Malawi - National Rural Development Program (NRDP) review

Document of i_ The World Bank FOR OFFICIAL USE ONLY Report No. 3895-MAI MALAWI NATIONAL RURAL DEVELOPMENT PROGRAM (NRDP) REVIEW June 30, 1982 Eastern Africa Projects FILE COPY Southern Agriculture 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.  MALAWI NATIONAL RURAL DEVELOPMENT PROGRAM (NRDP) REVIEW Currency Equivalents Currency Unit Malawi Kwacha (MK) US$ 1.00 =MK 0.90 MK 1.00 IJS$ 1.11 Abbreviations ADD - Agricultural Development Division ADMARC - Agricultural Development and Marketing Corporation CADO - Chief Agricultural Development Officer DA - Development Area DAD - Department of Agricultural Development DLVW - Department of Lands, Valuation and Water DPS - Deputy Permanent Secretary DRIMP - District Roads Improvement and Maintenance Program EDF - European Development Fund EPA - Extension Planning Area FHA - Farm Home Assistant IDA - International Development Association ISNAR - International Service for National Agricultural Research KRDP - Karenga Rural Development Project LLDP - Lilongwe Land Development Project LRDP - Lakeshore Rural Development Project MANR - Ministry of Agriculture and Natural Resources MFNR - Ministry of Forestry and Natural Resources MHC - Malawi Housing Corporation MOA - Ministry of Agriculture MOE - Ministry of Education MOF - Ministry of Finance MOH - Ministry of Health MOWS - Ministry of Works and Supplies MU - Management Unit (in each ADD) NRDP - National Rural Development Program NSO - National Statistical Office PCR - Project Completion Report PM - Program Manager OPC - Office of the President and Cabinet RDP - Rural Development Project SAL - Structural Adjustment Loan SVADP - Shire Valley Agricultural Development Project USAID - United States Agency for International Development Fiscal Year April 1 - March 31  FOR OFFICIAL USE ONLY MALAWI NATIONAL RURAL DEVELOPMENT PROGRAM REVIEW Table of Contents Page No. Summary and Conclusions ..................... I. .. ..................... v I. The National Rural Development Program in Perspective ......... 1 A. Structure and Performance of the Agricultural Sector 1 B. Brief History of Rural Development in Malawi. 3 C. The National Rural Development Program (NRDP). 8 II. NRDP - Performance, Issues and Recommendations .10 A. Multi-Sectoral, Multi-Component Approach . . .10 B. Government Structure .... 16 C. Investment Costs - Housing ............................... . 23 D. Manpower Availability and Requirements . . .27 E. Recurrent Costs . . ...28 III. Comments and Recommendations for Specific NRDP Activities 34 A. Extension... 34 B. Research ....40 C. Livestock ............................................... 42 D. Training ....42 E. Credit ....44 F. Monitoring, Evaluation and Reporting . . . 45 IV. Conclusions .45 A. Directions for NRDP .45 B. Future Preparation/Appraisal and IDA Lending For NRDP ..... 46 This report was prepared by L. Effron (Economist, Coordinator), A. Bose (Financial Analyst), A. Nyberg (Economist), M. Cecere (Education Specialist), all IDA, E. Boerema (Agriculturalist), RMEA, and Consultants, R. Lanier (Architect), A. Shaba (MOA Agricultural Economist), F. Kangaude (MOA Agriculturalist). 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.  - iii - MALAWI NATIONAL RURAL DEVELOPMENT PROGRAM REVIEW List of Tables in the Text Table 1 - Housing Construction Costs Table 2 - MOA Revenue Account - Actual and Required Expenditure, 1977/78 - 1985/85 Table 3 - Projected Gap between MOA Requirements and Available Funds on Revenue Account, 1982/83 - 1985/86 Table 4 - Agricultural Extension Staff and Supervisory/Technical Staff by ADD, 1981/82 Table 5 - Extension Service Cost per Section TA Table 6 - Research Efforts in Relation to Crop Importance  - iv - MALAWI NATIONAL RURAL DEVELOPMENT PROGRAM REVIEW Annexes Annex 1 - Comparative Costs of Rural Development Projects in Malawi Annex 2 - Terms of Reference for the NRDP Review Annex 3 - "Non-Agricultural" Components of Rural Development Projects Annex 4 - Organization and Management of MOA Annex 5 - Housing and Construction Costs in NRDP Annex 6 - Manpower Availability and Requirements Annex 7 - Recurrent Costs of NRDP Annex 8 - Extension Services Annex 9 - Agricultural Research Annex 10 - Livestock Annex 11 - Training Annex 12 - Monitoring, Evaluation, and Reporting MAP IBRD-15831R   v - MALAWI NATIONAL RURAL DEVELOPMENT PROGRAM REVIEW Summary and Conclusions (i) Investment in agricultural development by the Government of Malawi has taken the form, over the last nineteen years, of area development projects aimed at the smallholder population. These rural development projects (RDPs) were multi-component, involving several sectors (agriculture, livestock, forestry, health, water supplies, roads, and marketing facilities), intensive in terms of staffing and levels of investment per benefitting farm family and for a time, managed by a cardre of largely expatriate personnel, in management unit distinct from government civil service. These characterisitics - multi-component, intensive, and discrete and separate from governnent - reflected both a particular philosophical approach and a recognition of the practial problems of rural development. (ii) In general, the record of RDPs in Malawi is good. On infrastructure, the targets were usually met, sometimes exceeded, construction was completed in a timely manner, often within cost estimates, and were of a high standard. The handover of infrastructure for operation and maintenance to parent ministries (roads, health, water supplies) created problems, however, and in some cases roads reverted to tracks or bush, health facilities were underutilized, and water supplies were not maintained. On crop production, the record is mixed but where data exist several themes emerge: when farmers adopted extension recommendations and packages, yields were as high as or exceeded appraisal estimates; where farmers did not adopt packages it was usually because they preferred to plant subsistence crops to ensure a food supply or because the crop packages were financially relatively unattractive. (iii) The National Rural Development Program (NRDP) was launched in 1978 with the main objective of spreading the benefits of rural development more rapidly. All RDPs were to come under the umbrella of NRDP and while maintaining the multi-component nature of investment, NRDP was to be less intensive in terms of staffing and investment and to be integrated into government's civil service. The agricultural department divided the country into ecological zones and reorganized its staff to reflect this new structure. While NRDP has been as intensive in investment as previous RDPs it has been successful in integrating RDP management into government and in reorganizing the agricultural department.  - vi - (iv) After 3 years of NRDP, it seemed appropriate to review progress and recommend improvements. The purpose of the review was to examine how investment in smallholder agriculture could be made more effective and efficient. The review was done in recognition that larger issues, such as proper pricing policy and price incentives and adequate input delivery and crop marketing facilities also affect the success of NRDP. The problems examined in the review were: (a) the multi-component nature of NRDPs; organization, (b) management and staffing for NRDP; high investment, (c) particularly housing, costs; (d) resources available for NRDP, both human (staffing) and (e) financial;content of specific activities: agricultural extension, research, livestock, and monitoring, evaluation, and reporting. (v) While the multi-sectoral approach in RDPs has been based on the belief of the complementarity of investments, this has been more an ideal than a reality. Because of the difficulties encountered in the operation and maintenance of facilities or structures built by agriculture on behalf of other departments or ministries, the investments were not as effective as intended. Several departments and ministries have now developed or are planning a national program for rural investment which has the advantage of establishing national priorities and of planning investments with regard to staffing and resource availability. Thus, forestry has a wood energy project, which takes into the account the areas of greatest wood deficits; the Ministry of Works has a rural roads program which is establishing a local maintenance capability; Ministry of Health is planning a national program to take account of available staffing. These national programs are properly the responsibility of their respective ministries, and allow RDPs to focus on directly-related agricultural investments. This trend could be reinforced by encouraging the Veterinary Department to consider a national plan for animal disease control and to consider policy issues such as cattle tax and marketing incentives in this framework. (vi) Closer integration of agricultural activities under NRDP, including agricultural extension, land husbandry, agricultural research, and animal husbandry, would be fostered by a reorganization of the research and veterinary departments under the Ministry of Agriculture to coincide with the new organization of the Department of Agriculture. Several other measures should be taken to reduce staff turnover, improve morale, and ensure selection of high-quality key staff. Now that other ministries are taking charge of rural investments, coordination will be required at the planning as well as implementing stage and this should be  - vii - done by strengthening the senior, non-technical ministry, to take a more active role in planning various national programs. During implementation, the Ministry of Agriculture should take a leading role in resolving problems among ministries. (vii) After reviewing the costs of housing construction under NRDP, the mission found that although some cost savings could be realized by using private contractors rather than direct labor, these savings would depend on location of the construction - in remote rural areas private contractors are not available - and is probably a temporary solution, which depends on market conditions. The greatest scope for cost savings is in design changes which replace imported items with local ones but which do not compromise the housing standards. Technical assistance should be provided to develop lower cost designs which should be adopted as standard NRDP housing. (viii) On the resources available to Malawi, the mission found that, in spite of vacancies, the manpower available for ongoing and future NRDP phases seems adequate at the field extension worker and professional levels; the constraint would be at the intermediate (TO) or supervisory levels. To date, the vacancy rate for TOs is about 25 percent and expected to decrease gradually, even with an increase in NRDP activity, over the next eight years or so. Even these vacancies should not impede NRDP implementation, however, provided the ministry takes steps to upgrade selected staff and to shift TOs from heavily staffed areas to those in greater need. (ix) The estimates of recurrent resources required from Malawi's budget to finance ongoing NRDP projects compared to estimates of likely resources available for this show large deficits: if agriculture continues to receive the same proportion of government's recurrent budget as in the past, by 1985/86 the resources will be roughly half of what is required. This means that ongoing projects must be scaled down, staff transferred to new development areas, and costs cut on vehicle replacements, while treasury must increase the share of recurrent resources to agriculture. Future NRDP phases should keep staffing ratios and vehicles to minimum practical levels. (x) It is recommended that in the future NRDP focus investments on agricultural extension and research, credit for inputs, land husbandry, water supplies, animal husbandry, marketing facilities, farmer and staff training and evaluation. On agricultural extension, the basic framework exists for an effective system but could be further strengthened by focusing on packages for subsistence crops in the context of a farming systems approach. In addition, regular visits on a fixed time/fixed place would improve farmer contact.  - viii - Finally, annual evaluations of extension recommendations should be done jointly by the extension, research, land husbandry, animal husbandry, and evaluation staff. Agricultural research should have a complete review of its national program, with particular attention to rationalizing the use of its research stations and of establishing priorities for crop research which relate to the importance of the crops in Malawi. The research department should also be restructured to focus more effort on verification trials and smallholders' particular constraints. Malawi government has agreed that the International Service for National Agricultural Research would carry out this review, which might then lead to proposals for funding a revamped research organization. The training network is well-established in NRDP and functions relatively efficiently. Costs per manday of training are not high compared to neighbouring countries and capacity utilization of the centers is fairly good. Residential training could be improved by having shorter courses, which would allow greater coverage, by regular post-course evaluation and by providing in-service training for trainers. Demonstration plots at day and residential centers should be comparable to what exists on farmers fields. Costs of training should be somewhat defrayed by requiring farmers to contribute in cash or in kind (food, transport) for specialized courses on for example, ox-training. A thorough inventory and evaluation of the training program should also be done. Credit has worked well in Malawi, but future NRDP projects could be improved by consolidating some of the credit operations into a single fund. A good system for monitoring and evaluation of NRDP has been set up but could be strengthened by transferring responsibility for surveys and data collection to the National Statistics Office and encouraging evaluation officers and program managers to focus on the interpretation and use of the information. Reporting, which is very detailed and frequent, should be trimmed considerably to reflect only essential information. (xi) For the future, NRDP should concentrate on improving smallholder agricultural production by developing and advising on acceptable technical packages for subsistence crops which can be integrated into the farming systems. If IDA and the government can agree on the basic design and content of future NRDP phases, then IDA should consider providing sector loans to Malawi to finance several rural development projects. An abbreviated preparation/appraisal effort could be done, focusing on the particular research and crop packages appropriate for the area under consideration and on calculation of the economic viability. Such sector lending would serve the interests of both Malawi and IDA.  MALAWI National Rural Development Program Review Chapter One: The National Rural Development Program in Perspective A. Structure and Performance of the Agricultural Sector 1.01 Agriculture is the most important sector in Malawi: it employs about 85 percent of the population, contributes about 43 percent of GDP and accounts for almost 90 percent of export earnings. These export earn- ings come from tobacco (63 %), tea (20 %), sugar (8 %), groundnuts (3 %) and other crops (6 %).1/ 1.02 Malawi's agricultural production derives from two subsectors, smallholder agriculture and estate agriculture. The smallholder subsector accounts for over 85 percent of all agricultural production, meets the country's demand for food staples (maize, beans, groundnuts, sweet potatoes and rice), and provides some export surplus. However, the bulk of the country's agricultural export come from the estate subsector which has functioned as a principal earner of foreign exchange and an engine of growth for Malawi's development. The estate subsector contributes 15 per- cent of total agricultural production but accounts for nearly 70 percent of all agricuiltural exports. Estate production is mainly centered on flue-cured tobacco, burley tobacco, tea and sugar. The Government's aims in agriculture since independence have been to maintain self-sufficiency in food staples, expand agricultural exports significantly and improve rural incomes. 1.03 Evaluating the past performance of the smallholder subsector is somewhat difficult because of the shaky data base on yields and produc- tion. It is estimated, however, that smallholder output has been increasing at about 3 percent per annum in real terms in recent years. This performance is less than satisfactory considering the sizable public investments in smallholder agriculture since the late sixties. Better performance by the smallholder subsector has been constrained by (a) the transfer of some 140,000 persons and 470,000 ha of the best cultivable land to the estate subsector; (b) the limitation by Government regulation of smallholder participation in the production of high-value, high-cost agricultural export crops such as flue-cured and burley tobacco and tea; (c) the limited coverage of past agricultural projects; and (d) low incen- tives provided by low producer prices to smallholders. 1.04 The estate subsector is very efficient and has been growing at about 17 percent in real term since 1968. It is currently heavily concen- trated on two commodities: tea and tobacco, accounting for about 80 percent of all exports from the estate subsector (tobacco accounts for approximately 50 percent.) The future real growth rate of the estate sub- sector is projected to decrease to four or five percent per annum, and it cannot be expected to generate the same growth of employment and foreign exchange as in the past. If the estate subsector is to maintain its role as the country's principal foreign exchange earner, it must diversify into other crops. 1/ All figures in this paragraph are from 1978.  - 2 - 1.05 The greatest constraint to agricultural growth in Malawi is a limited supply of arable land. In the past decade, increased food requirements for the country's growing population have been met through expansion in cultivated areas. As only 38 percent of the total land area is considered suitable for cultivation under present technological limita- tions, however, and since 36 percent is already under cultivation, the possibilities for further expansion with present farming practices to meet food requirements is limited. Even if the Government holds the estate acreage constant at the 1978 level (13 percent of arable land), arable land for smallholders will be exhausted by about 1982. The depletion of arable land would force smallholders to extend their cultivation into non-arable areas or to shorten the fallow period. Hence, without additional investments in land, the average yields would fall. 1.06 Malawi's agricultural development could also be constrained in the future by the high cost of transport and by shortages of fuelwood. Because of its position as a land-locked country, Malawi has always had relatively high transport costs. Given the probability of further increases in the real price of fuels, these transport costs are likely to escalate in the future. In recent years, the Malawian economy has been adversely affected by severe transport disruption, delays, and damage in transit through Mozambican ports and trade routes because of the political and economic situation in that country. Second, expansion of fire-cured and flue-cured tobacco is constrained by the shortage of fuel- wood required in their production processes. Malawi's consumption of wood is much greater than its production of wood, and as the depletion of tra- ditional forests continues, the price of wood will increase. The rising price of wood will make growing fire- and flue- cured tobacco -- three-fourths of Malawian tobacco production -- less attractive in the future. This could mean the loss of an important source of foreign exchange.2/ 1.07 In the past decade the Government has been directing sizable por- tions of public investments into the smallholder subsector. The estate subsector -- which is viewed as the principal source of the country's foreign exchange and a generator of wage employment -- has been left largely in the hands of the private sector and Government has limited its role to providing a favorable investment climate. Beginning in 1968/69, integrated rural development programs were established in four specific areas. These four comprehensive programs are concentrated in the more densely populated areas and presently affect one million persons or 20 percent of Malawi's population. In the mid-1970's, it was realized that such intensive and costly capital investments as provided under the major projects could not be replicated in the rest of the country within a rea- sonably short time. A rethinking of the strategy resulted in the concept of the NRDP which aims to extend and consolidate various development pro- grams over the entire country within 20 years. It is designed to increase the level of smallholder production through the provision of agricultural inputs and farm services. Under NRDP, cultivation of new land is dis- 2/ Malawi's general economic development will also be constrained by over-reliance on tobacco for export earnings, which makes the country vulnerable to international price fluctuations. Malawi should diversify its agricultural base.  - 3 - couraged and emphasis is placed on improving the productivity of already cultivated areas. B. Brief History of Rural Development in Malawi The Design of Four Area Projects 1.08 Between 1968 and 1972 four area rural development projects (RDP) were started. Lilongwe Land Development Project (LLDP, 1968), Lakeshore Rural Development Project (LRDP, 1968), Shire Valley Agricultural Develop- ment Project (SVADP, 1969), and Karonga Rural Development Project (KRDP, 1972). The Lakeshore project was financed initially with German aid and now by the European Development Fund (EDF); the others with IDA funds.3/ 1.09 Each project is now in its third or fourth phase and all have had similar basic organization and content. They were all multi-component: they included agricultural extension and land husbandry activities, farmer and staff training, credit for agricultural inputs, animal disease control, sometimes dairy and stallfeeder programs, forestry activities, crop marketing facilities, and research. Although the projects were carried out by the Department of Agriculture in the (then) Ministry of Economic Affairs, subsequently in the Ministry of Agriculture and Natural Resources (MANR) and now in the Ministry of Agriculture (MOA), the projects included financing for activities that would have otherwise been the responsibility of other ministries. Rural roads were constructed, sometimes by the Ministry of Works and Supplies (MOWS) and sometimes by construction teams under the Department of Agriculture; health centers were also built by agricultural staff. Rural water supplies were installed by different departments, including MOWS, the Geological Survey Department in MANR, and now the Department of Lands, Valuation and Water (DLVW) in the Office of the President and Cabinet (OPC).4/ Staff and sup- porting infrastructure from other ministries such as community development were sometimes financed under these rural development projects. 1.10 Arrangenents varied somewhat for the operation and maintenance of the infrastructure, but generally the department or ministry responsible was expected to take over the buildings and civil works. Thus, MOWS was to maintain roads, even when they were built by agricultural staff; Ministry of Health (MOH) was to staff and operate the health centers; the Agricultural Development and Marketing Corporation (ADMARC), the official statutory marketing organization, was to take over (and sometimes con- struct) the marketing facilities. Financial responsibility for maintain- 3/ Except the second phase of the Karonga project, which was a third window loan. 4/ DLVW has absorbed the geological survey departrment from MANR.  - 4 - ing rural water supplies rests with the Ministry of Local Government, who in turn called on the geological surveys department of MANR (now in DLVW) to provide the service. 1.11 These projects were intensive in terms of the numbers of activi- ties and services financed, the amounts of financing, and the staffing levels. Investment per benefitting farm family ranged from IJS$315 (in 1967 terms - for the Shire Valley) 5/ to US$2,700 (in 1978 terms - for Karonga-Chitipa).6/ The projects' aims were to bring about increased standards of living and incomes to the farmers through concentrated efforts and the staffing ratios reflected this intention: the Shire Valley first phase project, for example, aimed at an extension worker to farm family ratio of 1:10 in the early yeras, to be reduced to 1:30 in the later years; the Karonga project planned for ratios ranging from 1:30 to 1:100 over a five year period. These ratios are very intense, particularly compared to ratios in non-project areas of as high as 1:1,500-2,000. 1.12 Finally, the four RDPs were all discrete entities as far as government was concerned. Project management and staff formed distinct units within MANR, separate from the organization of the rest of MANR's operations; in the first ten years of the projects, higher levels of pro- ject management were staffed almost without exception by expatriates. Only when a project was in its latter phase and was to be absorbed into Malawi's budget (as in the case of LLDP) was consideration given to reabsorbing project management into the regular MANR structure. 1.13 The characteristics of these RDPs -- multi-component, intensive, discrete and separate from government -- reflected both a certain philosophical approach to rural development and a recognition of the prac- tical problems. At the time, emphasis on integrated projects was popular, where poverty would be attacked and constraints to development removed on many fronts (a sort of "big push" on a small scale). The benefits from one sector would spill over and provide benefits to another: thus, for example, providing rural water supplies would improve health and reduce the time fetching water, more labor would be available for agricultural production at the same time that provision of extension advice and credit would increase yields. Organizing all these investments under one manage- ment unit was not only necessary to ensure proper coordination but was also dictated by the inability of other ministries and departments to plan and carry out their own projects in rural areas. At the time, MOWS was concerned with main roads, which reflected its own priorities for invest- 5/ In 1978 terms, US$695. 6/ The investments per farm family are not entirely comparable across projects, as some projects include lake transport and other large infrastructure items while others focus more heavily on agricultural investments.  ment, and had no capacity to plan, build or supervise construction of rural roads; similarly for the MOHand the Departments of Forestry, Veter- inary Services, and Geology (water), who were not prepard to establish national priorities and plan investments for rural areas. If integrated rural development was to go ahead, the investments had to be managed by a discrete unit staffed by people with sufficient qualifications and experience. 1.14 Over time, problems emerged as a result of the design and approach of the area RDPs. Some of these were addressed in the design of NRDP started in 1978. NRDP is discussed below (paras. 1.27-1.30). Impact of the Rural Development Projects7/ Infrastructure 1.15 In general, the record of the four RDPs shows an impressive performance on infrastructure - the physical targets were usually met, sometimes exceeded, construction was completely in a timely manner, often within the cost estimates, and were of a high standard.8/ 1.16 Some rural roads were built or improved by MANR/MOA construction units under the project management. Occasionally, MOWS built or supervised construction of roads. Over the years, hundreds of kilometers of access, feeder, district, secondary, and main roads were built or upgraded to good all weather standards, although the feeder and district roads were often narrower than the MOWS standard. Culverts, bridges, and embankments were also built. Under some phases, targets for miles of road construction were exceeded by as much as 50%. Maintenance of roads proved to be a problem, however. In some cases, MOWS refused to take over roads which were not built to its standards; in some areas it did take over the roads formally but hadn't the funds or staff to maintain them properly. 7/ Details of the IDA-financed rural development projects can be found in project completion reports and project performance audit reports (PPAR). For LLDP, see PPAR NO. 751-MAI (Credit 113-MAI) of May 23, 1975, PPAR No. 1597-MAI (Credit 244-MAI) of May 17, 1977, and PPAR No. 3414-MAI (Credit 550-MAI) of April 7, 1981. For KRDP, see PPAR No. 2576-MAI (Credit 282-MAI) of June 29, 1979 and for SVADP, see PPAR No. 895-MAI of October 22, 1975 and PPAR No. 2593-MAI (Credit 363-MAI) of June 29, 1979. The following discussion is based on information available in Washington; few data exist on LRDP, although some of the generalizations apply to this project as well. 8/ A few notable exceptions to these generalizations included the irrigation works in Karonga, costing four times the estimate, and the lakeshore transport component in the Karonga project, with large cost overruns, and both items having technical difficulties and delays in completion.  -6- As a result the roads built under the project, except where they were maintained by the MANR/MOA construction units themselves, tended to deteriorate and, in some cases, to disappear. 1.17 For health services, the projects built a range of facilities - maternity wards, health posts, sub centers, primary health centers, rehabilitation and expansion of hospitals - and provided funds for staff- ing new and existing facilities, for drugs and other supplies, for vehicles and their operating costs, and for specific programs to fight malaria and bilharzia. Aside from the construction of civil works, which usually met or exceeded appraisal targets, the health activities were the responsibility of MOH and in every project the health services suffered somewhat from understaffing, underutilization of services and/or inade- quate supplies of drugs. In some cases MOH appointed a Medical Officer to the project area only in the last year of the project and no decisions were made on using the funds provided for health facilities until the medical officer's arrival. 1.18 Boreholes were usually installed by the Geological Surveys Depart- ment in MANR; again appraisal targets were met or exceeded, but problems of maintenance became important and in some areas significant numbers of the installed boreholes were no longer operating by the fifth year of the project, either because of poor maintenance or because they had dried up. All installation and maintenance were directed from the capital in Lilongwe, with the Geological Surveys Department installing several bore- holes in one area, then moving on to another to install several more. Overheads were thus very high and because these were allocated to the installation cost, the unit cost of boreholes was also high and rose from MK2,500 (US$3,205) in 1976 to MK5,000 (US$5,555) in 1981. Local people were only minimally involved and although local councils were supposed to pay for maintenance from their budgets, they usually had insufficient funds. 1.19 ADMARC built and operated its input/crop marketing sheds and the only problem with meeting physical targets occurred when ADMARC was responsible for financing the markets (rather than Government) and didn't have enough funds to carry out its proposed program. 1.20 The training facilities - day and residential training centers - were built to good standards and have generally achieved their targets in terms of numbers (para. 3.28). 1.21 One of the most important achievements of the RDPs is the human infrastructure they have built up: throughout Malawi are Program Managers, Senior Extension Officers, Credit and Land Husbandry Officers, among others, who have developed qualifications and experience on the four major projects. Most senior positions, in MOA, in fact, are filled by Malawians and while posts in some subject areas are still filled by expatriates (finance, evaluation, computer work, mechanical workshop and maintenance), this investment in university and on-the-job training now has and will continue to have benefits that are both unquantifiable and beyond the immediate scope of each project.  -7- Production 1.22 The record of RDPs impact on production is mixed. With a few exceptions, appraisl targets have not been realized. In some cases yields have not been as high as projected, but more often it is the uptake rate of input packages (acreage) that has fallen short of expections. Where the data are clear, certain patterns do emerge: when farmers adopted pack- ages and extension's recommendations, yields were as high as or exceeded appraisal estimates. Where farmers did not take credit packages, it was usually for those crops with relatively unattractive returns or it was because farmers preferred to plant subsistence (and sometimes drought- resistant) food crops to ensure the family's food supply rather than planting cash crops. 1.23 Although much has been written on the impact of the Lilongwe pro- jects, the data on crop production are difficult to interpret, particularly for the first two phases. It appears that maize and ground- nut yields and incremental production did not reach the levels projected at appraisal. Reports on these two phases conclude that the quantifiable benefits are at best difficult to establish and the reports do not estimate rates of return. But the project completion report (PCR) on the third phase calculates, on the basis of assumptions on cropping patterns and without - project yields, a combined rate of return for the three phases of 25%. This PCR also has a breakdown of maize yields achieved in phase three for hybrid, synthetic, composite, and local maize which shows that maize yields did reach the levels expected. The failure in the first two phases to reach an average projected maize yield may be due to a slow uptake of the higher yielding hybrid and composite varieties; individual yields may have reached projected levels, but both the maize mix and the adoption rates projected may have been overly optimistic. 1.24 Two positive indicators of the first two phases of Lilongwe projects' irnpact on production are the use of credit and the differences in yields between "developed" and "undeveloped" areas. Credit use for fertilizer exceeded appraisal estimates (measured in quantitative terms), although cash sales were short of projections; the total fertilizer use in Phase II varied between 50-120% of appraisal targets, usually at 85% or better. The data on average maize yields for developed areas, which covered areas with project activity and so included non-adopting farmers, are on average 30% higher than for undeveloped areas; groundnuts yields in developed areas were not significantly different from undeveloped areas (the yields varied from 10% above to 10% below those in undeveloped areas). In addition, comparisons between developed and underdeveloped areas understates the project's impact because they cannot take into account the spillover effects from the project. The report on Phase III, which ran from 1976 through 1980, found more positive results on yields for maize, groundnuts and tobacco. 1.25 Of the three RDPs, Lilongwe is the most indefinite in terms of its impact on production. Both Karonga and Shire Valley projects show more positive results. Phase I in Karonga showed yields (for the last two years) significantly higher for farmers in the project schemes than for farmers in non-scheme areas. Again yields were short of appraisal  estimates, but incremental production was near or greater than appraisal estimates for the financially more attractive crops - rainfed rice and hybrid maize. For phase II of Karonga-Chitipa, yield informaton is not yet available, but credit uptake for rice and maize seed in Karonga and hybrid maize seed in Chitipa exceeded appraisal targets and estimated acreage in both districts under improved cultivation were greater than appraisal estimates. The first phase in Shire Valley was also successful in increasing yields, which were near appraisal estimates, as was incremental produdction. In Phase II a combination of severe drought and low cotton prices resulted in disappointing cotton production results, although the project was responsible for introducing guar beans which were very popular with smallholders because they were financially attractive and beneficial to the country because they were economically lucrative. Phase III has been no more successful in promoting cotton, although estimated total maize production has been near appraisal estimates and guar bean production has continued to increase. The performance on cotton is again attributed mainly to weather and prices. 1.26 What emerges from reviewing crop production data in the RDPs is that technical packages do exist for a number of cash crops (cotton, tobacco, hybrid maize, guar beans, rice)9/ and that these packages are adopted by farmers if (a) they assure themselves first of a food supply, (b) the returns are sufficient to justify the investment, and (c) they have adequate and timely supply of inputs and access to markets for the output. Successful implementation of the RDPs depends, therefore, on the relative prices of crops and on the marketing system. Technical packages are also needed for food crops used as subsistence so that yields can be increased and land can be released for other income-earning crops (see paras. 3.10 - 3.12 below). C. The National Rural Development Program (NRDP) 1.27 In 1978 the NRDP was launched with the primary goal of spreading the benefits of rural development more rapidly. Where the earlier RDPs had been intensive (para. 1.11), this would be more extensive in terms of staffing and presumably in terms of investment levels. Secondly, NRDP intended to integrate project management into government structure (para. 1.12), which in turn required a reorganization of the Department of Agricultural Department (DAD). NRDP continued to finance multi-component, cross-ministerial activities (para. 1.09). All RDPs, whether financed by IDA or by other donors,10! became part of this NRDP and were, in principle, to follow the same pattern. 9/ Some of which, like rice, are used both as subsistence and for cash. 10/ Other rural development projects are being financed by the African Development Bank (ADB), the Germans (KFW), EDF and the British (ODA).  - 9 - 1.28 While NRDP I (the first IDA-financed phase of NRDP, cofinanced with the Canadian International Development Agency, CIDA) was less inten- sive in terms of field staff and recommended an eventual average ratio of one extension worker per 750 farm families throughout the country, in terms of investment per benefitting farm family it was one of the more expensive rural development projects (see Annex 1). The relatively high investment cost was due partly to the build up of central services and management units and partly to the increased cost in real terms of housing and vehicle operating costs. 1.29 The organizational changes in MOA have led to considerable disrup- tions: problems in creation of posts, constant staff turnover, frequent vacancies, and consequent lack of management continuity or strong direc- tion have plagued the first three years of NRDP's implementation. This was to be expected from a major reorganization and, in spite of these problems, there has been progress at the field level in some areas in extending advice and seasonal credit to farmers and in establishing infra- structure (roads, houses). 1.30 NRDP was conceived and implemented in the context of prevailing assumptions. First, it was assumed that technical package were available, although it was recognized that research needed additional input. The United States Agency for Internaional Development (USAID) agreed to pro- vide this and now has a project underway. Second, it was recognized that pricing was important in providing sufficient incentives to farmers. Government agreed at NRDP negotiations to take a number of measures to ensure that pricing policies were clearly defined and examined for their likely impact on NRDP production targets and that these policies would be discussed regularly with IDA. Finally, NRDP's design was based on the belief that the complementarity of simultaneous investments in many sectors was an important aspect to developing agriculture. At the same time, however, those who designed NRDP were concerned with Malawi's absorptive capacity for investment in rural development and with certain institutional questions, so that under NRDP studies were to be carried out to examine (a) manpower requirements and availability for NRDP; (b) budge- tary implications of NRDP and methods of increasing revenues; and (c) credit to smallholders and possibilities for forming a national credit bank. The Review of NRDP 1.31 After three years of NRDP and about 13 years of the rural develop- ment effort it seemed appropriate to re-examine some of the assumptions and review progress on some of the issues highlighted in the NRDP documents. The manpower and budgetary studies mentioned above have not yet been carried out to a satisfactory degree, and in the near future both issues may become critical constraints to NRDP.11/ The review was 11/ The credit study by consultants was carried out, but Malawi Government is requesting additional information before it accepts the final study for consideration. Because credit works well in the RDPs, with very high rates of credit repayment, and because this study was being carried out, the issue was not examined in this review. See comments below, however, in para. 3.35.  - 10 - narrowly defined to examine how investment in smallholder agriculture could be made more effective and efficient; the review accepted the basic framework that this investment does have acceptable returns and that with sufficient access by smallholders to technical advice and to credit for purchasing inputs, with timely input delivery and access to markets for crops, and given proper incentives to smallholders, agricultural produc- tivity and farmers' incomes would be significantly increased. 1.32 Broader questions still surround NRDP. One is the role of the estates. It is possible that investment in estate production has high potential for Malawi's agricultural development; estate diversification studies have been carried out (financed by IDA) and the conclusions should be followed up. In addition, the issues of crop pricing and input delivery are critical to NRDP's success. Crop pricing is being reviewed in the context of the Structural Adjustment loans and input delivery, par- ticularly fertilizer supply, may be studied under the Bank's economic and sector work for Malawi. While all these questions are relevant for considering the suitability and the success of NRDP for Malawi, the task of the NRDP review was intentionally defined to deal with the practical problems of implementing a program basically accepted as being appropriate and worthwhile. 1.33 The terms of reference for the review are found in Annex 2. The exercise was a joint effort, with participation on the team by several Malawians. The conclusions reflect this joint effort but are not necessarily ascribed to by the Malawi Government or MOA. The Government has indicated its reactions to each issue raised in the report and these comments are included and discussed in the relevant sections. On those issues which do not have full agreement, compromises should be reached for the implementation of future NRDP phases. 1.34 Chapter two presents the mission's comments and recommendations on several issues. First, Section A reviews the multi-component nature of past RDPs and summarizes the mission's recommendations for the future structure of NRDP investment. Secton B discusses the problems of the structure, organization, and staffing of MOA that affect NRDP as well the problems of inter-departmental and inter-ministerial cooperation that will need attention for future investment in the rural areas. Then, in Section C, there is a discussion of the investment costs in housing under NRDP, with recommendations for lowering these costs; the last two sectons (D and E) discuss the resources likely to be required by and available to NRDP in terms of manpower and financing, respectively, and the applications for the size and phasing of NRDP. Finally, Chapter three presents the mission's recommendations for the content of the programs to be included in NRDP: extension, research, livestock, training, and monitoring,evalua- tion, and reporting. The last chapter draws conclusions on the future of NRDP. Chapter Two: NRDP - Performance, Issues and Recommendations A. Multi-Sectoral, Multi-Component Approach 2.01 The multi-sectoral approach of RDPs has led to a number of prac- tical problems for the operation and maintenance of non-agricultural facilities by non-agricultural departments (see paragraphs 1.16-1.19).  - 11 - Annex 3 contains a more detailed description of the "non-agricultural" activities under NRDP.12/ What follows here summarizes the annex. The philosophy of complementarity of investments has remained, in some cases, more an ideal than a reality. Health facilities were built but, in some cases,particularly for health posts, not staffed or used; even if they had been used, they often involved child care, whose benefits for agriculture would only be realized many years later and even then, not necessarily in the same area. Water supply was provided, but boreholes were not properly maintained so that the impact of a clean nearby water supply on villagers' health and use of their time was uncertain. Roads, providing access to extension advice, input deliveries, and markets, were not maintained and in some areas retured to bush. 2.02 A second practical problem of using the multi-sectoral approach is its inherent conflict in priorities. In the livestock and forestry sectors, for example, investments were made in disease control and afforestation, respectively, in areas that were not high priority for the country as a whole. Because one area was chosen as a priority for crop research and development, investments were made in other sectors that would otherwise have been made elsewhere. 2.03 Over the last few years, several departments and ministries have developed national programs to reflect their own priorities and criteria for investment and to be executed entirely by their own staff: rural roads, forestry, and in the future, rural health. What follows in this section is a discussion, on a component-by-component basis, of activities that either now have their own national programs or the mission feels ought to have national investment programs. These components should, in either the near future (next 3-5 years) or the longer term, be excluded from future NRDP phases. While the Government agrees in principle with this approach, there is some disagreement on the extent to which other ministries are ready now or in the near future to assume responsibility for national programs. These areas of disagreement are mentioned below. 2.04 MOWS is carrying out a district roads improvement and maintenance project (DRIMP), which involves covering the rural areas district by dis- trict, rehabilitating every district road where the estimated rate of return is at least 12 percent and providing local councils with sufficient training and equipment to maintain its roads. The mission fully supports this national program, but recognizes that RDPs cannot gain momentum without an adequate transportation network, so that coordinating the DRIMP schedule DRIMP schedule with the NRDP schedule is crucial. Because DRIMP can implement its program in any of the ten districts screened for economic acceptability,13/ it is flexible in its schedule and has to date been cooperative in coordinating its schedule with NRDP thus providing complementarity of investment in roads and in agriculture. 12/ Including forestry, fisheries, and research. 13/ Four districts have been covered under an earlier phase of the drimp program and the rest would be covered in later phases.  - 12 - 2.05 MOA has a separate roads construction unit, involved mainly in the construction of new roads rather than the rehabilitation of existing ones. Because DRIMP has also indicated its flexibility in constructing new rural roads where economically justified, the need for the MOA con- struction unit in the long term is not great. MOA has expressed the opinion, however, that DRIMP: (a) will not be available in the area needed for an RDP; (b) will not build sufficient new roads; (c) will not rehabilitate/build roads to an acceptable all-weather standard; and (d) will not be able to train and equip local councils to maintain the roads. MOA would, therefore, like to continue to operate its roads construction unit. 2.06 DRIMP has shown itself entirely flexible, however, in rearranging its schedule to fit in with the RDPs; it has changed the original plan to start work in Karonga to be followed by Chitipa in the north and has agreed to do the roads in Dowa district at MOA's request. In addition, DRIMP has agreed to build new roads where economic analysis shows the roads to have at least a 12 percent rate of return and appears to be building good roads, although this will be monitored after the rainy seasons; in any case, the standards can be improved if the roads prove unable to withstand the rains. While DRIMP may face a difficult task in training local councils to maintain their own roads, it seems the most sensible approach to long-term maintenance of local roads. Having a road construction unit under MOA certainly doesn't solve the road maintenance problem. 2.07 The mission therefore recommends that future RDPs rely on DRIMP to rehabilitate an acceptable road network which would include, where necessary, rural access roads. If certain roads have been excluded from the DRIMP program that MOA feels are important for their own investment, then MOA Planning Division should make the case to MOWS, showing that the roads to be built/rehabilitated would have an estimated rate of return of at least 12 percent. The existing roads construction unit under MOA should coordinate its present work program with that of DRIMP and should eventually be phased out or transferred to MOWS, whichever proves the more practical. Forestry 2.08 The forestry department, now part of the Ministry of Forestry and Natural Resources (MFNR), has a national Wood Energy Project (called NRDP II) which involves establishing plantations for eventual sale to the public of fuelwood and poles and a network of nurseries for selling seedlings to farmers. By approaching the wood energy problems from a national perspective, the project was able to concentrate the seedling nurseries as well as the plantations in areas identified as having larger wood deficits. The disadvantage of the national program run by a separate department is the need for careful coordination at the field level. The project is now in its second year, trying to improve the coordination by posting a forester in each Agricultural Development Division (ADD), who reports to the ADD Program Manager, and whose job is to supervise the wood energy nurseries activities, to work with the agricultural extension and land husbandry staff, and to make use of the ADD training facilities.  - 13 - While the appointment of these foresters is recent and their work programs and lines of authority have yet to be established, the mission supports this national program and the efforts at coordinating the activities at the ADD level. This approach is preferable to having separate forestry components under RDPs because it allows investment in priority areas and standard policies with respect to seedling and fuelwood prices.14/ In the near future, however, it may be that the program of rural afforestation will become the responsibility of MOA; if this proves the most effective means of reaching farmers, the mission would still recommend a national program which establishes priority areas and standard policies. The Government has stated a preference for having the scope to include individual forestry components under RDPs. The mission feels, however, that the current and possible future wood energy project is sufficiently flexible and has enough funds to accommodate changes (in nursery locations, in seedling species, in prices) appropriate to local areas. Health 2.09 The situation in the health sector is less clear. Health components under RDPs have suffered from inadequate staff and supplies and health facilities built under the projects have been underutilized. MOH needs to establish on a national basis its capacity to operate a network of health facilities. A second and equally important reason to develop a national planl5/ for health facilities is to establish realistic priori- ties for health care - which would also include geographical priorities. The problem this may raise for RDPs is that the priorities for crop development and for rural health services may not coincide, in which case RDPs would not include investment in health facilities. We must question, however, whether investment in health is a necessary element in the success of agricultural development. Although the Government maintains that complementarity between delivery of health services and increases in agricultural productivity does exist, data from the Lilongwe RDPs contradict this general belief and as noted above (paragraph 2.01), this complementarity under RDPs have remained more a principle than an accomplished fact. 2.10 MOR, with World Bank assistance, will develop during the next few years, a national health plan that would result in a national program of investment in delivery of health services. As in the case of the national rural roads program, the mission supports this direction. Once the national health program is underway, the mission recommends that future NRDP phases exclude health components so as to avoid duplication of effort. Until the national program is underway, each proposal for a health component under an RDP should be appraised on the basis of whether it fits in with any stated or perceived priorities and whether the proposed investments are realistic in the light of past experience. 14/ Or at least it allows for consistent policies that may differ because of local conditions. 15/ Malawi had had several national health plans but these have been statements of general policy priorities, and overall objectives and have not been translated into practical terms.  - 14 - Community Development 2.11 Community development activities exist in most rural areas and have not been included, with one or two exceptions, in RDPs. While most program managers recognize their usefulness in helping to motivate villagers, some RDPs have included their own rural development officers, whose functions have tended to duplicate those of community development. In addition, farm home instructresses (FHIls) under MOA, teaching health and nutrition, have also duplicated the functions of homecraft workers in community development. These functions should be streamlined, with FHIs focussing more on agricultural subjects (paragraph 3.14) and leaving self help and village participation to community development workers. Community development activities should not be a top priority of scarce funds, although this would be desirable in terms of cooperation and making full use of local organizations. In future projects, therefore, specific provision is not recommended for community development activities. In practice, of course, the project should be sufficiently flexible to permit the program manager to include support to community development if he judges this to be essential to the success of the project. Water Supplies 2.12 As in the case of roads, an argument can be made for the complementarity of rural water supplies and agricultural development - housing, training facilities, pilot schemes, research plots (as well as forest nurseries, livestock dips, health centers, schools, etc.) must all have access to water. Investment in water supplies should follow a master plan, however, based on a national survey of available water supplies, the capacity in terms of equipment and qualified staff as well as likely available funds of the Department of Lands, Valuation and Water (DLVW) to install water supplies, and recurrent cost implications of maintenance.16/ While the DLVW is drawing up an inventory of available water supplies, it does not have the planning capacity to develop a national investment plan. The mission feels that the Bank should encourage the DLVW in this direction and would hope that should a national program eventually emerge, the investments in water supplies would be coordinated with RDP's in much the same way as those under DRIMP have been. Until such time as a national program is drawn up, however, the mission recommends including funds for water supplies under RDP's. DLVW is developing a new approach to installing rural water supplies which includes village involvement, training, and establishing local maintenance ability, as well as low cost materials for both boreholes and shallow wells, so that the problems of high investment cost and inadequate main- tenance that have plagued the water components in the past should be greatly reduced. 16/ It is also time to give serious thought to developing a policy of recovering the recurrent costs of operation and maintenance of water supplies.  - 15 - Livestock 2.13 The investment in RDPs for livestock components have focused mainly on disease control (dipping tank construction or renovation and supply of drugs), marketing (building cattle markets), and to a smaller extent on animal production (breeding ranches, stallfeeder and dairy pro- grams, ox-training and grazing schemes). These investments have had mixed success: dip tanks have been built and operated and to the extent that data are available animal mortality has apparently decreased; stall- feeder programs have been very popular and limited only by the supply of cattle; but breeding ranches have not been profitable and grazing schemes have not been popular with farmers. The major issue for the NRDP review, however, was whether livestock investments under RDP's are the most appro- priate forum for developing the livestock sector. The problem has been that the areas for investments have been dictated by the RDPs and, particularly for disease control, have not followed what would have been national priorities. The mission recommends that the same principle be followed for investment in disease control as have been followed for the forestry program: national priorities should be identified for renovating old and establishing new dip tanks, consideration given to the staffing and recurrent costs of such investment and recommendations made for recovering these costs. Annex 10 gives a preliminary indication of what such a national program would involve. The results of the livestock study being carried out under a World Bank technical assistance loan should be taken into account in the elaboration of the program, with attention to any investments advised to improve livestock marketing. 2.14 The implementation of a national livestock program could be coor- dinated at the ADD level, as in the case of the forestry department's nur- series program under the Wood Energy Project, in consultation with land husbandry and extension staff so that dipping facilities would be properly placed. In addition, the mission recommends that other livestock activi- ties which involve animal production should continue to be integrated into RDPs so that extension and training can involve a farming systems approach to include livestock activities. The Government agrees with this approach. The Concept and Content of Future Rural Development Projects 2.15 The concept of rural development as a set of integrated activi- ties is still valid, but with the difference that the finance and imple- mentation of certain activities (roads, forestry, livestock) should be the responsibility of their respective ministries and departments and that the integration must be ensured through cooperation. For certain activities, however, the advantages of an integrated approach to investment in a particular area of the country must be weighed against establishing national priorities for those activities, such as health, forestry, and animal disease control, that might dictate a different order of investment. In a situation of restricted human and financial resources (see paras. 2.44-2.59 below), certain activities must be questioned realistically as to their priority and the degree of complementarity with rural development (this would include both community development activities and health).  - 16 - 2.16 What these comments and recommendations point to is that RDPs should focus on activities that directly support agricultural production, specifically on: research, particularly adaptive to farmers' needs; extension efforts, with attention to farming systems rather than single- crop recommendations; training for both farmers and staff; land husbandry; provision of seasonal and medium-term credit; establishment of markets through coordination with ADMARC; and, for the near future, provision of water supplies. MOA and MOWS should ensure that the road network is at an acceptable standard for rural development acticities and that for forestry and other livestock activities, coordination exists at the field level through the Program Managers (PMs). 2.17 To strengthen the integration of RDP activities within MOA, reor- ganization of MOA is recommended. This is discussed in the next section. In addition, to the extent that investment in rural areas will be and is already under the control of several ministries, the need for coordina- tion, both at the planning and the implementation stages, is great. Because each activity would have a somewhat different solution, no single arrangement would be appropriate for coordinating all rural investment. After reviewing the structure and staffing problems of MOA,the following section (paras. 2.33-2.36) presents some suggestions for dealing with the process of integration among departments and ministries. B. Government Structure Ministry of Agriculture: Organization and Management 2.18 In 1978, in an effort to integrate the management of NRDP into Government Structure (paragraph 1.26), a fundamental change was introduced in the demarcation of Malawi by the then MANR into units (known as exten- sion planning areas, or EPAs) by ecological and natural resources endow- inent; the EPAs were aggregated at two levels, development areas (DAs) and agricultural development divisions (ADDs)17/ and management structures were set up in the units (see Chart 1, below). NRDP envisaged the strengthening of the then MANR headquarters management capability (to pro- vide policy direction and specialized services), establishment of core management units (MUs) at the ADDs and the gradual delegation of powers to IADs and thence to DAs. The new demarcations applied to "agricultural" activities, although the reorganization proposed included research, veter- inary, forestry, and other MANR services. 2.19 Since 1978 when NRDP was launched, the Ministry has been pre-oc- cupied with the basic reorganization of its "agricultural" activities, which has been completed only recently. Headquarters reorganization has been more or less in line with that anticipated at the inception of NRDP (except that MANR has been split into two ministries: MOA and MFNR, with MOA responsible for NRDP), so that there are now specialist staff, in agriculture, for extension, training, crops, land and animal husbandry, 17/ There are 8 ADDs and approximately 40 DAs and 180 EPAs.  - 17 - credit, construction and women's programs, in addition to planning staff for preparation and evaluation. In the field, MUs have been established now at all the eight ADDs under Program manager (PMs) responsible for agricultural services. The line management runs from a Deputy Principal Secretary (DPS) through the Chief Agricultural Development Officer (CADO) to the PMs and respective supervisors of DAs and EPAs. Veterinary services and research are more or less programs parallel to agriculture but in spite of the proposals are still organized on the old district and regional lines. The organization and functions are discussed in detail in Annex 4 and a simplified, summary organization chart for MOA is given below: Chart 1 - Present Ministry of Agriculture (October 1981) Principal Secretary (PS) _ _7 Sr. Deputy P.S. (vacant) Deputy P.S Extension Agriculture Depart. of Agric. Veterinary Planning Aids Research 2/ Development Depart. 3/ Division ADDs 1 Asst CADO1 |Asst. CADO Program Admin. Technical Managers DAs 1/ Project Officers 1/ Rural Development projects usually cover one DA. 2/ Organized on crop-project basis. 3/ Organized on Regional basis. NOTE: CADO - Chief Agricultural Development Officer CARO - Chief Agricultural Research Officer CVO - Chief Veterinary Officer  - 18 - 2.20 In the process of reorganization, a number of staffing and management problems have emerged in MOA which relate to the success and efficiency with which NRDP can be carried out. First, reorganization has been slower than anticipated at appraisal of NRDP I. Second, the different structures of research and veterinary departments on the one hand and agriculture on the other have created problems of managerial coordination. Third, the stability of management has not been good and the quality of managerial and other key staff has been uneven. Finally, there may be overstaffing at some levels which mitigates against cost-effectiveness. Government agrees with this assessment. Each of these problems is discussed below, with some recommendations for changes. Reorganization 2.21 Progress on reorganization has been mixed, partly due to the "patchwork" of existing project organizations, but mainly because transi- tion from a civil service establishment to one capable of managing development projects is difficult under any circumstances. From this point of view, progress may be considered relatively good, even remark- able, given the sweeping changes planned. In addition, the ADD system seems to make sense, both from a technical, agricultural point of view and for management purposes. 2.22 Varying progress in establishing and staffing MUs relects the history of rural development in each ADD, including the relative importance of funding from "development" and "revenue" accounts. Overall, management in key areas has not become fully operative, which is reflected in weak but improving planning and control. A process of gradual delega- tion of responsibility from headquarters to ADDs has begun; at present the ADDs have different degrees of autonomy. Planning is still largely a headquarters function and systematic preparation of annual work programs at ADDs has yet to begin. The accounting function has been recently transferred to the MUs. Decentralization through to the DAs and EPAs will be a slow process because of the newness of the organization and the pace of recruitment of key staff, but once experienced staff are in post, delegation of responsibility, both for day-to-day operations and annual planning, should be encouraged. 2.23 Differences in the organization of agriculture on the one hand, and veterinary and research on the other, have created some understandable conflicts. At present the PMs are responsible, effectively, only for agricultural development even though research and veterinary activities are funded under NRDP. Thus, research and veterinary services are not formally integrated within the ADD managements nor within NRDP and colla- boration depends on personal relationships. Veterinary and research staff apparently consider it an anomaly that the PMs should report to the CADO which, to them, reflects the relative neglect of their disciplines for NRDP. Agricultural research is organized by crop or scientific project, which has led, in certain instances, to high costs of coordinat- ing trials and irrelevancy of some projects to local problems. Research management has proposed, accordingly, a reorganization of research by ADD for adaptive trials, together with certain national programs to be identi- fied, as proposed by the Bank mission (see paras.3.18-3.22). The problem  - 19 - of livestock development has been compounded by the division of responsi- bility for animal husbandry and veterinary services between the Department of Agricultural Development and Veterinary Department, respectively. 2.24 The mission recommended and MOA agreed that all three technical departments - agriculture, research, and veterinary services - should have the same organizational structure, with professional officers attached to each ADD under the management of the PM. This should permit better inte- gration of agricultural services. In addition, animal husbandry may be moved back to the veterinary department (where it had been located until 1962) if MOA feels this is appropriate. 2.25 The PMs, with their newly expanded responsibilities, would report to a technical deputy permanent secretary (DPS). With the reorganization of the other two departments, the PMs will manage staff from these depart- ments as well as from agriculture and, as such, they should be responsible to a non-department head (as opposed to the CADO). A second DPS post will, therefore, be created to serve as NRDP manager, much as foreseen at the appraisal of NRDP I and the heads of departments will retain their functions as chief technical advisors. This proposed organization is shown (simplified) in Chart 2. 2.26 While each department has agreed to make proposals for restruc- turing themselves into the ADD set up, the proposals should be carefully reviewed in MOA with respect to manpower and financial constraints. 2.27 Once MOA has firmly committed itself to the ADD structure and it is fully operating, other Ministries may be able to fit themselves in the ADD structure as well. Forestry Department has made a start in this direction by appointing a forester to each ADD to work with the PM on a program of farmers' woodlots. In the long term, a system may evolve where most, if not all, technical ministries have adopted an ADD structure with professional officers under the direction of a PM who would be attached to the OPC as the non-technical, senior Ministry. This system exists in other African countries and as a national organizational structure seems to work well.  - 20 - Chart 2 - Proposed Organization of Ministry of Agriculture Permanent Secretary Sr. Deputy P.S. DPS Technical DPS NRDP Coordinator Administrative |Administration Finance CADO c Extension Aid Planning < t t |~~~~~~~ Diviso ADDs Asst. CADO Research Veterinary Central PMs Technical 2/ Services 2/ Evaluation ! T T - r ~~~~~~~~~~~~~~~Unit 2/ Prof es- sional