Document of The World Bank Report No: 30930 IMPLEMENTATION COMPLETION REPORT (PPFI-Q0740 IDA-30360) ON A CREDIT IN THE AMOUNT OF US$ 21.8 MILLION TO THE UNITED REPUBLIC OF TANZANIA FOR THE SECOND AGRICULTURAL RESEARCH PROJECT ( TARP II) December 28, 2004 AFTS2 Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective June 2004) Currency Unit = Tanzania Shilling (Tsh.) I Tsh. = US$ .001 US$ 1 = 1,107 Tsh. FISCAL YEAR July 1 - June 30 ABBREVIATIONS AND ACRONYMS ADB/F African Development Bank/Fund ADRI Animal Diseases Research Institute ASARECA Association for Strengthening Agricultural Research in Eastern and Central Africa ASDP Agriculture Sector Development Programme ARF Agricultural Research Fund ARI Agricultural Research Institute CIMMYT International Centre for Maize and Wheat Improvement CGM Cassava Green Mite CMD Cassava Mosaic Disease COR Client Oriented Research CCPP Contagious Caprine Pleuro-pneumonia DANIDA Danish International Development Agency DLRT Department of Livestock Research and Training DRD Department of Research and Development DFID Department for International Development EZCORE Eastern Zone Client Oriented Research and Extension FAO Food and Agriculture Organisation of the United Nations FRGs Farmer Research Groups FSA Farming Systems Approach FSR Farming Systems Research FSZ Farming System Zone GIS Geographical Information System GLS Grey Leaf Spot GoT Government of Tanzania GTZ German Technical Assistance Agency ICRAF International Centre for Research in Agroforestry ICRISAT International Centre for Research in the Semi-Arid Tropics IFAD International Fund for Agricultural Development IFPRI International Food and Policy Research Institute IPR Internal Programme Review ISNAR International Service for National Agricultural Research LD Lethal Disease LGMC Livestock Germplasm Maintenance Centre M&E Monitoring and Evaluation MDG Millennium Development Goal MIS Management Information System MoU Memorandum of Understanding NAEP National Agricultural Extension Programme NALRP National Agricultural and Livestock Research Project NARF National Agricultural Research Fund NARP National Agricultural Research Programme NARS National Agricultural Research System NCD Newcastle Disease NCB National Competitive Bidding NGO Non-Governmental Organisation NRM Natural Resource Management NT/CT No-Till/ Conservation Tillage PAD Project Appraisal Document R&D Research and Development SACCAR Southern Africa Centre for Coordination of Agricultural Research SIDA Swedish International Development Agency SFI Soil Fertility Initiative SUA Sokoine University of Agriculture TARP Tanzania Agricultural Research Project TaCRI Tanzania Coffee Research Institute TORITA Tobacco Research Institute of Tanzania TCB Tanzania Coffee Board TDT Technology Development and Transfer TRIT Tea Research Institute of Tanzania TPRI Tropical Pesticides Research Institute TSZ Tanzania Short-Horn Zebu Tsh Tanzania Shilling UDSM University of Dar es Salaam VEO Village Extension Officer ZARF Zonal Agricultural Research Fund ZDRD Zonal Director of Research and Development ZILO Zonal Information Liaison Officer ZTC Zonal Technical Committee ZRC Zonal Research Coordinator ZRELO Zonal Research and Extension Liaison Officer Vice President: Gobind T. Nankani Country Director Judy O'Connor Sector Manager Karen Mcconnell Brooks Task Team Leader/Task Manager: Ladisy Chengula TANZANIA Tanzania TARP II CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 5 5. Major Factors Affecting Implementation and Outcome 15 6. Sustainability 16 7. Bank and Borrower Performance 18 8. Lessons Learned 20 9. Partner Comments 21 10. Additional Information 21 Annex 1. Key Performance Indicators/Log Frame Matrix 22 Annex 2. Project Costs and Financing 24 Annex 3. Economic Costs and Benefits 26 Annex 4. Bank Inputs 28 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 30 Annex 6. Ratings of Bank and Borrower Performance 31 Annex 7. List of Supporting Documents 32 Annex 8. Achievements from Research Programs 33 Annex 9. List of Selected Case Studies 35 Annex 10. Executive Summary of the Government ICR 41 Map IBRD 29232 Project ID: P002804 Project Name: Tanzania TARP II Team Leader: Ladisy Komba Chengula TL Unit: AFTS2 ICR Type: Core ICR Report Date: December 27, 2004 1. Project Data Name: Tanzania TARP II L/C/TF Number: PPFI-Q0740; IDA-30360 Country/Department: TANZANIA Region: Africa Regional Office Sector/subsector: Central government administration (72%); Agricultural extension and research (28%) Theme: Other rural development (P); Rural policies and institutions (P); Decentralization (S) KEY DATES Original Revised/Actual PCD: 10/30/1996 Effective: 03/31/1998 06/11/1998 Appraisal: 11/03/1997 MTR: 11/01/2000 03/05/2001 Approval: 01/29/1998 Closing: 06/30/2003 06/30/2004 Borrower/Implementing Agency: GOT/Min of Agric and Food Security Other Partners: STAFF Current At Appraisal Vice President: Gobind T. Nankani Callisto E. Madavo Country Director: Judy M. O'Connor James W. Adams Sector Manager: Karen Mcconnell Brooks Sushma Ganguly Team Leader at ICR: Ladisy Chengula Satish Kumar ICR Primary Author: Jane A. N. Kibbassa 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: Context: The Tanzania Agricultural Research Project phase II (TARP II), like its twin National Agricultural Extension Project phase II (NAEP II), was designed as a second phase of support for incremental strengthening of Tanzania's agricultural technology system. TARP II and its predecessor, the National Agricultural and Livestock Research Project (NALRP, Cr. No. 1970 ­ TA) supported agricultural research, and the National Agricultural and Livestock Extension Rehabilitation Project (NALERP, Cr. No. 1994 - TA) supported extension. When the project was designed in the mid 1990s, Tanzania's Agricultural Research System, like those of many African countries, was starting to overcome a legacy of inadequate funding, lack of clear priorities for research, weak links to extension and to ultimate users of the products of research, and low staff morale. The Ministry of Agriculture and Cooperatives (MAC), the governmental body at the time charged with overseeing agricultural development, recognized the need for improved technology, particularly for smallholders. As the National Agricultural and Livestock Research Project (NALRP, Cr. No. 1970 ­ TA) closed in 1997 the Government sought continued support through a new project for agricultural research, to run in tandem with the second phase National Agricultural Extension Project (NAEP II, Cr. No. 2899 - TA). NAEP II became effective in October 1996 and continued until December 2003. TARP II became effective in March 1998 and continued until June 2004. Although the Government sought continuity in support, TARP II was not intended simply as a linear continuation of earlier activities. The Government of Tanzania, with the support of various bilateral and multilateral donor agencies, re-evaluated the National Agricultural Research System (NARS) and requirements for its strengthening in 1997/98. As a result of this evaluation, the Tanzania Agricultural Research Project - Phase II (TARP II) was formulated in 1998 as part of the overall Government effort to strengthen the agricultural sector, and to make agricultural research more effective. The project's development objective (PDO) as stated in the Project Appraisal Document (PAD) of November 1997, was to "support the generation of technology to increase efficiency and productivity of crop and livestock production systems with sustainable use of natural resources, focusing on the needs of the smallholder sector". TARP II sought to assist in decentralization of financial management, operational decision making, and the formulation of the research agendas to Zonal Research Centers (ZRCs). The project also supported privatization of research as appropriate, and emphasized client-oriented adaptive research. Efforts were made to strengthen linkages with extension, and to move from research on individual commodities to attention to farming systems. Furthermore, TARP II was to assist in human resource development, improvement of research planning, zonal research management and monitoring and evaluation (M&E) procedures. The project's focus on changes in the system of agricultural technology to improve sectoral productivity was reflected in Tanzania's medium term development strategy. The latter sought to increase overall economic growth to at least six percent by the year 2000, and was substantially achieved. The project was consistent with the country's development objectives and the World Bank Country Assistance Strategy (CAS) that called for efforts focusing on raising agricultural productivity. It was envisaged that the basis for overall growth, and agricultural growth in particular, was to be through improved provision of economic services (research and extension, rural access roads, water resource development and links between social and economic services) for smallholders. In fact, much of the aggregate growth was achieved outside of the agricultural sector, but agriculture also made a contribution by growing at 4.3 % on - 2 - average over the period of TARP II's implementation, which was about a percentage point higher than growth achieved over the 1990s. 3.2 Revised Objective: Changes in the wording of the Project Development Objectives (PDO) were made during Mid-Term Review (MTR) in March 2001 to reflect an increased emphasis on institutional development aspects of the project in response to changes in the Ministry of Agriculture structure. Although the PDO wording was changed, the TARP II original PDO basically remained valid. Prior to the MTR, the Government split the Ministry of Agriculture and Cooperatives (MAC) into three ministries; namely, the Ministries of Agriculture and Food Security (MAFS), Water and Livestock Development (MWLD) and of Cooperatives and Marketing (MCM). The creation of the three ministries brought a revision of the mandate of the then Department of Research and Training (re-named Department of Research and Development, or DRD). The new DRD exclusively dealt with crops and related natural resource management. Research on livestock and range management was placed under the Department of Livestock Research and Training (DLRT) in MWLD. The restructuring of entities responsible for research and training created disbursement bottlenecks under the project, since expenditures had already been organized based on the prior institutional arrangement. Furthermore, extension staff were transferred to the districts and reported to a new Ministry - President's Office Regional Administration and Local Government (PORALG). Redeployment of people within the new structures also affected smooth implementation of ongoing research activities. Following these critical changes, the Bank and Government agreed to place increased emphasis on institutional development aspects of the project to ensure that the project development objectives could be met within the new institutional environment. The project development objective statement was subsequently changed to reflect the increased institutional development emphasis as follows " to increase the efficiency and productivity of crops and livestock production systems focusing on the small holder sector and institutional strengthening." Since the project development objective was not viewed as substantial, the task team did not seek Board's approval. In this regard, the ICR team assessment of the project outcomes is based on the original project development objective. The changes at the MTR brought greater clarity by introducing a more sharply defined set of indicators than those included in the PAD. The revised objective was to be achieved through: (1) increasing the efficiency and productivity of agricultural production (crops and livestock), as well as sustainable use of natural resources through generation and dissemination of appropriate agricultural technologies, with emphasis on the needs of smallholder farmers; (2) supporting the institutional development of the NARS, particularly the empowerment of the ZRCs, in financial resource management, operational decision making and the formulation of demand-driven, client-oriented zonal research agendas; (3) encouraging greater private funding of agricultural research as well as the strengthening of linkages between farmers, extension agents and research scientists and; (4) strengthening of DRD's capacity to manage and co-ordinate agricultural research through the improvement of research planning, monitoring and evaluation as well as human resource development, rehabilitation of research infrastructure, and procurement of essential research materials and equipment. Key performance indicators for outcomes and outputs of the project for these four key elements were also established during the MTR, and are indicated in Annex 1. These four elements are consistent with the initial PAD PDO. 3.3 Original Components: TARP II consisted of three components: (i) institutional development; (ii) research programs support; and (iii) resource development and management. The project had a total of fourteen sub-components as listed under each component below: - 3 - (a) Institutional Development (US $ 2.93 million). The project was to continue with the process of institutional reforms of the research establishment initiated during the preceding project. The main activities included: (i) redefining the roles of the then Department of Research and Training (later DRD and DRLT), the National Agricultural Research Council (NARC) and ZRCs with emphasis on stakeholder participation; (ii) establishment of zonal executive committees (ZECs) with substantial representation of research users; (iii) financial and operational empowerment of the zones; (iv) an effective arrangement of donor coordination and for coordinating regional and other collaborative research programs; (v) a phased consolidation of the research establishment; and, (vi) initiatives for privatizing research, as feasible. (b) Research Programs support (US $ 8.50 million). This component sought to: (i) provide incremental operating costs for the priority research programs in food crops, livestock and other programs (e.g. soil conservation and agro-forestry); (ii) strengthen research infrastructure at Sokoine Agricultural University (SUA) and provide support for research programs there; (iii) assist in establishing Zonal Agricultural Research Funds (ZARFs) through pilots in a few zones initially, including the establishment of guidelines for operating the ZARFs on a demand-driven basis; and, (iv) help in breeder seed production at the zonal research stations. (c) Resource Development and Management (US $ 11.55 million). This component aimed at: (i) strengthening human resource development; (ii) providing assistance to SUA for carrying out approved training programs; (iii) consulting services and short-term technical assistance; (iv) selective rehabilitation of research stations, equipment and infrastructure; and, (v) strengthening communications, documentation and library services, and research M&E. 3.4 Revised Components: After the MTR, the project retained the three original components, and the Institutional Development component was amended to include additional tasks, namely: re-integration of the Veterinary Investigation Centers (VIC) into the research system under the Animal Disease Research Institute (ADRI) as "Livestock Investigation Centers;" resumption of regular inter-zonal research and development meetings; and support to the establishment of pro-active programs with international research institutions. The Resource Development and Management component was also revised during the MTR. The changes were necessary because of unrealistic budget estimates and the need for additional works and goods that were initially not planned (such as the tissue culture laboratories for Uyole, Ukiriguru, Mikocheni, Tengeru and Temeke, and procurement of six units of Automatic Weather Stations). The Project Implementation Plan (PIP) was revised to incorporate the above changes. Similarly, the Development Credit Agreement (DCA) was amended to reflect the reallocation of Credit proceeds in Schedule 1 to accommodate the project's new priorities and focus. 3.5 Quality at Entry: The project was not subjected to a performance rating by the Quality Assurance Group for Quality at Entry. Ex-post, the quality at entry is judged to be marginally satisfactory; satisfactory because it incorporated lessons from other projects and emerging understanding of best-practice in design of systems of research, but only marginally so because the critical attention to monitoring of impact and returns to research was neglected. The design of TARP II was in line with the overall mandate of DRT (and later DRD and DLRT); namely, carrying-out of client-oriented, demand-driven and efficient research. The project was designed so as to: (i) incorporate lessons from previous investments in the sector, both in - 4 - Tanzania and other countries such as Uganda, Malawi and Kenya; (ii) accomplish and sustain the good work done during the predecessor project (NALRP), and NALERP. Some of the lessons drawn from these experiences were that: (i) zones, as the main action stations for research, should have final and operational autonomy to carry out their mandates; (ii) the stakeholders should have an effective voice in setting the research agenda; (iii) focus should be on financial sustainability of research; (iv) more pro-active measures were needed to foster farmer-research-extension linkages; (v) the research establishment should be consolidated for more cost-effective management. Despite these positive features, the project design lacked clearly-defined KPIs in the Project Appraisal Document (PAD). This inadequacy became apparent during the TARP II MTR and was subsequently addressed. A well-defined set of KPIs should have been developed during project preparation in order to provide benchmarks and to guide the M&E system as well as the final impact assessment. It should also be noted that the log frame approach was not in use during the project planning stage. In order to mitigate this anomaly DRT, with TARP II support as early as 1998, commissioned the Bureau of Agricultural Consultancy and Advisory Services (BACAS) of Sokoine University to undertake a detailed survey to collect baseline data on existing farming systems. As part of project M&E, time series surveys and studies were to be undertaken to measure and record the impact of new technologies developed by ZRCs. During the course of project implementation, however, many research projects were conducted in villages other than those where the BACAS baseline data were collected. This implies that the BACAS sampling procedures did not take into account the priority areas already identified and planned by the zones. As a result, the M&E data from the 2000 and 2003 time series surveys could not be compared to the baseline survey data of 1998 to determine the impact of improved technologies adopted by participating farmers. (the implementers have another view of this situation, stating in their comments that "areas covered by the baseline survey and the 2001 and 2003 surveys were consistently conducted in almost the same intervention areas within the same farming systems, same villages and farmer research groups. The problem with the survey was not its inadequacy but that is was partially analyzed.") Although the analyses of 2000 and 2003 time series data has shown that a significant proportion of participating farmers have adopted improved technologies, the failure to provide adequately for monitoring and evaluation during project design could not be compensated even by laudable efforts during implementation. For that reason, quality at entry is considered to be only marginally satisfactory. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: Overall, during TARP II and with the support of IDA funding, Tanzania undertook institutional reforms that strengthened the capacity for agricultural research, managed research programs of generally high quality, and developed technologies that, when applied on a pilot basis, proved successful in raising productivity and incomes. The institutional framework was strengthened, as noted in greater detail below. The generation and dissemination of technologies was commendable as evidenced by the number of new varieties developed and released. There was also increased private sector funding for cash crop research, particularly coffee and tea. Moreover, Tanzania's agricultural sector performed well on balance over the years of implementation of the project. Despite deteriorating international prices for key crops, such as coffee and cotton, and several years of poor rainfall, Tanzania has made significant progress towards the Government's target of sustained annual agricultural growth of 5 percent. The five-year moving average agricultural GDP growth rates were about 3.3 percent from 1990­2000, and 4.3 percent over 1999-2003. Over the latter period, productivity improvements, through yield changes, have contributed more to growth. On this basis, the overall project performance is satisfactory. Case studies of success are available even though it may be difficult to generalize from these to the performance of the project as a whole. Although - 5 - M&E survey reports have not included, the cost benefit analysis shows positive returns with likely income increases of smallholder farmers who participated in on-farm research (collaborating farmers) in various farming systems. In addition, the project had a substantial impact of crop (yield) loss prevention, particularly for maize. Strengthening of human development was clearly accomplished, but constraints in the public service system of remuneration limited the ability of the research system to retain qualified staff. The training must therefore be considered to have been only marginally successful in meeting the underlying objective. Finally, the institutional changes were commendable with regard to strengthening of the zonal research stations and consolidation of the system. At the same time, the ministerial reorganization reduced the effectiveness of the research and extension systems. Outreach and adoption varies significantly despite the positive development of appointment of the Zonal Research and Extension Liaison Officers (ZRELOs). Despite these noted shortcomings, the project is rated satisfactory as highlighted above. Moreover, as far as most of the project sub-components are concerned, significant achievements were observed. M&E surveys conducted in 2003 that compared the yields achieved by farmer groups (FGs) collaborating with research in on-farm adaptive trials with those obtained by other (non-collaborating) farmers indicate that collaborating farmers' yields were significantly higher (Table 3 on Annex 8). The project also made substantial achievements in other areas as summarized in a table of case studies (Annex 9), and as indicated in the outcome indicators in Annex I. Special care was taken by TARP II to address existing weaknesses in research-extension-farmer linkages. A total of 37 improved crop varieties and 10 different innovations for livestock production were developed and disseminated to collaborating farmers (farmers participating in various on-farm research activities) and non-collaborating farmers (farmers not involved in on-farm research) as indicated on Table 1, Annex 8. The most notable achievements include the development of loss preventing technologies, such as the Grey Leaf Spot (GLS) and Cassava Mosaic Disease (CMD) resistant varieties, as well as new tomato varieties which have significantly reduced post-harvest losses by increasing shelf life. Some of the achievements resulting from soil fertility initiatives include yield increases in cotton and rice of between 20% and 60% in a wide range of environments. Training targets were fully met: 21 PhD and 27 MSc. students from DRD and DRLT were trained, and about 943 staff attended various technical and managerial short-courses. Decentralization of ZRCs and development of competitive ZARFs were fully realized. The objectives related to strengthening of Farming Systems Research (FSR) and demand-driven client-oriented research (COR) were adequately addressed. Farmer participation in technology development was enhanced and involvement of the private sector increased through greater cost-sharing of industrial crop research (for Tea, Coffee and Tobacco). The design of the Agricultural Services Support Program (ASSP) currently under development as one of the ASDP pillars is based on the institutional framework and research programs management system established under TARP II. Despite the focus on the smallholder sector, the project also responded to medium ­ large scale or commercial farmers's needs through the client oriented research and management approach (CORMA) arrangements. Results from contract research financed by large plantations, input suppliers (seeds and agrochemicals) and the international research institutions indirectly benefited smallholder farmers. The project design also facilitated and enabled coordination with other donors actively supporting the Tanzanian agricultural research system. The design accommodated linkages within the NARS particularly, with SUA and with external research institutions, especially the Centers of the Consultative Group for Agricultural Research (CGIAR) such as ICRAF, CIMMYT, IITA, ICRISAT and CIAT. This collaboration also contributed substantial funds, physical facilities, technical support and training to the research effort at the ZRCs. - 6 - 4.2 Outputs by components: Component 1: Institutional Development (US $ 2.93 million) The main objective was to continue to support development of the NARS and assist in decentralizing financial management, operational decision-making and the formulation of the research agendas to the ZRCs. The project has largely achieved its core objective of developing and strengthening a more efficiently managed NARS in Tanzania. Overall, the project has completed its intended institutional reforms and laid a solid foundation for more effective client-oriented research in the future. Outputs by sub-components are listed as follows: Redefining the roles of the DRD Headquarters establishment, the National Agricultural Research Council (NARC) and of the ZRCs with emphasis on stakeholder participation: The reorganization of the DRD/DLRT to empower the zones to do research responsive to farmers' needs has been very effective. In order to develop strong institutional capabilities, the responsibilities of different institutions were re-defined with greater zonal (and client) empowerment in making decisions on the research agenda and resource utilization. Partnerships and linkages among the collaborating stakeholders were increasingly emphasized. Following the restructuring of MAC, a Memorandum of Understanding (MoU) between MAFS and MWLD was signed to ensure that functional linkages between all entities concerned with agricultural (crop) production, animal health and animal production research were maintained. The ZRCs remained intact and continued to carry out integrated research programs on crops, livestock and natural resource management. MoUs were also signed between ZRCs and respective Local Government Authorities (LGAs). However, these MoUs were not legally binding in terms of roles and responsibilities of both parties, including financial contributions and technical backstopping requirements. On the part of overall project coordination, the envisaged national advisory and coordination body was not established. It was envisaged in the PAD that NARC would be formed, but despite calls from various supervision missions, this did not happen. A stated reason for not doing so was that the Commission for Science and Technology (COSTECH) was able to coordinate all research activities in Tanzania. However, COSTECH's coverage is too broad to remain focused on agriculture related matters. Establishment of ZECs with substantial representation of research users: the establishment of autonomous zonal research committees such as the Internal Program Review (IPRs), the Zonal Technical Committees (ZTCs) and the ZECs in which zonal stakeholders are empowered to determine and influence the research agenda and budget in their respective zones, was accomplished and effectively operationalized in all seven zones. Zonal meetings of the IPRs, ZTCs and ZECs have strengthened the interaction between researchers and stakeholders, thus making the research agenda more focused, client-oriented and demand-driven. These efforts much improved the process of technology development and dissemination. During the project period these committees played a very important role in controlling the relevance and quality of the zonal research agenda. The committees met annually to review the progress of research activities and proposed research programs to ensure that they respond to the priorities as seen by stakeholders. The committees have been dominated by progressive farmers and researchers, and additional efforts are needed to reach out to poorer farmers. Financial and operational empowerment of the zones: The project was expected to facilitate devolution of financial management and control of resources to the ZRCs, and this was partially attained. Though the ZRC were given the mandate to plan and budget for all their activities, financing depended on disbursement by the Government, which has been erratic and inadequate throughout the TARP II implementation period. This resulted in major cuts in zonal priority research projects, since funding to the research activities was in many cases below the level approved by the ZECs. This jeopardized not only the size but also the quality - 7 - and continuity of the zonal research agenda. These problems are partly related to difficulties with the imprest­based financial management system. TARP II headquarters required financial reports from the zonal centers on monthly basis, and imprests were to be retired within 14 days after completion of the intended task in order to apply for Special Account replenishment based on Statements of Expenditures (SOEs). The ZRCs (which were inadequately staffed with accountants) repeatedly failed to meet the tight turn-around cycle and as a result, often ran out of funds. In addition, some of the important aspects of the project such as procurement, vehicle maintenance and servicing of machinery were not decentralized, and this posed implementation bottlenecks in the zones. In order to improve the financial operations the Bank supervision mission of March 2001 recommended the following actions: (i) Zonal managers should be instructed to work closely with accountants in making sure that all required financial information is made available as and when needed; (ii) urgently fill vacant accounting positions; and (iii) arranging for contractual technical backstopping support. An effective arrangement of donor coordination and for coordinating regional and other collaborative research programs: Effective donor coordination and partnership was realized. This is demonstrated by the involvement of several bilateral donors and other development partners in supporting various activities in the zones to complement efforts of TARP II. These included: the Netherlands Government, which successfully supported the piloting of COR in the Lake and Northern Zones (1998-2003); Ireland Aid's Eastern Zone Client Oriented Research and Extension project (EZCORE); and the NORAD-funded research project component at SUA, re-enforcing the TARP II-SUA assistance. Other donors included Sweden/Sida, which provided funding for the National Agricultural Research Fund (NARF), Germany/GTZ, supporting a coconut improvement program, Denmark/DANIDA providing funding for plant genetic resources and the European Union supporting industrial crops (tea, coffee and cotton) through the "STABEX" fund. During project implementation, several research activities were supported by regional networks, mainly the Association for Strengthening Agricultural Research in Eastern and Central Africa (ASARECA) and the IARCs. The project annual donor consultation meetings ensured harmonized processes, information sharing and joint evaluation of progress. TARP II/DRD/DLRT management was represented in the concerned Steering Committees and other project review fora. At zonal level, the progress of research activities supported by various donor agencies, regional networks, IARCs and other development partners was reviewed during IPR meetings to ensure that they were addressing priority problems of the zone. A phased consolidation of the research establishment: The process to consolidate the Tanzanian research establishment was initiated during the first phase (NALRP) in accordance with a staffing plan and NARS inventory developed under the World Bank-supported Agricultural Sector Management Project (ASMP) implemented during 1993-2001. The intent of the restructuring was to improve effectiveness and efficiency in the delivery of research services. NALRP facilitated the closure of several research sub-stations and facilitated staff redeployment. The TARP II project continued this process. A total of twenty five out of fifty research stations were closed during the period of implementation of NALRP and TARP II. Staff were redeployed to ensure a mixture of appropriate skills in response to the needs of a particular zone. In terms of institutional restructuring, this was a major and ambitious undertaking, and it was accomplished with minimal disruption. Initiatives for privatizing research: this sub-component aimed at rationalizing the functions of the publicly-financed NARS and at increasing private sector participation in research and development in the agricultural sector. Private sector involvement particularly in cash crop research was promoted through joint undertakings with agro-industries (e.g. cotton, sugarcane and cashew) or privatization of research such as occurred in tea, coffee and tobacco. Privately funded institutes (through the crop cess and with EU support) were established for Tea Research Institute of Tanzania (TRIT), the Tanzania Coffee research - 8 - Institute (TaCRI) and the Tobacco Research Institute of Tanzania (TORITA). This initiative was intended to reduce pressure on the Government budget and to free up resources to finance core public research programs. Private financing for tea and coffee research has led to: improved management, a more cost-effective technology development and transfer process, and improved service delivery which has in turn resulted in increased productivity. TRIT and TaCRI represent successful models for privately run agricultural research institutes, and succeeded in delivering significant outcomes and benefits to the intended clients cost-effectively. This has been possible through the horizontal integration of research, extension and farmer, as well as vertical integration between farmers and commodity processing firms, like tea and coffee factories. The main sources of operational funds for TRIT and TaCRI are cesses and levies (1.5% and 0.75% respectively), STABEX funds and crop sales (from Institute production). TRIT is also receiving support from the UK Department for International Development (DfID). TORITA depends solely on a research levy charged on tobacco "green leaf" (of Tsh 2.00 per kilogram). Although privately funded research in coffee, tea, and tobacco has been successful in terms of the quality of the research effort, there remain concerns about funding levels. The current levels of cess/levy appear too low to fully to support research (in the absence of EU support). TORITA in particular is experiencing problems associated with inadequate funding, human resources and a general lack of appropriate facilities. Moreover, the cess and levies for coffee and tea are collected and managed by the crop boards, which have a mixed history of performance. With prices of the products for which research was privatized depressed on global markets, the advisability of asking poor smallholders to pay for the research is questionable. Given the poverty of smallholder growers, the long term financial sustainability of these new institutions will depend on continued public support, either directly from the budget or indirectly through donors. TRIT and TaCRI seem to be doing well currently while still receiving STABEX funds. Although efforts to increase private funding for the remaining cash crops (e.g. cashew, cotton, sugarcane) are ongoing, the Government will need to consider continued public support for research undertaken by the privatized research institutions as long as the constituency they serve remains predominantly poor and the research outputs are public goods. Such an approach would be fully consistent with principles of best practice in design of research systems, in which public funding can support private delivery of services. Component 2: Research Programs Component (US $ 8.50 million) The focus of this component was to support research programs emphasizing development of appropriate technologies in order to improve production by small-scale farmers. Overall the achievement of this component was satisfactory. A large number of new research technologies including the release of new varieties, agronomic and management practices, post harvest technologies, natural resource conservation, livestock production and disease control were recommended for use by farmers. These have been adopted at moderate rates to date, and adoption can be scaled up. Mechanisms for ensuring greater farmer demand in the system have also been established. Outputs for each sub-component are highlighted below: Provide incremental operating costs for the priority research programs in food crops, livestock and factor programs (e.g. soil conservation and agro-forestry): Activities under this sub-component were to a large extent successfully accomplished. Development and dissemination of appropriate agricultural technologies was highly satisfactory. DRD researchers exceeded the MTR targets with the development and dissemination of 37 improved crop varieties. Other technologies developed by DRD were on crop protection, agronomic practices and post harvest technology. Recommendations on appropriate management of soil fertility were developed. Agro-forestry technologies were also developed and recommended to farmers. From an M&E survey of 2003, a general increase in the use of improved seeds and technologies was observed by collaborating farmers. In order to make a rigorous evaluation of the - 9 - impact of improved technologies, however, additional information beyond dissemination such as adoption rates and impact is necessary. With regard to DRLT, the research effort on production and use of thermostable New Castle Disease (NCD) vaccine I-2 has successfully reduced mortality of poultry. This is an important measure in poverty reduction and improved nutrition at the household level. Researchers also developed strategies to improve livestock productivity through proper feeding, improved management practices and disease control. Strengthen research infrastructure at SUA and provide support for research programs: Mainstreaming of SUA into the NARS was aimed at building the capacity of SUA to respond better to the needs of smallholder farmers. This was achieved through support from the IDA TARP II Credit and a grant from NORAD. A total of US$ 504,000 was allocated from TARP II for 5 years to support priority research activities, procurement of research materials, equipment and infrastructure development. Four research projects were conducted by SUA staff, and office equipment and a vehicle were procured from the IDA component. The NORAD component provided a total US$ 5,298,650 to support the TARP II-SUA project on Food Security and Household Income for Smallholder Farmers in Tanzania with emphasis on women, implemented in Southern Highlands and Eastern Zones, from 2000-2004. A total of 34 client oriented/demand driven research projects that involved researchers for SUA, MAFS and the Agricultural University of Norway were conducted in the project areas. Both the IDA Credit and the NORAD grant were fully disbursed as of June 30, 2004. Assist in establishing Zonal Agricultural Research Funds (ZARFs through pilots in a few zones initially) and establish guidelines for operating the ZARFs on a demand-driven basis: TARP II was successful in facilitating the process of establishing ZARFs including development of management guidelines, organizational set up and operational modalities. The ZARFs were established with contributions from district councils, development partners and other sources. These contributions received matching funds, initially at 100% and later at 50% from the World Bank, as part of TARP II financing. In addition TARP II supported zonal stakeholders meetings to create awareness on the value and relevance of the contribution of research to agricultural development, and concerning the successes and constraints faced by ZRCs. Currently, ZARFs are institutionalized in all seven zones and ZARF management committees are in place, although performance has varied by zone. The district councils were initially quite responsive and several councils made substantial contribution/commitments to ZARFs. The level of commitments and actual contributions to ZARFs during the project period varied across the zones and none of the zones attained their estimated targets. The total contributions to ZARFs was Tsh 572,368,535, mainly from district councils, bilateral donors and export crop industries (cashew); Tsh 321,000,000 was matched through TARP II. A total of 48 research projects were supported by ZARFs in the Northern, Lake, Southern and Central zones. The Southern Highlands, Eastern and Western zones experienced very poor responses to requests for ZARF contributions and did not support any research activities during TARP II implementation. Overall, the flow of district contributions to ZARFs during the project period has declined and poses a challenge for future funding. The contribution to ZARFs by district councils is likely to change in the future due to: (i) changes in fiscal arrangements at the district level, with local government capital grants and district agricultural development grants coming on-stream soon; and, (ii) increased emphasis on empowerment of farmers and participatory decision making at district level. Competitive grants and collaborative research under the new system of local governmental finance have a great potential for augmenting the ZRCs' resource bases. - 10 - Help in breeder seed production at the ZRCs: Involvement of research centers in breeder and foundation seed multiplication contributed substantially to availability of improved seeds to farmers. Breeder seeds of maize, phaseolus beans, rice, sesame, groundnuts, sorghum, millet, cowpeas, wheat and barley were produced and distributed. Certified seed of both rice and hybrid maize has also been produced. The amounts of seed of improved crop varieties multiplied through ZRCs in 2003 are presented on Annex 8. Although TARP II also facilitated improvement in the amount of breeder seed production by the research centers, this activity needs to be further scaled up. In addition, through TARP II, DRD in collaboration with the DANIDA-supported Agricultural Sector Program Support (ASPS), initiated informal seed multiplication and distribution ventures in order to produce "Quality Declared Seed" (QDS). In order to fill the continuing gap in seed availability, ZRCs may in the future need to involve the private sector and facilitate this activity being undertaken on a commercial basis. Component 3: Resource Development and Management Component (US$ 11.55 million) The objective of this component was to provide support for the development of human resources, essential research infrastructure and in improving critical support services and systems. Outputs from the sub-components were as follows: Strengthening human resource development: This sub-component was satisfactory; a significant number of research scientists and support staff received short and long-term training courses in various disciplines within Tanzania and abroad. The number of scientists trained at PhD level was nearly 100 % of that planned (62 against the target of 63). Staff numbers trained for MSc surpassed the target by 4.5 %. In addition, 21 new BSc. graduates were recruited so as to reach a current level of 67 graduates in this category. These figures imply that the targeted 1:3:1 ratio for PhD: MSc: BSc has essentially been attained. Staff members who completed their training and returned to their respective stations are performing well in their research work. The main challenge is retain the scientists as well as ensure succession since about half of the senior staff are due to retire in the next five years. Providing assistance to SUA for carrying out approved training programs: This sub-component also successfully achieved its core objective. The project-supported training of 32 MAFS staff to MSc and PhD levels at SUA, and several researchers for short courses in various professional areas. In addition, the TARP II-SUA project: (i) enhanced the capacity of both SUA and MAFS/MWLD researchers to conduct client-oriented and demand-driven research; (ii) strengthened Farmer-Research-Extension linkages through an interactive participatory research methodology; (iii) enhanced exchange of experiences and expertise between SUA and MAFS researchers; and, (iv) developed an in-built M&E and impact assessment system which enhanced project management and coordination, and contributed to institutional impact for both SUA and MAFS/MWLD. Procurement of Goods and Services: Procurement mainly focused on replacement of items supplied during the first-phase project (NALRP) as well as consumables such as laboratory reagents, spares for vehicles and farm equipment. As of May 2004, a total of US $ 3.5 million were spent and this is equivalent to 96% of plan. This amount includes goods and services procured for TARP II-SUA (US $ 210,000). Major activities included: procurement of furniture for DRD/DLRT headquarters building (Kilimo II); contracts for ICB tenders for 53 vehicles, 35 motorcycles, 8 tractors plus implements, 20 photocopiers, 59 computers and accessories, miscellaneous laboratory chemicals & equipment, 15 scanners, 10 multimedia projectors and 6 trucks. It also covered the supply of water pumps; telephone switch boards, 90 bicycles, water tanks, automatic weather station and workshop tools. - 11 - According to the DCA, procurement was to be done by DRD/DRLT headquarters. Contracts for the procurement of goods and services were accordingly signed by the Permanent Secretary, MAFS, sometimes without adequate consultation or involvement of ZRC management. Since the zones were not fully involved in the process of overseeing consulting services, items procured in some cases did not fully suit their priorities and requirements. For example, the standard design of biotechnology laboratories was assumed to be meet the requirements of all four research stations, but was later on found not to be the case. Similar problems applied to maintenance of vehicles and servicing of equipment and machinery. The garages to be used by the zones were pre-selected by DRD/DLRT Headquarters and in some zones this was found not to be the best option. Difficulties were also encountered in procurement of goods and services in the initial stages due to staff shortages and unqualified staff. This resulted in considerable delays in procurement. The situation however, improved later after short term training, seminars and workshops on procurement. Selective rehabilitation of research stations, equipment and infrastructure: Overall the achievements of this sub-component include 80% of the targeted station rehabilitation and 50% of the planned irrigation works. The research station rehabilitation program involved the following centers: Ilonga, Uyole, KATRIN, Maruku, ADRI - Temeke, Mlingano, PRC - Kongwa, Selian, Tumbi and LPRI - Mpwapwa. The construction of bio-technology laboratories was carried-out at Mikocheni, Uyole, Ukiriguru, ADRI - Temeke and HORTI - Tengeru. Construction of screen houses at Uyole, Ukiriguru, Selian and Mikocheni was also part of the plan. Most of the contracts for civil works were awarded between the last quarter of FY 2002/03 and first quarter of FY 2003/04. In this regard most of the civil works could only be better assessed towards the end of the project. The February, 2004 IDA supervision mission rated overall quality of constructed civil works at most ZRCs unsatisfactory. The mission thus recommended to the Government to ensure that all defects are rectified prior to handing over the construction sites to the concerned stations and take to task the incompetent consulting engineers. However, very little action has been taken four months after project closure. Given the importance of these facilities for research work and for the sustainability of TARP II-supported activities, the Government will need to ensure that the contractors finish the work and rectify all defects prior to handing over the facilities to the zones. Furthermore, close contract supervision by the ZRCs management is considered important in order to ensure quality work and greater accountability. Strengthening communications, documentation and library services: Efforts were made to establish appropriate systems of information acquisition, management and dissemination. Integration and strengthening of the ICTs, documentation and library services has brought about an expansion and institutionalization of such services including an MIS, GIS, and other databases. DRD's Research and Training Newsletter continues to be produced, although irregularly at present. In addition, the DRD/DLRT activities have been widely publicized through websites. This has greatly improved dissemination of information by the DRD/DLRT, as well its public image locally and globally. Documents and databases from FAO and CABI in CD-ROMs format were acquired and disseminated to researchers and other stakeholders. These included a subscription to complete sets of "The Essential Electronic Agricultural Library (TEEAL)" that has enabled the scientists to have access to the full-text of over 140 of the most important journals in agriculture. Currently TEEAL is available at the DRD/DLRT Headquarters and all the ZRCs. The DRD/DLRT has also subscribed to a number of on-line databases including Global On-line Research in Agriculture (AGORA) that provides access to over 400 journals from major scientific publishers in the fields of food, agriculture, environmental science, and related social sciences. Information flow and communication has been improved between DRD head office and the ZRCs through e-mail and Internet access. Six ZRCs (Selian, Ukiriguru, Tumbi, Uyole, Ilonga and Naliendele) have full Internet access, while the rest have e-mail connectivity. - 12 - Research M&E: During the project period, an M&E process was established. A detailed baseline survey was conducted in 1998 to collect data on the existing farming systems and to serve as a bench-mark for the assessment of project impact. This was followed by collection of subsequent data from collaborating and non-collaborating farmers in selected target areas, using structured questionnaires developed on the basis of key parameters from the base-line survey. Two rounds of surveys were undertaken in 2000 and 2003, in which 840 and 1995 farmers were involved respectively. In addition, supervision missions, the MTR, senior managers' field visits and zonal annual review meetings and Project Steering Committee sessions supplemented M&E of research activities. Some adoption studies on specific technologies were also carried out to track the impact of research activities. Important routine supervision, internal monitoring and evaluation procedures were the responsibility of the zonal research managers and ZECs. The zonal research managers and lead scientists, however, were not allocated sufficient funds to accomplish these functions and the participation of ZECs in monitoring research programs remained low during the project period. Moreover, the zonal socio-economic units responsible for M&E lacked the required human, technical and financial capacity. A well-established M&E time series data and results of adoption studies should have been aggregated to measure the overall impact of various technologies introduced over the project period; this has not been possible. Therefore in future M&E should be made routine and in-built in the work plans at both national and zonal levels. 4.3 Net Present Value/Economic rate of return: Calculation of the net benefits of research interventions requires a comparative approach along two dimensions: "before and after the project" and "with and without the project". In the absence of baseline data, only rough approximations can be made. Specifically, the PAD's ex ante rate of return can be recalculated in light of ex post information on the project's yield impact. "With and without" yield improvements are documented in Annex 8, Table 3 of this ICR. All other assumptions in the PAD's ex ante analysis are maintained (Annex 3), including those relating to adoption rates (25% of farmers adopting in the first five years of the project). The adoption assumption must be maintained because of the absence of any solid monitoring information on adoption rates at this stage. The limitations of the analysis should therefore be kept in mind when interpreting the results ­ lower adoption rates than those assumed would obviously reduce the project's net benefit. This provides a further argument for urgent improvement in the monitoring of technology adoption, an issue raised elsewhere in this ICR. The table below contrasts average annual yield increases projected in the PAD's cost-benefit analysis with yield increases derived from data collected over the life of the project (as documented in Annex 8). It gives a mixed picture: showing changes that are in some cases close to or larger than those projected in the PAD (maize, cotton, coffee, cassava) and in some cases smaller (sorghum, paddy, cashews, pulses, wheat). Projected vs. actual annual yield increases (percentage) Crop Projected Actual Maize 8.2 7 Sorghum-millet 5.0 1 Paddy 7.8 4 Cotton 6.0 20 Coffee 6.7 16 Cashewnut 17.5 11 Cassava 7.2 14 - 13 - Pulses 10.0 7 Wheat 5.0 (2) Sources: projected increases come from TARP II PAD; aggregate increases reflected in Annex 8, Table 3 which compares actual yields of project collaborators with those of non-collaborators. The project's economic internal rate of return is 21% when computed under the above assumptions - essentially the same as the ROR computed in the PAD. The Net Present Value is $22 million, slightly larger than that reported in the PAD (a discount rate of 12% is used). These results reflect the offsetting effects of yield improvements for maize, cotton, coffee and cassava on yield reductions for other crops. While no data was available on yield improvements for a few commodities that were included in the PAD's analysis (e.g. tobacco, tea, cattle) some commodities not dealt with in the PAD have seen impressive yield increases under the project - notably oilseeds, potatoes and bananas. The return estimates may be an underestimation of benefits as much research was focused on loss prevention rather than yield improvement. For example - the grey leaf spot (GLS) tolerant maize variety (UH615) was developed in the Southern Highlands zone and this controlled a serious fungal disease which was causing up to 50% crop loss. Likewise, cassava mosaic disease (CDM)-tolerant clones with the potential of reducing yield losses by 80 - 100% were developed in the Lake zone. 4.4 Financial rate of return: Financial Rate of Return is not calculated separately. 4.5 Institutional development impact: The project has strengthened the NARS through maintaining ongoing research efforts and adding new ones, consolidating the system for greater cost-effectiveness, enhancing the human resources, and increasing participation of stakeholders, particularly through the devolution of functions to the zonal level. The roles and responsibilities of DRD were (re-)defined in line with ZRCs created during the first phase of the project (NALRP). Reorganization of the DRD/DLRT to empower zones to do more farmer-responsive research has been effective. ZARFs were established and supported in the context of de-centralization and empowerment of ZRCs. The establishment of ZARFs opened ways for local authorities and the private sector to contribute to research funding. This enhanced cost sharing and partnerships. Institutionally, the formation of autonomous zonal research committees (IPRs, ZTCs, and ZECs) was a positive step. This contributed to bringing researchers closer to the clients and involving the stakeholders in the zonal research planning process. In addition, the appointment of Zonal Research and Extension Liaison Officers (ZRELOs) enhanced research ­ extension ­ farmer linkages. Empowerment of farmers to manage new or improved technologies and dissemination of the technologies to non-collaborating farmers or villages was in many places quite impressive. Farmers themselves have assisted in training their fellow farmers and through "farmer's" and "field" days. The project has strengthened the human capital base, research facilities, infrastructure, including staff houses and institutional capacity of DRD/DLRT. TARP II enhanced the capacity of researchers to write scientific proposals to access competitive research grants and collaborative research. The project also assisted the zones in identifying potential sources of income that could augment the zonal basket ("Self Help Funds"). The self help resource window has played an important role in financing zonal research activities. Collaborative research activities with IARCs, networks, NGOs agro-industries and competitive - 14 - research grants have contributed substantial funds, research facilities, technical support and training to ZRC. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: In accordance with the DCA, the Authorized Allocation for the Special Account was limited to US $ 750,000. This amount was not sufficient. The situation improved in the third year due to an enhanced financial management system and an increase in the authorized allocation in the Special Account from US $ 750,000 to US $ 1,250,000. 5.2 Factors generally subject to government control: The Government contribution (of 10% counterpart funds) was delayed and inadequate in the first year (8%) and there were no disbursement (0%) in the second year; this eventually affected disbursements to the zones. Disbursement improved in subsequent years, amounting to 100% of the total budget allocation in the last three fiscal years. By the end of May 2004, the Government had disbursed a total of Tsh. 1,429 million, which is 16% more than the amount committed in the DCA. The Government Imprest system that had to be followed was found to be inappropriate for research activities. Transaction costs were high and it was not possible for researchers to account for allocated imprest within 14 days after completion of the task. Moreover, this matter was further complicated by the persistently low and delayed disbursements to the zones. Instead of using SoE to file withdrawal applications for the replenishment of the special account, DRD has proposed to use Financial Monitoring Reports (FMRs) system in the future support for agricultural services. The latter provides quarterly or bi-annual budget resources based on action plan and financing requirement forecast. The reorganization of the Ministry of Agriculture into three different ministries had a negative impact on the research effort and had no discernible positive effects in other dimensions of sectoral performance. While research administration remained under the MAFS, extension services were decentralized and overall coordination of project implementation in the districts was moved to PORALG. Although the decision to decentralize was appropriate and has yielded benefits even in the medium term, at the time local authorities were weak and confusion reigned. This disrupted the research-extension-farmer linkages which were critically important to achieve impact; and to enhance interaction of key players to ensure that available technology was transferred to farmers and adopted. 5.3 Factors generally subject to implementing agency control: The implementing agency should have ensured greater accountability and been more pro-active in dealing with those involved in the poor quality of civil works in the research stations. Apart from TARP II SUA, the overall project M&E procedures were not satisfactorily established. The base-line data were not adequate for tracking the impact of research activities. Although the project mitigated this deficiency through base-line surveys and special adoption studies, this information was not well aggregated to measure the overall impact of various technologies introduced over the project period. The disbursement bottlenecks (for four months during FY 2000/01) due to the splitting of research and training institutions following the ministerial reorganization could have been addressed in a more - 15 - expeditious manner. Lengthy tendering procedures coupled with limited skills and shortage of staff within the implementing agency affected the overall procurement efficiency. Low staff motivation was recognized by DRD/DLRT management, and proposals to provide special awards to the best performing researchers were developed. However due to funding limitations, cash awards were distributed in FY 2001/2002 only. A total of US $ 17,894.8 cash prizes were awarded to 56 researchers and 51 technicians. Staff morale is still low, mainly due to poor remuneration. As a result, there has been a serious brain drain of scientists: 52 staff members (19 PhD holders, 30 MSc and 3 BSc) have left for better jobs outside DRD (including to privatized research). A more efficient mechanism for staff motivation with adequate performance incentives need to be development in consultation with President's Office, Public Services Management (PSM) department. Despite the broad stakeholder involvement in the zonal committees, participation of a critical mass of actual smallholder farmers has been insufficient, and adoption of new technologies has suffered as a result. The weakness resulted in part from the confusion associated with the early stage of decentralization and unclear mandates at the local level. This weakness has been acknowledged by DRD/DLRT, and will be addressed by greater emphasis on farmer groups in the support for agricultural services under the context of the proposed Agriculture Sector Development Program (ASDP). During the course of the project implementation, mechanisms for regular interaction between research and extension were disrupted due to unclear division of labor between MAFS and the President's Office Regional Administration and Local Government (PO-RALG). This problem was addressed by signing of Memoranda of Understandings (MoU) between the ZRCs (still under Central Government) and District/Municipal Councils (under PO-RALG). In order to strengthen the working relationship and clarify tasks between Local Governments and MAFs, a position of a Zonal Research Extension Liaison Officer was created in the zones. It however, took much time and effort to make the new arrangements functional and effective. The implementing agency did not have control over the process of decentralization and the new division of labor at the local level, but a faster reaction to the changing circumstances would have enhanced performance. 5.4 Costs and financing: Total project costs were estimated at US$ 22.98 million over a five year period (1998/99 ­ 2002/3), out of which IDA would provide US$ 21.75 million and US$ 1.23 was to be contributed by the Government. In addition, bilateral donors provided parallel support in other priority research areas. The project became effective on March 03, 1998 with closure originally scheduled for June 30, 2003. The project actually closed on June 30, 2004 after a one-year extension. Following recommendations made by the last TARP II supervision mission (as mentioned in the Closing Aide Memoire from the June 7-11, 2004 Mission), the project had the standard grace period of four (4) months, up to October 31, 2004 to pay for outstanding bills for goods and services. By the end of May 2004, IDA had disbursed a total of US$ 20.67 million (97%) and the Government of Tanzania had disbursed a total of Tsh. 1,429 million (116%). The remaining IDA balance of US$ 753,317.23 was disbursed during the grace period, and by October 31, 2004 a total of US$ 21.39 million, equivalent to 99.8% was disbursed. 6. Sustainability 6.1 Rationale for sustainability rating: Institutional level: Overall project sustainability is likely. Support within the Government, including Ministry of Finance, to invest in agricultural technology is high. The reorganization of the DRD/DLRT to empower zones to do research more responsive to farmers' actual needs has been satisfactory. Overall, the - 16 - institutional framework in which agricultural technologies can be improved is now better understood and is incorporated in the Agricultural Sector Development Program (ASDP). ASDP is a medium term program for agricultural development that will be supported by donors jointly under a sector wide approach. One of the five strategic intervention areas of ASDP is to improve agricultural support services. Within this context, the Government has designed the agricultural services pillar of the ASDP around three key components: (i) farmer empowerment; (ii) provision of services; and (iii) program management. The next phase of support will integrate research and extension services as a continuation of the efforts made by TARP II in the institutionalization of FSA and the Client-oriented Management Approach (CORMA). In order to bridge the financing gap between the closing of TARP II and new support under the sector wide approach, MAFS has in FY 2004/05 set aside an incremental development budget of TShs 350 million and a recurrent expenditure budget of TShs 1,000 million for crops research and development activities. A livestock research budget of Tsh. 480 million has been allocated from MWLD. The Government has embarked on a recruitment process to fill the knowledge gap, particularly in socio-economics with promising responses. Finally, the decentralization process has matured enough that new roles are understood at the local level, and new arrangements for financing will become operative in the coming year. The local government capital grants and the district agricultural development grants can increase demand for knowledge about new and promising agricultural technologies, and provide resources for adoption, and complementary investments to increase returns to adopters. Thus the decentralization process that created impediments to good performance of the research establishment in recent years should be reversed as the capital and development grants roll-out. Farm level: The sustainability of achievements of the project at farm level is also likely. Farmers have learned from their involvement in dissemination of the technologies to non-collaborating farmers. Farmers are able to explain confidently the entire research process and results generated, and in some cases, farmer experimentation is ongoing. Farmers on their own initiative have assisted in training their fellow farmers and disseminated technologies through farmers days. In collaborating villages, Farmer Research Groups (FRG) were formed voluntarily and this is very important for empowerment and sustainability. Since the project only managed to cover a limited population of farmers in on-farm research, demand for improved seeds and other technologies is quite high among collaborating as well as in non-collaborating villages. Outreach, although commendable, could have been improved by greater use of publications and materials explaining improved technologies. According to farmers, their involvement in all steps of research trials has not only improved the relationship between researchers, extension staff and farmers, but also saved researchers time in the field. The project also contributed to promotion of indigenous knowledge through farmer study visits and leaflets. FRGs are quite committed to continue working with researchers on new technologies. 6.2 Transition arrangement to regular operations: There is no doubt about continuity of important activities as well as institutions created by TARP II since the Government is committed to continue financing research stations and maintain the already established linkages. The institutional framework accomplished under TARP II will be strengthened under ASDP. Already this financial year 2004/2005, GOT has disbursed funds to the zones for supporting ongoing research activities. Although the funds are at present not sufficient to cover all the activities in the zones, it is a good indication of the Government's commitment. Donors have pledged support to the roll out of the follow-on Agricultural Services Support Program. Public support for investment in agricultural technology is likely to rise with the new mechanisms for financing at the district level. District Councils will have more money with which to pledge to ZARFs as well as to continue supporting the Research ­ Extension ­ Farmer linkages through contract research. In order to ensure that improved research facilities are being utilized as intended, there is a need for the Government to insist that the pending rectification of the civil - 17 - works is effected. The Government has verbally committed itself that funds will be set aside so that the civil works are rectified and completed before handing the facilities to the research centers. Some current activities are supported through the Bank-financed Participatory Agriculture Development and Empowerment Project (PADEP). These will include rehabilitation and supply of equipment and reagents to four soils laboratories at Mlingano, Selian, Ukiriguru and Uyole; PADEP will also support personnel training on laboratory management. The Agriculture Sector Development Strategy (ASDS) is now in place and programmatic support for the ASDP, including agricultural services, is being formulated. 7. Bank and Borrower Performance Bank 7.1 Lending: The project was the second to be financed by IDA as part of a long-term support program for agricultural research. TARP II was designed to continue to support the institutional development of the NARS and assist in the decentralization process which was initiated by the predecessor project, the NALRP. As part of project preparation, the Bank assisted by among other things sharing of experience and lessons from previous Bank supported projects is the sector. For example, the lessons from previous investments in Malawi, Uganda and Kenya were incorporated into the project design. As noted above, however, quality at entry was only marginally satisfactory due to weakness in monitoring and evaluation. An effort dependent on public funding such as agricultural research is highly vulnerable if results cannot be shown rigorously. For that reason weakness in monitoring and evaluation is a serious deficiency in design. The Bank's contribution to the project preparation is therefore considered marginally satisfactory. 7.2 Supervision: Overall project supervision during the entire project implementation period is considered satisfactory. The project scored a "satisfactory" rating by Quality Assurance Group at supervision conducted in 2002. The Bank maintained regular supervision and a total of nine missions were fielded between 1999 and 2004. Moreover, for the final years the project was supervised from the field, so implementation support was provided on a regular basis in between formal missions. Throughout the project life, the supervision teams focused on the development objectives and overall progress of project implementation. The Team was thorough in assessing all project activities and discussed with the Borrower issues that required immediate actions. Some of the key issues which the Bank team identified and proposed solutions include: reduced interaction and inadequate coordination between staff from different zones; low disbursements to the Zonal Research Centers; and the poor quality of constructed civil works. Many implementation issues were addressed at MTR where revision to the Project Development Objective was also agreed upon in response to institutional changes within Government. The project had a moderate turnover of Task Team Leaders (TTLs)- (three throughout the project life). The project also managed to involve some of the same individuals in most of the supervision missions so that continuity to a large extent was maintained. The achievements concerning the Development Objective and implementation progress of the project have always been rated successful during all supervision missions. 7.3 Overall Bank performance: The overall Bank performance is rated satisfactory. The Bank was supportive in project preparation. During supervision missions, the Bank discussed and recommended solutions to implementation - 18 - bottlenecks. The Bank assisted the borrower to make necessary changes to the project implementation plan as needed. Borrower 7.4 Preparation: The performance of the Government during the preparation of the project is considered satisfactory. The Government participated actively in all steps of project preparation and produced the required documentation for appraisal and approval and in a timely manner. 7.5 Government implementation performance: The Government's implementation performance is marginally satisfactory. Project performance was affected by insufficient counterpart funds in the early period and by a lengthy procurement procedures during the first two years. The situation in these two regards improved thereafter. The reorganization of the agricultural ministries and fragmentation of responsibility for oversight of agricultural research reduced the effectiveness of the project for a significant period. Furthermore, separation of the extension staff from MAC affected the research ­ farmer ­linkage that the project was trying to build. Although this was solved by MoU between MAC and all the district councils involved, these MoU were not well known in the zones. As a result, the implementation and proposed funding arrangements were not followed. Consequently, researchers and extension staff have had difficulty coordinating activities that required financial contributions from both. As the decentralization has matured and roles at the local level are more clearly understood, the situation has improved. Non completion and poor quality of the civil works illustrates weakness on contracts management. In addition lack of involvement of the ZRC management in contract supervision contributed towards poor quality of civil works. The Government should have ensured proper utilization of resources by being more proactive in supervising and inspecting the contractors' work. Despite these noted shortcomings on the part of the Government, the project was able to meet its objectives, as noted above. Performance is therefore rated marginally satisfactory. 7.6 Implementing Agency: The performance of the implementing agency is rated satisfactory. The rating is based on the following assessments: despite the delays and low levels of disbursements in the first two years of the project, 97% of the project funds were disbursed by June 9, 2004 and used for intended purposes. The envisaged institutional structure was put in place, research activities continued at a high level of quality, scientists were trained, and decentralization of activity to the zones accomplished. Annual work plans, progress reports, including audits have been submitted in accordance with the agreements. The borrower complied with the legal covenants. Despite this generally satisfactory performance on the part of the implementing agency, several shortcomings are apparent. Due to poor the remuneration package offered by the public sector, the three deceased zonal accountants in the Lake, Central and Northern zones could not be replaced during the project time. The implementing agency was not efficient enough in seeking timely technical assistance for some important studies, such as the beneficiary assessment, cost benefit analysis, and analytical interpretation and maximum utilization of survey data. Management and supervision of contracts could have been better. 7.7 Overall Borrower performance: - 19 - Overall Borrower Performance is considered satisfactory, based on ownership and responsibilities in project preparation, the contribution of counterpart funding, and overall supervision of project implementation. 8. Lessons Learned Consensus needs to be reached on how research programs should be evaluated, and these goals need to be properly reflected in project objectives. The ultimate economic impact of investments in agricultural research depends on many factors beyond the control of the research system, such as the effectiveness of the extension/farmer advisory services, agricultural marketing policies, international prices, weather conditions, prevalence of diseases and pests, etc. In addition, much of the impact is likely to occur after the investment period of the program. Therefore, several dimensions of impact need to be measured, including progress towards effective and sustainable institutional structures, results on the ground, and farmer satisfaction with the relevance and responsiveness of the agricultural knowledge system. Key performance indicators should be clearly defined during project preparation in order to guide the monitoring and evaluation system. Baseline data surveys, special impact studies should be designed to focus on issues of concern in a way that ensures analytical evaluation of data. Project monitoring and evaluation should be routine and in-built in the work plans, and with adequate budget provision at both national and zonal level. Competent monitoring and evaluation is needed for management of research and to secure continued public support for it. Retrofitting of a good framework for monitoring and evaluation, even when intentions are good and resources available, is risky. Public support for research undertaken by non-public institutions is appropriate and in some cases necessary. Producers of commercial crops in higher income countries can pay most or all costs of research through levies or set-asides, and this approach to funding is appropriate. Where many producers are smallholders in poor countries, however, research will be under-funded if it is dependent on the ability of producers to pay. Privatized, or non-public research entities, such as TaCRI and TRIT, may be able to achieve gains in management relative to public institutions, but they are likely to require public financial support, either from the budget or from donors. Decentralization of management of research to the zones can result in improved performance and impact, but the decentralization must be coherent, including financial management and procurement, as well as technical decisions. The transition process, especially in terms of flow of funds, has been difficult, but the payoffs in terms of creating a system which can address local farmer/client needs in an effective, efficient way have been substantial. Zonal units must interact, and not become scientifically isolated from each other. Decentralization must be matched with good systems of supervision and knowledge sharing. Farmers enjoy and benefit from their involvement in applied research, and mechanisms to ensure greater farmer control over research resource allocation decisions improves research effectiveness. Involvement of farmers builds commitment and competence, and also helps leverage time of researchers. The use of farmer research groups (FRGs) in the implementation of on-farm research improved research-extension-farmers linkage and research cost sharing. Client Oriented Research Management Approach (CORMA) can effectively facilitate demand-driven research, and contribute to institutional sustainability. The ZARFs and associated ZECs provide a useful base to strengthen farmer and broader stakeholder demand within the research system. - 20 - The research system must maintain a critical mass of skilled staff, and this may require modifications in standards of remuneration. Maintenance of a sustainable critical mass of skilled staff within the research system is necessary for sustainability of research work. Further, the fact that there has been some erosion of human resources due to inadequate incentives calls for a need to improve the latter. This problem is being faced by many of the NARS in the region and cannot be solve if addressed by the agricultural sector alone. The Commission for Science and Technology (COSTECH), together with the national research institutes (health, forest, fisheries, wildlife, etc.,) and the Ministry of Public Service develop a special need to consider alternative remuneration packages for researcher scientists, as is the case for the public Universities and other higher learning institutions. There is a need to look at agricultural research and extension as one package for an effective dissemination and adoption of improved agricultural technologies. The separation of research and extension services support has not effectively addressed the challenge of dissemination of technologies generated by DRD. Although MOUs were signed to enhance the linkages of research (under MAFS/MWLD) and extension services (under PO-RALG), and ZRELOs were appointed for each of the 7 zones, greater harmonization and linkage would have improved effectiveness. Efforts should be made to integrate research and extension services in future agricultural sector support programs. 9. Partner Comments (a) Borrower/implementing agency: The Executive Summary of the Government ICR is provided in Annex 10. The Ministry of Agriculture and Food Security also provided comments on a draft version of the ICR, which have been incorporated. (b) Cofinanciers: (c) Other partners (NGOs/private sector): 10. Additional Information - 21 - Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate Strengthening agricultural research systems 100% 100% - Zonal Executive and Technical Committees established and meetings held; operational powers delegated to zonal research directors Rationalization of public research network - 100% 100% Public research network reduced from 50 to 25 research stations; research work decentralized to zones; SUA being mainstreamed; much of zonal programs now conducted on farm. Privatization of Research & Sustainable 100% 90% Research - Tea research privatized; both coffee and tobacco research being privatized; research in most industrial commodities now carried out using cess/levy funds or industry contributions. Human resource development - Personnel 100% 100% needs assessment done; Action Plan for training in place; 38 scientists deputed for training; 200 attended short courses. Strengthening Research-Farmer and 100% 90% Extension Linkages-Extension staff farmers and NGOs participation in IPR and ZEC meetings. New institutional arrangement for research/extension linkage being set up under MoU. Strengthening planning, monitoring 100% 95% evaluation - Zonal PM&E units in place; baseline survey carried out; PM&E formats & performance indicators revised and introduced. Improving research management - Improved 100% 95% financial mgmt. system introduced and fully operational; budget & accounting manuals developed and introduced; DRT headquarters connected to internet; INFORM database updated. Implementation of Procurement & Training Satisfactory Satisfactory Plans - Satisfactory Counterpart funding Satisfactory Satisfactory Disbursement Status. Satisfactory Satisfactory Impacts (adoption, yields, incomes) - PIP 50% 95% envisages measuring these impacts through regular follow-up studies and surveys at mid-term and project closing. Output Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate Improved varieties/innovations developed A minimum of 15 new crop varieties and 5 37 improved crop varieties released and more - 22 - different innovations for livestock will be than 10 different innovations for livestock introduced by years 2003. introduced in target areas Research-capacity improved The number of trained scientists at MSc will Trained 27 Msc. and 21 PhDs, which is be increased from 176 to 203 and at PhD 100% achievement. Thus MSc. and PhD Level from 43 to 64 by 2003. Train 297 staff increased to 201 and 62 respectively. A total in short courses. of 943 staff trained in short courses. Research infrastructure for the 7 ZRCs and 4 Achieved about 80% and 50% on station key institutes improved by 2003. rehabilitation and irrigation works respectively. Research facilities improved by year 2004. Achieved about 98% Planning, Monitoring and Evaluation (PM&E) Zonal units for PM&E established and The Socio-economic units established. About systems institutionalized operational. 70% of staff introduced on PM&E concepts. More involvement in surveys, farmer 60% and 40% of the research managers and technology assessment, conducting partial scientists trained on key PM&E benefit-costs analysis and monitoring of concepts in FY 1999/2000 and Y 2000/01, research activities. respectively. Crop and livestock yields increased By 2003 crop yield/livestock products will be These data are according to M&E survey of increased as follows: 2003. This year was badly affected by drought Rice: 1.4 t/ha -2.0 t/ha Rice: 2.040 tons/ha Maize: 1.5 t/ha - 1.8 t/ha Maize: 1.508 tons/ha Sorghum: 0.9 t/ha - 1.5 t/ha Sorghum: 0.627 tons/ha Millers: 0.7 t/ha - 1.2 t/ha Millet: 0.563 tons/ha Beans: 600 kg/ha - 750kg/ha Beans: 0.791 tons/ha Cassava: 2.2 t/ha - 2.8 t/ha Cassava: 1.169 tons/ha Tea: 0.8 t/ha - 1.4 t/ha (data not available) Cotton: 0.8 t/ha - 2.5 t/ha Cotton: 0.715 tons/ha. Poultry: Poultry: Commercial chicken 200-240 eggs/layer/year - Commercial chicken 225 eggs/layer/year - Local chicken 53 eggs/layer/year Local chicken 45 - 60 egg/layer/year Local: 750 lits/year/lactation Milk production: (in intervention villages) Local from 600 - 1500 litres/year/lactation Improved: 1950 lits/year/lactation Improved from 1,600-2400 litre/year/lactation Beef cattle: Carcass weight increase from 150-180kg Carcass weight: 160 kg Animal diseases and pest incidences At least technology for the diagnosis of tick- minimized borne diseases established by year 2003 2.2 At least 2 IPM methods developed by Year Two IPM technologies for the control of 2003 banana weevil and striga weed in cereals recommended. Incomes of participating farmers Improved Increase by a minimum of 20% by 2003 data not available relative to year 1998 Management Information Systems improved Management Information Systems (MIS) Database using INFORM-R developed; 5 institutionalized and expanded at all levels by zonal farming systems, agro-ecological, and year 2002, e.g. CD-ROM, INFORM, GIS soil maps available using GIS; DRD website scheduled meetings, reports etc. in place; Various agricultural technologies easily accessed in CD-ROM format. 1End of project A rating of 100% signifies complete accomplishment of the indicated action - 23 - Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Actual/Latest Percentage of Estimate Estimate Appraisal Component US$ million US$ million 1. Institutional Strengthening 2.93 3.10 12.8 2. Research Programs 8.50 8.09 36.9 3. Resource Development and Management 11.55 11.64 50.3 Total Baseline Cost 22.98 22.83 Physical Contingencies 0.00 Price Contingencies 0.00 Total Project Costs 22.98 22.83 Total Financing Required 22.98 22.83 Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) 1 Procurement Method Expenditure Category ICB NCB 2 N.B.F. Total Cost Other 1. Works 0.00 2.26 0.29 0.00 2.55 (0.00) (2.04) (0.26) (0.00) (2.30) 2. Goods 2.60 0.45 1.09 0.00 4.14 (2.60) (0.45) (0.98) (0.00) (4.03) 3. Services 0.00 0.00 8.75 0.00 8.75 Training and Research (0.00) (0.00) (8.75) (0.00) (8.75) Grants 4. Operations and 0.00 0.00 7.54 0.00 7.54 maintenance (0.00) (0.00) (6.67) (0.00) (6.67) Total 2.60 2.71 17.67 0.00 22.98 (2.60) (2.49) (16.66) (0.00) (21.75) Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent) 1 Procurement Method Expenditure Category ICB NCB 2 N.B.F. Total Cost Other 1. Works 0.00 3.06 0.40 0.00 3.46 (0.00) (2.20) (0.30) (0.00) (2.50) 2. Goods 2.75 1.00 0.42 0.00 4.17 (2.39) (0.90) (0.41) (0.00) (3.70) 3. Services 0.00 0.00 3.90 0.00 3.90 Training and Research (0.00) (0.00) (3.90) (0.00) (3.90) Grants 4. Operations and 0.00 0.00 11.30 0.00 11.30 maintenance - 24 - (0.00) (0.00) (11.30) (0.00) (11.30) Total 2.75 4.06 16.02 0.00 22.83 (2.39) (3.10) (15.91) (0.00) (21.40) 1/Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies. 2/Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. Project Financing by Component (in US$ million equivalent) Percentage of Appraisal Component Appraisal Estimate Actual/Latest Estimate Bank Govt. CoF. Bank Govt. CoF. Bank Govt. CoF. 1. Institutional 2.93 0.00 2.90 0.20 99.0 0.0 Development 2. Research Programs 7.53 0.97 7.39 0.70 98.1 72.2 3. Resource Development 11.29 0.26 11.11 0.53 98.4 203.8 and Management - 25 - Annex 3. Economic Costs and Benefits Adoption of agriculture technologies is a long term process hence difficult to measure the impact over a short term. On the other hand evaluation of the contribution of research to increases in output is a big challenge since other variables such as materials input, soil conditions, agronomical practices, supply constraints, etc. have an affect on adoption and output. In the above context it can be correctly assumed that increases in yield and adoption rates observed during TARP II may have been a result of predecessor projects including the National Agriculture and Livestock Extension Rehabilitation Project (NALERP). The incremental adoption analysis of the potential impact of TARP II research investment relied on the major production activities of the agricultural sector where research has direct impact (which is measurable) to estimate the incremental benefit values over a twenty year period. The analysis approach models the gradual effect of the project on individual farmers, as they adopt new technologies, and obtain increased output. The objective of the "incremental adoption" and impact analysis was to see how the new agricultural technologies and practices would bring about increased agricultural productivity across a series of selected crops and livestock activities in order for the investment to be justified. The analysis compared the incremental net benefit streams for changes in on-farm economic value against the incremental off-farm cost streams invested in research. The ICR follows the incremental adoption analysis to recalculate the Economic Rate of Return (ERR) over a 20-year period from 1998. The analysis indicated that the Net Present Values (NPV) of the proposed investment would be positive at Tshs. 12.6 billion or US $ 21.0 million at a discount rate of 12%. The ERR was estimated at 21% while the Cost Benefit Ration (BCR) was 1.6. The "incremental adoption" analysis indicated that the above ERR and BCR could be obtained by the research-extension system which introduces/transfers technologies and farm practices to about 25% of the farmers cultivating the selected crops (and raising the selected animals) in the period from the fourth to the ninth year of the project. Since the crops represent some 70% of the area cultivated, (and if area cultivated is evenly shared amongst the country's 3.5 million smallholders), this implies that the invested research funds would bring an additional 875,000 farmers to use improved production techniques over the project and the post project period. Therefore the project proposed had potential to improve the country's agricultural GDP growth rate (which by then was about 2.5% p.a.) by about eight percentage points. In estimating costs, benefits and rate of return a number of assumptions were to be made to guide the framework of analysis at both the activity (farm) and aggregate levels as elaborated below: (i) Target production activities: As research planned would be country-wide, a range of activities were incorporated in the analysis to reflect diversity and representation of different farming systems. In addition, only activities with demonstrable (measurable) impacts were incorporated in the analysis. Consequently fourteen production activities (6 cash crops, 6 food crops and 2 livestock activities were selected for the analysis. These activities account for over 70% of the total cultivated area in a given season and 20% of the cattle herd. Some cash crops were adopted for analysis as they form an essential part of a smallholders' farming system. Further, some of the cash crops supported by donor funds and by cess/ levies imposed on marketing of these crops. These analysis however, excluded the cash (industrial) crops. (ii) Benefits from Research: Benefits from research (i.e. increased yield values) are derived from new and efficient technologies and farm practices which are adopted by farmers. In this analysis, production benefits were estimated net of on-farm input and labor costs. The project impact (net benefits) is compared, - 26 - theoretically, with what would have happened had the project not been implemented (i.e. the with and without project situation). The benefits 'with project' are assumed to accrue from the fourth year as the first 3 years are assumed to be the gestation period for the new technologies to be adopted by the farmers. Impact of some technologies is taken from even later years. It should be noted however that agricultural productivity would continue to benefit from the 'without project' situation due to past research investments and related factors such as mechanization, but such increment would be very small (e.g. about 0.15% as opposed to the estimated 8% growth rate p.a. with the project). (iii) There are several other benefits of the project which are not quantified by this study. The increased farm productivity derived from the project will enhance food security of the smallholder households. Increased incomes may be invested in education, and in improving nutrition and health standards. The project would promote sustainable use of natural resources and help arrest the degradation of environment. As over half of the smallholder farmers in Tanzania are women, the project will directly contribute to improving women's incomes and living standards of the families. The project will also contribute to capacity building through the planned training of some research personnel. Such personnel would improve their ability to undertake agricultural research through which they would contribute to the country's objectives of sustainable farming and food security. - 27 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, 1 FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation 01/1997 5 TEAM LEADER (1); PROCUREMENT (1); FIN. (1); PROG. ASST. (1); AGRIC. MGT AND TRAINING (1) Appraisal/Negotiation 08/1997 7 TEAM LEADER (1); RESEARCH/TECH. (1); ECON. ANALY. (2); PROJECT COST (1); PROCUR. (1); FINANCIAL MGT. (1) Supervision 11/1999 11/1999 11/1999 5 TEAM LEADER (1); PROC. S S SPECIALIST (1); RURAL DEV.SPEC. (1); FINANCIAL MGT. SPEC. (1); AGRICULTURIST (1) 05/2001 7 TTL (1); RURAL S S DEV.SPECIALIST (1)NRM SPEC. (1); SENIOR AGRIC (1); FINANCIAL SPECIALIST (1); FINANCIAL ANALYIST (1); PROCUREMENT SPEC. (1) 10/2001 5 TEAM LEADER (1); RURAL S S AGRIC. SPEC. (1); SR. AGRIC. (1); SR.FINANCIAL ANALY. (1); PROCUREMENT SPEC. (1) 05/2002 5 SR. AGRONOMIST (1); RURAL S S AGRIC. SPEC. (1); SR.AGRIC (1); PROCUREMENT ANALYST (1); SR, FINANCIAL ANALY. (1) 12/2002 7 TEAM LEADER (1); SR S S FIN.ANALYST (1); PROC.ANALYST (1); M&E (1); FIN. MGT.SPEC. (1); NRM SPEC (1); SR.AGRIC. (1) 06/2003 7 TEAM LEADER (1); RURAL S S DEVELOPMENT SPEC (1); SR FIN.ANALYST (1); SR.PROC.SPEC (1); M&E (1); - 28 - FIN. MGT.SPEC. (1); NRM SPEC (1); SR.AGRIC. 02/2004 6 TEAM LEADER (1); RURAL S S DEVELOPMENT SPEC (1); SR FIN.ANALYST (1); SR.PROC.SPEC (1); M&E (1); FIN. MGT.SPEC. (1); SR.AGRIC 06/2004 5 RURAL DEV. SPECIALIST (1); S S PROCUREMENT ANALYST(1); FINANCIAL MNGT SPEC. (1); M&E SPEC. (1) ANALYST (1); SR. AGRICULTURIST (CON (1) ICR 10/2004 3 RURAL DEV.SPECIALIST S S (1); OPERATIONS OFFICER (2) (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 45.6 183.1 Appraisal/Negotiation 17.1 46 Supervision 119.34 492.07 ICR 10.7 21.37 Total 192.74 742.54 - 29 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Macro policies H SU M N NA Sector Policies H SU M N NA Physical H SU M N NA Financial H SU M N NA Institutional Development H SU M N NA Environmental H SU M N NA Social Poverty Reduction H SU M N NA Gender H SU M N NA Other (Please specify) H SU M N NA Private sector development H SU M N NA Public sector management H SU M N NA Other (Please specify) H SU M N NA - 30 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending HS S U HU Supervision HS S U HU Overall HS S U HU 6.2 Borrower performance Rating Preparation HS S U HU Government implementation performance HS S U HU Implementation agency performance HS S U HU Overall HS S U HU - 31 - Annex 7. List of Supporting Documents Project Appraisal Document United Republic of Tanzania, Agricultural Research Project, Report No. 17175-TA, November 25, 1997, Agriculture Operations Eastern and Southern Africa, Africa Region Aide Memoire/Supervision Reports Appraisal Mission (September 11, 1997), Aide Memoire Project Progress Review Mission ( September 29, 1999) - Statement of Mission Objectives Supervision Mission/Mid Tem Review (March 2001), Aide Memoire Supervision Mission (October 2001), Aide Memoire Supervision Mission (May 2002), Aide Memoire Supervision Mission (December 2002), Aide Memoire Supervision Mission (June 2003), Aide Memoire Supervision Mission (February 2004), Aide Memoire Closing Mission (June 2004), Aide Memoire Studies and Reports 1. Project Implementation Plan (PIP) 2. Survey Report of Farmers Participating in On-farm Research, Department of Research and Development (DRD), Ministry of Agriculture and Food Security, Tanzania, February 2001 4. Final Report on Baseline Survey on the Agricultural Research Systems under the Department of Research and Training, Bureau for Agricultural Consultancy and Advisory Services (BACAS), Sokoine University of Agriculture, Morogoro, Dar es Salaam, March 2000 5. Survey Report of Farmers Participating in On-Farm Research, DRD, May 2004 6. Review of Privitazed Agricultural Research, DRD, May, 2004 7 Review of the National Agricultural Fund in Tanzani, DRD, May 2004 - 32 - Additional Annex 8. Achievements from Research Programs Table 1: Varieties Released During TARP II Period Crop/Commodity Number of Varieties Name of Varieties Maize 6 UH 615, Lishe H1, Lishe H2 Lishe K1, Situka-1, Situka M1 Wheat 1 Chiriku Rice 3 TXD306, TXD88, TXD85 Sorghums 4 Macia, Pato, Wahi, Hakika Millets 2 Shibe and Okoa Phaseolus beans 5 Wanja, Uyole 98, Uyole 03, Urafiki Cowpea 2 Vuli1, and Vuli 2 Pigeonpea 2 Tumia and Mali Soyabean 1 Uyole Soya-1 Groundnuts 2 Sawia and Upendo Cassava 3 Naliendele, Mumba, Kibaha Sweet potato 5 Simama, Jitihada, Vumilia, Sinia, Ukerewe Cotton 1 Mkombozi Table 2: Improved Seeds multiplied by Research Centers in Year 2003 Crop Variety Research Institute Amount of Seed (tons) Situka Selian 1.58 Maize Lishe Selian 2.90 UH 615 Uyole 162 Various Ukiriguru 3.7 Rice (Various) Katrin and Dakawa 140 Wheat Chiriku Selian 0.3 Phaseolus bean (Various) Selian and Uyole 48.90 Sunflower Record Ilonga 2.5 Soyabeans Bossier Ilonga 1.5 Cotton Mkombozi Ilonga 0.96 UK91 Ukiriguru 24 Cassava Various Kibaha, Hombolo and 2,077,044 cuttings Ukiriguru Banana Various (imported Selian 8,000 plantlets from Kenya) Coffee Clonal coffee Ukiriguru 103,940 seedlings Sweet potato Various Ukiriguru 443,854 vines Table 3: Crop Yield (kg/ha) from farmers participating in on-farm adaptive trials - 33 - (Research Collaborators) and others (Non-Collaborators) Crop Collaborators Non Collaborators t-test at 5% Maize 1635.63 1290.18 S Rice 2166.70 1861.53 NS Sorghum 635.48 601.48 NS Sunflower 1348.73 1094.20 NS Beans 862.75 675.00 NS Cassava 1326.85 858.00 S Coffee 691.13 418.80 NS Ground nuts 775.43 625.43 S Sweet potatoes 3268.45 1570.60 S Bananas 9242.03 6085.83 NS Cotton 811.85 453.05 S Sesame 559.38 327.00 S Round potato 13630.50 9870.40 NS Cashew 694.03 478.18 NS Pigeon pea 872.63 1086.78 NS Wheat 1582.58 1725.55 NS Cowpea 603.00 828.00 NS Source: M&E Survey of 2003 - 34 - Additional Annex 9. List of Selected Case Studies of TARP II Technical Problems Impact Existing or Pipeline Technologies in Tanzania Maize ­ Grey Leaf Spot Widespread in high altitude GLS-tolerant and high (GLS) Disease (>1000 m.a.s.l) parts of yielding varieties (UH Tanzania. The disease causes 615, UH 6303) up to 50% crop loss and developed at ARI-Uyole declined quality. Maize is a and now available on the staple food for majority of market. people in Tanzania, and therefore GLS threatens national food security. Maize is also source of cash income to resource-poor peasants. Cassava Mosaic Disease Cassava is an important food Promising CMD-tolerant (CMD) security crop due to its clones MM 96/4684, adaptability to marginal soils MM 96/46/9, SS4, TME and erratic rainfall conditions. 14, I91/00063, I91/2324, Major biotic stress has been MH 91/0067, and CMD, particularly the I91/0057 have reached Ugandan Variant (CMD-UgV) on-farm stage of that reduces potential yields by evaluation 80-100%, limiting availability of cassava leaves, an important and nutritive green consumed by many households. More than 500,000 farmers reached through direct and indirect distribution (multiplier effect) of CMD-tolerant varieties. Post-harvest losses in Tanzania is among the world Improved processing roots/tuber crops e.g. cassava top 10 largest cassava equipment tested, producers. More than 80% of recommended and is now total production is consumed available. as food while the remainder is used as livestock feed, alcohol and starch making, and some is exported. Most traditional products have poor quality, and thus fetch low market values and some is completely - 35 - spoilt and wasted. More than 200,000 beneficiaries have been trained in the past seven years. It includes about 6 fabricators, more than 20 processors, 400 farmer groups and the rest are individuals farmers. Bean seed availability Beans ­ important crop for Improved varieties Uyole household food security, 94, Uyole 96, Uyole 98 general food chain and an and Kabanima are high important commodity for crop yielding and tolerate merchants. However, the crop diseases. They are is faced with low productivity therefore included in the and tolerance to diseases. High Quality-Declared Seeds quality seeds are not readily (QDS) production available. system. Bean Improvement Program Producer and basic seeds has produced producer and produced and is available basic seeds to a tune of 6 for sale. tones/year, and seed multiplication is going on in a number of regions Bean Stem Maggot (BSM) BSM affect bean seedlings Control measure: below the ground resulting in application of Murtano or yield losses of up to 50-100%. Marshall at 30 mg/100 Kg seeds. Also spraying bean Awareness creation has been seedling with Thiodan, carried out in Mbeya and DUSBAN, Sumithion, Mbozi districts about the Selecron or botanicals 4-5 BSM; its insect pest, life cycle, days after germination. mode of action, symptoms and control measures. Earthing-up and soil fertility improvement has been demonstrated to reduce damage. Farmers have started domesticating Tephrosia and Vernonia because they also have some other useful elements of soil fertility improvement and control of livestock ectoparasites. Yield increase of up to 50% has been recorded when BSM is controlled on - 36 - time. Short shelf life of tomatoes Tomato is one of the most Varieties Tengeru 97 important vegetable crop (indeterminate) and grown and consumed in Tanya (determinate) are a Tanzania. Though fairly high result of vigorous yielding, the earlier varieties selection of breeding lines were intolerant to diseases and obtained from the Asian lacked fruit firmness to Vegetable Research and withstand long-distance Development Centre transportation and good shelf (AVRDC). Seeds for life. planting now readily Tanya and Tengeru 97 tomato available in the market. varieties have increased yields and subsequently incomes by farmers, and have changed dependence on foreign markets for seeds. Cashew Powdery Mildew Cashew powdery mildew · Sulphur application at Disease (CPMD) disease reduces cashew yields 0.25 Kg/tree/ round (5 by up to 70% if not controlled. rounds) at 21 days Cashew is mainly grown by intervals, smallholders (approx. 280,000 · Organic fungicide households) and is main cash Anvil, Bayfidan, and Topas crop in Mtwara, Lindi and for control and found more Coast regions. effective for cure and protection · 20 cashew clones identified as potentially high yielding and tolerant to diseases and pests Management of nematodes Banana is an important food Chemical: and weevils in bananas in crop in most parts of Tanzania. · Carbofuran 5G control Kagera Region In 1978, reduced banana yield weevils and nematodes at observed from 30 to 4.5 60gm/stool, thrice/year tones/ha due to weevils and · Neem extract sprayed nematodes infestation. It around stools created food insecurity in · Soaking in Furadan dip Kagera region forcing families Improved varieties: to relocate. Yangambi KM5 and After interventions, banana Bluggoe are known to yields increased from 4.5 to 20 resist nematodes/weevils tones per hectare per year. Cultural: Increased production and Weevil trapping; deep consequently income and planting of stool; Chopping improved livelihoods in Kagera of pseudo-stems after - 37 - region. harvest; intercropping; practicing sequential uprooting and replanting of run-down fields. Deforestation/environmental Introduction of tree species for · Introduced and trained degradation in tobacco fuel wood is significantly farmers how to manage and growing areas conserving environment by use/harvest indigenous fruit reducing pressure on trees and fruits indigenous miombo forests. · Introduced suitable tree Some farmers have own species for provision of woodlots. fuelwood, soil fertility improvement, etc. Trees and shrubs suitable for soil fertility popular among farmers and many have plots for soil fertility improvement in fallow systems. On-farm Introduction of About 98% of total cattle Mpwapwa breed cattle Dual-Purpose Mpwapwa production in Tanzania introduced in pilot Cattle constitutes Tanzania villages and milk Short-Horn Zebu (TSZ) which production increased is poor dairy cattle. Earlier from 1-2 liters to 5-7 efforts to improve milk liters/day under farmer production of TSZ were management. through crossbreeding at Mpwapwa that resulted into Mpwapwa breed by 1958. Rigorous systematic on-farm testing and evaluation of Mpwapwa started in 1980s and Mpwapwa breed was found to perform best. LPRI Improved mean birth Mpwapwa developed an weights range from 21 to elaborate program "Open 30 Kgs, and demand for Nucleus Breeding Scheme" to bulls has increased assess Mpwapwa cattle under tremendously. farmer conditions. Apart from evaluation, a program to improve growth rates and body weights using Mpwapwa bulls in pilot villages was also introduced. There has been an increased - 38 - demand for Mpwapwa bulls by livestock keepers following their performance in the participating locations. For example Longido are in need of 78 bulls, Ngorongoro 120, Simanjiro 40 and Manyoni 16 bulls. Given the small number of Mpwapwa cattle at LPRI, JICA agreed to contribute a total of TShs. 126,000,000.00 through ZARF for the multiplication of the breed. Improved Dual-Purpose Tanzania has 12.5 million Milk production from the Goats in Central Tanzania goats raised under introduced goats ranged agro-pastoral system with from 1.0 to 3.0 liters/day average weight 18-25 Kg. (vs 0.2 liters/day by local Efforts made to improve both breeds). The goats meat and milk production by weighed about 2.9 and crossing with exotic breeds 13.5 Kgs at birth and 16 into Blended goats with live weeks of age, weights of up to 45 Kgs and respectively. 1.5 liters /day/doe. These were later put under evaluation under farmer conditions under crop-livestock based production system. 25 farmer groups trained on management of dual-purpose goats in 4 districts in 2001, and 500 goats and 50 bucks were distributed to farmers. Goat population increased from 550 to 1856 and farmer groups increased to 43. An APEX body formed to coordinate all primary groups. Improvement of Rural Tanzania has a total of 27 Thermostable Vaccine Chicken Productivity million chickens under (I-2) available for control Through Vaccination Against scavenging production system of NCD in rural areas. New Castle Disease by smallholder farmers. New Castle Disease (NCD) hampers local chicken industry and has caused mortalities of up to - 39 - 100% where vaccination is not done. Earlier efforts to control disease were difficult because conventional vaccines require cooling facilities such as refrigerators that are not available in rural localities. Laboratory tests on thermostability of the thermostable vaccine (I-2) were followed by on-farm testing on the vaccine efficacy, with over 14 million doses of vaccine having been produced for evaluation and on-farm testing. Results have shown a reduction of mortality from 80-90% to 10-20%, and the vaccine remain viable when kept under room temperature for 1 month, 6 months in a refrigerator, and for some years when frozen. - 40 - Additional Annex 10. Executive Summary of the Government's ICR EXECUTIVE SUMMARY 1 Background The Government of Tanzania, with the support of various bilateral and multilateral donor agencies, appraised the National Agricultural Research System (NARS) in 1997/98 following a successful implementation of the National Agricultural and Livestock Research Project (NALRP), implemented between 1989/90-1996/97. The appraisal identified problems of inadequate funding, lack of clear research orientation and priorities that are demand-driven. Subsequently it was resolved to design a follow up phase project that would provide a framework for addressing the above mentioned problems. The second phase of the Tanzania Agricultural Research Project (TARP II) was formulated to continue providing support to institutional development of the NARS, implementation of priority research programs as well as resource development and management. 2 Project Objectives: The Second Phase of the Tanzania Agricultural Research Project (TARP II) was implemented for 6 years from July 1998 to June 2004. The Department of Research and Development (DRD) of the Ministry of Agriculture and Food Security (MAFS) executed the project, in collaboration with the Department of Livestock Research and Training (DLRT) of the Ministry of Water and Livestock Development (MWLD) and Sokoine University of Agriculture (SUA). Specific Objectives of TARP II were to: · increase efficiency and productivity of agricultural production(crop & livestock), as well as sustainable use of natural resources through the generation and dissemination of appropriate agricultural technologies, with emphasis on the needs of smallholder farmers; · support the institutional development of the national agricultural research system, particularly the empowerment of the zonal research centers in financial resource management, operational decision making and formulation of demand-driven, client-oriented zonal research agendas; · encourage the privatization of agricultural research as well as the strengthening of linkages between farmers, extension agents and research scientists and; · strengthen DRD's capacity to manage and co-ordinate agricultural research through the improvement of research planning, monitoring and evaluation as well as human resource development, rehabilitation of research infrastructure, and procurement of essential research materials and equipment. - 41 - TARP II had three components namely institutional development, research program and resource development and management The total project cost amounted to US $ 22.98 million. Out of this the World Bank/IDA was to provide US $ 21.75 and Government of Tanzania US $ 1.23 million. Other bilateral donors provided parallel financing support to the project. These included The Netherlands US $ 3.817 million for Lake and Northern zones; Norwegian Government (through NORAD) that provided a grant amounting to US $ 5.298 million to TARP II-SUA and Ireland Government (through Irish Aid) -committed a total of Euro 2.3 million to support Eastern Zone Client Oriented Research and Extension Program (EZCORE). 3. Major Achievements: 3.1 Institutional Development Under this component the focus was on redefining the roles of the Department of Research and Development, establishing zonal research committees, zonal empowerment, right sizing of the research establishment and supporting initiatives for privatizing research particularly on industrial commodities such as tea, coffee, sugar cane etc. The seven Zonal Research Centers, which were set up during NALRP, continued to enjoy semi-autonomy status in day-to-day operations. The local stakeholders in the zones determined the research priorities of the zone. DRD and DLRT at headquarters level provided guidance and technical backstopping and disbursed funds regularly according to the zonal needs, priorities and governing policy. In order to enhance zonal effectiveness zonal research committees were established and empowered. These are Zonal Internal Program Reviews (IPRs), Zonal Technical Committees (ZTCs), and Zonal Executive Committees (ZECs). DRD/DLRT continued to benefit from a wide range networks and collaborative partnerships with various International and Regional Agricultural Research Centers and Organizations. These include ASARECA, SACCAR, CIMMYT, ICRISAT, IFPRI, ISNAR, and others. There were also collaborative activities with local institutions of higher learning, notably SUA and UDSM. The mainstreaming of SUA into NARS has been rated as very successful. Following the decentralization of extension services to the Local Government Authorities (LGAs) (district and municipal councils), the close link that existed between research and extension was somehow disrupted. In order to restore this linkage the Zonal Research Centers signed memorandum of understanding (MoUs) with the LGAs in the respective zones. To-date, all 120 LGAs have signed the MoUs. To further strengthen linkage between research centers and all stakeholders in the zone, the creation of liaison offices and appointment of Zonal Research and Extension Liaison Officers (ZRELOs) has greatly - 42 - improved the situation. The annual participation of the DRD/DLRT and the zonal centers in the National and Zonal Agricultural Shows( Nane Nane), field days, quarterly training workshops, farmer filed visits, etc, has enhanced linkage and dissemination of research findings. TARP II implementation also involved promoting private sector participation in funding and managing agricultural research particularly industrial commodities such as tea, coffee, tobacco, cotton, cashew, pyrethrum, sisal, sugarcane and barley. Modalities were developed whereby, the private sector (particularly the commodity boards) were to provide operational funds for actual research including payment of work related allowances to supplement the staff salaries. Research on cashew, sugarcane, cotton and pyrethrum is currently operating in this mode. Research on tea, coffee and tobacco was fully privatized between 1998 and 2001. A comprehensive review of the experience with privatization of research on tea, coffee and tobacco has been done. The Tea Research Institute of Tanzania (TRIT) and Tanzania Coffee Research Institute (TaCRI) have established themselves as viable and effective institutions. The Tobacco Research Institute of Tanzania (TORITA) is still evolving and needs further review and facilitation (refer separate review report). 3.2 Research Programs This component focused on provision of incremental operating costs for priority research programs in the DRD/DLRT research network as well research infrastructure at Sokoine University of Agriculture (SUA) and supporting its research programs. In addition, this component supported the establishment of Zonal Agricultural Research Funds (ZARFs) and also facilitated breeder seed production. The key performance indicators which were revised during MTR and emphasized on increasing yields of crops and livestock products, and enhancing research ­ extension ­ farmer linkages, among others 37 new improved crop varieties were released against the target of 15. The results from time series data survey for 1998, 2000, and 2003 indicated that the proportion of farmers who used improved varieties has been increasing over time. For example farmers who used improved maize varieties in 1998 were 25 percent and increased to 38 % and 56% in 2000 and 2003 respectively. On the other hand, ten improved innovations on livestock production were developed and disseminated by the end of 2003(the target five innovations) resulting in substantial improvement in livestock yields. In the Central Zone milk for instance, production has increased from 2 to 5 liters per day per cow following the multiplication and introduction of the Mpwapwa breed. In both Central and Northern zones, cross breeding of Mpwapwa bulls with indigenous TSZ has increased body weight gains from16 to 30 kg. - 43 - Achievements on control of livestock diseases have been recorded and include a reduction of chicken mortality rates by more than 70 % through the use of the Australian thermostable vaccine (I-2) to control Newcastle Disease. The cost of accaricide use has been reduced by 50 to 70 % following applications of East Coast Fever vaccine. Mass production of sterile male tsetse flies continued for wider usage at prone areas of Mafia Island and Kiteto District in the Northern Zone. Furthermore, TARP II supported research interventions on soil fertility improvement through the use of organic farming through application of improved land husbandry practices such as leguminous cover crops, tie-ridging, contour farming and minimum tillage. Some of the achievements after the above interventions include the yield increases of cotton, maize and rice to about 20 - 60% in a wide range of environments. Consequently, farmers' demand for industrially manufactured inputs (fertilizers and chemicals) has been reduced by at least 10 to15 percent. Several trees and shrub species (i.e. sesbania sesban, Gliricidia sepium, Tephrosia vogelii, Lablab purpuriens, clitoria ternatea) which are N ­ fixing have been recommended and have reached over 3000 farmers in the Western Zone. Three early maturing trees (i.e. Acacia leptocarpa, Acacia julifera and Acacia crassicarp) have also been recommended. These species take about 5 years to mature compared to indigenous ones that take over 30 years. Application of 5t/ha mulch from legume species (e.g. Leucaena ) increased maize yields from 0.5t/ha to 3t/ha. TARP II also supported the identification, propagation, and promotion of processing of indigenous tree fruits in the Western Zone. Some of the suitable species identified are Vitex mombassae ("Ntalali") Strychnos caculoides ("Ntonga"), Parinary curastellifolia ("Mbula"), Flacourtia India ("Mbuguswa") and Vitex doniana ("Mfuru"). Products made include jams, juice and wine. Over 1000 farmers have adopted processing technology and sale of processed products has significantly increased the household income, especially among women. Labour saving technologies in land tillage and soil conservation were also introduced to farmers, in the Southern Highlands and Lake Zones. The technologies introduced included ox-drawn weeders, ridgers, ripper tines, sub-soilers, carts and rice peddlers. Labour input using improved implements has been reduced by 178% (i.e. from 21.65 hrs/ha using conventional mould-board plough to 7.80 hrs/ha using conservation tillage implement). In addition, the implement have enabled farmers to increase area under cultivation and crop productivity due to carrying out farm operations in time and efficiently. Other equipments evaluated and fabricated include cassava graters and sweet potato chippers that have significantly reduced the labor requirement by 200% in post harvest processing. Such devices have also improved the quality of the products. The Client Oriented Research Management Approach (CORMA) was piloted in the Northern and Lake Zone since 1998 with the financial support from the Netherlands - 44 - Government encourages demand-driven research, represents important complementary moves towards decentralization and involvement of private sector in research and promotes inter-agency and inter-disciplinary research teams. Emphasis has been in the areas of human resources management, financial, linkage and collaboration, research and quality output production. It was during the subsequent external review of the performance of CORMA conducted in mid 2003, it was recommended to spread the approach in other zones and also reconstitute and rename it the Client Oriented Research and Development Management Approach (CORDEMA) to cover also development issues during the up scaling process. 3.3 Resource Development and Management: This component focused on: (i) Strengthening human resource development; (ii) providing assistance to SUA for carrying out approved training programs; (iii) consulting services and short-term technical assistance; (iv) selective rehabilitation of research stations, equipment and infrastructure; and (v) strengthening financial management and accounting, information and communication services, documentation and library services, and research monitoring and evaluation. The scope of work for the research station rehabilitation program was to cover the rehabilitation of residential houses, laboratories, offices, and access roads. The irrigation works aimed at ensuring the continuity of research activities even during dry season and hence shortening the duration of the research period. Irrigation works at Uyole has been completed, while at Mpwapwa and Ilonga are over 90% completed. Whereas the irrigation works for Naliendele and Tumbi did not start due to tedious tendering procedures and none responsiveness of bidders; irrigation systems at Ukiriguru collapsed due to poor design and subsequently the contract was suspended in early 2004. Procurement under TARP II focused on replacement of items supplied in the in mid- 1990s (vehicles, tractors, computers and furniture) as well as consumables such as laboratory reagents and spares for vehicles and farm equipment. The project procured almost all (95%) the planned items. The procurement of laboratory equipment and selected chemicals was shelved due to delayed delivery prior to project closure at the end of June 2004. Forty-eight research scientists were enrolled for postgraduate studies at M Sc and Ph D degree levels. Out of the planned 27 MSc levels, 25 staff successfully completed their MSc degree courses and returned to their respective work stations. One MSc candidate died and the other did not succeed because of ill-health. Of the planned 21 doctorate degrees, three have successfully completed their courses and returned to respective workplaces. The remaining PhD students are expected to finish during 2004/2005. - 45 - A total of 943 staff against the planned 297 attended short courses in various fields and disciplines. The number of staff trained exceeded the target because DRD received assistance from DFID - East Africa Natural Resources in order to train staff on research management skills. Before 1997, FSA training was offered specifically to researchers in zonal FSR teams. During TARP II, the goal was to train all researchers in the DRD/DLRT network. A total of 372 researchers (268 male and 104 female) acquired FSA skills between 1997/98 and 2003/04. A total of 56 research scientists and 51 technicians were awarded cash prizes totaling US $ 17,894.8 due to outstanding performance in their fields. Some of the problems encountered during TARP II implementation included the continuing erosion of the human resource base in search of greener pastures as well as deaths. Other bilateral donors provided parallel support to TARP II. These included Government of Netherlands (GoN) for the Lake and Northern zones, Norwegian Government (NORAD) for TARP II-SUA and Government of Ireland for EZCORE in the Eastern Zone. The DRD/DLRT started several initiatives to develop extra-budgetary sources to cover non-salary recurrent expenses related to agricultural research. The establishment of Zonal Agricultural Research Fund (ZARFs) was one of the efforts to ensure research-funding sustainability. The World Bank/IDA through TARP II provided matching funds to already contributed local stakeholder funds. This has integrated aspects of systems perspective in all research projects. The establishment of the Agricultural Research Fund (ARF) in Tanzania was an attempt to ameliorate the problem of insufficient funds for research. Initially ARF was expected to provide a transparent financing mechanism for the DRD to support and facilitate collaborative, contract and research grants for agricultural and livestock priority research programs. 4. Impact and Sustainability of TARP II interventions During TARP II implementation period, enormous improvements in research approaches and application of innovative methodologies have been realized. Farmers were effectively involved in the process of Technology development and Transfer (TDT). The results of time series studies conducted in 1998, 2000 and 2003 revealed that about 21%, 35% and 51% of farmers respectively were involved in the TDT process. A good number of gender balanced farmers groups participated on the assessment of proven research technologies. Despite of unfavorable weather condition during the project period, farmers who collaborated with researchers have substantially increased the yields of their crops The establishment of semi-autonomous zonal research committees (IPRs, ZTCs and ZECs), whereby zonal stakeholders are empowered to determine and influence research agenda and budget in their respective zones, has enhanced the move to strengthen client oriented and demand-driven research and development agenda. In addition, the appointment of Zonal Research and Extension Liaison Officers (ZRELOs) has helped to strengthen research, extension and farmer linkages. - 46 - The establishment of ZARFs has proved to be very instrumental in the quest to promote sustainable research funding mechanisms. Such initiatives have created awareness, such that some LGAs and the private sector have come to appreciate the need to contribute to research funding. This has enhanced partnership and cost sharing in research endeavors. The use of Farmer Research Groups (FRGs) in the implementation of on-farm research has greatly improved research-extension-farmer linkages and research cost sharing. Most research scientists have been exposed to participatory research methodologies Due to acute shortage of quality seeds in the country research centers were involved since 2001 in the multiplication and distribution of good quality seed to farmers particularly maize, sorghum and millet, paddy, and some vegetable seeds. In 2003 alone Uyole produced some 162 tons and Seliani 220 tons of quality maize seed. Dakawa and Katrin Ifakara produced a total of 140 tons of rice seed during the same period. Privatization of research on industrial commodities such as coffee, tea and tobacco has relieved the Government on the burden of funding the broad research spectrum. Treasury funds are increasingly used to cover research on food crops, livestock production and natural resources management aspects. - 47 - - 48 -