EXTENSION SERVICES 616.10 445.22 1,061.32 574.13 414.89 989.03 41.9 22.1 TRAINING 112.89 175.53 288,43 105.20 163.58 268.78 60.9 6.0 RESEARCH 100.83 113.66 214.49 93.96 105.92 199.88 53.0 4.5 HORTICULTURE 30.45 26.96 57.42 28.38 25,13 53.51 47.0 1.2 LAND HUSBANDRY 133.58 134.01 267.59 124.48 124.88 249.36 50.1 5.6 LIVESTOCK AND VETERINARY 45.94 102.86 148.79 42.81 95.85 138.66 69.1 3,1 CREDIT SUPERVISION 57.01 36,39 93.40 53.13 33.9l 87.04 39.0 1.9 CREDIT FUND 143.81 323.04 466.84 134.01 301.03 435.04 69.2 9.7 MARKETING 79.74 103.05 182.79 74.31 96.03 170.34 56.4 3.8 WATER SUPPLIES 336,722 316.77 653.49 313.78 295.19 608.97 48,5 13.6 MONITORING AND EVALUATION 58.