Tanzania - National Agricultural and Livestock Research Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 19078 IMPLEMENTATION COMPLETION REPORT TANZANIA NATIONAL AGRICULTURAL AND LIVESTOCK RESEARCH PROJECT (CR. 1970-TZ) March 24, 1999 Rural Development Operations Eastern and Southern Africa 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 Dec.-88 US$ I = T Sh 122 Dec.-89 US$ 1 = TSh 178 Dec.-90 USS 1 = TSh 196 Dec.-91 US$ I = T Sh 233 Dec.-97 US$ 1 = T Sh 625 FISCAL YEAR OF BORROWER Government of Tanzania July 1 - June 30 ABBREVIATIONS AND ACRONYMS ADF African Development Fund ARF Agricultural Research Fund COR Client Oriented Research DRD Directorate of Research and Development (from mid-1998) DRT Department of Research and Training (until mid-1998) FSR Farming Systems Research GoT Government of Tanzania IARC International Agricultural Research Centre IDA International Development Association MAC Ministry of Agriculture and Co-operatives MTR Mid-Term Review NALRM National Agricultural and Livestock Research Master Plan NALRP National Agricultural and Livestock Research Project NARC National Agricultural Research Council NARS National Agricultural Research System SAR Staff Appraisal Report SPAAR Special Programme for African Agricultural Research SUA Sokoine University of Agriculture TAJAS Tanzanian Journal for Agricultural Sciences TALIRO Tanzania Livestock Research Organisation TARD Tanzanian Agricultural Research Data Base TARO Tanzania Agricultural Research Organisation TARP II Tanzania Agricultural Research Project II TPRI Tanzania Pesticides Research Institute UAC Uyole Agricultural Centre ZARF Zonal Agricultural Research Funds ZTC Zonal Technical Committee Vice President: Mr. Callisto Madavo Country Director: Mr. James W. Adams Sector Manager: Ms. Sushma Ganguly Task Manager: Mr. Satish Kumar  FOR OFFICIAL USE ONLY IMIPLEMENTATION COMPLETION REPORT TANZANIA NATIONAL AGRICULTURAL AND LIVESTOCK RESEARCH PROJECT (Credit No. 1970-TZ) Contents Preface ....................................................................i Evaluation Summary .................................................................... ii PART I: PROJECT IMPLEMENTATION ASSESSMENT .....................................................1 A. STATEMENT/EVALUATION OF OBJECTIVES ..............................................1 B. ACHIEVEMENT OF OBJECTIVES .....................................................................2 C. MAJOR FACTORS AFFECTING THE PROJECT ............................................6 D. PROJECT SUSTAINABILrTY ........................................................... ..........8 E. BANK PERFORMANCE ............................ .........................................9 F. BORROWER PERFORMANCE ..................................................................... 9 G. ASSESSMENT OF OUTCOME .................................................................... 10 H. FUTURE OPERATIONS .................................................................... 10 1. KEY LESSONS LEARNED .................................................................... 10 PART II: STATISTICAL TABLES .................................................................... 13 Table 1: Summary of Assessments .................................................................... 13 Table 2: Related Bank Loans/Credits .................................................................... 14 Table 3: Project Timetable ..................................................................... 15 Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual ........... 15 Table 5: Key Indicators for Project Implementation ............................................... 16 Table 6: Key Indicators for Project Operation ........................................................ 17 Table 7: Studies Included in Project .................................................................... 17 Table 8A: Project Costs .....................................................................18 Table 8B: Project Financing .................................................................... 19 Table 9: Economic Costs and Benefits .................................................................... 19 Table 10: Status of Legal Covenants .................................................................... 20 Table 11: Compliance with Operational Manual Statements ................................. 21 Table 12: Bank Resources: Staff Inputs .................................................................... 21 Table 13: Bank Resources: Missions .................................................................... 22 APPENDICES A. Aide Memoire B. Borrowers Contribution to the ICR C. Supplemental Information D. Map 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.  IMPLEMENTATION COMPLETION REPORT TANZANIA NATIONAL AGRICULTURAL AND LIVESTOCK RESEARCH PROJECT (Credit No. 1970-TZ) Preface This is the Implementation Completion Report (ICR) for the National Agricultural and Livestock Research Project (NALRP) in Tanzania for which Credit 1970-TZ in the amount of SDR 6.0 million (US$8.3 million equivalent) was approved by the International Development Association (IDA) on December 13, 1988. The Credit became effective on August 2, 1989. The project was co- financed, under parallel funding arrangements, by the African Development Fund (ADF) for which a loan of US$8.2 million was approved. Grants were also provided by the Kingdom of the Netherlands (US$2.7 million), the United Kingdom (US$2.85 million) and the Federal Republic of Germany (US $1.65 million). A Mid-Term Review was to be undertaken by October 31, 1991 but due to delays in project start-up, it was postponed until April 1993. The Credit was closed on December 31, 1997, nine months after the original Closing Date of March 31, 1997. The ADF Loan is still open and due to close on June 30, 1999, by which time it is anticipated to be fully disbursed. The bilateral grants have all closed and were fully disbursed. The ICR was prepared by staff of the FAO/World Bank Co-operative Program, on behalf of the Africa Rural Development Operations Group 1 (AFTRI) of the World Bank (WB). The report is based on a review of the project files and supervision reports, on discussions with WB staff and on the findings of an ICR mission that visited Tanzania from November 14 to December 4, 1998. The mission' visited the eastern and northern research zones and Sokoine University of Agriculture (SUA) and held discussions with Govemment of Tanzania (GoT) staff associated with the project, particularly staff of the Directorate of Research and Development (DRD)2 and of the extension services of the Ministry of Agriculture and Co-operatives (MAC). The borrower prepared its own assessment of the project performance which was issued in a report dated April 1998. A copy of the Aide Memoire of the ICR Mission and of the Govemment's Completion Report are attached as Appendices A and B respectively. This ICR was prepared at a time when the follow-up Tanzania Agricultural Research Project II (TARP 1I) was already effective and being implemented. Many of the issues reported in this ICR are being addressed under the follow-up project; some of the recommendations made could usefully be taken into account in ongoing implementation. Guy Evers (Mission Leader), Martin Maurer (Econonlist-APO) and Michael Macklin (Agricultural Services Consultant) 2 The Department of Research and Training (DRT) was created in 1989. In 1998 the training function was separated and DRT re-named the Directorate of Research and Development (DRD).  IMPLEMENTATION COMPLETION REPORT TANZANIA NATIONAL AGRICULTURAL AND LIVESTOCK RESEARCH PROJECT (Credit No. 1970-TZ) Evaluation Summary Introduction 1. From independence through the late 1970s World Bank lending to the agricultural sector of Tanzania was channelled through 21 agricultural and integrated rural development projects, constituting about 30% of the overall lending program. The emphasis of previous assistance was on crop and livestock specific projects implemented through parastatal institutions and regional development authorities. Despite this investment, agricultural production was falling, indicating that the fundamental problems facing agriculture were of a sectoral nature. The National Agricultural and Livestock Research Project (NALRP) was identified around 1985 after a hiatus in agricultural lending and in an environment of continuing macro-economic adjustment. At the beginning of the project, the national agricultural research assets comprised a substantial scientific manpower and station infrastructure which were fragmented and largely unproductive. NALRP was approved by IDA on December 13, 1988 and closed on December 31, 1997, after disbursing fully the Credit. Project Objectives 2. The main project objective was to implement the first stage of a long-term support program for developing Tanzania's agricultural research system. The project was intended to lay the foundation for establishing an effective national research system, to be implemented in two phases. The first phase was to focus on institutional re-structuring and development, staff training, strengthening planning and management capacity, preparation of a research master plan, strengthening and rehabilitating physical infrastructure and limited support for research in three specific programs. The second phase was to continue the institutional strengthening process and establish an Agricultural Research Fund to provide funding to additional research programs. The project was viewed as a precursor to a more investment-oriented second stage once the basic institutional and human capacity is developed. 3. Project design was appropriate and based on a good diagnosis of critical issues facing the NARS at the time of appraisal. Project objectives were correctly focused on fundamental needs for institutional consolidation and restructuring, staff training, and improved research planning and management. It was recognised that the full achievement of project objectives would be a long-term process requiring several investment phases, and that the first phase should basically strengthen research institutions and physical infrastructure  - ii - Implementation Experience and Results Achievement of Objectives 4. Project start-up and implementation was slow during the initial years. The Credit disbursement lag was 74% at the time of the Mid-Term Review (MTR). Following a strengthening of project management and re-allocation of funds after the MTR, implementation improved significantly. Overall, the project achieved its core objectives of reorganising and developing a more efficiently managed NARS and in laying a good foundation for client oriented research in the future. Significant improvements in generation and adoption of demand driven technologies have, however, still to be realised. 5. The project was successful in completing a major restructuring exercise in terms of the dissolution of three of the four agricultural research parastatals and their integration within the Department of Research and Training (DRT) of MAC. A. consolidated research structure, comprising seven research zones and a national headquarters, was established. The planned downsizing of the system has been partially completed with 15 of the 50 original stations closed or being handed over, and another 13 downgraded to field stations or trial sites. Support staff have been reduced by 25%, whilst the number of scientists has remained constant, around 350. 6. A comprehensive National Agricultural and Livestock Master Plan (NALRM) was funded by collaborating donors under the Special Program for African Agricultural Research (SPAAR). The NALRM defined a national agricultural and livestock research program, identifying ranked commodity priorities for the short, medium, and long terms, and specifying staff and physical resource requirements. Launched in 1992, NALRM was instrumental in initiating better resource allocations on a prioritised basis and in developing a framiework for demand driven zonal research. Building on this, DRT subsequently undertook a further research prioritisation exercise at zonal level, injecting a greater degree of client involvement. Farming of Systems Research (FSR) and Client Oriented Research (COR) methodologies are taking root within the NARS and are being integrated into all commodity research programs. This represents a significant step towards demand driven agricultural research. 7. The capacity building component was largely successful. Staff development objectives were achieved through a well directed and executed program that provided staff appropriate short- and long-term training especially in management and research methodologies. The procurement of vehicles and equipment and civil works improved capacity at selected research stations, even though the civil works program only completed about half of intended construction. 8. An Agricultural Research Fund (ARF) was established to finance collaborative and contract research with the intention that it would eventually attract incremental research funding from other sources, developing into a self sustaining "consolidated funding mechanism". This did not occur and ARF performance was not fully satisfactoiy. Implementation was delayed, review procedures proved time consuming, and research output was limited. Despite these shortcomings, the ARF succeeded in exposing DRT scientists to collaborative research, especially with Sokoine University of Agriculture. It also served as a useful scientist-training tool by demanding high quality research proposals from applicants. It served as a useful pilot activity and the lessons learned from it were applied under the phase 2 project namely the Tanzania Agricultural Research Project (TARP II) to improve ARF management and to launch Zonal AgricuLtural Research Funds.  - iii - 9. Extension-Research linkages improved during the project with measures such as joint monthly workshops; annual zonal research-extension workshops for joint research and extension program planning; on-farm demonstrations/trials; farmer field days; activities under the FSR and establishment of farmer research clubs in one of the zones. These were helped by allocations made under the extension projects (particularly NAEP II) for supporting researcher participation in several field activities, e.g., farmer days, field days, demonstrations and on-farm trials. Project Cost and Financing 10. Total project costs were estimated at US$25.3 mnillion, which was largely in line with the actual expenditure of about US$26.4 million'. IDA contributed US$8.6 million; ADF, US$8.2 million; the Netherlands, US$3.1 million; the United Kingdom, US$2.1 million; Germany, US$1.7 million; and GoT, US$2.8 million. Government recurrent budgets to research operations were to have been maintained in real terms at the pre-proj ect 1987/89 level. GoT failed to comply with this requirement as inadequate provision of non-salary operating funds prevailed throughout the project. However, the chronic operating fund crisis was mitigated by project management actions that included stations' revenue retention, cesses on eight industrial crops (coffee, tea, cotton, cashew, barley, sugarcane, tobacco and sisal), contract research and collaboration with international institutions. These initiatives are expected to continue under TARP II which would further reduce research dependence on Government budget. Project Sustainability 11. NALRP was intended to be a first phase of a long-term process of support to agricultural research, and focussed on institutional strengthening and provision of research infrastructure. The main structural and institutional reforms have been internalised with a strong degree of local ownership. Adequate institutional arrangements are therefore in place (further improved under the successor project TARP U) to ensure institutional sustainability. As anticipated in the SAR, short- to medium-term project sustainability will be assured largely by continued IDA support under TARP H , through anticipated support from several bi-lateral donors and from extra-budgetary resources raised from revenue retention, cesses and industry contributions. The long-term financial sustainability of the NARS will, to a large degree, depend upon: the growth of secure cess funding arrangements; increased collaborative and contract work; and through increased cost recovery on services, and is considered likely. 12. The major reasons for unsatisfactory performance until project mid-term were: weak proj ect management; inadequate provision of research operating costs; and poor terms of services resulting in low staff morale. Useful guidance through Bank supervision and a willingness to try new ways of fundraising on the part of Government were instrumental in turning the project round after 1993. The main factors influencing this change were: a strengthening of project management and improved remuneration of key managers; allocation of IDA funds for stations rehabilitation in accordance with NALRM recommendations; agreements with agro-industries to fund commercial crop research through cesses; agreement by Treasury to establish revenue retention funds at research stations; and, eventually, improved flow of funds from ADF. l Assuming full disbursement of ADF loan which is due to be closed in June 1999.  - iv - Bank and Borrower Performance 13. Bank performance was considered satisfactory. Appraisal was well done and design did not shy away from addressing difficult issues directly. Supervision input was intense and especially productive at MTR and in the latter years. Borrower performance was mixed, initially poor but eventually good as management improved and innovative! decisions began to take hold. Summary of Findings 14. The overall outcome of the project is rated satisfactory as it achieved its primary objectives of structural consolidation and institutional reform. A new de-centralised zonal structure has been adopted and a more client oriented research approach internalised. The SAR did not attempt to quantify the expected project impact and an economic rate of return was not calculated. During project implementation several new technologies were generated, some of which were reportedly widely adopted by farmers. The DRD has reported the following on release of varieties and adoption rates in the surveyed areas. Important crop varieties developed and released include four in wheat, ten in maize, four in beans, three in barley, six clones of highly resistant coffee and 13 vegetable varieties. Dual-purpose Mpwapwa breed of cattle continued to be introduced in Dodoma region. Control techniques for trypanosomiasis were further improved and introduced in Unguja region of Zanzibar where the incidence of disease reduced from 25.5 % to zero percent. Released wheat varieties are adopted by 80% of the large-scale farmers anld 25-30% of the small-scale farmers; maize varieties are being used by about 40% of the farmers; regarding beans 60% farmers in 6 surveyed districts had adopted Lyamungu 85 and 90 varieties. I'he research also isolated cause of "patchy stunting" disease in wheat which was a major production constraint for 23 years. The recommendations on application of fertilizers, herbicides and pesticides have also received substantial adoption. Technology generation was, however, biased in favour of crop breeding and variety releases. Only a few technologies were released that address crop and livestock management issues. The improved capacity of DRT was also reflected in increased collaboration with IARCs and regional research and development networks. Future Operations 15. Many of the issues reported in this ICR are being addressed under TARP II including steps to further decentralise research to zones and significantly empower farmers through establishment of Zonal Executive Committees, Zonal Agricultural Research Funds, Client Oriented Research and by stipulating that 50% of resources would be expended on-farm research. Suggestions are made in this report, which could usefully be taken into account in on-going implementation, in order to: increase flexibility in adjusting skill-mix of staff establishment to current and future needs; maintain strong inter-zonal co-ordination to avoid duplication and to drive synergies; continue NARS downsizing and rationalisation; improve cost recovery on research services and operations; secure reliable cess contribution from industry; review the mandate of the national ARF as new Zonal Agricultural Research Funds are developed; mitigate low staff morale issues; and better integrate socio-economic and marketing inputs in COR to improve the relevance and adoption of the research product.  Key Lessons Learned 16. Key lessons learned are as follows: With its new "development" content, DRD may develop strategies for a broader R&D function. The new "development" function needs to be clearly defined and explained to DRD staff. It is also important that DRD gives increasing emphasis to socio-economic and marketing issues in the final development of its product, and to forging stronger links with all stakeholders, including farmers, NGOs, private sector and others. Though COR has been formally intemalised by DRD, its approaches are still to be fully integrated in commodity and cross-cutting programs and there is need for further training and support in this area. Livestock research, especially small-stock, is still under-represented in some zones and deserves strengthening as this sub-sector can have significant impact on national priorities for improving nutritional standards and in poverty alleviation. NALRP has demonstrated that chronic financial constraints can generate improved and innovative management approaches to mitigate funding problems. In the case of NALRP, the financial crisis became so severe that GoT was driven to adopt flexible financial arrangements enabling scientists to access non- government funding sources. After a very difficult period, a healthier attitude has emerged such that the research stations regard raising resources outside the budget as an important management function. This offers hope that a more cost-effective and sustainable NARS can develop. Overall impact of technical assistance (funded mainly by bilateral donors and amounting to some 20% of total project cost) does not appear commensurate with this level of investment. Expenditures on TA should be very selective and their quality and utility carefully monitored. Decentralisation of management and administration to the seven agro-ecological zones is strongly supported; but because the zones are far from homogenous and similar agro-ecological situations are shared, imaginative inter-zonal coordination would be needed which helps avoid duplication and derive synergies but does not dilute stakeholder control of zonal research agenda. Rationalisation and consolidation of the NARS has resulted in a significant reduction in the number of research institutes, centres and stations, and a cut in support staff. This must however be regarded as a dynarnic process rather than a one-off project activity and DRD may require further structural adjustment to improve efficiency and effectiveness as conditions change. As far as practicable a statutory framework should be established for cess-funded commodity research. Options to secure levy or cess funding for food crop and  - vi - livestock research should also be investigated through the food processing and marketing industries. Retention of competent and motivated management within DRD is essential for NARS effectiveness. Consideration may be given to including a management overhead element in DRD cost recovery operations to finance performance supplements for key researchers and research managers. The mandate of the National ARF will require review in the light of lessons learned from the Zonal Agricultural Research Funds (ZARF) established under TARP 1I. The national ARF could concentrate on researchable problems of a cross-zonal nature. The research-extension-farmer linkages improved overtime and were helped by allocations made in the extension projects (particularly NAEP II) for supporting researcher participation in several field activities. With the transfer of responsibility for implementation of the extension program to the district councils and reorganization of the regional administration there is need to review extension-research linkage mechanism to respond to the above changes.  IMPLEMENTATION COMPLETION REPORT TANZANIA NATIONAL AGRICULTURAL AND LIVESTOCK RESEARCH PROJECT (Credit No. 1970-TZ) PART I: PROJECT IMPLEMENTATION ASSESSMENT A. STATEMENT/EVALUATION OF OBJECTIVES 1. The National Agricultural and Livestock Research Project (NALRP) was appraised after a period of curtailed International Development Association (IDA) lending to the agricultural sector and at a time of macro-economic reform. The National Agricultural Research System (NARS) was in a run-down state and unproductive. The main project objective was to implement the first stage of a long-term program of support to agricultural research in Tanzania. The project was intended to lay the foundation for establishing an effective research system that could effectively contribute to the sustainable generation of technologies, food security, improved national nutritional standards and the restoration of agricultural exports to earlier levels. To achieve this, the project was to provide financial and technical assistance to support: (a) Consolidation and downsizing of existing services to establish a new organisational structure for the national agricultural and livestock research system (NARS) within the Ministry of Agriculture and Co-operatives ( MAC); (b) Preparation of a National Agricultural and Livestock Research Master Plan (NALRM) to identify specific research activities ranked in order of priority and define the human, physical and financial resources required; (c) Training of research staff to meet needs for research management and methodology and to support NALRM priority research activities; (d) Provision of vehicles, equipment and civil works to meet immediate research needs; (e) Funding of cotton, root and tuber and farming systems priority research programs which had been identified and prepared before proj ect approval; and (f) Provision of unallocated research funds and establishment of an Agricultural Research Fund (ARF) for financing the crop and livestock research programs for which priorities would be established in the proposed NALRM. The project was to have two main implementation phases. The first phase was to focus on institutional development and rehabilitation of research infrastructure with limited support to the three research programs. In the second phase institutional and infrastructure development would continue, an ARF would be established to support research in terms of the priorities established in the NALRM.  - 2 - 2. Project design was appropriate and based on a good diagnosis of critical issues facing the NARS at the time of project appraisal. Project objectives were correctly focused on fundamental needs for institutional consolidation and restructuring, staff training, and improved research planning and management. Institutional reform objectives were ambitious and it was realistically recognised that their full achievement would be a long-term process requiring several investment phases. The Staff Appraisal Report (SAR) anticipated that the proj ect would result in major improvements in the organisation and management of the NARS, leading to more efficient use of limited resources. In addition the project would enhance the quality of experimnental work, sharpening the focus on farmers problems through improved research-extension-farmer linkages. There were no major diversions from principal objectives during implementation but several critical changes in project organisation and management and fund allocations were made. B. ACHIEVEMENT OF OBJECTIVES 3. Overview. Despite implementation delays and limited progress in the first four years, the project largely achieved its core objective of reorganising and developing a more efficiently managed NARS and in laying a good foundation for more effective client oriented research in the future. Significant improvements, however, in the generation and adoption of demand-driven technologies which would contribute to achieving the policy objectives of improving food security, nutritional standards and the restoration of agricultural exports have still to be realised. 4. Organisation and Management. Major restructuring in terms of the dissolution of three of the four agricultural research parastatals and their integration within the Department of Research and Training (DRT) of MAC was largely completed during 1989/90. However, staff re-deployment and re-appointments took longer than anticipated, creating uncertainty and exacerbating the already low staff morale. A consolidated research structure, comiprising seven research zone4 and a national headquarters in Dar Es Salaam, was established following recommendations of the NALRM. The planned downsizing of the system has been partial and is still ongoing. Out of the original 50 research stations, 23 are retained as priority DRT research stations. 13 others have been downgraded and retained as field stations or trial sites and the process of closing or handing over the remaining 15 is underway (see Table 1, Appendix C). Support staff numbers have been reduced by 25%, whilst the number of scientists has remained constant (i.e., some 350), giving a scientist to support staff ratio of 1 to 6.5. 5. Following the establishment of DRT, the project was to support the formation of financial, planning and administrative units at headquarters. This has been achieved and accounting, administrative and project management services are in place and performing satisfactorily. Procedures for the development of annual work plans and budgets were established, and subsequent annual plans and budgets have been regularly submitted to IDA for review. 6. As planned, a National Agricultural Research Council (NARC) was established to be the principal advisory body on agricultural and livestock research to MAC. Attempts to provide the Council with statutory powers were not realised and, though the Council did function, there is little evidence that it has had significant impact on the development of new agricultural research policies. Its future mandate in the context of a strongly de-centralised NARS would need to be reviewed. I Northern, Central, Eastern, Southern Highlands, Western, Lake and Southern (see Map, Appendix D).  - 3 - 7. National Agricultural and Livestock Research Master Plan (NALRM). The NALRM was funded by collaborating donors as a pilot program under the Special Program for African Agricultural Research (SPAAR) and co-ordinated by ISNAR!. It was intended to develop a comprehensive plan that reviewed and defined a national agricultural and livestock research program, identifying ranked commodity priorities for the short, medium, and long terms, and specifying staff and physical resource requirements. A draft NALRM was circulated for internal and external review in early 1991, finalised in December 1991 and launched at a national workshop in March 1992. Concurrently an Organisation and Management (O&M) Study was undertaken with support of the UK, which inter alia provided recommendations on a new zonal research structure, respective zone and centre functions, research management, and staff terms and conditions of service. Most NALRM recommendations were endorsed by GoT including those for re-structuring and downsizing the NARS. IDA funds were eventually distributed to concerned research centres for implementation of agreed priorities in late 1993, eighteen months after finalisation of the NALRM. The subsequent response from bilateral donors was, however, disappointing as few pledged incremental funding to support NALRM proposals. Subsequently DRT undertook a further research prioritisation exercise for each zone, injecting a greater degree of client involvement. The project succeeded in preparing a comprehensive NALRM which was instrumental in initiating better resource allocation on a rational and prioritised basis and in developing a framework for demand driven zonal research. Whilst NALRM undoubtedly proved a useful foundation tool, it was expensive and superseded by the zonal planning and prioritisation exercise. 8. Staff Development and Training. The project largely met its training targets. Some 21 staff attended long-term training, and about 660 staff attended short courses which were mainly conducted locally and in the region. The short-term training focused on research, support services, monitoring and evaluation, and financial management in line with the project objectives. In addition some 25 postgraduate scholarships and 50 short term courses were also offered by various other donors and organisations. This component was well planned, covered the priority training needs identified at appraisal for scientific, management and support staffs, and was successfully executed. 9. Civil Works. Provisions were made for constructing a new DRT Headquarters and for research station rehabilitation and construction needs which were to be recommended by the NALRM. In all, IDA funded the new DRT HQs (1,700 m2) in Dar Es Salaam (completed by end 1997), and infrastructure improvement at 9 research stations3. About 50% of the planned IDA-funded construction program was completed at a total cost of US$2.94 million. Partial completion was due to serious cost escalation arising from delays in tender and contract approvals, inadequate consultancy services and delayed fund allocation after NALRM completion. The quality of completed works seen appeared satisfactory. Infrastructure improvement of 6 other stations, funded by ADF, was only initiated at the end of 1997 and is expected to be completed by mid-i 999. 10. Vehicles and Equipment. This component was intended to meet the immediate requirements of vehicles and equipment, prior to completion of the NALRM and O&M studies. About 60% of IDA vehicle and equipment procurement was completed within the first two years of I International Service for National Agricultural Research. 2 1 BSc, 16 MSc and 4 PhD, mostly overseas, finded by ADF. In addition, The Netherlands funded training in Farming Systems Research (FSR) which was not anticipated at appraisal. 3 Including water and electricity supply, the rehabilitation and/or construction of 38 houses, two workshops, one warehouse, one laboratory and three office blocks, all of which were completed by the end of the project.  - 4 - project implementation, the remaining funds were utilised from March 1993 onwards'. Although the objective of rapidly procuring vehicles and equipment was only partly met, this overall provision greatly facilitated implementation of research activities recommended under the NALRM and in the subsequent zonal research prioritisation. 11. Priority Research Activities and Research Results: One central and seven zonal FSR teams were established to assist researchers to better identify and address farmers' priority needs; to strengthen farmers-extension-research linkages and to facilitate linkage between the commodity programs within DRT. Most crop commodity scientists, and some livestock and soils scientists, attended training on FSR methodologies. Increased participation of farmers in technology generation was promoted through the Client Oriented Research (COR) methodology and on-farn research. The cotton and root and tuber programs were to be funded by ADF and the lUK (cotton only): however, ADF disbursement delays and withdrawal of UK support to cotton resulted in little research activity or output. Nevertheless the zonal stations continued to carry out crop and livestock research in terms of the NALRM and zonal priorities with the limited resources that were available during the proj ect period. The DRD has reported the following on release of varieties and adoption rates in the surveyed areas. Important crop varieties developed and released include four in wheat, ten in maize, four in beans, three in barley, six clones of highly efficient coffee, three each in sorghum/millets, sesame and cassava, one variety in cotton, one cultivar in tobacco, three clones in pyrethrum and five varieties in sugarcane. Dual-purpose Mpwapwa cattle breed was further improved and introduced in Dodoma region. More effective control techniques for trypanosomiasis were developed and introduced in Unguja region of Zanzibar where trypanosomiasis incidence reduced to zero from 25.5%. Released wheat varieties are adopted by 80% of the large farmers and 25-30% of the small farmers; maize varieties are being used by 40% of the farmers; 60% farmers in 6 surveyed districts had adopted the Lyamungu 85 and 90 varieties; 15% farmers in the Southern Zone have adopted the vegetative propagation technique in cashew; barley varieties have received almost cent percent adoption by large and small farmers. The research also isolated cause of "patchy stunting" disease in wheat which was a major production constraint for 23 years. The recommendation on the appropriate use of fertilizers and other soil improvement techniques as well as application of pesticides and herbicides including integrated pest management, and on several cultural practices have been widely adopted in various zones. 12. Unallocated Research Funds. US$5.7 million was provided by IDA and ADF to develop capacity at selected stations to undertake priority research programs identified by the NALRM and secondly to provide seed funding for a national contract and collaborativeAgricultural Research Fund (ARF). After the 1993 MTR, IDA funds were allocated from this component for research station development, leaving only US$148,000 for contract and collaborative research under ARF. 13. The ARF constitution was issued in 1992 and the first projects started in 1994. The aggregate fund amounted to US$210,0002. In addition, GoT provided US$30,000 to meet ARF overhead costs. A total of 16 collaborative research projects were approved for a total budget of about US$380,000, leaving a balance of US$170,000 to be funded by other sources such as TARP II. ARF performance was not fully satisfactory. Implementation was delayed, review procedures proved I Overall, some 70 4-WD vehicles, 50 motorcycles, three water pumps, 15 tractors, 25 computers, 4photocopiers and miscellaneous equipment and material were procured with IDA funding. 2 i.e., US$148,000 from IDA and US$62,000 from ADF.  - 5 - time-consuming and, as a result, research output was limited at project completioA. Competitive contract research was not tested under the project as anticipated. It had been hoped that the ARF would attract incremental research funding from other sources and develop into a self-sustaining "consolidated funding mechanism", but this did not occur under NALRP. Despite these shortcomings the ARF succeeded in exposing DRT scientists to collaborative research, especially with SUA, significantly improving overall DRT/SUA linkages. It also served as a useful scientist-training tool by demanding high quality research proposals from applicants; and a useful pilot activity for testing ARF management and procedural options. 14. Management Information Systems, Documentation and Communication. The project has started to improve scientific information and documentation services through the procurement of computer facilities, the establishment of a Tanzanian Agricultural Research Data Base (TARD), and access to a number of national and international information networks. The INFORM package, a research MIS developed by ISNAR, was introduced at DRT Headquarters in 1992/93 to assist in preparing work plans, budgets, resource allocations and summaries of on-going research. INFORM has also been established at the zonal centres. However, further INFORM and TARD software upgrades and more regular data base updates are essential to sustain effective operations. Communication between zonal centres and headquarters was substantially improved through telephone/fax and Email systems, though not completed and some stations still have poor communication facilities (i.e., Ilonga, Tumbi and Mpwapwa). 15. Extension-Research Linkage. Formal links between research and extension were built into project design. These included appointment of research-extension liaison officers, joint monthly workshops and annual zonal research extension workshops for joint research and extension program planning. In 1992, Zonal Technical Committees (ZTCs) were formed to further enhance client influence on research and extension planning. Because operating fund provisions under both research and extension programs were inadequate to conduct the proposed activities, their combined impact has been limited. Better linkages were realised under the FSR program which was well funded and succeeded in testing FSR approaches involving greater farmers' participation in DRT commodity research programs. 16. Donor Co-ordination. The involvement of SPAAR was intended to enhance and institutionalise donor co-ordination. Tanzania was also selected as a pilot for implementing the "Framework for Action" intended to establish a common funding mechanism for DRT. Although there were some co-ordination problems amongst SPAAR donors in initiating the NARLM, better co- ordination was generally achieved during project implementation through the establishment of a GoT- Donor Consultative Group and regular exchange of information. A common funding mechanism was not established as donors preferred to fund identifiable research activities which fitted with their own aid program strategies and priorities. 17. Financial Objectives. The IDA Credit of SDR 6 million (US$8.3 million at appraisal) was fully disbursed by December 31, 1997. There were substantial disbursement delays until 1993, when the disbursement lag was 74%, but thereafter the pace of disbursement improved. After re- allocations, total revised IDA project costs as per category of expenditure are: (i) civil works SDR 2.06 million (US$2.94 million); (ii) vehicles and equipment: SDR 1.37 million (US$1.96 million); (iii) consultancies and training SDR 1.28 million (US$1.84 million); (iv) non-salary operating cost (new category of expenditure): SDR 0.38 million (US$0.56 million); (v) ARF: SDR 0.2 million Initially, many proposals submitted to ARF were of poor quality and had to be reformulated. In early 1998, ARF review procedures were simplified, and deadlines for each step established.  - 6 - (US$0.28 million); (vi) project operating costs: SDR 0.44 million (US$0.65 million); and (vii) the PPF: SDR 0.21 million (US$0.26 million). Actual IDA project costs are US$8.56 million. 18. It is anticipated that the ADF loan will be fully disbursed by its expected closing date in June 1999. The total proj ect costs are therefore estimated at US$26.4 million, largely in line with the SAR estimates. Project financing in summary is: GoT: US$2.8 rnillion, IDA: US$8.6 rnillion, ADF: US$8.2 million, the Netherlands: US$3.1 million, UK: US$2.1 million and Germany: US$1.7 million. 19. Whilst GoT funded all local staff salary and allowance costs there were major shortfalls in the Govemment's counterpart funding for non-salary research operating costs. Indeed GoT provisions of non-salary operating costs have continued to decline to reach a low point of TSh 59 millions in 1996/97, the last full fiscal year of NALRP (see Appendix C, Table 3). However, project management took a number of commendable actions to mitigate the chronic operating fund crisis, in particular: - Revenue Retention: this concept was approved by the Treasury for fiscal year 1994/95, and subsequently implemented at nearly all stations. Although revenue generation potential varies considerably between location, this initiative has been instrumental in generating income to fund essential research operations and maintenance of stations. According to DRD1, some US$2 million were collected between 1994/95 and 1997/98. Revenues have been mainly generated from production sales, laboratory services and rental of surplus land and buildings. - Cesses: in 1995/96, agreements were reached with the major agro-industries to contribute to research funding through cesses or annual allocations for eight cash crops, namely coffee, tea, cotton, cashew, barley, sugarcane, tobacco and sisal. Cesses are also used for providing incentives to the involved research staff. Tea research was fully privatised in 1996. There are comparable privatisation initiatives being discussed for tobacco and coffee. In 1996/97, DRD estimated that the contribution from agro-industries vvas US$3.18 million; and - International collaboration: additional operating funds (and staff incentives) were increasingly generated through contract and collaborative research with Intemational Agricultural Research Centres (IARCs) and their regional networks, and from consultancy income. C. MAJOR FACTORS AFFECTING THE PROJECT 20. Research Operating Funds. The failure of GoT to fund research operating costs at agreed levels was a persistent issue raised by almost every supervision mission. As a result priority research programs in cotton and roots and tubers did not perform as anticipated, with little evidence of improved technology generation. After the MTR, IDA re-allocated SDR 0.42 million to a new disbursement category for research operating costs to support priority programs identified under the NALRM. This was helpful but insufficient to have significant impact on overall NARS technology generation under NALRP. The chronic lack of working funds worsened already poor staff morale, and negatively affected technology generation output. However, this financial crisis provided the In 1998, the training function of DRT was separated and DRT re-named the Directorate of Research and Development (DRD).  - 7 - impetus to stimulate introduction of several positive revenue generation innovations on the part of GoT and a more entrepreneurial management approach amongst many scientists and research managers. 21. Low Staff Morale. Low staff morale was another persistent issue raised throughout implementation --and is still unresolved. GoT salary scales for qualified agricultural scientists are very low, promotional opportunities limited, and allowances modest. No monetary performance incentives are in operation, and remedies are not available under current civil service regulations. Whilst options for improving the conditions of service for scientists working on cess-funded industrial crops are available, similar funding sources have not yet been identified for food crop, livestock or cross-cutting research. However, TARP II has established a system of recognition (carrying a monetary reward) for outstanding research results produced by a scientist. The value of reward is higher if research finding has national or regional level applicability. 22. Civil Works. Delays resulting from inadequate civil works program management and late allocation of IDA funds and release of ADF funds for station rehabilitation led to the construction program being compressed into the last four years of the project. By this time construction costs had escalated considerably and works had to be scaled down to about 50% of original plans in line with available funds. 23. Project Management. Unsatisfactory management performance in the first four years of the project was exacerbated by delayed appointments of management staff, appointments on acting conditions for long periods and by generally slow re-deployment of staff to and within DRT from previous research parastatals. Furthermore the Project Implementation Unit did not operate effectively as DRT organisation lacked clear lines of responsibility and accountability. Effective implementation of a complex project such as NALRP required highly motivated managers able to devote 100% of their time to the project. This proved difficult when key managers operated under normal civil service conditions. Both the Bank and Borrower skirted this issue, and only after the MTR, the required adjustments were made and management significantly improved. 24. Technical Assistance and Studies. Technical assistance focused on supporting the preparation of the NALRM (funded by Germany), the O&M studies (funded by the UK) and the FSR program (funded by The Netherlands). IDA funded TA to strengthen research management, including inter alia the establishment of INFORM (16 person-months) and support for financial management and assets verification (24 person-months). Whilst some TA has had a positive input in facilitating NARS restructuring, not all has provided adoptable outputs. Overall, output does not appear commensurate with the estimated investment of US$5.3 millions. In particular, few recommendations of the O&M study have been intemalised. The MTR recommended three additional studies be carried out on: (i) TPRI restructuring; (ii) the review of alternative source of recurrent funding; and (iii) a review of private and public tea research. The first two studies were not completed, and the third study, prepared with FAO assistance, contributed to the privatisation of tea research. 25. Mid-Term Review and DCA amendment. The MTR was thorough and played an important role in assisting MAC address several outstanding implementation issues. However, in hindsight it was unfortunate that the MTR was delayed by more than a year (until mid-1 993) because of the delayed credit effectiveness and slow implementation. Had it been possible to let it take place nearer the originally planned date, in 1991, decisions on adjustments to project management and the  - 8 - re-allocation of Credit proceeds to implement research station rehabilitation in accordance with NALRM recommendations could probably have been acdvanced by one year. D. PROJECT SUSTAINABILITY 26. Institutional sustainability. The significant institutional reforms introduced under NALRP appear to have been intemalised with a strong degree of local ownership and therefore the Mission is confident they will be sustained. The number of scientists in the NARS is constant and, despite poor terms of service, leakage of professional staff to the private sector or other organisations appears insignificant. No progress has been made in developing improved service conditions for agricultural scientists and it is unrealistic to expect that exceptional conditions will be approved within the civil service structure. Providing opportunities and incentives for scientists to generate non-government source financing to augment their operating funds and bolster income is one way of improving rewards and morale and of enhancing NARS sustainability. The maintenance of preferential salaries and allowance for key NARS managers is critical in the medium term to maintain implementation pace under TARP II, but long-term solutions to the general issue of sustaining quality managers in the public sector NARS have still to be foun.d. 27. Financial Sustainability. In the short to medium terms project sustainability will be ensured by the continued IDA support under TARP II and through anticipated support from a number of bilateral donors. The future health and long term sustainability of the NARS will to a large degree depend upon: the growth of secure cess funding arrangements; increased collaborative and contract work with regional and international organisations; and through increased cost recovery on services. Cess funds have under-pinned research on industrial crops since 1995 and future research on these commodities will be fully dependent on the support of their respective industries. GoT is likely to remain the main source of funding for most food crop and livestock research. But with decreased demands on Government budget by exclusion of the budget support to industrial/commercial crops, increasing trend in resources raised from cessAevies, contract and collaborative research and funds availability from the zonal agricultural research funds (being established under TARP II), and the above initiatives underpinned by a strong institutional stricture the overall sustainability is considered likely. 28. Vehicle and Equipment Replacement. The lack of a reliable internal capital replacement scheme for vehicles and equipment is a matter of concem. GoT funds are rarely available for vehicle and equipment replacement which has become a donor dependent exercise. Whilst not an immediate problem, donor support cannot be guaranteed in the future and an affordable intemal replacement scheme has to be devised to ensure long-term operational sustainability. E. BANK PERFORMANCE 29. Project design was good and directly addressed key organisational, management and institutional issues. It was though ambitious in expecting that specific research activities could be implemented concurrently with major institutional change and optimistic in its assumption that GoT would adequately finance research operating costs. Project supervision was intensive, with an average annual coefficient of 25 staff weeks over 17 missions. Supervision was unable to prevent serious implementation delays and resolve problems of weak project management in the first four years. Despite intensive supervision the somewhat generic issues of inadequate GoT funding of research operating funds and low staff morale persisted. The MTR was thorough and important in improving subsequent implementation. At that time Banik support helped to strengthen the Project  - 9 - Implementation Unit developing clear lines of responsibility and accountability and providing key managers better terms of service. Thereafter management and supervision performance significantly improved and, in the second half of the project, the pace of implementation and disbursements picked-up'. Bank staff also assisted MAC introduce important initiatives to ease the operating fund crisis (i.e., crop cesses and retention scheme). SPAAR played a positive role in encouraging donor co-ordination and assisting research management. Overall, the Bank performance is considered satisfactory. F. BORROWER PERFORMANCE 30. Borrower performance can be separated into two main phases. Performance in the first phase (1989-1993) was less than satisfactory and characterised by implementation delays and weak management. Progress, however, was made in the dissolution of three of the four research parastatals and their integration within DRT; in completion of NALRM and O&M studies; and through start-up of the Netherlands-funded FSR component. There were however, significant delays in rationalisation of the research network and in staff redeployment and actual capacity building at zonal research institutes and centres was minimal. Borrower performance improved markedly from 1993 through Credit closing on December 31, 1997. Three main factors facilitated this change; (i) the strengthening of the Project Implementation Unit within DRT; (ii) DCA amendment allocating funds for implementing NALRM priority programs at zonal level; and (iii) ADF funds started to become available. Thereafter, Borrower performance improved and implementation accelerated both at headquarters and in the zones. Although there were delays in meeting legal requirements, the Borrower eventually complied with all main DCA covenants. The Borrower however failed to comply with the agreement to maintain the 1987/88 value in real terms of its annual contributions to the recurrent research budget. Overall, Borrower performance was satisfactory. G. ASSESSMENT OF OUTCOME 31. The overall outcome of the project is rated satisfactory as it achieved its primary objectives of structural consolidation and institutional reform. A new de-centralised zonal structure has been adopted and a more client oriented research approach internalised. The SAR did not attempt to quantify the expected project impact and an economic rate of return was not calculated. During project implementation several new technologies were generated, some of which were reportedly widely adopted by farmers. The DRD has reported the following on release of varieties and adoption rates in the surveyed areas. Important crop varieties developed and released include four in wheat, ten in maize, four in beans, three in barley, six clones of highly resistant coffee and 13 vegetable varieties. Dual-purpose Mpwapwa breed of cattle continued to be introduced in Dodoma region. Control techniques for trypanosomiasis were further improved and introduced in Unguja region of Zanzibar where the incidence of disease reduced from 25.5 % to zero percent. Released wheat varieties are adopted by 80% of the large-scale farmers and 25-30% of the small-scale farmers; maize varieties are being used by about 40% of the farmers; regarding beans 60% farmers in 6 surveyed districts had adopted Lyamungu 85 and 90 varieties. The research also isolated cause of "patchy NALRP was a high cost operation spanning thirteen years from project identification to Credit closing. It involved 393 weeks of Bank staff time and cost US$913,000 to administer --equivalent to over 10% of the value of the IDA Credit. This cost is largely explained by the depth and complexity of the structural and institutional reforms involved; this led to intensive inputs for project identification through negotiation and for supervision.  - 10- stunting" disease in wheat which was a major production constraint for 23 years. The recommendations on application of fertilizers, herbicides and pesticides have also received substantial adoption. Technology generation was, however, biased in favour of crop breeding and variety releases. Only a few technologies were released that address crop and livestock management issues. The improved capacity of DRT was also reflected in increased collaboration with IARCs and regional research and development networks. H. FUTURE OPERATIONS 32. Many of the issues reported in this ICR are being addressed -under TARP II including steps to further decentralise research to zones and significantly empower farmers through establishment of Zonal Executive Committees, Zonal Agricultural Research Funds, Client Oriented Research and by stipulating that 50% of resources would be expended on-farm research. Suggestions are made in this report, which could usefully be taken into account in on-going implementation, in order to: increase flexibility in adjusting skill-mix of staff establishment to current and future needs; maintain strong inter-zonal co-ordination to avoid duplication and to drive synergies; continue NARS downsizing and rationalisation; improve cost recovery on research services and operations; secure reliable cess contribution from industry; review the mandate of the national ARF as new Zonal Agricultural Research Funds are developed; mitigate low staff morale issues; and better integrate socio-economic and marketing inputs in COR to improve the relevance and adoption of the research product. I. KEY LESSONS LEARNED 33. Summarized below are important lessons learned. These include lessons which are the basis of several initiatives included in TARP II to effect improvements. Appropriate suggestions as to next steps are made in respect of several of the summarized lessons which the DRD can consider for appropriate implementation: - With its new "development" content, DRD may develop strategies for a broader R&D function. This is fully consistent with increased emphasis on COR and on- farm research, and the creation of de-centralised, zonal research funds under TARP II. The new "development" function needs to be clearly defined and explained to DRD staff. It is also important that DRD gives increasing emphasis to socio- economic and marketing issues in the final development of its product, and to forging stronger links with all stakeholders, including farmers, NGOs, private sector and others. - Though COR has been formally internalised by DRD, its approaches are still to be fully integrated in commodity and cross-cutting programs and there is need for further training and support in this area. - Livestock research, especially small-stock, is still under-represented in some zones and deserves strengthening as this sub-sector can have significant impact on national priorities for improving nutritional standards and in poverty alleviation. - NALRP has demonstrated that chronic financial constraints can generate improved and innovative management approaches, even in a NARS that has been historically  - 11 - dependent upon, and used to in-adequate government funding over many years. In the case of NALRP, the financial crisis became so severe that GoT was driven to adopt flexible financial arrangements enabling scientists to access non-government funding sources. After a very difficult period, a healthier attitude has emerged that the research stations regard raising resources outside the budget as an important management function. This offers hope that a more cost-effective and sustainable NARS can develop. NALRP implementation experience illustrates the importance of an adequately remunerated and motivated project management operating in an environment with unambiguous lines of responsibility and accountability. Technical assistance, funded mainly by bilateral donors, amounted to some 20% of total project costs (para 24). Whilst the product of several studies proved useful, especially the NALRM and the revised financial management system, overall TA impact does not appear commensurate with this level of investment. Relatively few studies' recommendations have been adopted and internalised. Further, their utility quickly wanes as they are over-taken by new events. Expenditures on TA should be very selective and their quality and utility carefully monitored. DRD staffing is dominated by crop-related scientists' and under-represented in disciplines such as natural resource conservation, socio-economics, marketing and livestock. As client oriented research is increasingly undertaken, the subjects of emerging research priorities may not be matched by the existing skill mix. Therefore, there is a need for flexibility in adjusting existing staff establishment and in future recruitment, and to increase contract and collaborative research with outside bodies, especially in areas where DRD skill capacity is limited. Decentralisation of management and administration to the seven agro-ecological zones is strongly supported; but because the zones are far from homogenous and similar agro-ecological situations are shared, imaginative inter-zonal coordination would be needed which helps avoid duplication and derive synergies but does not dilute stakeholder control of zonal research agenda. To optimise use of specialist scientific skills and laboratory facilities across zones, further staff re-deployment and increased use of common laboratory services may be necessary. Rationalisation and consolidation of the NARS has resulted in a significant reduction in the number of research institutes, centres and stations, and a cut in support staff. This must however be regarded as a dynamic process rather than a one-off project activity and DRD may require further structural adjustment to improve efficiency and effectiveness as conditions change; As far as practicable a statutory framework should be established for cess-funded commodity research. This would help avoid uncertainty on the cess rate, fund inflow and their continuation. Options to secure levy or cess funding for food crop and livestock research should also be investigated through the food processing and It is estimated that 75% of staff work on crop related research.  - 12 - marketing industries (e.g., millers, abattoirs and markets; canneries and processors, etc). Retention of competent and motivated management within DRD is essential for NARS effectiveness. As it is unlikely that exceptional terms of service would be acceptable within a common civil service structure, innovative solutions have to be found. Consideration may be given to including a management overhead element in DRD cost recovery operations to finance performance supplements for key researchers and research managers. The mandate of the National ARF will require review in the light of lessons learned from the Zonal Agricultural Research Funds (ZARF) established under TARP II. The national ARF could concentrate on researchable problems of a cross-zonal nature. The research-extension-farmer linkages improved overtime and were helped by allocations made in the extension projects (particularly NAEP II) for supporting researcher participation in several field activities e.g., farmer days , field days and on-farm demonstrations/trials. Some zones established farmer research clubs where extensionists regularly participated. With the transfer of responsibility for implementation of the extension program to the district councils and reorganization of the regional administration there is need to review extension-research linkage mechanism to respond to the above changes.  - 13 - PART II: STATISTICAL TABLES Table 1: Summary of Assessments A. Achievement of objectives Substantial Partial Negligible Not Applicable (v) (v) (v) (v) Macro policies L i I I] Sector policies a EI En E Financial objectives ED E7 En [I] Institutional development E71 En Eii En Physical objectives En En En Poverty reduction En En En En Gender issues En En En IL] Other social objectives [I] Fi [I] En Environmental objectives En m Li] En Public sector management En E En EnI Private sector development En m EE En Other (capacity building) FI En En En B. Project sustainability Likely Unlikely Uncertain (v) (v) (v) E E E Highly C. Bank performance Satisfactory Satisfactory Deficient (v) (v) (v) Identification En En Preparation assistance En F[ En Appraisal En En En Supervision En En En  - 14- Highly D. Borrower perfonnance Satisfactory Satisfactory Deficient (v) (v) (v) Preparation I mE] Implementation' [Z E O Covenant compliance 0 m E Operation (if applicable) E7 ] Highly Highly E. Assessment of outcome Satisfactory Satisfactory Unsatisfactory unsatisfactory (v) (v) (v) (v) El El .l E Table 2: Related Bank Loans/Credits Loan/credit title Purpose Year of approval Status Preceding operations None Following operations 1891-TZ Agriculture Export 1988 Completed Rehabilitation 1994-TZ. National Agriculture & 1989 Completed Livestock Extension Rehabilitation 2050-TZ Cashew and Coconut 1989 Completed Tree Crops Development 2537-TZ Agricultural Sector 1993 Ongoing Management 2899-TZ National Agricultural 1996 Ongoing Extension II 2900-TZ River Basin 1996 Ongoing Management and Smallholder Irrigation Improvement 3036-TZ Agricultural Research II 1998 Ongoing 1/ Performance up last 3 years. Deficient in first 3 years. Improvement for later years (3-4).  - 15 - Table 3: Project Timetable Steps in Project Cycle Date Planned Date Actual/ Latest Estimate Identification August 1985 August 1985 Preparation November- December 1986 November- December 1986 Appraisal May 1987 May 1987 Staff Appraisal Report August 1987 July 1988 Negotiations November 1987 April 1988 Board presentation December 1987 December 1988 Signing Mid-1988 February 1989 Effectiveness October 1988 August 1989 Mid-Term Review October 1991 April - May 1993 Project Completion September 1996 December 1997 Loan closing March 1997 December 1997 Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual (US$ millions) FY89 FY90 FY91 FY92 FY93 FY94 FY95 FY96 FY97 FY98 Appraisal 0.80 2.4 4.0 5.4 6.7 7.5 8.1 8.3 8.3 8.3 estimate Actual -- 0.63 0.67 1.37 1.76 2.29 4.23 6.11 7.94 8.56 Actual as % 0% 26% 17% 75% 26% 31% 52% 74% 96% 103% of estimate Date of final disbursement: April 20, 1998.  - 16 - Table 5: Key Indicators for Project Implementationl I Estimated T Actual I. Re-structuring of Agricultural Research 1. Consolidated Research & 1 2 (DRT and TPRI which was Training Department maintained as a parastatal) 2. Zonal Research Centres 6 -7 7 3. Downsizing of research staff 12 (3% reduction) 4. Downsizing of support staff 775 (25% reduction) 4. Constmrction of DRT HQ 1,700 m2 1,700 m2 5. Long-term Technical 60 person-months 33 person-months Assistance for NALRM IL Training (DRT staff trained) Long-term 1. B.Sc 0 1 2. M.Sc 14 23 3. PhD 8 5 Short term Courses 713 655 II. Civil works (IDA funding only) l 1. New & rehabilitated 9,625 m2 4,560 m2 buildings 2. Rehabilitated office 10 7 buildings & supply systems I IV. Vehicles for research and re-structuring of research 1. Vehicles 4-wheel, pick-ups & lorries 104 125 2. Motorcycles & bicycles 168 187 3. Tractors 17 15 V. Agricultural Research Fund (ARF) 1. Funding US$ 753,000 US$ 382,000 2. Number of projects - 16 Except for training, performance indicators were not defined in the SAR.  - 17 - Table 6: Key Indicators for Project Operation Key Operating Indicators Estimated Actual Not Applicable Table 7: Studies Included in Project Purpose as Defmed Study At Appraisal/Redefined Status Impact of Study Defined at Appraisal 1. National Agricultural Definition of medium- and Completed Substantial, served as a and Livestock Research long-term national agricul- basis for NARS Master Plan tural research strategy & restructuring; but programme prioritisation; super-seeded by follow-up Zonal prioritisation 2. Organisation and Plan for NARS institutional Completed Limited, as few recom- Management Study restructuring and research mendations were inter- management; nalised in the consoli- dated DRT Recommended atMTR 3. Restructuring of TPRI Review and recommend Not done NA new mandate and institu- tional set-up 4. Review of alternative To identify ways to over- Not done Proposal overtaken by source of recurrent come the chronic lack of events, as DRT funding for DRT GoT non-salary operating established Retention funds Scheme and cesses 5. Tea Research To carry out a review of Completed by FAO Contributed to the Privatisation public and private tea privatisation of tea research with a view of full research. privatisation of tea research  - 18 - Table 8A: Project Costs Appraisal estimate Actual/latest estimate (US$M) (US$M) Local Foreign Total Local Foreign Total costs costs costs costs Item A. Research Programmes Roots & Tuber 0.4 1.4 1.8 0.5 0.7 1.2 Cotton 0.6 2.4 3.0 0.6 0.5 1.1 Farming Systems 0.6 1.9 2.5 0.5 3.1 3.6 Sub-Total Research Programmes 1.6 5.7 7.3 1.6 4.3 5.9 B. Organisation & Management 1.7 5.6 7.3 0.6 7.8 8.4 C. Unallocated Research Fund2 Agricultural Research Fund 0.03 0.3 0.3 NALRM Research Priorities 0.6 3.9 4.5 Sub-Total Unallocated Research Fund 0.8 4.0 4.8 0.7 4.2 4.9 D. Vehicles, Equipment 3.0 3.0 2.8 2.8 & Materials Fund3 Total Base Costs 4.1 18.3 22.3 2.9 19.1 22.0 Physical Contingencies 0.2 0.2 0.4 - - - Price Contingencies 0.5 2.0 2.6 - - - TOTALI 4.8 20.5 25.3 2.9 19.1 22.0 Difference of US$ 3.3 million between estimation and actual (US$ million): 4.4 outstanding ADF contributions (now in process of disbursement) 0.8 withdrawal of UK contribution for Cotton Research Programme - 0.4 increased contribution from The Netherlands for Farming Systems Programme - 0.3 higher contribution from IDA in $-terms (SDR remained constant) - 1.2 increased domestic contributions 3.3 TOTAL. 2 Appraisal estimate including contingencies = US$ 5.7 million; unutilised funds refer to ADF. 3 Initial disbursement for urgent needs: US$ 0.6 million (only by IDA); the balance was eventually re- allocated for regular vehicles and equipment purchase under the various project components.  - 19 - Table 8B: Project Financing Appraisal Estimate (US$M) Actual/Latest Estimate (US$M) Local Foreign Total Local Foreign Total costs costs costs costs Source IDA 8.3 8.56 ADF1 8.2 3.8 Netherlands 2.7 3.14 United Kingdom 2.9 2.1 Germany 1.6 1.6 Domestic contribution 1.6 2.8 TOTAL 25.3 22.0 Table 9: Economic Costs and Benefits There was no net present value (NPV) or economic rate of return (ERR) estimated at appraisal. Therefore the NPV and ERR were not re-estimated by the ICR mission. An assessment of the project outcome is presented in the main report. As of October 98, US$ 4.4 million (U.A 3.5 million) remainundisbursed; credit closure has been extended until end-June 1999.  - 20 - Table 10: Status of Legal Covenants Original Revised Covenant Present Fulfilment Fulfilment Description of Agreement Section Type Status Date Date Covenant Comments Credit 3.03 (a) 3 NC Borrower to maintain GoT encountered serious 19700 TA armual contribution budgetary problems and did not to research budget at maintain 87/88 level in real term. 87/88 level in real terms. 3.03 (b) 3 CP Annual review of Annual review was done, but lack research budget with of agreement on feasible GoT IDA and donors. budget, though in 1996/97 and 1997/98 GoT increased it's budget on research. 3.03 (c) 3 C 12/1990 3/1992 After NALRM Complied with, and further preparation, review adjustment were made to better of budget in reflect zonal prioritisation. accordance with recornmendations. 3.04 (a) 9 C 05/15/89 05/15/89 Complete study to Completed. identify the site for the DRT Headquarters. 3.04 (b) 9 CD 12/31/89 12/31/89 Commence Head- Started September 95, completed quarter Construction. December 1997. 3.05 (c) 5 C 6/30/89 6/30/89 Establish National Complied with, but mandate to be Research Advisory adjusted following Committee. decentralisation. 3.06 5 CD 12/31/89 12/31/89 Integrate UAC into Completed. DRT. 4.01 (b) 1 C Have records, Complied with. including those for Special Account audited. Furnish audited accounts to IDA, not later than 9 months after end of financial year. Para 2 9 C 9/30/89 9/30/89 Submit Work Plan Complied with. (Sched. 4) annually by 9/30. Para 3 (b) 9 CD 10/31/91 10/31/91 Borrower to under- Complied with, but 18 months (Sched. 4) take mid-term review delay. with IDA and other donors. Para 4 (a) 5 C Borrower to Complied with. (Sched. 4) consolidate TARO, TALIRO and UAC into DRT. Status: NC: Not complied with CP: Partially complied C: Complied with CD: Compliance after delay  - 21 - Table 11: Compliance with Operational Manual Statements Statement Number and Title | Describe and Comment on Lack of Compliance Basically, there was compliance with the applicable Operational Manual Statement Table 12: Bank Resources: Staff Inputs Stage of Planned Revised Actual Project Cycle Weeks US$ Weeks US$ Weeks US$ Preparation to N/A N/A N/A N/A 85.2 201,000 appraisal Appraisal N/A N/A N/A N/A 59.3 117,000 Negotiations through N/A N/A N/A N/A 21.5 16,800 Board approval Supervision N/A N/A N/A N/A 227.5 578,800 Completion N/A N/A N/A N/A 9.7 44,500 TOTAL N/A N/A N/A N/A 403.2 958,100  - 22 - Table 13: Bank Resources: Missions Performance Rating Project Cycle Month! of DaysN Specialised Ismplemen- Development ProjecteayderMonth/ of Days in Staff Skills tation Objectives Types of Year Persons ield Represented Status Problems Identification 8/85 2 na na Preparation 11/86 4 20 ag, fa, ar, rm Appraisal 5/87 6 20 ag, ar, fa, Is, nr.- Post-appraisal 2/88 2 na ag, rm Negotiation 4/88 3 - DCA 2/89 Supervision 1 7/89 2 8 ag, ls S HS EF 2 12/89 1 2 ag TA, PM 3 3/90 2 8 ag, as U S AC, SM, LC, PR, 4 1-2/91 4 12 ag, rs, rel, as U S PM 5 6/91 2 14 tr, as AC, SM, PNI 6 9/91 2 7 ag, as U S 7 3-4/92 2 10 ag, as u S AC, SM, PM, LC, AL 8 7/92 3 10 ar, mi, pr S SM, PM, TA 9 11/92 9 11 ar, pr, ag, fa, es, as S S PM, SM 10 (M- R) 10/93 10 18 ar, ps, fa, as, Is, rm S S LC, SM, PM, CW 11 11/93 2 4 ar, es S S OF, SM, PM. RO 12 5/94 8 12 ar,pr,as,ls,fa,n S S OF, SM. REL 13 11194 3 4 ar, mn S S OF, SM, RRC 14 5-6/95 5 8 as, ar, ae, pr, nn S S SM OF. RRC 15 9/95 4 4 as, es, ae, S S SM, OF, RRC, AR 16 3/96 2 17 as, ar S S OF, AR, AP 17 11-12/96 4 13 as, ar, as, co S S OF. ADF 18 12/97 1 6 as S S OF, ADE OF, ADF Completion 11-12/98 3 14 ar, as, ae S s Abbreviations: Mission composition: ar: agricultural research; Is: livestock specialist rel: research-extension liaison rm: research management pr: procurement specialist specialist as: agricultural services fa: financial analyst tr: training specialist ag: agriculturalist es: extension specialist co: comnmunication specialist. Type of problems: AL: Late submission of audit report EF: Delay in project effectiveness OF: lack of operating funds RO: Lack of research output AC: Absence of Comnmissioner for REL: poor research-extension AR: non-compliance with covenant to Research and training linkage undertake annual review with donors SM: Poor Termrs of services and low RRC: unsatisfactory progress on and IDA staff morale research centres rationalisation AP: Accounting problems PM: Project Management ADF: Slow disbursement of ADF loan.  - 23 - TANZANIA NATIONAL AGRICULTURAL AND LIVESTOCK RESEARCH PROJECT (Cr. 1970-TA) APPENDIX A AIDE MEMOIRE   - 25 - TANZANIA National Agricultural and Livestock Research Project (IDA Credit 1970-TA) Implementation Completion Report APPENDIX A Aide-Memoire Introduction 1. A mission from the FAO/World Bank Cooperative Programme visited Tanzania from November 19 to December 4, 1998 to prepare the Implementation Completion Report (ICR) for the National Agricultural and Livestock Research Project (NALRP). The missionl worked closely with staff of the Ministry of Agriculture and Cooperatives (MAC), particularly the Department of Research and Development2 (DRD). Visits were made to Eastern and Northern Zones, and detailed discussions held with staffs of research institutes and centres,3 the Dean and staff of Sokoine University of Agriculture (SUA) and extension staff. In Dar Es Salaarn the mission also met with several industrial crop boards and representatives of The Netherlands, the United Kingdom (UK) and the World Bank (WB). 2. In April 1998, the Government prepared its own ICR in accordance with the WB procedure. This Aide-Memoire outlines the mission's preliminary findings; it was discussed at a wrap-up meeting chaired by the Permanent Secretary of the MAC on 3 December 1998. The mission would like to express its appreciation for the cooperation and assistance shown by the staffs of MAC and other organisations met. It would also like to record its appreciation for the excellent facilitation and organisation provided by project management. Project Objectives and Components 3. The overall objective was to implement the first stage of a long-term process of developing a re-organised and efficiently managed agricultural and livestock research system. To achieve this, the proj ect was to provide financial and technical assistance to support: a) Consolidation and downsizing of existing services to establish a new organisational structure for the national agricultural and livestock research system (NARS) within MAC; 1 Guy Evers (Mission Leader), Martin Maurer (Economist) and Michael Macklin (Agric. Services, Consultant) 2 The Department of Research and Training (DRT) was created in 1989. In 1998, the Training Section was separated and DRT re-named as the Directorate for Research and Development (DRD). Ilonga Agricultual Research Institute (AR1), Mlingano ARI, Tanga Livestock Research Centre, Tsetse and Trypanosomiasis Research Centre (Tanga), Selian ARI, Lyamungu Coffee Research Centre andlrengeru Horticultural Research Centre.  - 26 - b) Preparation of a National Agricultural and Livestock Research Master Plan (NALRM), in accordance with national policies, to identify specific research activities ranked in order of priority and define the human, physical and financial resources required for each activity; c) Training of agricultural and livestock research staff to meet needs for research management and methodology and to support priority research topics; d) provision of vehicles, equipment and civil works to meet immediate research needs; e) Cotton, root & tuber and farming systems priority research programmes which had already been identified and prepared; and f) Establishment of a research fund for crop and livestock research programs for which priorities would be established in the proposed NALRM. 4. The Staff Appraisal Report anticipated that the project would result in major improvements in the organisation and management of the NARS, leading to more efficient use of limited resources through introduction of prioritised programme planning. In addition the project would enhance the quality of experimental work, sharpening the focus on farmers problems through improved research-extension- farmer linkages. Project design was based on a good diagnosis of critical issues facing the NARS at the time of project appraisal. Project objectives responded directly to these and were correctly focused on fundamental needs for institutional consolidation and restructuring, staff training, and improved research planning and management. Institutional reform objectives were ambitious and it was realistically recognised that full achievement of objectives would be a long-term process requiring several investment phases. Project Timing and Financing 5. The project was identified and prepared by the WB in 1985/86, and appraised in 1987/88. A Project Preparation Facility (PPF) was approved in 1986. The project was co-financed, through parallel funding arrangements, by the African Development Fund (ADF), the Netherlands, the United Kingdom (UK) and Germany. The bilateral donor funding was pledged under the umbrella of the Special Programme for African Agricultural Research (SPAAR). The IDA Development Credit Agreement (DCA) was signed in February 1989 and became effective six months later. Bilateral donor support became effective during the second half of 1989, but the ADF loan only around mid-1990 with first disbursement about two years later. A Mid-Term Review (MTR) was launched in April, 1993. There were 9 IDA supervision missions before the MlR, and 8 afterwards. The DCA was amended three timesl. The project initial closing date was March 31, 1997 and was extended to December 31, 1997, with last disbursement in January, 1998. The ADF Loan is still open and due to close June 30, 1999. Towards the end of NALRP implementation, a phase 2 IDA-funded project (Tanzania Agricultural Research Project - TARP II) was appToved and became effective in June 1998. in October 1993 (following the Mid-Term Review, with retro-active effectiveness from July 1993), Febraary 1995 and June 1996.  - 27 - Achievement of Objectives 6. Despite considerable implementation delays and limited progress in the first four years, the project largely achieved its core objective of reorganising and developing a more efficiently managed NARS. Overall, the project has completed its intended institutional reforms and laid a good foundation for more effective client oriented research in the future. However, significant improvement in the generation and adoption of demand driven technologies has still to be realised. Implementation Experience and Results Restructuring of the National Agricultural Research Services 7. Major restructuring in terms of the dissolution of three of the four agricultural research parastatalsI and their integration within the Department of Research and Training (DRT) of MAC was largely completed during 1989/90. However, staff re-deployment and re-appointments took longer than anticipated. Stemming from NALRM recommendations, the consolidated research structure comprised seven research zones2 and national headquarters in Dar Es Salaam. The planned downsizing of the system has been partial and is still ongoing. Support staff numbers have been reduced from around 3,025 to 2,250 (25% cut), but the number of scientists has remained more or less constant (i.e. some 350), leaving a scientist to support staff ratio of 1 to 6.5. Out of some 50 original stations, 22 institutes/centres were identified as priority DRT research stations and 13 others were retained as field centres or trial sites. The process of closing or handing over the five remaining sites is underway. In some cases, resource distribution between zones and institutes has remained in-equitable with marked variations in both physical and human resource endowments. 8. Following the establishment of DRT, the project was to support the formation of financial, planning and administrative units at headquarters. This has been achieved and accounting, administrative and project management services are in place and perforning satisfactorily. Procedures for the development of annual work plans and budgets were established, and subsequent annual plans and budgets regularly submitted to IDA for review. 9. As planned, a National Agricultural Research Council (NARC) was established to be the principal advisory body on agricultural and livestock research to MAC. Attempts to provide the Council with statutory powers were not realised and, though the Council did function, there is little evidence that it has had a significant impact on agricultural research policy development. Its future mandate in the context of strongly de-centralised zonal research services would need to be reviewed. National Agricultural and Livestock Research Master Plan (NALRM) i.e. Tanzania Agricultural Research Organisation, Tanzania Livestock Research Organisation andUyole Agricultural Centre. The Tanzania Pesticide Research Institute was not dissolved/integrated within DRT. 2 Noriem, Central, Eastern, Southern, Western, Southern Highlands and Lake Zones.  - 28 - 10. The NALRM funded by the SPAAR donors and coordinated by ISNARI, was intended to develop a comprehensive plan that reviewed and defined a national agricultural and livestock research program, identifying ranked priorities for the short, medium, and long terms, and specifying staff and physical resource requirements. In early 1991, a draft NALRM was circulated for intemal and extemal review, finalised in December 1991 and launched at the March 1992 workshop. The plan established national research priorities on a commodity basis, which determined research resource allocations under the project. Concurrently an Organisation and Management (O&M) Study was undertaken with support of the UK, which inter alia provided recommendations on the new zonal research structure. However, IDA resources from the "Research Fund Component" to support NALRM implementation were only allocated and distributed to concerned centres in 1993, after completion of the MMR (i.e. 18 months after issue of the NALRM). Most NALRM recommendations were endorsed by GoT including those for restructuring and downsizing the NARS. There was, however, a disappointing response from donors, as few pledged incremental funding to support the Master Plan proposals. 11. The project succeeded in preparing a comprehensive NALRM, in better directing resources on a rational and prioritised basis and in developing a framework for zonal research. Subsequently DRT commendably undertook a further research prioritisation exercise for each zone, injecting a greater degree of client involvement into research planning. Whilst NALRM undoubtedly proved a useful foundation tool, it was superseded by the zonal planning and prioritisation exercise. Civil Works 12. The SAR made provision for constructing a new DRT Headquarters and for the station/institute rehabilitation and construction needs which were to be recommended by the NALRM. IDA funded the new DRT HQs (1,700 m2) in Dar Es Salaam (completed by end-1997), and the infrastructure improvement of 9 research stations2. About 50% of the planned IDA-funded construction programme was completed at a total cost of US$ 3.1 million; partial completion was due to delays in tender procedures and contract approvals, and inadequate consultancy services, which together resulted in substantial cost escalation. Infrastructure improvement of 6 other stations, funded by ADF, was only initiated at the end of 1997 and is expected to be completed by mid-1999. Staff Development and Training 13. The project largely met its training targets set by the SAR. Some 21 staff attended long-term training3, and about 660 staff attended short courses which were conducted mainly locally and in the Region. The short-term training focused on O&M, in line with the project objectives. In addition, during NALRP implementation, some 25 postgraduate scholarships and 50 short-term courses were also offered International Service for National Agricultural Research. 2 including water and electricity supply, the rehabilitation and/or construction of 38 houses, two workshops, one ware- house, one laboratory and three office blocks, which were completed by the end of the project. 1 BSc, 16 MSc and 4 PhD, mostly overseas, funded by ADF. In addition, The Netherlands funded training in FSR which were not anticipated at appraisal.  - 29 - by various donors and organisations. This component was well planned covering priorities identified at appraisal for scientific, management and support staffs, and was successfully executed Vehicles and Equipment 14. This component was intended to meet immediate requirements of vehicles and equipment, prior to issue of the recommendations of the NALRM and O&M studies. About 60% of IDA procurement was completed within the two first years of project implementation, and the remaining funds were utilised from March 1993 onwardsl. Although the objective of rapidly procuring vehicles and equipment was only partly met, this overall provision greatly facilitated research activities as recommended under the NALRM and in the subsequent zonal research prioritisation. Priority Research Activities on FSR, Cotton, and Root and Tubers 15. Funded by the Netherlands, one central and seven zonal FSR teams were established to assist researchers to better identify and address farmers' priority needs through an integrated approach, to strengthen farmers-extension-research linkage and, linkage between the commodity research activities within DRT. Most crop commodity scientists, and some livestock and soils scientists, attended training on FSR methodologies. Increased participation of farmers in technology generation was promoted through the Client Oriented Research (COR) methodology and on-farm research. On project completion, the zonal FSR teams were disbanded as step towards intemalising the FSR and COR approach in all DRT research activities. The FSR team is to be replaced in each zone by a small socio-economic backstopping team and technical staff redeployed into commodity programmes. The cotton and root and tuber programmes were to be funded by ADF and the UK (cotton only) from the on-set of the project, without waiting for the NALRM recommendations. However, ADF disbursement delays and withdrawal of UK support to cotton resulted in little research activities or output. Unallocated Agricultural Research Fund 16. US$ 5.7 million was initially provided under the project (IDA and ADF). These resources were primarily intended to develop research capacity at selected stations to undertake priority research programmes identified by the NALRM and secondly to provide seed funding to a national contract/collaborative Agricultural Research Fund (ARF). After the 1993 MTR, IDA funds were allocated for expenditures on civil works, equipment and non-salary operating costs, leaving US$ 148,000 for contract/collaborative research. 17. The ARF constitution was issued in 1992 and the first projects started in 1994. The aggregate fund amounted to US$ 210,0002. In addition, GoT provided US$ 30,000 to meet ARF overhead costs. A total of 16 collaborative research projects were approved for a total budget of about US$ 380,000. At Overall, some 70 4-WD vehicles, 50 motorcycles, three water pumps, 15 tractors, 25 computers, 4 photocopiers and miscellaneous equipment and material were procured with IDA fumding. 2 i.e. US$ 148,000 from IDA and 62,000 from ADF.  - 30 - IDA Credit closing date (December 1997), ARF had released US$ 183,000 for research projects' implementation, leaving a balance of US$ 170,000 to be funded by other sources such as TARP II. ARF performance was not fully satisfactory. Implementation was delayed, review procedures proved cumbersome and time consuming and, as a result, research output was limited at project completionl. Competitive contract research was not tested under the project as anticipated. It had been hoped that the ARF would attract incremental research funding from other sources and develop into a self sustaining "consolidated funding mechanism", but this did not occur under NALRP implementation. Despite these shortcomings the ARF succeeded in exposing DRT scientists to collaborative research, especially with SUA, which significantly improved overall DRT/SUA linkage in the process. It also served as a useful scientist-training tool by demanding high quality research proposals from applicant scientists; and a useful pilot activity for testing fund management and procedural options. Technical Assistance and Studies 18. Technical assistance focused on supporting the preparation of the NALRM (funded by Gernany), the O&M studies (funded by the UK) and the FSR programme (funded by The Netherlands). IDA funded TA to strengthen research management, including inter alia the establishment of INFORM (16 person-months) and support for financial management and assets verification (24 person-months). Overall, TA has had a positive input in facilitating the NARS restructuring, but not allTAs performed satisfactorily. Total investment in TA are estimated at around US$ 3 million. The mission was concerned that the TA output did not appear commensurate with this high level of investment, in particular few recommendations of the O&M study have been approved. The MTfR recommended three additional studies be carried out on: (i) TPRI restructuring; (ii) the review of alternative source of recurrent funding; and (iii) a review of private and public tea research. The first two studies were not completed, and the third study, prepared with FAO assistance, contributed to the privatisation of tea research. Extension-Research Linkages 19. Formal links between research and extension were built into project design. These included appointment of research-extension liaison officers, monthly workshops at which research and extension staff would interact on research recommendations and feedback from extension and annual zonal research extension workshops for joint research and extension programme planning. In 1992 Zonal Technical Committees (ZTC) were formned to further enhance client influence on research and extension planning. The combined impact of these initiatives has been limited. Operating fund provisions under both research and extension programmes were inadequate to conduct the proposed joint activities. Better linkages were realised under the FSR programme under which operating funds were available. The programme succeeded in testing FSR approaches involving greater farmers' participation with DRT scientists in their respective zones. At the beginning, many proposals submitted to ARF were of poor quality and had to be refonnulated. In early 1998, the review procedures were simplified, and deadlines for each step established.  - 31 - Donor Coordination 20. The involvement of SPAAR was intended to enhance and institutionalise donor coordination. Tanzania was also selected as a pilot for implementing the "Framework for Action" intended to establish a common funding mechanism for DRT. Although there were some coordination problems at proj ect inception amongst SPAAR donors in initiating the NALRM, better coordination was generally achieved during project implementation through the establishment of a GoT-Donor Consultative Group and regular exchange of information. The establishment of common funding mechanisms did not take place. Management Information Systems, Documentation and Communication 21. The project has started to improve scientific information and documentation services through the procurement of computer facilities, the establishment of a Tanzanian Agricultural Research Data Base (TARD), and access to a number of national and international information networks. The INFORM package, a research MIS developed by ISNAR, was introduced at DRT Headquarters in 1992/93 to assist in preparing work plans, budgets, resource allocations and summaries of on-going research. INFORM has also been established at the zonal centres. However, further INFORM and TARD software upgrade and more regular data base updates are essential to sustain effective operations. Communication between zonal centres and headquarters was substantially improved through telephone/fax and Email systems, though not yet at all zonal centres and stations (i.e. Ilonga, Tumbi and Mpwapwa zonal centres). Financial Aspects 22. The IDA Credit of SDR 6 million (US$ 8.3 million at appraisal) was fully disbursed by December 31, 1997. There were, however, substantial delays until 1993, at which time the disbursement lag was 74%. Thereafter, there was significant improvement in the pace of disbursement. IDA contributed to institutional restructuring, construction of HQs building, infrastructure construction and/or rehabilitation of nine research stations, and provision of vehicles, equipment and materials. In accordance with the NALRM recommendations, initial civil works, vehicle, consultancy and training Categories of Expenditure were increased, through allocation from the "Research Fund" component. Total revised project costs as per category of expenditure are as follows: (i) civil works SDR 2.27 million; (ii) vehicles and equipment: SDR 1.36 million; (iii) consultancies and training SDR 1.24 million; (iv) non-salary operating cost (new category of expenditure): SDR 0.42 million; ARF: SDR 0.1 million; (v) project operating costs: SDR 0.36 million; and (vi) the PPF: SDR 0.2 million. Actual project costs for IDA are US$ 8.6 million, in line with SAR estimates. 23. It is anticipated that the ADF loan will be fully disbursed by its expected closing date (June 1999). The total project costs are estimated at US$ 26.2 million, largely in line with the SAR estimates. Project financing can be summarised as follows (in US$ millions): GoT: 2.8, IDA: 8.3, ADF: 8.2, the Netherlands: 3.1, UK: 2.1 and Germany: 1.7.  - 32 - 24. There were major shortfalls in the Government's counterpart funding, particularly for non- salary research operating costs (see para 27). 25. Project management took a number of commendable actions to mitigate the chronic operating fund crisis. The concept of Revenue Retention was approved by the Treasury for fiscal year 1994/95, and subsequently implemented at nearly all stations. Although business and revenue generation potential vary considerably between location, the scheme has been instrumental in generating income to maintain essential operations and maintenance of the stations. According to DRD, some TShs 412 million were collected between 1994/95 and 1996/97. Revenues have been mainly generated from production sales, laboratory services and rental of surplus land and buildings. Additional operating funds (and staff incentives) were increasingly generated through contract research with development projects, collaborative research with International Agricultural Research Centres (IARCs) and their regional networks, and from consultancy income. 26. In 1995/96, agreements were reached with the major agro-industries to contribute to research funding through cesses or annual allocations for eight cash crops, namely coffee, tea, cotton, cashew, barley, sugarcane, tobacco and sisal. Cesses are also used for providing incentives to the concerned research staff. Tea research was fully privatised in 1996. There are comparable privatisation initiatives being discussed for tobacco and coffee. In 1996/97, the contribution from agro-industries was estimated by DRD at US$ 3.2 million. Project Sustainability 27. At this stage, project sustainability will be ensured largely through the already effective TARP II and through continuing support from bilateral donors. Throughout NALRP implementation, a critical shortage of non-salary operating funds was a major issue and GoT was unable to meet its' commitment to maintain operating cost provisions at 1987/88 levels in constant terms (as per SAR). The ability of GoT to provide adequate operating cost funding for the NARS in the future remains an issue. Although increased use of industrial crop cess funds has propped up research on these commodities since 1995, current arrangements are established through various arrangements such as Memoranda of Understanding, legal instruments and informal agreementl. To ensure sustainability of cess funding, it is desirable that all industrial crops be covered by statutory arrangements. GoT is likely to remain the main source of funding for most food crop and livestock research for which there are no cess opportunities. 28. Through force of circumstances and pro-active encouragement of project management, researchers have adopted innovative means of directly raising alternative external and internal funds. Latterly better managers have been successful in doing this, and dependence on central revenue funding has been reduced to some extent. Rigorous financial control and auditing is essential to ensure transparency and to protect them from possible abuse and criticism. Current occasional and irregular Govemment audit practice appears inadequate for this purpose and alternative stricter controls are I crops already covered by legal statutory agreement are: tea, coffee, cashew and cotton.  - 33 - needed. Cost recovery on laboratory services has improved but there is potential to further rationalise and expand income generation as most laboratories are operating at levels below their capacities. 29. A reliable internal capital replacement scheme for vehicles and equipment is lacking. Vehicles are rarely replaced from GoT sources and replacement is normally donor dependent. Whilst not an immediate problem this is a long term issue. Bank and Government Performance. 30. The project overall design was adequate, though the phasing appeared over-ambitious, especially with respect to priority research activities which were to be initiated concurrently with major institutional restructuring and reforms. The MTR, though delayed, provided an in-depth analysis of the problems and close follow-up of an agreed action plan contributed to improving project implementation afterwards. Bank Supervision was generally intensive, maintained momentum and provided useful guidance. However, it was unable to resolve the persistent issues of inadequate operating funds and low staff morale raised by almost every supervision mission, or prevent implementation delays in the first four years of the project. SPAAR played a positive role in encouraging donor coordination and assisting in research management. 31. After a slow start, MAC performed satisfactorily in implementing an ambitious and difficult restructuring of the NARS. Project management was weak during the early years of implementation because of lack of day-to-day leadership, including the delays in replacing the Commissioner for Research and Training. After creation of an effective project implementation unit in 1993, there was marked improvement in overall implementation. There were significant delays in the rationalisation of the research station network and reluctance by staff toredeploy to their new station. The separation of the Training Section from Research, recommended at the MTR, was only effective after project completion. Government has failed to provide adequate non-salary operating costs and to establish attractive and performance related staff terms of services. However, MAC introduced innovations to alleviate financial constraints through retention and cess funds. Assessment of Overall Outcome. 32. Overall the project satisfactorily achieved its primary objectives although the restructuring exercise and downsizing of the system is still to be completed. The project was also successful in introducing several un-planned but commendable innovations, in particular for securing alternative source of funding. Issues of financial sustainability and staff morale and motivation persisted throughout the project and remain. As a result the considerable human resource base of 350 highly qualified research scientists is not yet fully utilised and research productivity can be considerably increased. 33. The SAR did not attempt to quantify the expected project impact on agricultural production. An ERR was not estimated because the project was primarily concerned with institutional restructuring and capacity building. During project implementation new technologies were generated, of which some were reportedly widely adopted by farmers. Technology generation was biased in favour of variety  - 34 - releases for most commodity programmes (e.g. maize, beans, wheat, cotton, sugar cane, tobacco and pyrethrum). Only a few technologies were released that address crop and livestock management issues. An increasing level of on-farm participatory research (COR) has been initiated over the last years. The improved capacity of DRD was also reflected in increased collaboration with IARCs and regional networks. DRD is also publishing a newsletter, and recently jointly launched with SUA the Tanzanian Journal for Agricultural Sciences (TAJAS), whose first issue was published earlier this year. Future Operations. 34. This ICR is being prepared at a time the follow-up TARP II is already effective. Many of the issues reported under NALRP are being addressed under TARP II for which a detailed Project Implementation Plan has been prepared. This includes steps to strongly decentralised research to zones and significantly empower farmers through establishment of Zonal Executive Committees, Zonal Agricultural Research Funds, COR and through direction of 50% of activities to on-farm research. In addition, the mission has developed the below recommendations which may be considered under TARP II implementation; these include: a) Increased cost recovery should be sought by: (i) further rationalisation of laboratory services (soils, crop protection, food science, etc.); (ii) full privatisation of industrial estate crops research such as tobacco, coffee, sisal and sugar; (iii) contribution from cess funding to the cost of personal emoluments of their commodity scientists with the savings retained by DRD to fund other staff performance incentives and/or operating costs (e.g. for food and horticultural crop, and livestock research); b) Until a statutory framework for cess funded research is established, there will be uncertainty on the cess rate, fund inflow and their continuation. A legal framework to cover all industrial crops should be developed as soon as possible. c) Initial success in developing a more risk taking and innovative research management style should be fostered with direct fund raising from non-govemrnent sources being an important function of decentralised, zonal centre management. In this context caution may be exercised in the design and operation of centrally managed performance incentive schemes which are notoriously hard to manage obj ectively within the public sector. d) The mandate of the National ARF will require review in the light of the proposal to create Zonal Agricultural Research Funds (ZARF). e) The recent reorganisation of extension services (i.e. transfer to local governments) is likely to render the strengthening of currently weak research - extension - farmer linkages more difficult, at least in the transition period, and innovations to mitigate this risk should to be sought. f) With its new "development" content, DRD may develop strategies for a broader R&D function rather than just research. This is fully consistent with increased emphasis on COR and on-farm research, and the creation of de-centralised, zonal research funding mechanisms. The new "development' function needs to be clearly defined and explained to DRD staff. It is also important that DRD gives increasing emphasis to socio-economic and marketing issues, and to forging stronger links between all stakeholders, including farmers, NGO's, private sector and others through collaborative and co- funded research.  - 35 - g) Though FSR and COR have been formally intemalised by DRD, these approaches are still to be fully integrated in commodity and cross-cutting programmes and there is need for further training and support in this area. h) Livestock research, especially small-stock, is still under-represented in some zones and deserves strengthening as this sub-sector can have significant impact on national priorities for improving nutritional standards and in poverty alleviation. Lessons Learned 35. The DRD skill structure is dominated by crop-related scientists and is lacking in disciplines such as socio-economics, marketing and livestock. As COR methodology is increasingly adopted, emerging zonal research priorities may not be matched by the existing skill mix at either zonal or national level. Therefore, there is a need for greater flexibility in adjusting existing staff establishment and future recruitment, and through increased contract and collaborative research with outside bodies (e.g. SUA, Forest Research Institute, NGOs, etc.). 36. Adversity and financial crisis has spawned more innovative, and risk-taking management, suggesting that "shock treatment' may sometimes be necessary to induce changes in entrenched management structures that have historically been fully dependent on a centrally budgeted (usually inadequate) funding source. 37. There has been a significant reduction in the number of research institutes, centres and stations, and cut in support staff. However, the rationalisation and consolidation of DRD should be seen as a dynamic process, requiring further adjustments to improve efficiency and effectiveness. This process will require further staff re-deployment and increased use of common laboratory services. Strong inter- zonal coordination will be essential to avoid duplication and achieve synergies. 38. Given the size of the country, transport and communication costs are high. Further investment in management information and communication systems will facilitate coordination and collaboration between the zonal centres for commodity and cross-cuffing programmes, and facilitate access to international and regional research and development networks. 39. The promotion of COR, which includes active dialogue between farmers, advisory services and researchers in all stages of the R&D process, from the identification of problems and causes, to the selection, on-farm testing, adaptation, monitoring and evaluation (including labour and economic aspects) of possible solutions should result in higher greater adoption of generated technologies. Follow-up 40. The mission will prepare a preliminary draft ICR to be forwarded to IDA by mid-December, 1998. This will include further analysis and fine tuning of data. The final draft version of the ICR will be submitted to the Bank by early January 1999, after incorporating comments from the FAO-CP intemal review process.   - 37 TIIE IINITED REPUBLIC OF TAN7ANM MINISTRY OF A GRICUL TUIRE AND COOPERA TI VES Department of Research aid Traninii,gt NATIONAL AGRICULTURAL AND LIVESTOCK RESEARCII PROJECT (NALRP) IMPLEMENTATION COMPLETION REPORT (ICR) April 1998   - 39 - NAT)IONAl, A(CRICUClIlTUE ANI) IVE,STOCK RIFSEAIRCII IPROJECT (NALRP) I MP 1' L1FN IENTATION CO'iM PLETION RE PORT (ICIR) OUTLINE Page Accrotnyits .......... i 1. itrodluction: .I * (icral.. I * Rationale ot NAI ..................................................................................... I * Majior ('otilnponiclus of' NA I, I . ......................................... 2 2. Review of Progress Made in the Implementation of NALRP: .. 4 2.1 C tonsolidation and Restructuring of the DRT ............................................... 4 2.2 Priority Setting ............................................... 6 * National Level ............................................... 6 * Zonal Level ............................................... 7 2.3 Proctiremc nt of Essential Goods and Services ...........................................8 * Procurciiement of Goods ......................................................8 * C'ivil Works ...................................................... ') 2.4 1 luman Resources Development and Training ........................................... I * Staff Rcdeploymcnt ........................................................1. I0 * Staff l'r-aiiiiiig ....................................................... I (0 2.5 Finianlcial Managemenlt and Researcih Funding Aspects .............................. 1 2 * Projects Accounts, Disbursements and Audits .................................. 1 2 * Establishment of Sustainable Funding Initiatives .............................6 I 2.6 Informatics and Documentation ................................................20 * Information and Documentation ............................................... 2(. * Management Information System .............................................. . '1 2.7 Linkages and Partnershiips .................................. .23 2.8 g ajor Conistrainsts.- .. 3. Forward Outlook ........... 28  - 40 - 4. Athahmenilai, ts: r: Status of Implementation of NALRP Components and Achiievements ........................................ 29 II: Rationalized Institutes and Centres of the DRT ............................. 34 III: Status of Procurcnilcnit of Goods ...................................... 35 IV: Scope of Civil Works and Implementation Status .......................... 36 V: Sulmmary of Slort and Loiig-Tcmi Training ................................. 38 VI: Disburseimient or l'roject Funds ............................................. 40 VII: List of Approved ARF Projects .................................. 41 Vill: Crop Levies and Estimates of Income Generated .......................... 43 I'1.,.  - 41 - ,ICRON lIS AINDI ARlBRIJ'1ATIONS AAIRNET - Animal Agricultuve Research Network A C - Assistant Commnnissioner, MoAC A 41)1 - African DC1evLelopme('nt Bank AMOl ' - Af/ican Dev'elopmnent Funtd ADRIJI - Alniaii I)iseases Research Institute, Tanzania AG(;RIS' - Inlterna(ltionll lin rinaitin Sisicinfi o Agricultiral Sciences and(l T'chnologiv Al' - Agricultural Research l'mnd A SIRlE( A - Association, for Strengthening Agricultural Research Al?! Agricultl lr Rese 'arch Institute AR'77 - Agricultural Research and( 7aining Institute .4 IFP - Animal Woork Plan RARAVESA - Banana Research Net-work for Eastern anid Soutther-n Africa CAlI - CA B-International (formerlv 'o Cnmmonwealtlh Agricultural Bureau) C'A() - C'ief Adotinistrtwive Officer CALD - CoYmniissionerfo-r A4griculttrae and Livestock Developmenit CA Rl15 - Ctu rrent Agricultural Research Information System C'47,5 - C'entre for Arid Zonie Studies C'D -ROM - Compact-Disc Read Only Memory C'S)8/IISI - ( Compqmterized OCumnentatian Service/Integrated Set of informati(ol Services ( 1 ( ' - (oininissioi of' tile Eurapean Co(maiunities ('TAI - (Cansolidalted Fund(ling Mechanism ('C14 R - C'onvultative Group o Itnternational Agricultural Research CIR/ I D - International C'entre for Agricultural Researclh and Development. Falice C()"'l'ECII - C nommissionfiw- Science andtd 7echnjology C7Ti - Technical Centrefink Agricultural andc Rural Cooperation CTB? - Central Te'ndler Board 1)11 '1) - 1)(epartmen:t 6I 111ter1n1a1tiona(l Oevelopment )I?lT - I)e(pcrtinci o/ I? (ir'Irc and 'Irnining, AfoA( 1T)TC( - l)epartmental Tender Technical Coniunittee ECA RSA A! - East and( C'entral Afiican Reseacrch Network on Sorghumir and Millets Ecahren - East aned Central Alfirican Bean Research Network ERP - Economic Recoverv Programinme FAQ0 - Food and Agriculture Organization of the United Nations FSA - Fariming Systems Appproach FISR - Farming ,Systems Research FY - Financial Year GDP - Gross Domestic Product GIS - Geographical Intlrination System GoT - Government of Ta7,-ania (;7 G (,ermanvy Technical Axsistanmce Agencly IA R( ' - International Agric rlowal Research C(eln tre C/(1 - International Comiptitive Bidding 11)4 - International Dev'lwinment Association I)11 - Information ind l)ocumnenwation Unit-l)RT, MoAC ll)RC ' - International Dev'lopnment Researilch Centre INFORM - Inforiniation for Agr icuiltura(ll Research Manaiagemilent IAFORA !-R - Info rmation.for A gricultural Research AManagemient- Revised ISNAR - International Service for National.4gricultural Research LC - Letter-s of Credit AIIS - Management 111finfmation Sy-stems lvo,'l C - A'tinisty qo'Agriti ttIre and CooPeratives  - 42 - MoU - MeImorandum of Understanding MT - Mfanamgeme'nt Tleam AI TTC - Ministerial Tec/hiic(l Tend(er- Committee MTR? - Miid-Terin Reviewi' NALRM - National Agricultural andl Livestock Resealrch Masterplan NALRP - National Agricultural and Livestock Research Project NARF - National Agricultural Research Ftud NARC - National Agricultural Research Council NARS - National Agricultural Research Syvem NGO - Non-Governmental Organixation NORAD - Norwegianl Agencyfor International Development NRI - National Resourses linstitute, UK ODA - Overseas Developiment Adlministration, UK O&M - Organisationi and Management PMEU - Planning, Monitoring and Evaluation Unit PME - Planning, Monitoring and Evaluation SACCAR - Southern African Centrefor Cooperation in Agricultural Research and Training SADC - Souitherni Alfiican Developmenit Community SARRNE T - SADC Regional Rootcrop Research Network SCOPO - Stand(ing Conmnitteefor Parasiatal Organisations SDI - Selective Dissemination of Information SIDA - Swedish International Development Agency 15 - SPAAR ln/i'irmnation S)lstein SKR - Swvedish Kr-oinor- SPAA R - Special Prvgranunefor Afiican Agricultural Research SUA - Sokoine University oJ Agriculture, Tanzania SUN - SUN Accounting Systemn TA FORI - Tanzania Forestmy Research Institute TAI.IRO - Tanizania Livestock Resear ch Organization TARD - Tanzania Agriculiural Researcih Database TARO - Tanzania Agricultural Research Organization TARP 11 - Tanzania Agricultural Research Project - Phase 11 TDT - Technology Development and Transfer TPRI - Tropical Pesticides Research Institute UAC - Uyole Agrictiltural Centre UDSM - University of Dar Es Salaam UNESCO - United Nations Educational, Scientific and Cultural Organization VEO - Village Extension Officer ZAC - Zonal Advisory Committee ZTC - Zonal Technical Committee ZOPP - Ziel Orientierte Projekt Plannung (Goal Oriented Project Planning)  - 43 - NA 7710NAL. AG(iRI('CtI.T7'11IIAIL AND) 1.11'I1.'S7TCK RESEA RCHJ PROJEC(T (NI,LRII) IMPLEMENTATION COMPLETION REPORT (ICR) INTIRODUCTION 1.1 GENERAL Tlanzania cconiomily depends lieavily on the agricultural sector, which accounits for about 60%) of the GDP, over 61 % of the export eamiings and about 84% of rural employment. The majority of the peopic (>90%) live in the rural areas and are cigag(gd in subsistence farminig, withi an average of 1.2 liectares per household. Alilotigil the national population lhas growil by about 2.8% per annum in the last 1() - 15 years the overall agricultural production has kept pace with the increasing populilation growth, with exccption of drought years when food is imported lo mieet shortage. This has been attributed to horizontal expansion of the cultivated arca and improvements in vertical productivity of several food and commercial crops. Notable gains reported in the last 10 years (1985/86 - 1995/96) include pro(LIctivity increases (anilual averages over the 10 years) in maize (4%), rice (3%',`), cotton (4%1,) and tea (6%