Rwanda - Agricultural Research Project

Docmen of The World Bank FOR OFICLAL USE ONY Rhpot N. 12590 PROJECT COMPLETION REPORT WA~ AGRICULTURAL RESFARCH PROJECT (CREDIT 1546-RN) DEBCBER 10, 1993 MICROGRAPHICS Report No: 12590 Type: PCR Agriculture Operations Division South-Central and IndiAn Ocean Department Africa Region Tks document has a resticted distibudon and may be used by recpiens oiny In dte peforman of their officid duts. Its contents may not othwbise be dilosed witdho Wold Bank aunwoidaon.  CURRENCY EQUIVALENTS Currency Unit = Franc Rwandais (FRw) FRw = US$1.00 1985 - 101.3 1986 - 87.6 1987 - 79.7 1988 - 76.5 1989 - 80.0 1990 - 82.6 1991 - 125.1 10i592 - 137.0 WEIGHT AND MEASURES Metric System ABBREVIATIONS CAT Interntional Center for Tropical Agriculture (Centre international d'agronomie tropicale) CIP tional Center for Potatoes (Centre intematonal de la pomme de terre) EEC European Economic Community GOR Government of Rwanda GIZ German Agency for Technical Cooperation (Agence de copEration technique - Allemagne) ICRAF Interational Center for Research in Agro-festry IL'RC International Development Research Center IFAD lnemational Fund for Agricultural Development ILCA International Livestock Center for Africa ISAR Rwandese Institute for Agricultmal Research (Insfftut des sciences agronomiques du Rwanda) ISNAR International Service for National Agricultural Research PRAPAC Regional Program for the Improvement of Potatoes in Central Africa (Programme rdegional d'am6lioration de la pomme de terre en Afrique Cerle) UNR National University of Rwanda (Universite nationale du Rwanda) USAID United States Agency for International Development FISCAL YEAR OF BORROWER 1 January - 31 December  liOR OMCLAL USE ONLY THE WORLD BANK WashbVton, D.C. 20433 U.S.A Offie f DOrd"or4nal Op.rMone Evalualbn December 10, 1993 MEMORANDUM TO THE EXECUTIVE DIRECITORS AND THE PRESID SUBJECfl Project Completion Report on Rwanda AMdcultural Rsarch Project (et 1546RW ) Attached is a copy of the report entitled Project Completion Report - Rwanda - Agricultural Research Project (Credit 1546-RW)" prepared by the Agriculture Operations Division, Country Department Im Africa Regional Office. Part II was prepared by the Borower. The project outcome is considered satisfactory, because it reached its principal objective of developing a national research plan while strengthening the institutional and physical infrastructure for a follow-on, comprehensive research operation. Sustainabilit is likel assuming successful impementaton of the follow-on project approved in October, 1993. Expansion of the cadre of trained scientists, coordination of research activities among donor organizations, and release of proven new technologies were less than projected. hes objectives were over-optimistic, and are carried forward into the new operation. The PCR is of satisfactory quality. The project may be audited upon completion of the follow-on project Attachment This docunen has a resd distributon and may be used by reclpiQnts only In the pefmnc o th offica dute Us cornwets may not oUthis be diclosed wht World Bank authruon  FOR OMCIAL USE ONLY PROJECr COIPLETION REPORT RWANDA AGRICULTURAL RSEARCrH PROJECT (Cr. 1546-RW) TABLE OF CONTENTS PREFACE . .................................................. i EVALUATION SUAMMARY ............................................... i PART k PROJECr REVIEW FROM THE BANK ERSPETE ........ 1 PtoiextIdendt ~~~~~............ .": .......,1 ProjeIdentity ....................... 1 Brojeut Objecti.es and escip .n............1........... Project Objeegn and Otmieption .............................. I Peroect esinp and Organization ..................... 2 Prject .s.Its..6.................... I?roject Sustainabsi ............ 6 Proectstaeinmabnit ................................... 6 Bank's Performance ...................... 7 Boffower's Performnance........................ 7 Consultnte' Performance ................................. 7 PART II: PROJECr REVIEW FROM THE BORROWER'S PERSPE ......... 8 PART m: STATISTICAL INFORMATION .................................. 16 Table 1: Related IDA Credits ........... ............................ 16 Table 2 Project Tunetable ............. ............................ 17 Table 3: Credit Dsbursements .......... ............................ 18 Table 4: Project p entation.................... 19 Table 5: Projec Costs and Fmancing .................................. 22 Table 6: Status ofCovenants ............. ........................... 23 Table 7: Use of Bank Resources ..................................... 24 A4NNEXCI ..... . 26 Map BRD 18308 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bak authod2ation.  i PROJECr COMPLETION REPORT RWANDA AGRICULTURAL RESEARCH PROJECr (Cr. 154-RW) PREFACE This is the Project CDompletion Report (PCR) for the Agricutural Resarcb Project, Phase L for which Credit No. 1546-RW in the amount of SDR 11.7 million (US$11.5 milion equivalent) was approved on Februay 12, 1985. The credit was closed on December 31, 1991, after a three year exensio; The last disbursement took plae on April 30, 1992, at which time the undisbured balance of SDR 2.3 mfllion (US$2.3 million equvlent) was cancelled. The PCR (Parts I and l) was finalized by the Agriculture Operations Division, South- Central and Indian Ooean Department of the Africa Region. The draft report was prepared by the Food and Agriculture Organization/World Bank Cooperative Program, folowing a field viSit in October 1992. It is based on project documentation such as the Staff Appraisal Report; the Development Credit Agreement, project correspondence, and supervison reports. A copy of the report was sent to the Borrower on April 14, 1993, with a request for comments. These comments are presented as Part I of this repor.  PROECT COMPLETION REPORT RWANDA AGRICULTURAL RESEARCIH PROJECr (Cr. 1546RW) EVALUATnON SUMMARY BACKGROUND 1. Plcy contaext Smce the 1970s, the Government of Rwand (GOR) has had three basic objectiv for the agricultur sector. food self-sufficiency, import sibstituton, and expo promotion and diversication. These objetves tansated into the following development prorties: (i) intensification of crop and livestock production; (i) soil consewation and improvement of soil fertDity, (imi) improved management and rational exploitation of agricultural :and; (iv) regional specialization in production as a function of agro-ecological potential; and (v) implementation of agricultural interventions within the framework of a sectoral or crop specfic approach. To achieve these objectves, the GOR has reLied heavily on donor financed rral development projects, which were successful in the field of road and infaucture developmet and soil conservation, but could not prevent a decline in productivity due to inreasing land pressure, and related losses in sofl fertility, and a lack of new technologies to address these isues 2. Faced with the need for productivity increases, the GOR initiated, witd support from the International Service for National Agricultural Research (ISNAR), a review of the agriculural research system in the early 1980s and took the first steps to strengthen the Agricultural Research Institute of Rwanda (ISAR). The resulting project was conceived as the first phase of a long- term development program for a problem solving oriented agricultural research systea. Ojecdvs and Componeml 3. The project aimed at: (i) preparing a research master plan as the basis for the long-term development of agrcultural research; (ii) streamlining ISAR's researh and financial management system though the introduction of program based research and research financing and (iii) establishing a long-term manpower development plan Major project components included: (i) the preparation of the research masterplan; (ii) the reorganization of ISAR; (iii) intucton and/or strengthening of multffdisciplinary research teams in crop and liveb * research and the establishment of three regional Farming Systems Research Groups; (.,l rehabilitation and extension of ISAR's infrastructure; (v) vehicles and research equipment; and (vi) operating eosts. The IDA credit of SDR 11.7 million was exptd to cover about 65% of total project costs and was to be disbursed over a period of three years (1985-1988). 4. The project objectives, components and time-frame proved to be too ambitious and the credit agrement had to be extended twice (1988 and 1989) until December 31, 1991 when the credit closed. The undisbursed balance in the amount of SDR 2.3 milion (20% of the credit) was cancelied on April 30, 1992. The main reasons for the long implementation delay were: (i)  lv the extremely difcut budgetay situadon, aggavated by an uncontroled increase in mainly subordinate staff () the vy slow ple tion of al component rlated to research prgamming and management, including the preparation of the resar mteplan and (ifi) pmcement problems The latter, however, were overcome as soon as ISAR staff had been familarized with IDA prourement procedures, as denstrated by the smooth entati of the infastructue and equipment components in Years 2 to 4 of project impl tation Actual project costs amounted to US$17.2 million (96 percent of estimates), with Goernment contributing about US$5.1 milion (78 percent of estimates) and IDA USI21 milion (105 percent of estimates). ResIII 5. The main project objective - to pave the way for the establishent of a log-term development plan for agricultural research - has been achieved. The-research masterplan has proven to tX a solid basis for the preparation of the second phase project at present being processed by IDA. The proposals for institutional s ning of ISAR - progam based planning, budgeting and evaluation of research activities, etablihment of four Research Departments, decentralization of research management - were changed in the course of project implementation and new proposals were approved by GOR as part of second phase project preparation. 6. Manpower development achieved mixed results. The number of research staff doubled over the project period but postgraduate training is still largely insufficient and only some 30% of the research staff have a postgraduate degree. Postgraduate, particularly ong-term, taining is mainly provided by bilateral sources and the kind of training depends more on the availability of grant scholarships than on a long-term plan to develop manpower based on ISARs needs. 7. The project was not designed to support individual research programs except for livestock research where the project proposed substantial investments in infrastructure and technical assistance. This program, however, never got off the ground and progress in livestock research during the project period was negligible. Other research programs benefitted from project investments (e.g. inErastructure, vehicles, equipment etc.) and from support to operating expenses. Successful research results, particularly in the development of disease resistant bean and root crop varieties, have been extended to farmes & During the project period the number and quality of publications in journals and conference proceedings have increased and annual reports of a h-gh standard have been published since 1989. S _ nabil 9. Sustainablity of agricultural research under the economic conditions of Rwanda is still the most crucial issue and long-term agricultunl research development can only be assured ifi (i) there is a cear commitment by the GOR to agricultural research; (ii) funding of research on industrial crops is at least partly provided by the respective industries; and (iii) the donor community agrees to a long-term support program for ISAR's research activities. Thee issu are addressed in the second phase project  v Leons lAmmed 10. The important lessons from this first phase Agricultural Research Project are: (i) the establishment of a national strategy for agricltura research and of related proposals for the reorganization of a national agricultral research system cannot be achieved in a short lived project but needs time to be developed and accepted by all decision makers; (ii) investment in strengthening research services is jeopardized if the necessary level of recurrent funding is not maintained to keep them operating effectively; (iii) under a long-term constrained budget situation like in Rwanda agricultural research should take full advantage of technologies generated by regional and international research organizatons and concentrate on testing and - if needed - adapting them to local conditions; (iv) the construction of new research infrastructure should not precede the approval of a long-term research development plan, neither should a training program be executed without the benefit of a human resource development plan; (v) large increases in research and support staff have not resulted in increases in research efficiency or productivity; and (vi) donor coordination is of vital importance for a sustained research development. A coordinating funding mechanism would be required through which all funding for research expenditure would be screened. This does not necessarily mean a pooling of all Government and donor funds but rather a pooling of research efforts and programs which leaves Governments the choice of requesting individual donor fnding for any set of activities within the parameters of the research masterplan. These lessons have been taken into account in the design of the follow-on project,  PROJECr COMPLEIION REPORT RWANDA AGRICULTURAL RESARCH PROJECI (Cr. 1546-RW) PART E PROJECr REVIEW FROM THE BANK'S PERSPEtIV 1. Prect Idenoe Project Name : Agricultural Research Project Credit No. : 1546-RW RVP Unit : AF3AG Country : Rwanda Sector Agriculture Sub-sector Agricultural Reseach v B }*MM 2.1 Polc cantexL Since the 1970s, the Govenmment of Rwanda (GOR) has had three basic objectves for the agricultural sector. food self-sufficiency, import substitution, and export promotion and diverification. These Government objectives tlated into the following development priorities: (i) intensification of crop and livestock pruduction; (ii) soil coefvation and improvement of soil fertlity, (iii) improved management and rational exploitation of agriultural land; (iv) regional specialization in production as a function of agroecological potential; and (v) implementation of agricultural interventions vithin the frmeork of a sectoral or crop specific approach. To achieve these objectives, the GOR has relid heavily on donor financed rural development projects, which were successful in the field of road and infrastructure development and soil conservation, but could not prevent a decline in productivity due to increasing land pressure, and related losses in soil ferdlity, and a lack of new technologies to addres these issues. 2.2 Faced with the need for productivity increases, the GOR initiated with support from the International Service for National Agricultural Research (ISNAR), a revew of the agricultural reserch system in the early 1980s and took t'1e first steps to strengthen the Agricltural Reseach Institute of Rwanda (ISAR). The resulting project was conceived as a first phase of a long-term development of a problem soling oriented agricultural reseach system. 3. Ptoj Objectv and Desupton 3.1 The project aimed at (i) preparing a research masterplan as the basis for the long-term development of agricultural research; (ii) streamlining ISARs research and financial managment system through the introduction of program based research financing, and (iii) establishing a bng- term manpower development plan. Major project components included: (i) establishment of a research masterplan, (ii) strengthening of research management programming, evaluation and  2 review through the reorganization of ISAR a.d establishing appropriate procedures for the formuladon of research programs; (iWi) introduction and/or strengthening of multi-disciplinary research teams in crop and livestock research (five mu uz-disciplinaty crop research teams dealing with 13 crop research programs, a central animal prodixtion systems research unit and livestock research groups at three stations) and the strengthening and coordination of three tegional Farming Systems Research (FSR) programs, (iv) establishment of a documentation system; (v) rehabilitation and extension of ISAR's infrastructure; (vi) vehicles, and research and eperimental equipment; and (vii) operating costs. 'Te project was to be implemented over three years (1985- 1988). 4. Ptiect Despl and Q%Wo 4.1 Poj pc m Prnject identfication was based on an ISNAR anays of Rwanda's research system (1982) and a national seminar on agricultural research held in early 1983. ISAR, with ISNAR and IDA support, prepared working documents on crop, lvestock and farming systems research, infstucture rehabilitation and construction, equipment requirements and research documentation, which were consolidated in a preparation document by an IDA/ASNAR mssion (late 1983). The project was appraised in early 1984 with support from ISNAR and the International I.vestock Center for Africa (ILCA), and negotiations took place late 1984. 4.2 Desin Considering the limited time-frame and the weak insttutonal capability of ISAR, the project objectives and components proved to be over-ambitious and not in line with the Lmited availab;e human resources (19 mainly inexperienced research staff). 4.3 Ihe infrastructure component of the project - rehabilitation of existing and construcion of new buildings - was well prepared with detailed architectural designs available at appraisaL The inclusion of substantial new construction under the project (some 3,850 m2 of housig and 2,730 m2 of servioe buildings) was, however, premature considering that research priorities and regional distribution of research actmities were to be defined in the masterplan to be prepared under the project (eg. six staff houses constructed on the Karama research station brng the total number of staff houses to 38, which is more than needed to accommodate research and technical staff). 1/ 4.4 Project implementation was the responsibility of the Director of ISAR. ISAR was to be reorganized (establishment of an internd research committee and a reseach programming and evaluation unit; delegation of the day to day reseach management to the four heads of the research departments (crops, livestock, farming systems, forestry); and of the accounting and financial management to the head of the Fmiancial and Administrative Service) as part of the projecL 5. Projec lmentadon 5.1 Genera The credit agreement was signed in March 1985, and became effective on November 11, 1985, a delay of five months. The project was originally expected to be implemented over three years with a planned closing date of December 31, 1988, but the closing j/ Govenment and DA do not agree on this issue (see para 4.3, Part II).  3 date had to be extended twice, fiSt to December 31, 1989 and subsequently to Dcmba 31, 1991. The main rteaons for these extenions wer (,) the extrwly difficult budgetaiy sitdut aggravated by an uncontol increase in mainly subordiate sta4 (i) the vey sow implementation of all components telated to research prog-ammn and includig the prparon of the Research Mastetplan; ad (iii) pcement problems The latter, hower, were oveacome as soon as ISAR staff had become famila ith IDA procuement procedures, as demons by the smooth implementation of the infrstructe and equipment components itn year 2 to 4 of the prject. 5.2 _ a dmd pQd im Ih mjot pa of the implemen_aton period of the projct coincided with a singulaly diicult pedod in Rwanda's economy, during which the financi and budgetary situation increasing deteoraed as a result of a large drop in coffee prices, the countr's main export, and an inadequate policy response. Ths prmpted the governm t to impose financial restrictions and undertake a range of adjustment policies including drastic reductions in public investment and opaating budgets These restidons were applied without setting priorities and seriously affected the avability of countepart funding for this project 53 Goverment's bud*etaiy difficulties caused excessive delays in a number of ways Frst, funds were not available as had been projected in the Staff Appraisl Report. In fact, Gover'ments contrbution to the project decesed substantially from FRw 86 milhon in 1986 to FRw 50 mion in 1991 (Table E, Annex ). TMi caused project activities not to be undertAer at the foreseen time and scale and was particularly important for the rrent cost funding as the IDA contribution to this category was conditional on the amount of Govenmment funding. In mid-1991, a revio of the Development Credit Agreement albowed the fincing of 100 pecnt of operating costs by the IDA credit. Tbe budgetary situation was eaerbated by an uncontroled increase in stafE, particularly in seasonal and daily lablr from some 1,000 in 1984 to more than 3,200 in 1989 and it was not until 1990 that ISAR management made an effort to reduce this categor, of staff by about one quarter. 7I Between 1982 and 1991, expenditures rose by about 290 percent, while income increased by 83 percent only. As a result, ISAR accumulated a deficit of about FRw 1,690 mIlion over the same period (Tables A and B, Amnex I) J 5.4 Project costs were orignally estimated at US$1&0 milion, of which US$6.5 millon was to be provided by Govenment and US$11.5 million by IDA. Actual project costs amounted to US$17.2 million (96 percent of estimates), of which IDA provided US$12.1 million (105 percent of estimates) and Government US$5.1 milion (78 percent of estimates). In SDR terms, IDA's contnrbution was only 80 percent of estimates. 5.5 Staff ATle number of national research staff increased sgficntly from 19 in 1984 to 41 in 1992 with five more staff to return from PhD and MSc training (Table 4, Part III). This is in reasonable r ,..eement with the requirements of 70 researchers by 1995, predickd in the research masterplan, Long-term training has been provided exclusively by bilateral sources, mainly the United States Agency for International Development (USAID), the Geman Agecy 2y The Government believes that the increase in staff was warranted by the increase in research activities (see parm 5.1 (d), Part II). w/ Government and IDA do not agree on this number (see par. 5.4, Part II).  4 for Technicad Cooperation (G1Z) and Swiss cooperation which also povided short-term taing. Only some 50% of project funds for training (equivalent to 71 short-term feilowshlps) were used, E rimari because the Government preferred grant financing for training activities. _ 5.6 Throughout project implementation, turnover of research staff was relatvely higL Particulary experienced research staff have been attracted to administrative posts with better career prospects in the central administration or in the pnvate sector. Ts problem is expected to be at least partially resolved with the introduction of a career development plan for research staff (Statut de chercheur) developed under the project. 5.7 Tecnicl sta (Table 4, Part m). Out of 19 man-years of log-term technical assistance, only 11 were used in the areas of research management, a tration and laboratory senices ISAR was not able to recruit a local head of the administration and finance section until early 1993, ne tating continued technical assistance. Ewith the reseach prog&amming and evaluation advisers were mixed at best; research planning and evaluation is still seriousl flawed and a concise research programming and evaluation system stili has to be developed. A technical assistant was hired in 1990 to assist in improving laboratory service. His performance has been satisfatory. Other long-term technical assistance was not used or was funded by bilateral projects 5.8 Twenty-one out of fifty-six man-months of short-term technical assistance have been used, mainl for the external evaluation of some 20 research programs. Most of this assistance was of goad qualty and has contnbuted sgnifiantly to the establishment of the research masterplam. Except for some programs (eg. soil conversation), where a more permanent link with a foreign institute could be established, the evaluations have had, however, only a limited impact on on- going research activities. 5.9 Research materpsa Published in 1990, the plan is comprehensive in terms of research priorties and staffing needs. The plan emphasizes the setting of research priorities and most of the conclusions remain applicable, but changes in the ec)noffic and technical envronment demand a peiodical updating of the plao. The research masterplan has not yet been used by the Govemment for its main purposes, to guide individual research proposals on priority topics and to coordinate the participation of donors in the field of agricultural research. 5.10 Research programs. The project was not intended to support specific research programs but to build-up a more client oriented research system through a reorganization of ISAR in four research departments (crops, livestock, FSR and forestry), the introduction of program based research financing and the introduction of a donor coordinating mechanism. The four research departments were established in the early stages of the project but have recently been replaced by research programs, each headed by a program leader under the supervision of a Scientific Director. This approach is in line with the program based research budgeting and cost accounting system presently being introduced at ISAR. In order to avoid the establishment of too many small units, individual programs are coordinated in broader program groups. 5.11 The project had only limited success in establishing multi-disciplinary commodity research teams, covering aU aspects of commodity development such as genetic improvement, plant 41 Govemment claims that at the most only 10 fellowships were awarded (see para 5.6, Part II).  5 protection, fertilization, cultural practces, integration in crop and variety mixes, acceptably for fanners etc. Most of on-going research is still concentrated on breeding while other aspects are widely neglectd and on-fiarm experimental work to test innovations under farm conditions is stm not a common feature. Examples like the CIAT supported legume program, where variety adaption is at least as important as varety development, should be followed in the future. On- farm participatory research is of vital importance for the development of client oriented research. 5.12 livestock researcb made little progress under the project -Te use of project investments and operatig costs in livestock research - mainly pasture inpovement and water supply in Karma making-up about half of all ifrastructure investments (Tabk 4, Part im) - were conditioned by ISAR enterng into a formal agreement with ILCA to provide guidance and support for al actvities of the livestock department and to post two lvestock research scientists at ISAR. Ihe Institute, however, never reached such an agreement; reportedly because of the incapa of ILCA to provide such semces. The civ works program was executed nevertheless. 5.13 Central services The library was reornized in 1987 and is flly operationaL -he central laboratories were equipped in 1991 but the problem of the reliability of electicity and water supply are not yet solved and almost no internal rehabilitation of the laboratories at Rubona had been indluded in the civil works program of the project. Laboratory operations have, howeer, improved with the arrival of a technical assistant in 1990. 5.14 Cooatio with the Univt. ISAR and the University of Rwanda established a formal agreement of cooperation in 1989 but no project related research activities were undertaken because of limited human resources at the Univesity and cumbersome procedures g ioverig the cooperation between two institutions falling under different Ministries 5.15 Donor The donor coordination mechanism set up under the project - an annual meeting of all bilateral and multilateral agencies interested in financing agricultural reseach to mview research concepts, staffig and financing - failed to produce coherence in research planning execution and financing A strikig example for the lack of donor cordinato is the farming system resarh progam where four different donors e td their projects independently using different research methodologies and where activities coflapsed as soon as the foreign financing ended and the technical assistants departed. 5.16 in ISAR maintained numerous linkages with regional and international research bodies and the strengthening of regional collaboration was (and still is) high on its agenda. A small national agricultural research system, such as the Rwandese, depends largey on technology generated elsewhere. Its function is primary to screen such technology and to adapt it to prevailing local conditions. / his approach, already actively developed at ISAR for some commodities (beams wth the Intemational Center for Tropical Agriculture (CLAI), potatoes wth the Regional Program for the Improvement of Potatoes in Central Afica (PRAPAC) and the Internatioal Center for Potatoes (CIP), agro-forestzy with the International Center for Research in Agro-forestry (ICRAF)) needs to be etended to other research programs. 5.17 linkages with extension semices, development projects and NGOs were hampered by two major problems: (i) research themes were mainly defined by the researchers and did not alwas Y/ Goverment and IDA do not entirely agree on this issue (see para 4.5, Part II).  6 reflect farmers needs; and (ii) on-farm research, when practiced, was of variable quality and intensity. !/ A closer collaboration of farmers, extension staff and researchers is needed. 6. Project Resut 6.1 Considering the main objective of the project - to pave the way for the long-term development of agricultural research in Rwanda - the project has beea reasonably successful although it took about twice the initially planned time to achieve it TIhe research masterplan forms a solid basis for this goal and has been extensively used for the preparation of the second phase project, at present being processed by IDA. The amended proposals for institutional strengthening of ISAR-program based planning, budgeting and evaluation of research activities, establshment of a Scientific Directorate, and decentralization of research management through the establishment and/or strengthening of regional research centers will be implemented under this second phase. 6.2 The Progrmming and Evaluation Unit was not established and research prograng and evaluation of research progress and results remained unsatisfactory during the whole project period. ISAR's management remained excessively centralized with the result that program and station managers had no knowledge of research costs and the actual budget situation. Consequently, there was no relationship between the program based budget proposal and the actual budget execution. 6.3 Manpower development achieved mixed results. The number of research staff doubled over the project period but postgraduate training is still largely insufficient and only some 30% of research staff has a postgraduate degree. 64 As mentioned above, the project was not designed to support individual research programs except for livestock where the project proposed substantial investments in infrastructure and technical assistance to be provided by ILCA. This program, however, never got off the ground and progress in fivestock research was negligible. Other research programs substantially benefitted from project investments (eg. infrastructure, vehicles, equipment etc.) and from support to operating expenses Notable research results, particularly in the development of disease resistant bean and root crop varieties, are reported to be adopted by farmers. 6.5 The heavy investment in ISARs infrastructure has left it with excellent facilities which will facilitate the future etension of research activities. 6.6 During the project period the number and quality of publications in joumals and conference proceedings have increased and annual reports of a high standard have been published since 1989.