Dommut of The World Bank FOR OFFICIAL USE ONLY Report No. 6576 PROJEC'r COMPLETION REPORT BANGLADESH AGRICUIL'T'U!RAL KFSEARCH I PROJECT (CRFI)IT 928-BD) December 31, 1986 5outb Asia Project- Department G,eneral Agrictilture T TDivision This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank d,ithoriza.!onf WEIGHTS AND MEASURES 1 acre (ac) 0.405 hectare (ha) 1 square mile (sq mi) 640 ac (259 ha) 1 maund (md) = 82.27 lbs (37.3 kg) 1 metric ton (to-,) = 26.8 md AbBREVIATIONS BARC - Bangladesh Agricultural Research Council BARI - Bangladesh Agricultural Research Institute BAU - Bangladesh Agricultural University BJRI - Bangladesh Jute Research Institute BRRI - Bangladesh Rice Research Institute BWDB - Bangladesh Water Development Board DCA - Developmert Credit Agreement EC/NEC - Executive Committee of the National Economi.c Council FRI - Forestry Research Institute COB - Goverr,ment of Bangladesh 1YV - High Yielding Varieties IRRI - International Rice Research Institute PP - Project Proforma PWD - Public Wtorks Department RMB - Resident Mlission, Bangladesh SRI - Sugarcane Research Institute TRI - Tea Research Institute TWG - Techr.ical Working Group UISAID - US Agency for International Development FOR OFFICIAL USE ONLY THE WORLD BANK Washington. DC 20433 U S A Offe of Die,tow-GmneraI Opertain Ivauultan December 31, 1986 MEMORANDUM TO TIlE EXECUTIVE' DIRI'CIORS ADl) THE PEE:IDEN'I SUIBJECT: Project Completion Report: Bangladesh - A2ricultural Research T Prj ect (Credit 392-BID) Attached, for information, is a cPov oa a reT)ort mtitlod "Project Completion Report: Bangladesh - Agric2iltiral Research T Project (Credit 392-BD)" prepared Iv the South Asia Repional nffice. Under the modified svstem for oroject performance auditing, further evdluation of this prolec:t bv the Operations Fvaluatior, Department has not been -tace. Attachment This document has a restricted distribution and may be used by reuipients only in the performance :( their olficia! duties Its contents may not otherwise be disclosed without Worid Bank authorization j FOR OM(AAL USE ONLY BANGLADESH AGRICULTURAL RESEARCH I PROJECT (CREDIT 828-BD) PROJECT COMPLETION REPORT Table of Contents Page No. PREFACE i BASIC DATA SHEET SUMMlAY v I. INTRODUCTION .,, ............... ,,,*....,,,,,............... 1 II. BACKGROUNCDTIO . . . . . .. . . . . . . . .. . . . . . .1 A. Project Origin ... ............1.... B. Appraisal and Effectiveness .................... .... 2 C. Project Description and Detailed Features ........... 2 - Project Description .. ...... .................. ... 2 - Detailed Features ............................ off. 3 III. IMPLEMENTATION 4. .................... . 4 A. Start-up and Timing of Disbursements ............... 4 B. Revision ..... ................. 5 C. Reporting .. .............. 6 D. Procurement .. ... .. ... .. ..... 7 E. Cost ................................................ 7 F. Disbursement by Categories .................. .......... 7 G. Implementation of Specific Components ...* ........... 8 - BARI Regional Station Development ............ 8 - BARC Headquarters Complex ................ ....... 9 - Research Grant Subprojects ......... ....... 9 - Contract Research Results .............. *I ...... 11 - Development of National Agricultural Research Plan 13 - Training .... ........... , , , . . 13 - Technical Assistance ....... *................-.*, 14 - BARC Staffing ................ .................. 14 - Provision of Operatiig Funds ..................... 14 IV. PROJECT IMPACT AND INSTITUTIONAL DEVELOPMENT ............ 15 V. IDA PERFORMU47CE . ....... . ...... . . .. . . . . .... . 16 VI. CONCLUSIONS AND LESSONS LEARNED ...................... 17 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. TABLES 1 Project Cost 2 Local and Foreign Exchange Cost (Current) 3 Status of Disbursement 4 Appraisal Estimates and Actual Disbursement Schedule 5a Status of Expenditure Under Research Subprojects Approved by BARC (Summary) 5b Research Subprojects Approved by BARC (Soil and Water) 5c Research Subprojects Approved by BARC (Agriculture Economics) 5d Research Subprojects Approved by BARC (Coordinated Cropping System) 5e Research Subprojects Approved by BARC (Agricultural Engineering) MAP IBRD 13447 BANGLADESH AGRICULTURAL RESEARCH I PROJECT Credit 828-BD Prf ace This is a project completion report (PCR) for the Agricultural Research I Project in Bangladesh for which a credit of US$6.0 million (No. 828-BD) was extended to :he Government of Bangladesh in June 1978. Credit closing occurred on time (12/31/84) in spite of a two month delay in credit effectiveness. The final disbursement was made in September 1985 and the undisbursed credit balance of US$350,000 (6% of the credit amzunt) was cancelled. The PCR has been prepared by the General Agriculture I Division of the South Asia Region. Its preparation is based on the Staff Appraisal Report (No. 1974a-BD) dated May 16, 1978 and the Development Credit and Project Agreements dated June 16, 1978. Correspondence with the borrower and internal documents and memoranda on the project contained in the Bank's files, have also been consulted. Following a review of the PCR by OED staff, it was decided not to audit the project and the PCR was sent to the Borrower for comment on October 8, 1986. The Bangladeshl AgriculLural Researoh Council acknowledged receipt but had no additional cominent to make. -ii- BANGLADESH AGRICULTURAL RESEARCH I PROJECT Credit 828-BD Basic Data Sheet KEY PROJECT DATA Appraisal Actual or Column 2 as x EstimEte Current Estimates of Column 1 (1) Total Project (US$ million) 7.4 6.2 84 Credit Amounts (US$ million) 6.0 5.65 94 Date Physical Components Completed 6/30/84 L,30/84 Institutional Peformance Good CUMULATIVE DISBURSEMENTS FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY86 Appraisal Estimate (US$ million) - 1.5 3.7 5.5 5.8 6.0 - - Actual (USS million) - 0.62 2.13 4.14 5.01 5.51 5.64 5.65 Actual as % of Appraisal - 41 56 75 86 92 94 94 Date of Final Disbursement: September 18, 1985 Original PROJECT DAIES Plan Actual Negotiations 04/27/78 Board Approval 06/15/78 Credit Signing - 06/16/78 Credit Effectiveness 09/15/78 11/15/78 Closing Date 12/31/84 12/31/84 -iii- OTHER DATA Borrower Government of Bangladesh Name of Currency (Abbreviation) Taka (TO) Fiscal Year July 1 - June 30 Exchanpe Rate j/ Appraisal Year (1977) 15.25 Interv2ning Years' Average (1979/84) 19.86 Completion Year Average (1984/85) 25.50 Foilov-up Proiect Name Agricultural. Research lI Project Credi.t No. 14554-FD CrediL Amount (SgR MiX I .on) 23.4 Date of Board Apprcoval Karcb 29, 1984 Date of Credi t Si.gqing May 18, 1984 Date of Effectiveapss April 23, 1985 AJ The deili e in r&ki; vs;.ue over th? Ji.sbursemner!t period regpistered 40%. BANGLADESH AGRICULTURAL RESEARCH I PROJECT (CREDIT 828-BD) Hission Data Date No. of Manweeks Performance Rating Type of Mission Month/Year Persons in Field Specialization Status Trend Problem 11 2/ 31 5/ Identification Preparation 4/ Appraisal 11/12/77 4 iC, A. Eng. E. - Post Appraisal 1 6/78 2 2 A.E. 1 2 P Post Appraisal I1 10/78 2 _ 2 A.E. 1 1 M Sub-total 20 Supervision 1 3/79 2 1+RMB 6! A.E. 1 2 H Supervision II 11/79 2 l+RMB A.E. 1 2 H c Supervision III 5/80 2 1+RMB A.E. 2 3 M.F. Supervision IV 12/80 1 RMB E. 1 2 M Supervision V 6/81 2 2+RMB A.E. 2 1 M.F. Supervision VI 2/82 2 RMB E.E. 2 1 M.F. Supervision VII 7/82 2 2+RMB A.A. 2 1 M.F. Supervision VIII 12/82 1 1 A 2 2 M.F. Supervisiir IX 5183 1 RMB A 2 2 M.F. Supervision X 11!83 1 1 A 2 2 M.F. Supervision Xi 5/84 1 RHIB A 2 1 m Supervision XII 11/84 I 1 A 2 1 H Sub-total 10 Total 30 jJ Specialization A - Agriculturist; Eng. - Engineer; E. - Economist; Ar. - Architect. _2J Performance rating, status: I problem free/minor problems; 2 = moderate problems; 3 major problems. 3! P-rformance rating, trend: 1 improving; 2 = stationary; 3 = deteriorating. 4/ Carried out by Resident Mission (RMB) and GOB. 5/ Type of problem; M management; F - financial; P = political. 6/ 'Supervision whollv and partly undertaken by Resident Mission (RB) not included in man-weeks. BANGLADESH AGRICULTURAL RESEARCH I PROJECT (CREDIT 828-BD) PROJECT COMPLETIOW REPORT EVALUATION SUMMARY 1. The project is the result of the reconv.tendations contained in the Bank's Land and Water Resource Sector Study (IBRD Report PS-13) completed in December 1972 which recommended inter alia an US$80 million ten-year research program for strengthening of agricultural res. rch in Bangladesh. The project was prepared by a Government of Bangladesh (GOB) team consisting of scientists from the Bangladesh Agricultural Research Council (BARC) and the Bangladesh Agricultikral Research Institute (BARI), and was appraised by IDA in November/December ;977. The Credit agreement was signed on June 16, 1978, and became effective on November 15, 1978. The Credit closed as originally plannec, on December 31, 1984. The final disbursement was US$5.65 M and an amount of US$350,000 was cancelled. Objectives 2. The project's objective was to strengthen agricultural research, specifically the research facilities, management, and scientific manpower of BARI; and the planning, coordination and monitcring functions of BARC. The project included funds for: (a) land development, physical facilities, machinery and equipment at four BARI regional research stations and an office complex at BARC headquarters in Dhaka; (b) research subprojects at various research institutes and universities, to be administered by BARC; (c) train- ing of scientific and extension staff; (d) short-term technical assistance; and (e) incremental operating costs for BARC. 3. BARC was responsible to guide and direct the preparation of a national research plan and ensure coordination between research institutes and donors. Physical constructioLn and procurement were the responsibility of BARC and BARI. Selection, supervision and evaluation of research subprojects was the responsibility of BARC's technical comm:ctee. Formal approval of research subprojects for financing was to be the responsibility of BARC's Executive Committee, which was interdisciplinary in composition. Implementation Experience 4. Though project activities started on time, difficulties were experienced in procurement of equipment and provision of local funding. -vi- There were considerable cost overruns under civil works and equipment com- ponents. A vevision of categories was agreed to in late 1981, resulting in increased spending under civil works and equipment, made up by using funds under "unallocated" category and some reduction under training, contract research, and technical assistance. The reduction under the latter categories was made up by additional funding from the United States Agency for International Development (USAID). Despite the above, the credit closed on schedule with 94% of funds disbursed. Implementation was completed by December 1984, six years after start up. j. The appraisal report was too specific in itemizing some of the equip- ment. Once this was included in the project proforma, flexibility was lost and purchases were made without reference to the updated project require- ments. Procurement of Rquipment was also delayed due to a weak procurem!nt cell at BARC and complex (,OB clearance procedures. 6. Project reporting was poor. Under the project, BARC was responsible to coordinate the work of different institutions. As BARC did not have executive authority over the institutions, coordination was weak. The SAR did not suggest regular meetings chaired by the Secretary, Ministry of Agriculture to review project progress with the Executive Chairman BARC and the Director of BART. As a result, reporting requirements were not monitcered. 7. Separate accounts were noL maintained Eor individual research sub- proiects. There was considerable delay in providing audited accounts due to weaknesses in the geri,ral auditing procedures. Ag of September 18, 1985 (date of fin.al disbu-sement), an amountt of US$5.65 mil;i.ni had heen 3iqburFed ind LtS$350,OO0O cancelled, against the total credit ot US$6.0 miliion, Ihe mairn shortfal.l. was due to the suspension of research station development at Ramgarh and 51.ow progress of the contract research propram. Results 3. The mlain feature of the project was to acceleraLe thie institu- tion-huLLding process for agricultural researcht in Bangl.adesh. As a result of the project, BARC's role as an apex institution for planning, coordinAt- ing, and mocaicortng research has gained increased recognitiorn. The strengthening of research facilities at regional research stations located in three out of the four main agroecological zones has helped in identification and addressing uf farmers problems under different conditions. Action has been initiated by BARC to prepare a National AgricuLtural Researc'i Plan aimed at developing research policies, strategy and prioritCes. -vii- Findings and Lessons 10. The main findings are the following: (a) there has been considerable improvement in research planning and institution building; (b) while the contract research program has been we l implemented, its objectives are not clear to al' concerned; and (c) an appropriate coordination mechanism was not set up to ensure better monitoring of project activities and resolve implementation problems. 11. Main lessons learnt are: (a) provision of equipment at regional research stations should be based on the research programs to be implemented; (b) advance planning should be done to moke lull utilization of training funds; (c) technical assistance funded by different agencies should be properly coordinated to ensure specific responsibilities; (d) contract research funding should be made to supplement normal research programs; (e) composition of supervision missions should be such that expert support is available to help solve specific implementation problems; and (f) a coordina- tion mechenism is essential to bring together different organizations respon- sible for project implementation, review progress, and resolve problems. BANGLADESH AGRICULTURAL RESEARCH I PROJECT PROJECT COMPLETION REPORT I. INTRODUCTION 1.01 Before the implementation of the IDP-assisted Agricultural Research I Pr-ject, agricultural research in Bangladesh was widely dispersed over 12 autonomous research institutes which were under nine ministries. There was no overall coordinating body to provide broad guidelines for research activities to promote relevant research appropriate for Bangladesh conditions or to prevent duplication of research effort. The Agricultural Research I Project was designed to both strengthen adaptive research capacity as well as institutionalize coordination among the dispersed research institutes. 1.02 The Agricultural Research I Project was not the first research project in Bangladesh. In 1976, USAID had initiated their first agricultural research project which concentrated on supporting the Bangladesh Agricultural Research Institute (BARI) at Joydebpur, developing the Ishurdi Centre for work on wheat, and providing funds for technical assistance, training and contract research. Also, in early 1976, the Bank initiated a review which eventually led to the Extension and Research I Project (Credit 729-BD); this project became effective in January 1978. A small research component was included which covered development of five research sub-stations, the financ- ing of contract research through a research grant scheme of about US$200,000 under the dangladesh Agricultural Research Council (BARC) and demonstration of simple farm implements. 1.03 While these above developments were steps in the right direction, the Agricultural Research I Project was the first attempt to establish a coherent agricultural research system in Bangladesh. II. BACKGROUND A. Project Origin 2.01 The origin of the project lies in the World Bank's Land and Water Resuurce Sector Study (IBRD Report PS-13) completed in December 1972 which recommended strengthening of research institutes, focusing of the investiga- tions on specific agricultural problems, and establishing efficient linkages between the various institutes concerned. The Agricultural Research I Project was prepared by a GOB team consisting of BARC and BARI scientists, assisted by staff of the Bank's Dhaka office. -2- B. Apraisal and Effectiveness 2.02 The project was appraised by IDA in November/December 1977 and a report issued on May 16, 1978 - Bangladesh Agricultural Research Project, Report No. 1974a-BD. 2.03 The appraised project's objectives concentrated on strengthening BARC, strengthening the research system in general to enable it to better undertake applied and adaptive research relevant to local conditions, and to assist GOB in the preparation of a National Agricultural Research Plan (NARP). The project was approved on June 15, 1978; the Credit agreement was signed on June 16, 1978 set to become effective on September 15, 1978. However, effectiveness was delayed by two months due to the lateness of issuing the project proforma (pn) by COB. By November 15, 1978, the out- standing conditions were met and the project became effective. C. Project Description and Detailed Features Project Description 2.04 The project was designed to provide funds for: (a) essential buildings and land development a:, BARI regional research stations located at Jamalpur, Jessore, Hathazari, and the highland crop-cum-soil-conservation station at Ramgarh; (b) essential cropping and harvesting machinery, and workshop and laboratory equipment for the above BARI research stations; (c) the construction of a modest 16,000 sq ft first stage office complex for BARC at an existing site in Dhaka; (d) research subprojects to be funded by BARC, for execution by local research institutions and universities; (e) special studies to assist BARC in developing a long-term agricultural research plan; (f) overseas and local training for scientific staff and local training of extension staff; (g) about 36 man-months of short-term technical assistance in: Ci) land development at BARI research stations; (ii) sttdies to define research programs and research plans; (iii) periodic monitoring of BARCs research grant system; and -3- (h) incremental operating costs of BARC and the four BARI stations to make effective use of facilities, equipment and scientific staff provided under the project. Detailed Features 2.05 BARI regional station development. Four research stations located in four distinct agro-ecological zones were selected for development. The stations were to be provided with essential office and residential buildings; land development, including levelling and shaping; irrigation facilities; services including provision of drinking water and electricity, and essential storage and workshop space. Land development involved land levelling, land shaping and soil conservation work at Ramgarh, and improved drainage at all stations. Irrigation facilities were to be designed to cope with peak water requirements of the crops to be researched. At each station a brick road was to connect the main road with the office, residential, and storage blocks. The farms were to be fenced and provided with water storage tanks, on-farm piped water, and electricity. A workshop/machine shed was to be built. The project was also to provide machinery and equipment. For land development work at each station, a scraper, a leveller and a land plane were provided to be drawn by two tractors (45 and 65 hp) supplied for cropping purposes. Also, associated accessories and implements plus basic workbhop facilities and tools were to be provided. The provision of basic laboratory equipment and vehicles was also part of the project. 2.06 BARC headquarters complex. To enable all five divisions of BARC to be housed together and to enable BARC to organize periodic scientific semi- nars, a functional office of 16,000 sq ft was to be constructed. Design and layout were to allow for subsequent expansion. The design was to be prepared by a local consulting firm with qualification and experience satisfactory to IDA. 2.07 BARC financed research subprojects. All research institutions in Bangladesh were eligible to receive funds for research subprojects. BARC was to develop coordinated efforts in cropping systems research, soil and water management, appropriate technology in agriculture, agricultural economics, and crop protection. Other priority research area were to be agreed between BARC and IDA during implementation. 2.08 Manpower Development. The project was to fund 25 Ph.D. programs, 20 one-year M.Sc. degree programs, about 30 short-term career development cour- ses, study tours for research managers, local training at research stations for extension workers and subject mattor specialists and local seminars. About 80Z of Ph.D. training was to be overseas but BARC was to attempt to arrange with the overseas university for chesis work to be done in Bangladesh. BARC was to encourage local Ph.D programs at the Bangladesh Agricultural University (BAU). Also, BARC was to promote 20 M.Sc. degrees at -4- BAU and encourage active interaction between BAU and other local research institutions. BARI was to organize about 1,200 man-weeks of training spe- cially for extension staff so as to promote the links between research and extension. 2.09 Research plan. BARC was to develop a long-term pLan for agricultural research covering all commodities and problem area. Short-term technical asaistance hired by BARC was to be used for this purpose. 2.10 Technical assistance. As most technical assistance was already being provided to the research system by USAID, about 36 man-months of supplemen- tary technical assistance was to be commissioned (15 man-months for BARI and 21 man-mcnths for BARC). It was agreed that it would be a condition of disbursement against research subprojects that about 6 man-months of short-term expert services would be used annually to monitor the effective- ness of BARC's research grant system. USAID agreed to provide additional techni^ial assistance for contract research monitoring. 2.11 Incremental operating costs. Both BARC and the four BARI project stations were to be fully staffed by December 1979 according to a staffing plan satisfactory to IDA. During project c?velopment, incremental operating costs of BARC and four BARI stations were Lo be financed under the project. 2.12 Project cost and financing. The project was estimated to cost US$7.4 M inclusive of US$0.85 M for taxes and duties and US$1.6 M for physi- cal and price contingencies. Civil works (US$2.6 M base cost) and machinery and equipment (US$0.48 M base cost) together made up about half of all costs. Overseas training, local training, contract research futnds, BARC incremental operating costs and technical assistance made up the remainder. The IDA credit was for US$6 million which was to finance about 90% of project costs, net of duties and taxes. The credit was to cover all foreign expenditure (US$2.8 million) and about 70% of local costs of Tk 70.4 million (US$4.5 million). Detailed appraisal. cost estimate and actual expenditure for the project are in Table 1. III. IMPLEMENTATION A. Start-up_and Tim_ingof Disbursements 3.01 Project implementation started promptly with agreement over architec- tural and land developmen. consultancies, but considerable difficulty was experienced over equipmert procurement, where there was very little technical assistance to an inexperienced procurement cell. Implementation was com- pleted by December 1984, just over six years after start up, and final :lis- bursement on September 18, 1985. -5- 3.02 Implementation proved to be quite difficult in the early years, owing to considerable cost overruns in equipment procurement, in the main building components, and latterly with difficulties experienced over the provision of local funding. 3.03 Moreover, all major disbursements, apart from training and contract research, were scheduled in the SAR to be completed by the end of Year 3, June 30, 1982. As the PPs were based on the SAR, there was no authority for implementing agencies to incur further investment expenditure although sta- tion development had not been fully completed. A short-term solution was found by the utilization of PL 480 Title II funds for certain specified areas of day-to-day expenditure. 3.04 Notwithstanding the above and with a revision of categories in 1981 (see para 3.05), the credit closed as scheduled with 94% of funds disbursed, following more or less the original implementation schedule. All major legal covenants were complied with, although audit reports were submitted with some delays. The performance of the project compares favorably with other agricultural development projects in Bangladesh. B. Revision 3.05 The BARI and BARC PPs conformed entirely with the SAR but, due to a lack of coordination between these two institutions and some confusion over the apportioning of funds between categories concerning land development and research stubprojects, cost of buildings and equipment diverged from the SAR and led to early indications of overspending. A revision of the categories was agreed in late 1981. This increased spending on buildings and machinery, and reduced spending on local training, contract research and technical assistance. Additional funding for these latter activities was provided to BARC from USAID which by that time was strongly supporting BARC affairs. 3.06 Table 3 shows both the original and the revised DCA allocation. The following categories were reallocated additional funds: US$ Equipment and machinery 200,000 Construction 93,000 Overseas training 200,000 BARC operating costs 200,000 These increases in allocation amounted to US$693,000 and came largely from the 'unallocated' category in the original DCA. The 'unallocated' funds were reduced from US$600,000 to US$70,000 releasing US$530,000. The remaining US$163,000 were made available by reductions under local training, technical assistance, and contract research. -6- C. Reporting 3.07 According to the SAR, quarterly reports, which were to be based on the monitoring of physical execution of project works, were to be provided by the office of the D)irector of BARI and the Executive Vice Chairmtn of BARC. They were to include action programs for the following quarter and were to compare actu&t achievement with targets. Regular reviews of BARC's research grant system were also to be carried out. 3.08 Reporting requirements of the project were not met. Apart from regular reports on disbursement, there was no regular progress reporting during the implementation period. A review of research grant subprojects was carried out by BARC with the assistance of USAID-funded TA and procedures were found to be satisfactory. A seminar on the results of specific sub- projects was held in late 1984 and the proceedings published in March 1985. Ad hoc reports on particular experiments have been published on a variety of subjects such as an agro-eco.omic study of 'country bean' (Dolichos lablab) production, and on cropping systems research but these have not been pub- lished in a regular manner with due cognizance given to the research grant system. The only other regular reporting during implementation was that done by the twice yearly Bank supervision missions. 3.09 The non-compliance in reporting may be attributed to three main causes: (a) Although a format and methodology was established, it was not apparent to wl-om reports should be made; (b) the fact that BARC di.d not have executive authority over research workers on institutes to demand reports; and (c) the complexity of the reporting format. Repurts were to be based on the progress of the procurement and construct.;on cells wnich in the evenc were not properly structured and were lt( '-v relatively inexperienced officers. 3.10 No regular meetings were foreseen in the SAR as, for example, a prescribed regular meeting at the office of the Secretary for Agriculture with the Director of BARI and the Executive Vice Chairman of BARC to review progress. Ad hoc meetings were called by BARC but were often poorly attended. Regular reporting during the early stages of implementation might have prevented the early overspending which was parcly responsible for a cut back in local training. 3.11 BARC accounts were only audited up to FY82 and those of BARI up to FY83. These were not made available under 1984. Accounts for the .otal contract research fund was maintained by BARC, but no separate iccounts for -7- specific contract research projects were maintained in sccordance *th Project Agreement, clause 3.J2. The slowness in proiiding audited accounts indicated weakness in the accounting procedures of BARI and BARC and in the Auditor Ceneral's Office. D. Proc ir .neri 3.12 A prociren2nt ceLl -7as set up in BARC to hanile all purchases. This Jnit was sa.ed :y reLatiVely inexperienced officers wK h .10 access to the technical expeccisd necessary to interpret the list cf equipment in the SAR. Difficulries weru also experienced over complex G03 clearance procedures on Leader sele:i.n which in turn led to letter of credit extensions and extra C o) S t.s 3.13 A L;horc, lrA inpuC to reviow the full list or elilJLpirlen1 drawn up by the procurement cell, before tenders were called, would have minimized such Problems a, ±cilrchas~iig .ractors with mismatched imple.nents and 3pecifying overly sophi.ssi(7ated laboratory equipment for regional stacions. The SAR was coo spvcific in iternizing equipment which was included in tne PP aild pur- chaesed wdloi- rererence to che proposed work programs. C> i K'ines and block alloc.ioni jihould have been shown which could then liave beel, applied accord- ir.g to the exigencies of the project within the Development Credit Agreement (DCA) cateF,ori.s and the nacional budgets. i.. Cost 3.14 'The cotaL project costs (including l.ocal funds), given in Table 1, show that ,he actual cost in 'T'aka was 84% of the SAtR estimate. The main reabn DefindJs .'iis lower tnan estlmated cotal cost appears to be the devalua- rtion of the Taka by 40% 'Detween 1978 and 1984. The lower total cost, does noL, however inmean tha: chers were no cost overruns in specific categories. rn Taa tms h e cost: Stf BA. buildings, land development, the purchase of ma-thi .erv and equipnein, - t:in>ig and BARC's operating costs all appear Lo lave irvoIved cost overrin.;. r D,sc)ars'uua :>:voived. incr cm en,:Ia' costs, mainly tor 3ARC, were heavily ., ; '.r-ing r:n1; were on -..rge,, ii bn>h ap,rlicat;on of Lraini ng -8- funds was overly biased in favor of overseas training. Consultancy funds were somewhat underspent due to an abundance of TA expertise already avail- able to BARC under the USAID-funded Agricultural Research II Project. 3.16 Disbursement, however, was orderly, following the original estimates reasonably closely. Some difficulties were created by the depreciation of the Taka and delays in the release of local funds but completion was achieved with a satisfactory utilization. 3.17 Overall, the project's disbursement remained somewhat below expecta- tions due mainly to the suspension of station development at Ramgarh, a slowing down of the contract research program because of administrative difficulties and the near absence of expenditure on local training. G. Implementation of Specific Components BARI Regional Station Deve!opment 3.18 Facilities were created on only three of the four stations selected. For socio-political reasons, it was agreed between the Bank and COB not to proceed with the development of Ramgarh, which is situated almost on the border between India and Chittagong Hill Tract District. Only a small amount of building material had been purchased when work was suspended. 3.19 Most of the improvements anticipated at appraisal were completed on the remaining stations apart from fencing where it was decided that the specified design was not strong enough to deter human and livestock incur- sion. Concurrently, much shorter lengths of 'pucca' wall were constructed in the most critical areas, leaving the stations relatively unprotected. 3.20 Deep tubewells were sunk on each station. Outputs were unexpectedly high and exceeded the delivery capacity of the lined channels, which had no control structures or proper gates making it difficult to irrigate accurately small plots. The buildings were designed by local architects and displayed heavy brick overhangs and arched roofs. The designs were to provide cool working conditions but reduced practical efficiency. Workshops were created without inspection pits, processing sheds are high and open sided making it difficult to work in the rainy season and blocks of flats have been built with design faults in kitchen arrangements and electrification. 3.21 Equipment was not well specified in the SAR. Further complications arose due to the fact that the equipment list from the printed Project File was not available to the procurement cell. The main items of workshop equip- ment, such as 6 ft. lathes and table drilling machines and some of the equip- ment in the cropping and harvesting list such as the reversible ploughs would have been more appropriate to a commercial farm than for a research station. Technical assistance in agricultural engineering was provided to BARC and BARI by USAID but too late to change equipment procurement lists. -9- 3.22 Lack of critical analysis of scientific equipment requirements led to the purchase of highly techn;.cal laboratory equipment which had little application in the present relatively unsophisticated atmosphere of a regional station. In fact, sophisticated laboratory analysis is presently centralized at BARI headquarters in Joydehpur and regional stations do not play a major role in this regard. 3.23 Notwithstanding the above shortcomings, it should be emphasized that the project has substantially improved facilities at the regional stations. The availability of buildings, equipment, and proper access roads allows staff to move out from the BARI central complex at Joydehpur and to par- ticipate in decentralized adaptive research. 3.24 At the BARI Jamalpur regional station, the mission was told that during the rainy season the station was more suitable for rice research than for any other crop due to the flooding of almost two-thirds of the station. However, BARI does not conduct rice research. Creater use could have been made of facilities if the station were used by more than one institute but that was not fe,sible under the project. In Jessore, research was largely concentrated on wheat. Fodder crops and legumes should be given more atten- tion in future programs. BARC Headquarters Complex 3.25 Originally BARC functioned in a small rented office with few staff and no room for expansion. Under Credit 828-BD, a suitable headquarters building for BARC was constructed on a site already identified by GOB in the center of Dhaka. 3.26 Considerable difficulty was experienced by IDA in clearing the design prepared by local consultant architects. Finally, an imposing building was constructed with considerable cost overrun. The additional costs were accom- modated in the 1981 revision of categories. The building is functional although space allocation could have been better designed especiclly in view of the possibility of further expansion. The building nevertheless has increased the status of BARC and has facilitated the activities of the Council. Research Grant Subprojects 3.27 This component was foreseen as an instrument for BARC to have an increased impact on the agricultural research system. Since 1976, BARC had been operating a research grant scheme to direct research efforts into areas of priority. The first funding came from GOB and the USAID-funded Agricultural Research I Project, and some 40 studies had already been initiated. About 90 further research projects were proposed under Credit 828-BD at a total cost of about $1.8 million. Under the constraints -10- imposed by the closing of the PP which, since 1982, only allowed the servic- ing of ongoing projects, 41 subprojects were actually funded during implemen- tation at a total cost of less than US$0.50 million. 3.28 Contract r.s;earch project headings and individual costs are shown in Tables 5a to 5e. In summary form and under the headings proposed in the SAR research grant, subprojects covered the following areas during implementation. Priority_Research Area Number of Subprojects Approved Soil and Water Management 19 Agricultural Economics 7 Cropping Systems 8 Agricultural Engineering 7 Total 41 3.29 The weakest area of subproject activity was agricultural engineering where two selected studies had to be dropped. Of the total funds finally 'u.ilized, agricultural engineerin.g took up just over 7% which reflects a ge.neral weakness in agricuiLural engineering research and also a weakness in BAr.RGs st aff capability in this field. On regional. stations, too much emphasis was put on plant breeding and agronomy and too little oni agricul- Z.ral economics and engineering. This reflects the make up of research Sc.-nc.ists in Bangladesh; there are many breeders and agronomists but very few economists and agricultural engineers. 3.30 A formaL seminar was held at BARC at the end of 1984 when highlights of the program were presented to an invited audience in che form of short papers. The seminar proceedings ;.2re dublished in March 1985 and illustrated some .f the research results achieved to date. Unfortunately, the semnirar 'eai.t wlth all contract research, irrespective of funding, and grouped lresentations which meant that results from individual contracts could not be _]oLatcd. Further reporting of specific contracC results is expected and vi-ere appropriate pubiicationsi should be made for utilization by the exten- SioTI agency. 3./il -ihe provision of research grant funds has increased BARC's influence on the research system. It has alLowed the Council to direct funds into areas oi priority and to encourage the participation ot institutions outside the nmainstream of agricultural research such as, for example, the i,oi-agricultural universities. The Council has also been able to assist rescarch programs requiring inter-institutional and inter-disciplinary p;Rrat On. -11- 3.32 Four universities, the Institute of Nuclear Agriculture, four agricultural research institutes, the Ministry of Agriculture and BARC itself were all involved in specific grant-funded research subprojects. Four of the projects involved inter-institute cooperation but many of the other projects displayed inter-disciplinary cooperation which conformed to the original concept of research grant funding. 3.33 In any future assessment of the research system, the role of contract research should be clearly defined. Contract research is a facility to promote priority research in areas not already covered by existing research programs. It should not be seen as a replacement for normal research programs. Contract research is a valuable tool for promoting research coor- dination between institutions. However, the workload of administering con- tract research and particularly in reviewing work programs is considerable. At present staffing levels contract research threatens to overburden BARC's administrative capability. Contract Research Results 3.34 The proceedings of the workshop on contract research held at BARC in late 1984 dealt with the results to date of all contract research; it did not separate work under the project from other contract research. Further, reporting was organized on the basis of participating institutions and not according to individual subprojects. In future each subproject must be isolated for reporting purposes. The proceedings give many detailed trials results some of which are summarized here. 3.35 National Coordinated Cropping Systems Projects were conducted by BAU, BARI, BJRI and SRI and funded from contract research. Combinations of rice, wheat, jute and potatoes were tested and specific varieties identified for their rotation cropping potential. Great inter-seasonal differences in varietal response were apparent. Sugarcane was intercropped with a range of crops including vegetables, mustard and rice. Vegetables performed well and significantly enhanced sugar profitability but technologies need multiloca- tion testing before being released. BARI had identified traditional cropping patterns at many sites and had managed to combine high yielding rice with several crop system3. Zinc and sulphur deficiency was a serious limiting factor in rice ptoduction. This finding was confirmed in micronutrient studies; zinc deficiency was found to be widespread in calcareous soils and sulphur deficiency was a serious problem around Debiganj. Recommendations for sulphur and zinc application were produced and confirmed in several independent subcontracts. Based on the above work, BARC produced a manual titled "Fertilizer Guide for most crops in Bangladesh 1985", which includes site and crop specific recommendations to reach those yield levels. 3.36 Monitoring of costs and returns for each cropping sequence in the Cropping System Projects for a sample of 45 farmers was conducted by the Agricultural Economics Units of BARI, BWDB and BJRI. The work was undertaken in cooperation with the Agronomy Division. Descriptive staten,ents on farm -12- and farm family characteristics and cost benefit analysis added to the study. Information was also collected on yields, returns, and farmer uptake of improved technology. Socio-economic studies were carried out on assessment of improved technology for wheat, rice, potato and jute. Studies were car- ried out on the management of deep and shallow tubewell system. All this information should be fed back to scientists for their use in modifying research objectives and trials' procedures. 3.37 BARI and BRRI had worked on the efficiency of fertilizer applica- tions. A single basal application of urea gave higher rice returns than split broadcast applications particularly in boro but not in aman. 3.38 A study of soil and water management was begun in 1980 and scheduled for completion in 1986. Crop water responses of wheat, potato, cotton, sugarcane, mustard and maize were determined. No significant responses to irrig tion of lentils and chickpeas were observed. Benchmark socio-economic studies are almost complete at several sites and will be combined with irrigation management and cropping pattern results to produce crop recommen- dations for farmers. Interesting work was undertaken on biological nitrogen fixation under irrigated and non-irrigated conditions. Seven local and two exotic strains of rhizobium had been collected from riungbean. Dhaincha showed promise as a green manure and 31 root modulc forming bacteria are under study. Eleven native and three exotic strains of Azolla are under study. Biomass studies with Azolla had revealed problems with high tempera- tures; shade and phosphate application increased biomass. Low cost techni- ques where developed to reduce seepage losses in farm-level irrigation. Studies on brackish water management are continuing. 3.39 One contract research subproject had investigated .he possibility of using bamboo instead of galvanized pipes for hand tubewells. The bamboo tubewells discharged from 8-11 gpmn and could irrigate between 0.3 and 1.0 acre depending upon the crop. Well life was estimated at 3-6 years. This work is continuing with different types of hand pump. A multipurpose seed drill for jute, wheat and soyabean was developed. After some adjustments, it would be tested under farmers conditions. 3.40 A consultant was employed by BRRI using contract research funds tc undertake a survey of deep-water aman paddy in Dhaka, Tangail and Mymensilgh Districts; no authentic records were available previously. Yield and proauc- tion constraint information had been gathered and recommendations for research produced. The survey showed the enormous potential for increased production by improving varieties, by better agronomic practices and by introducing an improved cropping pattern. The survey identified major crop- ping patterns, flood depths and farmers' practices and constraints. Over 100 varieties of paddy were identified in Dhaka. This particular study was valuable and illustrated the benefit from close supervision of an independent experienced local consultant. -13- Development of a National Agricultural Research Plan (NARP) 3.41 Prior to Lhe nroject, .angladesh had no Research Plan. Fulfilling one of the main objectives of the project, BARC published a preliminary NARP in December 1979 which drew up some of the perspectives and guidelines for the nation,al research systcm. It also presented research priorities. The NARP was a first effort and mostly involved BARC staff; other Ministries were not involved. Much more work will have to be done to translate the plan into a national program with priorities, manpower and resources clearly spelled out and matched to the national development effort. The development of a fully satisfactory NARP will take time and many versions as more institutions contribute to tiue document. However, the project's impact in starting this process has been important. The strengthening of the agricltural research system initiated througll this project is being sustained and expanded through a Second Agricultural Research Project (Cr. 1455-BD) which became effective in April 1985. Training 3.42 In general, the overall program was not particularly well executed although 3± Ph.D. programs were arranged overseas in appropriate disciplines. The only training in Bangladesh was 515 man-weeks of seminars arra;iged by BARI. 3.43 Initially, the BARC training cell was headed by a relatively junior scientific officer who had considerable difficulty in identifying suitable candidates, overseas placements and planning suitable programs in Ph.D. courses. The practical difficulties of studying in foreign universities were not appreciated, like the need for a good command of the English language, which led to most courses of study being extended. Much assistance was given by USDA, particularly with regard to U.S. universities which helped to over- come some of the difficulties. 3.44 The main problems appear to have been, first, the weakness of the BARC training cell (a Director cf Training was eventually appointed in 1981), second, the lack of suitable curricula at BAU, and, third, almost a total lack of communication between BARC and BAl' on priorities and training needs in research training disciplines. BARC advertised the availability of fund- ing for local Ph.D.s at BAU but received no applications. Good candidates were only interested in overseas studies. The weakness of the BAU Ph.D. curricula is reflected in the fact that only four Ph.D. degrees have been awarded by the University between its creation in 1961-62 and the end of 1984, although currently 20 Ph.D. students are under supervision. 3.45 In late 1981, when it appeared that there would be a considerable overspending on buildings and equipment, and when it was agreed that locaL training programs were running into problems, GOB and IDA agreed to limit manpower devElopment to that already underway, namely 31 overseas Ph.D. programs, and to suspend attempts to promote local training courses. -14- 3.46 Of the 31 overseas place.ents, 23 have returned with a Ph.D. degree, 2 have returned with M.Sc. degree. Five remaining Ph.D. candidates will complete their courses shortly. A recent BARC survey of returnees has shown that without exception they have benefitted from their studies and most have assumed senior positions in their parent institutes. The project has created a greater awareness of the need to strengthen in-country training and has prompted BAU to make better use of available funds by preparing adequately structured curricula for advanced level studies. Technical assistance for training was provided by USAID and a manpower training plan was produced during the project period. Technical Assistance 3.47 Thirty-six man-months of technical assistance were proposed covering station development, studies to assist BARC in research programming, and monitoring of the research grant system. Assistance was utilized for station development and for preparing a manual for contract research, but little was needed for research planning in view of the TA available liberally to BARC unecr the USAID Agricultural Research II Project. 3.48 The assistar.ce provided was adequate but, as already mentioned, considerabLe advantages could have bee;, derived from providing outside exper- tise on procuremnent matters where there was a marked weakness. BARC. Staffing 3.49 Staffing lceves at BARI regional stations and at BARC headquarters were built uip satisfactorily except at the most senior levels where expertise is short throtighout Bangladesh. BARC's improved credibility and standing depended largely on the status of its Executive Vice Chairman. Considerable difficulty was experienced in finding well qualified and experienced can- didates for the key posts of Member-Directors for the different BARC Divisions. The position of hember-Director Engineering was not filled during implementation and this no doutt reflected on poor equipment and machinery specifications and procurement. Three of the Member-Director positions were still vacant at project completion and three of the nine positions were filled by t mnporary staff. Provision of Operating Funds 3.50 Project proforma for the release of funds were prepared on a timely basis. Or June 30, 1982, a crisis developed when the PP came to an end. Theoretically, COB should have continued to provide all operating costs and support for facilities developed under the project. However, as this occurred during a period of severe stringency in all GOB budgets, all pay- ments to staff on the development budget and normal servicing tf running -15- exercises were postponed, leading to a severe cutback in activity, par- ticularly on the regional stations. kesearch stations were disrupted and all travel was stopped. 3.51 The crisiS was partly resolved wien GOB and USAID reached agreement over the utilization of PL 480 Title II funding for specified areas of expen- diture such as operating costs of the BARI research station. This arrange- ment is c-ntinuing but is obviously not a satisfactory long-term solution. IV. PROJECT IMPACT AND INSTITUTIONAL DEVELOPMENT 4.01 The project was set up to strengthen the agricultural research system in Bangladesh and to enable it to improve adaptive research relevant to local conditions. The project was also to support an ongoing GOB program of improving BARI research facilities, management and staffing, and to strengthen BARC to permit it to fulfill its role in planning, coordination and monitoring agricultural research on a national basis. 4.02 As has already been mentioned, the project has enabled BARC to move into a substantial new headquarters complex with meeting room and seminar facilities to assist research coordination. Also BARC has established con- siderable credibility through its management of the research grant system and the creation of a forum to discuss national needs in agricultural research. A start was made on the development of a National Research Plan. 4.03 BARI now has improved facilities on regiona! stations where little existed in the past. BARI is now involved in serious internal discussion over moving more of its research programs from the central complex at Joydehpur to the regional stations; this would be an important positive move. Staffing levels have been increased with adequate interdisciplinary representation. 4.04 The project has also created a greater awareness of national require- ments in agricultural training at all levels. Curricula are being upgraded at BAU and guidelines are being drawn up to better combine overseas institu- tions with local training. 4.05 During the project period and largely as a resul: of BARC's increas- ing influence, the Jute Research Institute and the Institute for Nuclear Agriculture were brought into the central system. In the words of its Chairman, BARC is slowly moving the Bangladesh agricultural research system towards being "functionally separate but structurally integrated". In effect, the main crop research institutes would maintain their "functional" autonomy but fully cooperate with BARC in establishing "structurally" coor- dinated development aims and planning a national strategy. Before the project, BARC was almost non-existent, now it is a significant central force in agricultural research. -16- 4.06 The project was principally of the institution-building type and has long-term objectives. It is not p(ssible at present to assess all of its major benefits. Nevertheless, overall the project has assisted COB in strengthening its national agricultural research system through improved facilities, better coordination, and improved management. 4.07 When considering the impact of the project, it has to be appreciated that it is part of an ongoing program assisted by a number of donors (mainly IDA and USAID) aimed at developing a mature research system. The ultimate success of the project will depend on whether BARC can continue extending its coordination efforts among the autonomous institutes and whether it will be able to carry the national agricultural research plan through to a series of properly funded and managed complementary research programs. The main problem presently besetting BARC io that the increasing volume of resources at its disposal may outst-ip its administrative capability. Also, exces- sively complicated procedures as exemplified in the screening of research proposals could indeed hamper the promotion of research activities. BARC must be seen as a research facilitator, not another bureaucratic hurdle. In order to maintain its position as the apex institution in agricultural research, BARC must make every effort to attract suitably qualified and mature candidates to assulme the positions of Mermr~ Directors of its divi- sions. V. IDA PERFORNANCE 5.01 Project design has generelly proved satisfactory and overall cost estimates were not far from. those actu \ / ) / caJ\2 25' I~'_ o L~~~~~ ~~~~ - - , (o.-. SYN Ts aut 7 ifi V J r-2i?urttA e \< 0 Ne.rrJktGno 21'Y NSiNG a ) t >la 0GK ~>c 8 1l ~ w;* T tudhEp,- Mht ' N ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ j~~~~~~~~~I<~~~~~~~~~~~~~Ksborqonlt RAJS kHAJHydr TAM;A' - _ C) 'T.bwnio PA OB NW, \yheipnk 2 ) 4;t'. Sbanw nx1HI V~~~~~~~~~~ Ii D-AAr RIDPL)R~~~~~~~~oidki 0-_ d-, ~ ~ ~ ~ x- -5 '1~~~~~~~~~~~~~~~~~~~~~~~~~~~adod ILL Haji ~'V. 2 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-~~~~~~~~~~~~r i 'Ni / ( Rohmrutjur0 0t 0kt 2 A RBARISAL Si Chund ANGA AT ( ( ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~) 0, PAT!JAKHA, CHIT r_.s y.s , _... f.., ..o.. -'. ,' A.. ..A * .... n _ ifS _ f t1de o e f . ..... *._.'. ,,.. an' - a 1 *,',d S,e y o r I'AK-: := =_.. _ Cox;s Razor rR 4r 2 p4----.-----,, Ra,!woys r A ' RNers v' r'BuRMA In*ernatfonaIl boutoaries BANGLADESH B 0 0 10 20 30 40 50 s 0 21, 13D * g fJoz of fJengo/ o 20 40 o0 so SRtI IANV.A co _ _ ____- ____