Nepal - Agricultural Projects

Document of The World Bank FOR OFFICIAL USE ONLY MICROFICHE COPY Report No. 10661-NEP Type: (PPR) Report No. 10661 JAMES, E / X31756 / T9057/ OEDDI PROJECT PERFORMANCE AUDIT REPORT NEPAL AGRICULTURAL EXTENSION AND RESEARCH PROJECT (CREDIT 1100-NEP) HILL FOOD PRODUCTION PROJECT (CREDIT 1101-NEP) CASH CROP DEVELOPMENT PROJECT (CRED-T 1339-NEP) MAY 20, 1992 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS Napalese Rupee (NRa) Year US$1.0: NR. 12.00 1980 NRs 14.55 1983 NRs 21.23 1986 NRs 27.19 1989 NR. 42.30 September 1991 WEIGHTS AND MEASURES 1 hectare (ha),. 2.47 acres (ac) 1 kilometer (km): 0.62 miles 1 metric tone (m ton): 2.205 pounds 1 kilogram (kg): 2.2 pounds HMON (BORROWER'S) FISCAL YEAR July 16 - July 15 FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Directo-General Operations Evaluation May 20, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report - NEPAL Agricultural Extension and Research Project (Credit 1100-NEP) Hill Food Production Project (Credit 1101-NEP) Cash Crop Development Proiect (Credit 1339-NEP) Attached, for your information, is a copy of a report entitled "Project Performance Audit Report on Nepal - Agricultural Extension and Research Project (Credit 1100-NEP), Hill Food Production Project (Credit 1101-NEP) and Cash Crop Development Project (Credit 1339-NEP)", prepared by the Operations Evaluation Department. Attachment 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. FOR OFFICIAL USE ONLY ACRONYMS AND ABBREVIATIONS AA Agricultural Assistant AADO = Assistant Agricultural D-evelopment Officers ADBN - Agricultural Development Bank of Nepal ADO - Assistant District Officer AERP - Agricultural Extension and Research Project AIC - Agricultural Inputs Corporation APROSC- Agricultural Project Service Center ARPP - Agricultural Research and Production Project ASC - Agricultural Service Center (HFPP) CCDP - Cash Crop Development Project DADO - District Agricultural Development Officer DFMS - Department of Food and Marketing Services DLDAH - Department of Livestock Development and Animal Health DOA - Department of Agriculture DTO - District Technical Office EPAD - Evaluation and Program Analysis Division FAO/CP Food and Agriculture Organization - World Bank Cooperative Program FIWU - Farm Irrigation and Water Utilization Unit FSP - Farming System Perspective GDP - Gross Domestic Product HFPP - Hill Food Production Froject HMGN = His Majesty's Government of Nepal ICP - Integrated Cereal Project (USAID) IDA - International Development Association IFAD - International Fund for Agricultural Development IPM - Integrated Pest Management JCF - Janakpur Cigarette Factory JT - Junior Technician JTA - Junior Technical Assistant LAC - Lumle Agricultural Center M&E - Monitoring and Evaluation MFA - Ministry of Food and Agriculture NARC - National Agricultural Research Council NARSC - National Agricultural Research Service Center NGO - Non-governmental Organization OED = Operations Evaluation Department PCR - Project Completion Report PLAA - Panchayat-level Agricultural Assistant PPAR - Project Performance Audit Report RDA - Regional Director of Agriculture ROO - Research Outreach Officer RRS - Regional Research Station SAR - Staff Appraisal Report SMS - Subject Matter Specialist T&V - Training and Visit Methodology TDC - Tobacco Development Company UNDP - United Nations Development Program USAID - United States Agency for International Development VDC - Village Development Council 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 outhorization. PROJECT PERORMANCE AUDIT REPORT NEPAL AGRICULTURAL EXTENSION AND RESEARCH PROJECT (CR. 1100-NEP) HILL FOOD PRODUCTION PROJECT (CR. 1101-NEP) CASH CROP DEVELOPMENT PROJECT (CR. 1339-NEP) TABLE OF CONTENTS PREFACE . . . . . . . . . . . . . . . . . . . i BASICDATASHEET . . .ii EVALUATION SUMMARY . . . . . . . . .0. .. . . ix I. PROJECT BACKGROUND . . . . . . . . . . . . 1 Natural Zones. . . . . . . . . . . . . . . . . 1 Environmental Concerns . . . . . . . 2 Production Concerns . . . . . . . . 3 Agricultural Strategy * . * . . . . . . . . . 3 Bank Support for Agriculture . . . . . . . . . . 3 Sectoral Institutions 9 # * * * . . . . . . 4 II. PROJECT DESIGN . . . . . . . . . . 7 Agricultural Extension and Research Project . . 7 Objectives . . . . . . . . . . . . . . . . . . 7 Extension . . . . . . . . . . . . . . . . . . 8 Research . . . . . . . . . . . . . . . . . . . 9 Hill Food Production Project . . . . . . . . . 10 Objectives . . . . . . . . . . . . . . . . . . 10 Organization ......... . . . . . . . 10 Cash Crop Development Project . . . . . . . . . 11 Objectives . . . . . . . . . . . . . . . . . . 11 Organization...... . . . . . . . . . . . 12 III. IMPLEMENTATION EXPERIENCES . . . . . . . . 13 Agricultural Extension and Research Project . . 13 Extension . . . . . . . . . . . . . . . . . . 13 Farm Trials and Demonstration . . . . . . . . 15 Research . . 9 . . . . . . . . . . . . . . . 16 Other Components . . . . . . . . . . . . . . . 17 Hill Food Production Project . . . . . . . . . . 17 Infrastructure . . . . . . . . . . . . . . 18 Staff Numbers and Performance . . . . . . . . 18 Research-extension Linkage . . . . . . . . . 21 Other Components . . . . . . . . . . . . . . . 22 Cash Crop Development Project . . . . . . . . . 22 Sugarcane . . . . . . . . . . . . . . . . . . 23 Tobacco . . . . . . . . . . . . . . . . . 23 Mustard . . . . . . .. .. . . . . . . ... 24 Other Components . . .* . . . . . . . . . .* 24 Credit . . . . . . . . . . . . . . . .o . . 25 TABLE OF CONTENTS (cont'd) Pg&e IV. PROJECT OUTCOME . . . . . . . . . . . . . 26 Agricultural Exteneirn and Research Project . . . 26 Production ImpaGz of Research and Extension . . 26 Awareness and Adoption of Tech.ology . . . . . . 28 Hill Food Production Project . . . . . . . . . 30 Production Impact . . .. . . . . . . . . . 30 Awareness and Adoption of TechnioQgy . . . . 31 Cash Crop Development Project . . . . . . . . . . 32 Impact . . . . . . . . . . . . . . . . . . . 32 Sugarcane . . . . . . ....... 33 Tobacco . . . ........ . . . . . . . . 33 Mustard . . . . . . . . . . . . ... . . . . 34 Other Componenta . . . . . . . . .... . . . . 34 V. FINDINGS AND ISSUES . . . . . . . . . 35 Changes in Cropping Pattern and Productivity . . . 35 Changes in Production Factors . . . . . . . . . 36 Improved Varieties . . . . . . . . . . 36 Inorganic Fartilizer... ..... . . . . . 36 Other Production Factors . * . . . . . . . . 38 Institutional Performance . . . . . . . . . . ..38 Sustainability . . . . . . . . . . . . . 44 Recent Institutional Developments . . . . . . . . 46 Concepts for Improved Institutional Performance . 47 Responsiveness to Farmers' Conditions . . . . . 47 Farming System Perspective . . . . . . . . . . . 48 Reference Groups . . . . . . . . . . 48 Problem Census . . . . . . . . . . . . 0 . . 51 Research-extension Linkage . . . . . . . . . . . 52 Demonstrations . . . . . . . . . . . . . . 52 Mass Media . . . . . . . . . . . . . . . . . . 52 Institutional Implications and Requirements . . . 53 Extension . . . . . . . . . . . . . . . . . . 53 Research . . . . . . . . . . . . . . . . . . . . 56 Environment . . . . . . . . . . . . . . . . . . 57 Bank Performance . . . . . . . . . . . . . . . . . 57 Design , . . . . . . . . . . . . . . . . . . . 57 Supervision ................ . . 59 TABLES . . . . . . . . . . . . . . . . . . . 60 Table 1: Nepal - District-level Organization and . . . . 60 Operations of Agricultural Extension Systems Table 2: Agricultural Extension and Research Project - . 61 Performance Indicators Table 3: Hill Food Production Project - . . . . . . . . 63 Performance Indicators Table 4: Cash Crop Development Project - . . . . . . . . 65 Performance Indicators . . . . . . . . . . Table 5: Crop Yields in Cash Crop Development Districts 67 Table 6: Budgets to District Agricultural Offices . . . 68 Table 7: Agricultural District Budgets and . . . . . . . 68 A6ricultural GDP Table 8: Agricultural Research Budgets . . . . . . . . . 69 Table 9: Agricultural Research Budgets and . . . . . . 69 Agricultural GDP TABLE OF CONTENTS (cont'd) Page FIGURES ................ . . 70 Figure 1: Wheat Yields Indices: 1972/73 - 1989/90 . . 70 Figure 2: Rice Yields Indices: 1972/73 - 1989/90 . . . 71 Figure 3: Responsive Agricultural Researcher - . . . . 72 Extension - Farmer Interactions ANNEXES . . . . . . . . . . . . . . . . . . 73 Annex 1: Description of the T&V Extension Service . . 73 Designed for Agricultural Extension and Research Project - The Plan Annex 2: Implementatic- of the T&V Methodology in . . 76 the Terai - Agricultural Project Service Center Reports Annex 3: The Wealth Ranking Techniques and its . . . . 78 Use in Extension ATTACHMENT Comments from the Borrower..... . . . .80 NAPS: IBRD 23594 IBRD 23595 IBRD 15049R1 PROJECT PERFORMANCE AUDIT REPORT NEPAL AGRICULTURAL EXTENSION AND RESEARCH PROJECT (CR. 1100-NEP) HILL FOOD PRODUCTION PROJECT (CR. 1101-NEP) CASH CROP DEVELOPMENT PROJECT (CR. 1339-NEP) PREFACE This is a Project Performance Audit Report (PPAR) on three projects in Nepal which had a common major objective to increase agricultural production thvou6h improved services in agricultural research and extension - Agricultural Extension and Research Project (AERP, Credit 1100-NEP), Hill Food Production Project (HFPP, Credit 1101-NEP), and Cash Crop Development Project (CCDP, Credit 1339-NEP). The former two credits were approved in February, 1981 and closed with a three-year delay in July, 1990. The CCDP was approved in March, 1983 and closed on schedule in December, 1988. Cancellations represented 25% of the US$17.5 million credit for the AERP, 33% of the US$8.0 million credit for HFPP, and 24% of the US$6.0 million credit for the CCDP. The PPAR is based on the Project Completion Reports (PCR) for the three projects 1, the eval%.ation reports prepared by the Agricultural Projects Service Center (APROSC) in Nepal for the project agencies of His Majesty's Government of Nepal (HMGN), the Staff Appraisal and President's Reports, the credit documents, project files and discussions with Bank staff in Washington, D. C. and in Kathmandu. An OED mission visited Nepal in September 1991 for discussions in the field and with project-related agencies in Kathmandu. The kind cooperation and assistance of HMGN personnel are gratefully acknowledged. The PCRs provide a comprehensive description of the projects, and the PCR on the AERP and HFPP provides some comparative analysis of the forms of extension currently used in Nepal, including the Training and Visit (T&V) system which was the methodology utilized in the projecLs. The PPAR elaborates on particular aspects of the design, implementation and impact of the projects, and examines the implications for the further development of agricultural research and extension services in Nepal. Following standard OED procedures, copies of the draft PPAR were sent to HMGN. Comments were received from the Project Coordinator of the on-going Agricultural Extension Project which followed the projects under review; these are included as an attachment to the report. I Nepal: Agricultural Extension and Research Project (Credit 1100-NEP) and Hill Food Production Project (Cr. 1101-NEP). Report Number 9994, October 16 1991. Nepal: Cash Crop Development Project (Cr. 1339-NEP). Report Number 9659, June 14, 1991. - i1i - PROJECT PERFORMANCE AUDIT REPORT hEPAL AGRICULTURAL EXTENSION AND RESEARCH PROJECT (CR. 1100-NEP) BASIC DATA SHEET A. Key Project Data Actual as 2 Appraisal of Appraisal Item Estimate Actual Estimate Total Project Cost (US$ million) 20.85 15.70 75 Credit Amount (US$ million) 17.50 13.06 75 Date of Board Approval 2/81 Date of Effectiveness 7/81 Date Physical Components 12/86 12/89 Completed Closing Date 7/87 7/90 Institutional Performance Mixed B. Cumulative Disbursement (US$ million) IDA Fiscal Year 1982 1983 1984 1985 .1986 1987 1988 1989 1990 1991 Appraisal 4.1 8.4 10.4 13.4 16.4 17.5 - - Estimate Actual 2.8 5.9 6.9 7.3 8.3 9.2 11.5 12.3 12.9 13.1 Actual as 68 70 66 55 51 53 66 70 74 75 5 of Estimte Date of Final Diaburasment: January 2, 1991 - iv * C. Staff Input i' Scs as of Project Pre. Throuah Appraial * 1.6 49.0 - * * * * * * * * * 30.8 App.at I tdxough 0.3 * 18.1 22.0 - A . 4 . - . . . . £0.4 Bad approval Board approval - - - - . . . . . fattte supervision * * 2.9 7.0 10.0 4.9 9.2 4.8 42.7 48.2 12.0 7.3 . 151.2 0.3 1.8 47.1 24.9 7.0 10.0 4.9 0.2 6.8 42.7 46.2 12.0 7.5 0.1 241.4 SStaftt wakes Local H. and consultants are discounted at Wel D. Mission Data No. of Days in Specializatiq; Performaneo . Type of mission Type MonthfYear Persons Field Represented - _atg Problems I Identification 7/79 2 18 A,Ex Preparation 12/79 4 27 A,E,EX,S I - Appraisal 6/80 3 25 A,E,EZ - . Start-up 3/81 2 26 A,Ex 1 - Mission Supervision 1 12/81 3 19 A,Ez,ME 1 M,T Supervision 2 3/82 1 3 A I M Supervision 3 10/82 2 15 A,A I T Supervision 4 7/83 2 I8 A9Ex 2 M Supervision 5 12/83 1 2 A 2 M Snpervision 6 3/84 1 18 A 2 M Supervision 7 12/84 2 12 A,Ex 2 FIM Supervision 8 6/85 2 9 A,A 2 H,F Superviion 9 2/86 2 18 A,Ex 2 - Supervision 10 10/86 3 14 A,E.Ex 2 - Supervision 11 4/87 1 3 A 2 - Supervision 12 10/87 2 8 A,A 2 - Supervision 13 2/88 2 15 A.A 2 - Supervision 14 10/88 2 20 A,Ex 2 - Supervision 15 3/89 2 9 A,A 2 Supervi2ion 16 12/89 3 22 A,E,Ex 2 - Supervision 17 3/90 2 23 AsA 3 - V Date of return to headquarters. it A a Agriculturalist, E s Economist; Ex a Exteneionist; HE - Monitoring and Evaluation Specialist; S - Sociologist 1 s Problem free or minor problem; 2 * Moderate problem; 3 a Major problem. r w - Financial; N - Managerial; T - Technical. V Due to change in supervision formats, information not available. -v - NEPAL HILL FOOD PRODUCTION PROJECT (CR. 1101-NEP) BASIC DATA SHEET A. Key Proieo T)ata Appraisal Actuals as % of Item Estimate Actual Appraisal _Estimate Total Project Cost (US$ million) 9.70 6.71 69 Credit Amount (US$ million) 8.00 5.39 67 Date of Board Approval 2/81 Date of Effectiveness 5/81 Date Physical Components 12/86 12/89 Completed Closing Date 7/87 7/90 Institutional Performance Poor B. Cumulative Disbursement (US$ million) IDA Fiscal Year 1982 1983 1984 1985 1986 1987 198C 1989 1990 1991 Appraisal 0.3 1.4 3.4 5.4 7.0 8.0 - - - . Estimate Actual 0.4 1.1 1.2 1.6 2.0 2.9 3.6 4.3 5.1 5.4 Actual as 33 79 35 ;0 29 36 45 54 64 67 Z of Estimate Date of Final Disbursement: August 29, 1990 0. staff tunut 1 apptl • •P29 2.8 • • ••.- • • 4. approval Boa.approval - • • - Total • • 42.3 27.1 7.0 13. .1 7.I 6.4 10.3 7. 1.4 i ftaff veckas Local EL ad 4onsuts are dtsouned at 441 D. 1isLon Data mo. of Day. In Specialization Performance Typa of lsgion Tne Month/Year 1/ Porgona .igjd Rogresented Z/ Rating I Problema 4/ Identification 3179 3 24 E,I,A - Preparation 10179 5 25 E,E,I,A.L - - Appralcal 6180 4 26 A.A,A,I,- Start-up Miecion 3/81 1 6 A 1 - (prior to credit öignature) Supervislon 1 6/81 1 5 A 1 7 Supervision 2 2!82 1 5 A 1 F Supervislon 3 6/82 1 8 A 2 i,0 Supeviaion 4 12/82 2 11 A,E 2 N,0 Supervision 5 5/83 1 14 E 2 Supervilon 6 10/83 1 12 A 2 ä,j Supervision 7 9/84 1 10 A 3 ä,T Supervision 8 5/85 2 9 A,A 2 M,T Supervision 9 12185 1/ 3 9 A,A,Ac 3 p Supervigion 10 2186 1 8 A 3 -/ Supervioion 11 6/86 1 15 A 2 - Supervlsion 12 10/86 1 7 A 2 - Supervision 13 6187 2 27 A,Ex 2 - Supervision 14 12/87 1 21 A 2 - Supervieion 15 6188 2 22 A,A 2 - Supervision 16 3/89 2 9 A,A 2 - Supervi@lon 17 12189 3 22 PO,A,A 3 - Supervilon 18 4/90 2 12 A,A 3 - a4 arUt M - LAo mtaum 9 oniE - 94~ c teculca specIallet X X &Umspa s L - " ok Spectallse; 10 - project Operarlms OfIcer i trea or a~inr le3 i wa~te problemo 3 - fajor problem. #I I Irl~al m,* w~seaswiaT - .ahpleal. A trgG»r Me . D evaIaM,on ~ ~ estue revtwed the Ta coupone~t ta Decmber 195. u a nupervio formats. lnformation 0ot available. - vii - NEPAL CASH CROP DEVELOPMENT PROJECT (CR. 1339-NEP) BASIC DATA SHEET A. Key Proiect Data Actuals as % Item Appraisal of Appraisal Estimate Actual Estimate Total Project Cost (US$ million) 7.36 6.44 88 Credit amount (US$ million) 6.00 4.59 76 Date of Board approval 3/83 Date of Effectiveness 8/83 Date of Physical Components 7/88 7/88 Completed Closing Date 12/88 12/88 Economic Rate of Return 31-64 Negative/ Very low Institutional Performance Poor B. Cumulative Disbursement (US$ million) IDA Fiscal Year 1984 1985 1986 1987 1988 1989 Appraisal 0.2 1.0 2.5 4.1 5.4 6.0 Estimate Actual - 1.6 2.4 3.6 4.3 4.6 Actual as 0 160 96 88 80 76 % of Estimate Date of Final Disbursement: January 1991. - viii - C. Staff Input j/ IDA Fiscal Year FY81 FY82 FY83 FY84 FY85 FY86 FY87 FY88 FY89 FY90 TOTAL Through Appraisal 5.2 8.9 - - - - - - - - 14.1 Appraisal through - 21.9 34.8 - - - - - - - 56.7 Board approval Board approval - - - - - - - - - - - through effectiveness Supervision - - 1.7 8.7 5.9 15.4 9.4 18.7 4.1 .2 64.2 Total 5.2 30.8 36.5 8.7 5.9 15.4 9.4 18.7 4.1 .2 135.0 D. Mission Data No. of Days in Specialization Performance Type of Mission Type Month/Year I/ Persons Field Represented 2/ Rating 2/ Problems A/ Pre- 2/81 2 8 A,A - - Identification Identification 7/81 2 10 A,E - - Preparation 3/82 2 16 A,E - - Appraisal 5/82 5 30 A,C,AtT,E - - Start-up 5/83 1 5/ 17 A - - Mission Supervision 1 9/83 1 10 A I M Supervision 2 4/84 1 10 A 2 M Supervision 3 9/84 1 5/ 8 A 2 M Supervision 4 12/85 3 51 21 A,Ac,E/R 3 M Supervision 5 2/86 1 8 A 3 M Supervision 6 6/86 2 25 AgAg 3 M,T Supervision 7 10/86 1 S1 7 A 3 V Supervision 8 6/87 3 5/ 27 A,E,To 3 M,T Supervision 9 12/87 3 15 A,E 3 M,T Supervision 10 6/88 3 20 A,A,A 3 M,T 1/ Date of return to headquarters. 1/ A * Agriculturalist, Ac a Accountant; C * Credit Specialist; E - Economist; E/R - Extension/Research Specialist; T - Tea Specialist To * Tobacco Specialist; Eg - Engineer. 3/ 1 - Problem free or minor problem; 2 = Moderate problem; 3 - Major problem. W/ F - Financial; M - Managerial; T - Technical. I/ Multiple purpose missions. - ix - PROJECT PERFORMANCE AUDIT REPORT NEPAL AGRICULTURAL EXTENSION AND RESEARCH PROJECT (CREDIT 1100-NEP) HILL FOOD PRODUCTION PROJECT (CREDIT 1101-NEP) CASH CROP DEVELOPMENT PROJECT (CREDIT 1339-NEP) EVALUATION SUMMARY Introduction on schedule in December, 1988 with 24% 1. Agriculture dominates the of the credit canceled. HFPP, which economy of Nepal. In the 1970s the attempted to improve extension in four Bank Group supported the agricultural Hill districts, and CCDP, which sector in development of large-scale concentrated on specific cash crops in irrigation schemes in the Terai (which selected districts which were mostly is part of the Gangetic plain and is in the Terai, are rated as located along the border with India) unsatisfactory. There is some and rural development and forestry evidence to suggest that AERP had an programs in the Hill zone (which is 'on increasing the rate of the higher, dissected area, rising to adoption of technology in some aspects the Himalayas). The Training and of production in the Terai. it is, Visit (T&V) extension methodology was therefore, rated as satisfactory, but used in the Terai irrigation schemes. the services instituted under the Three projects were initiated in the project are in need of substantial 1980s which had a common major modification and improvement if they objective to increase agricultural are to be sustainable. production by expanding the coverage of T&V services and supporting 3. The PPAR records the agricultural research - Agricultural implementation experience and outcome Extension and Research Project (AERP), of each project and then examine the Hill Food Production Project (HFPP), institutional implications of these and Cash Crop Development Project experiences. Recent institutional (CCDP). The latter project also developments are also analyzed to included substantial credit and arrive at recommendations on how the production components. This PPAR lessons from the projects can be reviews the three projects which had a practically applied in the development total approved credit amount of of agricultural services in Nepal. US$31.5 million. Objectives 2. The AERP and HFPP were closed in July 1990 after three annual 4. The common objectives of extensions and had 25% and 33% of the increasing the effectiveness of credit canceled, respectively, at agricultural services were reflected account closure. The CCDP was closed similarly in the formulation of AERP -x - and HFPP. Both projects provided moat cases. Slow release of budget infrastructure, equipment, additional funds and administrative problems were staff and staff training for the the main causes of the procurement extension service, and the procurement inefficiencies. Government (HMGN) of fertilizer to ensure this key substantially met the quantitative production input was available. In targets in incremental staif addition, AERP provided specific recruitment. Many of these recruits, support for research at key stations however, were in temporary posts which in Terai; HFPP addressed the comon resulted in high staff turnover and problem of access in the Hills through low morale. construction of trails and suspension bridges, and also provided for 3,000 8. The T&V methodology was ha of small-scale irrigation installed in the Terai districts, and development. appeared to obtain reasonable coverage of farmers, at least for food crop 5. CCDP relied on the T&V services production. Selected farmers were in AERP or other on-going projects to used as extension workers (AA) to provide improved extension to carry out the programmed farm visits sugarcane, tobacco and mustard (under the guidance of the lowest growers, as well as giving support to level civil service employee in relevant research stations dealing extension - the JTA) with the with these cash crops. Smaller incentive of a small monthly schemes for groundnut, sericulture and allowance. The planned transfer under ginger research and development were CCDP of extension responsibility for also included. The objective in sugarcane and tobacco to DOA was selecting the particular crops was to delayed for three years, and was enhance the level of self sufficiency subsequently poorly implemented by in the production of raw materials DOA. required for agro-based industries serving domestic needs and potential 9. The off-station, adaptive export markets. The AERP and HFPP research capacity at the Terai were not crop-specific, but it was research stations was introduced but assumed that food crops would receive did not develop sufficiently to result most attention, especially in the in the intended attention to latter project. technology which would be relevant to the real constraints experienced by 6. The Department of Agriculture most farmers. Research emphasis in (DOA) of the Ministry of Food and both food and cash crops continued to Agriculture (MFA) was the implementing focus on more favorable environments agency for both research and with the use of relatively high levels extension. of cash inputs, and on crops which had traditionally received most research Implementation Experience attention. Research-extension linkage arrangement were instituted, but were 7. Infrastructural investments in not organized to produce the most the projects were largely completed effective interaction between but with considerable delays and were researchers and extension personnel. of poor quality in many instances. The procurement of equipment and vehicles 10. In the Hills, the planned T&V were also subject to serious delays in system proved more difficult to - xi - implement due to the conditions of 13. In AERP, where paddy and wheat access and to a scarcity of relevant were the two major crops, the project technology for the many different had no apparent impact on wheat production environments in this zone. yields. There was some evidence that The use of farmers as AAs did not a response in paddy yields may have function as well as in the Terai occurred due to the project, which conditions, and supervision by JTAs raises the possibility that other suffered due to a large number of unmeasured improvement&; in the local vacanciec associated with low staff farming systems may also have been morale. This resulted in a much generated. Despite the emphasis given poorer coverage of farms under HFPP. to use of inorganic fertilizer, it was In the final years of this project, a applied to wheat crops by less than U.K.-funded entity in the DOA, the 60% of farmers, and to paddy by only Lumle Agricultural Center (LAC), was about 45% of farmere. The rate of brought into the project to improve fertilizer application was relatively the research-extension linkage and uniform among users regardless of farm increase the relevance of technology size or contact with extension; this being recommended to farmers in the rate was about half that recommended Hills. by the research and extension services. Use of crop protection 11. External training programs were measures was very low. carried out, but many trainees left the sponsoring institutions on their 14. In HFPP, the average district return to Nepal due to dissatisfaction yields of the main crops - paddy, with working conditions. The scarcity maize and millet - did not increase. of qualified staff with experience in Apart from improved varieties for practical agriculture was a chronic wheat and maize, adoption of the deficiency throughout the projects. recommended practices was very low. This was especially important for the The technology adoption in the HFPP extension subject matter specialists Hill districts was no higher than (SMS) and research outreach officer other districts. The irrigation (ROO) positions because of their investment (on 3,200 ha) contributed critical role in the regular training to higher production due to increased of the front-line extension agents and cropping intensity and yield, but this formulation of recommendations. represented only 18.6% of project costs. Results 15. The CCDP did not have the 12. A number of base-line and expected impact on growth in the cash monitoring studies were contracted to crop sub-sector. Tobacco and mustard a semi-government agency (APROSC), but yields did not increase and tobacco the necessary impact evaluation area substantially decreased. Although studies were not undertaken, with the a yield increase occurred in sugarcane exception of CCDP. Nevertheless, some in the second and third years of the quantitative assessment of project project, this was raised to only pre- outcomes was possible using APROSC project levels and was likely data and the official agricultural associated with the increased use of statistics. fertilizer which was facilitated by the project; there was no upward trend in the 1981-1988 yields. In - xii - general, farmers reacted rationally to potentials of the majority of farmers, use of their resources in their mixed and for a marked improvement in cropping systems, which often did not personnel policies to increase the coincide with the crop-specific level of researcher competence and objectives of the project. The output. The projects supporting groundnut pilot was a failure; the research in the Terai (AERP and CCDP) ginger component had negligible initiated some improvement in the development impect; and the former concern, through emphasis on sericulture program serious under- the outreach (off-station) research achieved its objectives. programs and research-extension linkages. The results were variable, Sustainability however, and the measures will need much more development to justify 16. Public sector agricultural continued and expanded support. research and extension services are likely to be sustained if they can 19. Within the extension attract continued support from institution, the APROSC studies budgetary authorities and if generally reported a favorable satisfactory professionalism and impression of the T&V system amongst morale are maintained by staff in the lower level staff, but dissatisfaction services. Normally, these two factors with wcrking conditions, operational are interconnected and depend on the support and benefits. Many higher institutions being able to provide the level extension staff, however, have type of services needed by been conscious of some major agricultural producers and to deficiencies in the system. Prominent demonstrate the benefits obtained from among these were the development of a the services. largely top-down information flow system which reduced the relevance of 17. Throughout the implementation recommendations to farmers' needs, period of the project, HMGN despite the contrary objectives of the demonstrated its appreciation of the program; the apparent ineffectiveness importance of the agricultural sector of the contact farmer concept; and and of the potential role of research special difficulties in operating the and extension by providing reasonable system in the Hill zones. This led to annual budgets, albeit with some an internal review of the extension delays in release of funds. service in 1989 and again in late Nevertheless, over the last four years 1991. In Deceml-r, 1991, the concepts the budgets allocated to both research of contact farmtr and of using a and extension have experienced some contracted farmer as the main front- decline in nominal terms and over 40% line extension worker were abandoned, in real termas at the same time the and a modified group approach to rate of the research budget to extension contact was adopted. agricultural GDP has declined from 0.29% to 0.19% and of extension from 20. While the principles of research 0.782 to 0.51%. and extension supported under the projects continue to be the objective, 18. Within the research institution, the extension mechanism is in the many professional staff have brin process of change, and the introduced aware of the need for greater research measures need substantial responsiveness to the constraints and improvement before they can serve - xiii - their purpose. Coordinated support by future direction of these institutions donors for further development of the (para 5.32-5.36). Based on the national research system, and Nepalese experience and that of introduction of further improvements countries in similar circumstances, it in the extension methodologies then develops a series of concepts to (combined with their careful address the known constraints to management and a greatly enhanced institutional performance (para 5.37- training program), may permit the 5.48), and outlines how these concepts gradual development of an effective can be incorporated (para 5.49- technology generation and transfer 5.64). system which is able to demonstrate to the Ministry of Finance its 23. The concepts include: substantial contribution to agricultural sector development. enhanced awareness of production Fiscal constraints will persist, constraints by the research however, and there will be great system through the wide adoption difficulty in obtaining substantially of a farming system perspective, higher performance from the existing which will alter the composition public sector institutions as a whole. of commodity and discipline This suggests that further development research programs to more of agricultural research and extension closely meet the requirements of services will have to be conservative farmers; and realistic in their objectives and scale of operations if they are to be the identification and use of sustainable. existing groups of farmers with similar resources and Findings and Lessons technological requirements (reference groups) as the 21. The scarcity of evaluation of extension target, using data limits the conclusions which can identification techniques be drawn from this group of three already demonstrated in Nepal projects. With few exceptions, (wealth ranking technique); and however, the agricultural technology adoption and production response of the use of a problem census farmers in the project districts technique with reference groups appeared to be very similar to that of to serve as a basis for non-project farmers in the same programming extension activities production zones in Nepal. which meet farmers' needs. 22. The experience of the projects, 24. In extensiont the use of these and also of projects supported by concepts has impo..;ant implications other donor agencies in the 1980s, for the degree of responsibility for enable some constructive district-level staff, and necessitates recommendations for improvement in the considerably more emphasis on staff agricultural research and extension training. Not only training in the services. The PPAR analyses the simple application of sociological and performance of the main institutions economic principles will be necessary, involved in the projects (para, 5.11- but also a greatly improved program of 5.25) and considers recent practical technical training. The developments which must impact on the investment in demonstration plots - xiv - would be affected significantly and research management to the development the mass media programs of the DOA of those principles will be required, (which were ignored under the three and the adoption of personnel policies projects) could be made much more to support and reward staff on the relevant and effective without basis of merit will have to be additional investments. instituted. The recent creation of the semi-autonomous National 25. These indicated modifications Agricultural Research Council (NARC) would only be feasible if the concepts provides the opportunity to address are given the full support of the most of the issues constraining management of DOA and of the Ministry research performance. Nevertheless, of Agriculture, and personnel policies as for extension, research strategy are introduced to provide an incentive and investment decisions should be for staff to make a career in based on realistic projections of agricultural extension (para 5.56). public funding and improved In addition, fiscal and institutional institutional performance; this constraints dictate that implies priority being given to adaptive research which maximizes the - further development of the face- use of technology developed elsewhere to-face extension service should in similar environments, and to focus on areas where relevant measures which improve the quality of technology exists or can be staff. readily generated, and where it is believed that such services 27. Lessons for future Bank will be necessary to bring about participation in projects concerned an acceptable rate of technology with agricultural research and adoption; extension services include: - the improved program should only It is unwise to assume that a be instituted at the rate specific extension methodology determined by the availability which is apparently beneficial of trained personnel; and in a particular situation will be uniformly applicable to other - private sector firms involved in situations with different bio- agroprocessing and input physical and socio-economic supplies should be encouraged to characteristics. provide specialized technical services to their farmer It is more appropriate for the clients. Bank and the Borrower to agree on extension objectives, 26. In research, the increased principles and concepts, and emphasis to be given to demand-driven then to allocate adequate programs (in the form of increased resources to designing extension farming system diagnostic and outreach systems which are the most work, more responsive commodity and appropriate for each situation. discipline research, and more effective linkages with extension) - The availability of trained mostly represents a major extension of personnel should determine the principles already initiated in the rate at which improved extension systemo The complete commitment of systems can be implemented, and - xv- assessment of training needs and dictate a strongly top-down organization of the required direction in all communication, training program should receive the imposition of a methodology priority in the preparation which runs contrary to this will phase. not be institutionalized unless special measures are taken to - It is essential in extension change these traditional project preparation to determine attitudes. whather there is adequate improved technology available - In attempting to promote a which is applicable to the significant change of production constraints faced by orientation in a research mcst farmers (improved practices system, the potential impact of being used b- the more partial support for operations successful farmers in the in a select number of research various bio-physical and socio- facilities can be severely economic production limited by established attitudes circumstances should be and expectations within the reviewed); if not available, the institution as a whole. generation of relevant technology for a deficient - In supervision of research and production environment has first extension service projects, it priority. is essential that priority attention is given to the - The adoption of improved systems relevance of technology and to to provide services which farmer responses, while also require more dedication and being aware of how the interest of staff will not be methodologies to provide the effective unless personnel services are being implemented. policies which provide incentives for good performance - Where serious implementation are installed. problems persist in a project, a major review is essential to - In dealing with an institution revise its content with the that has traditionally been Borrower, or cancel the bound by hierarchies which loan/credit. PROJECT PERFORMANCE AUDIT REPORT NEPAL AGRICULTURAL EXTENSION AND RESEARCH PROJECT (CREDIT 1100-NEP) HILL FOOD PRODUCTION PROJECT (CREDIT 1101-NEP) CASH CROP DEVELOPMENT PROJECT (CREDIT 1339-NEP) 1. PROJECT BACKGROUND 1.1 Nepal has a per capita Gross Domestic Product (GDP) of only US$170 and estimated 40% of its population live in absolute poverty. Agriculture dominates the economy and over 90% of the population depend on it. The sectoral share of GDP was about 60% at the time of preparation of the projects and has been about 55% in recent years. Enhanced performance in the agricultural sector is essential to improve the economy and the welfare of the population; since the mid-70s, however, agricultural production has not kept pace with the high rate (2.6%) of population growth. Natural Zones 1.2 The country is most commonly divided into three broad ecological zones - Terai, Hills and Mountains - which differ greatly from one another in altitude, topography, climate, agricultural productivity and population density. They account for 14%, 43% and 43%, respectively, of the country's total area (Map IBRD 23595) 1.3 The Terai, part of the Gangetic plain, is a narrow strip of land, never more than 50 km wide, stretching along the border with India. Elevation ranges from 60 m near the Indian border to 275 m nrar the Hills. The Terai has a humid subtropical climate and has in most part only recently been settled because of the prevalence of endemic malaria untij. the late 1950s. More than one half of the cultivated area in the Terai receives some form of irrigation and this zone has nearly 802 of the net irrigation command area in Nepal. This is the major agricultural production area accounting for about 53% of the country's gross cropped area and produces 73% the national rice harvest, 61% of the wheat and from 80% to 100% of the important cash crops represented in oil seeds, sugarcane, tobacco and jute, mostly cultivated on small holdings.' Since its long southern boundary does not constitute a natural frontier, movement across the Indian - Nepalese border under normal conditions is essentially unrestricted. The Terai economy is, therefore, crucially affected by the virtually free flow of people and commodities between the two countries. 1.4 The Hills form the central of the three ecological zones. This zone is made up of high ridges and steep valleys embracing numerous streams and rivers. The deep gorges of the major and many smaller rivers have an important I Average holding size of the eight districts of the AERP project area is 1.3 ha. - 2 - effect on the economy of the Hills in limiting transport and communication and in c,ntributing to the creation of the numerous micro environments that are characteristics of that belt. The hill elopes are generally carved into terraces which have been built over hundreds of years. Each Hill area is a micro environment with varying conditions for plant growth depending inter alia on elevation, direction and steepness of slope, solar radiation and cloud cover. Farms are small and heavily subsistence crier.ted.2 A typical farmer operates a number of plots which usually differ in agro-ecological characteristics. While making for more difficult farm management, this spreads crop risk. The rugged topography and the resulting poor accessibility of these areas limit availability of modern inputs and farmer support services. Irrigated land is mainly in the valley bottoms, in small patches along river banks, or on small terraces where stream diversion provides some seasonal irrigation during or after the rains. The Hills account for 38% of gross cropped area which is mostly below 2,000 m in elevation. Maize is the most important crop accounting for about 70% of national production; it is usually relayed3 with finger millet, and grain legumes are also important. Livestock un;its per household are about double that of the Terai. The scarcity of suitable agricultural land and accessible fodder resources for livestock severely limit the potential for expansion in agriculturally based livelihood and there is a continuous migration from the Hill zone to the Terai. The Kathmandu valley is one of the largest valleys in this zone and, owing to its proximity to the capital and a good road network, tt is one of the most-developed parts of the country. For all intents and purposes, it may be considered a separate geographic entity.' 1.5 The Mountain zone comprising the high Himalayas parallels the northern border. It is characterized by high peaks, steep slopes and narrow valleys. Mountain agriculture is livestock oriented, as neither climate nor terrain provide much opportunity for cropping outside the valley floors and the lower slopes. The cropping system is maize-based, and is usually followed by buckwheat, barley or potatoes. The zone accounts for less than 10% of the country's gross cropped area. Environmental Concerns 1.6 Strong external support provided health care and malaria control in the 1950s. This resulted not only in a movement to the Terdi from the higher zones, but also a big inc -ease in the overall population growth from 1.4% to 2.6% 2 75% of holdings in the four districts of the HFPP project area are below 0.5 ha and occupy only about one-third of the lowland irrigated area in these districts; the remaining holdings which are larger than 0.5 ha have two- thirds of the irrigation. "Relaying" means that a second crop is planted in the first crop before the first crop is harvested. 4 The Kathmandu valley has a warm-temperate climate which occurs between elevation levels of 900 m and 1,400 m in Nepal. -3- annually. With 902 of the population being rural, and mostly in subsistence production, the rapid population growth increased the demand on forests for fuelvood, timber, fodder and material for compost, and on land for expanded crop production. Extensive deforestation occurred in both the Terai and the Hills. A greater sediment load in rivers with siltation of reservoirs and irrigation canals in the Terai became obvious, but studies had not distinguished the extent to which these types of environmental degradation were man-made or occurred due to natural forces in the geological formation of the Himalayas which, in geological terms, are relatively recent. Production Concerns 1.7 In the seventies, the total production, area cultivated and yields of maize, barley, potato and the cash crops, ajute and tobacco were either stagnant or declined. Production of rice and millet and the cash cropil, oil seed and sugarcane increased but these increases were associated more with an expansion of area rather than in yields. Wheat was the exception and increased in both production at about 5.5% per annum and in yield at about 3% per annum; this was associated with the spread of high-yielding varieties introduced in the sixties and an expansion in irrigation in the Terai in which wheat was an attractive second crop after rice. Chemical fertilizer use was extremely low at an average of about 6 kg nutrients/cropped ha (compared to 25-30 kg/ha in neighboring India). Agricultural Strategy 1.8 The generally poor performance of the agricultural sector was attributed to inefficient use of the public sector irrigation schemes in the Terai, a low use of production inputs (especially fertilizer), and poor services to assist farmers to adopt technology which would increase productivity and incomes. The Sixth National Development Plan (1980/81 to 1984/85) placed high priority on developing agriculture. This included continued efforts to realize the Terai's considerable potential for increased production of both food grains and cash crops; improved management of the irrigation infrastructure was encouraged as well as the adoption of better services in extension, research and input supply. Revitalization of the Hill zone was planned through improved services for food production to meet local requirements and for crop diversification into other activities such as small-scale and cottage industries. Reforestation programs were also given priority in order to provide fuelwood and fodder as well as to reduce soil erosion, especially in the Hill zone. This reflected some recognition of environmental concerns, at least in the area of deforestation, but was not indicative of a general development strategy which focussed on resource management. Bank Support for Agriculture 1.9 Consistent with the strategy of HMGN, the Bank group in the seventies had supported the agricultural sector in development of large-scale irrigation schemes in the Terai, and rural development and forestry programs in the Hill zone. The irrigation projects in the Terai introduced intensive agricultural extension services using the Training and Visit (T&V) methodology.' During the Sixth Plan, the Bank approved the three credits which are the subject of this PPAR - the Agricultural Extension and Research Project (AERP, Cr. 1100-NEP), the Hill Food Production Project (HFPP, Cr. 1101-NEP) and the Cash Crop Development Project (CCDP, Cr. 1139-NEP);' they were designed to assist HMGN in improving services and production in significant parts of the Terai and the Hill zone. The AERP covered eight of the 20 districts in the Terai, while the HFPP covered four of 37 districts in the Hill zone. The CCDP was located in parts of 12 districts, nine of which are in the Terai,' one in the Kathmandu Valley and two in the Hill zone (Map IBRD 23595). Sectoral Institutions 1.10 There were four ministries concerned with agricultural development programs at the time of preparation and implementation of the three projects: the Ministry of Food and Agriculture (MFA - later changed to the Ministry of Agriculture, but MFA is used throughout this report), the Ministry of Water Resources, the Ministry of Forestry and Soil Conservation and the Ministry of Land Reform. The MFA included the Department of Agriculture (DOA) which was responsible for agricultural research, extension and training and a Department of Livestock Development and Animal Health (DLDAH) responsible for livestock research, extension and health services; it also had responsibility for the Agricultural Development Bank of Nepal (ADBN), the Agricultural Inputs Corporation (AIC), the Nepal Food Corporation and the Farm Irrigation and Water Utilization Unit (FIWU) which was responsible for farm irrigation development. 1.11 The DOA had separate Deputy Directors General responsible for research and extension. Research was organized along commodity (rice, wheat, maize, citrus and potato) and discipline lines, and had a central research complex at Khumaltar (in the Kathmandu Valley), seven research stations and some 5 T&V is a hierarchically organized and time-bound method of managing extension, designed to deliver selected and timely technology to farmers with strict regularity. The system encourages links with agricultural research institutions, emphasizes a professional approach to extension, and requires an exclusive devotion to extension work. In principle all extraneous activities such as those related to the physical handling of inputs and credit applications are avoided. 6 The agricultural development program in Nepal has been heavily dependent on external aid. Bilateral donors (e.g., USAID, Federal Repuilic of Germany, Switzerland, Australia and the United Kingdom) and FAO/UNDP and the Asian Development Bank (ADB) have also supported projects designed to enhance agricultural services and rural development in addition to the financing provided by the Bank. 7 Five of the Terai districts under CCDP were also covered by the agricultural service improvement under the AERP. -5- 30 farms to complement the stations. Extension had a D.rector (RDA) in each of the four administrative regions of Nepal,' who supervised district level extension and development programs managed by District Agricultural Development Officers (DADO). In each district, the DADO was assisted by Assistant Agricultural Development Officers (AADO). Below this level, non-graduate staff in the form of Junior Technicians (JT) and Junior Technical Assistants (JTA) were the main extension contacts with farmers. An innovation in the seventies was to select farmers for a minimal training program and employment on a part-time basis as Agricultural Assistants (with an allowance of NRs 50 to 150 per month) to make visits to farmers in his local area under the guidauce of the JT and JTA. As noted in the previous paragraph, livestock research, extension and health services were under a separate and parallel system, as were services to the forestry sector which were the responsibility of the Ministry of Forestry and Soil Conservation. 1.12 The ADBN was the single major source of institutional credit in the sector. It provided short-, medium-, and long-term credit to individual, cooperatives and companies for agricultural development and operations. About 60Z of farm families were indebted, but the average level of indebtedness per household was low. The debt level was highest with larger farmers, as was the amount borrowed annually, and the debt level was higher in the Terai than in the Hill and Mountain zones. 11 Institutional credit made up about 20% of debt and was concentrated on production inputs, and horticulture, irrigation and machinery investment. Lind and livestock purchase, payment of wages and consumption 8 Subsequently, these were modified to five regions; all regions extend from the Indian border to the Himalayas and contain parts of the three natural zones. 9 The allowance of NRs 50 was equivalent to about US$4.00 per month in 1980; only in the districts with exterually-assisted projects was the allowance increased to NRs 150 per month. 10 The ADBN was established in 1968 and took over the assets and liabilities of the Cooperative Bank which had been established in 1963. In 1973, the Land Reform Savings Corporat,on (which served to mobilize rural savings) merged with the ADBN. 11 FARM SIZE ALL LARGE MEDIUM SMALL MARGINAL Proportion of Farm Households Hill 61 80 56 52 33 Indebted (1976/77) - Z: Terat 63 69 63 66 58 Debt per Farm Households Hill 691 1028 580 320 476 (1976/77) - Res Terai 994 1 4913 1069 736 329 NRe 12.5s US$1.00 (1976/77) -6- expenditure were mainly covered by non-institutional sources. Professional money lenders made up less than 30% of non-institutional credit sources. Loan recovery from institutional credit was poor, at 401 to 50%. 1.13 A coope- -tive (Sajha) development program had been introduced in 1976 to revitalize the cooperative movement, mobilize savings and link economic development with political decentralization. Sajhas were to provide credit, supply inputs and market consumer goods, Borrowing from ADBN and the Sajha' s own equity made up the total financial resources of the Sajhas. Sajhas were the major distributors of fertilizers from the AIC, which used Cooperative Unions at district level as the primary recipients where a Union existed (less than half of the districts). 1.14 The AIC was the sole HMGN agency responsible for the supply of inputs. Its functions included (a) importation and distribution of chemical fertilizers and maintenance of a buffer stock;12 (b) collection, processing, storage and distribution of imp roved seeds; (c) procurement and distribution of agricultural chemicals; and (d) distribution of locally manufactured and imported agricultural machinery. 1.15 The MWR was primarily responsible for the implementation, operation and maintenance of irrigation projects over 50 ha in size. Resources for small- scale irrigation were scarce, but a number of different ministries had become involved in these works in the seventies, including MFA through its FIWU. At district level, small-scale irrigation activities were under a common District Technical Office (DTO) which was responsible for all engineering activities (HMGN buildings, village water, local roads and bridges) and was usually poorly equipped and staffed. 12 All chemical fertilizers were imported, and HMGN heavily subsidized the cost of fertilizers to farmers. The retail price was dependent on the prevailing price in neighboring India, and a uniform price structure was adopted for all delivery points throughout the country, regardless of transportation costs which were very high in the Hill and Mountain zones. is AIC received foundation seed from the research stations of DOA and multiplied this through contract growers before sale as certified seed. It also received and sold all certified seed multiplied on the research stations. - 7 - II. PROJECT DESIGN AGRICULTURAL EXTENSION AND RESEARCH PROJECT Obiectives 2.1 The main objective of the AERP14 was to achieve early and sustained improvement in agricultural production, particularly of food grains, in eight administrative districts in the Terai (Map IBRD 23594). These improvements were to derive from adoption of low-cost, labor-intensive agricultural practices and use of appropriate levels of fertilizer. The project would reorganize and strengthen the agricultural extension service and upgrade adaptive research to achieve this objective. It provioed for: (a) additional staff, housing, offices, equipment and training for the extension service in the framework of the T&V methodology; (b) additional staff, laboratory facilities, machinery and equipment at four regional research stations in a strengthened program of applied and adaptive research directed at addressing farmers' production problems; (c) overseas training of project staff and provision of consultants;1 and (d) provision of 22,100 tons of fertilizer to ensure its availability and sale on project districts. Regular monitoring and evaluation procedures and impact studies were also envisaged. This was to be the responsibility of the Evaluation and Program Analysis Division (EPAD) of MFA with support from the Food and Marketing Services Department (DFMS) and local consultants. 14 The AERP and CCDP were prepared by the Agricultural Projects Service Center (APROSC) with assistance from FAO/CP; the HFPP was prepared by FAO/CP. 15 The UNDP provided a US$0.65 m grant for this component; it envisaged training as short courses in international and national research and extension institutions in Asia, and technical assistance to research and extension institutions in Nepal as well as to ADBN and AIC. -8- Extension 2.2 The traditional extension service as described in para 1.11 was considered to have a low level of effectiveness. This was assumed to be due to lack of work programming and supervision, multiple risponsibilities (including coordination of input supply and credit) and extensive reporting which seriously diverted attention from the extension function, top-down targeting of specific crop areas and input use, poor training and technical backstopping of field staff, and a deficiency in technology to meet the specific needs and resource circumstances of large numbers of farmers. The T&V system had been introduced in six Terai eistricts in the seventies as part of Bank-supported irrigation projects (para 1.9), and encouraging results were reported at the time of project preparation from those which had been in operation for a number of years. It was, therefore, adopted as the extension methodology on the additional eight Terai districts included under AERP. 2.3 *The organization of the T&V service is described in Annex 1 and summarized in Table 1. A special characteristic of the T&V methodology adopted in Nepal was the contracting and training of selected farmers as the front-line extension staff (Panchayat-level6 Agricultural Assistants: PLAA) who would carry out the regular, programmed visits to contact farmers. The PLAAs would receive technical guidance from Junior Technical Assistants (JTA) who were civil service employees of the DOA, with each JTA being responsible for three to five PLAAs. The usual T&V components of regular fortnightly traii,ing and farm visit 16 An appreciation of Panchayat and some other aspects of Nepalese institutions is facilitated by an understanding of Nepal's political history: O Nepal was unified in the late eighteenth century by the Shah dynasty of the Kingdom of Gorkha in central Nepal. The Shah kings moved their capital to FV-thmandu. " From 1946 to 1950 Nepal was ruled by an hereditary dynasty of Prime Ministers - the Ranas. The Shah kings were mere figureheads. The state continued to be administered through intermediaries as an essentially feudal state until the Ranas were overthrown in 1950. O In 1950 the Shah kings were restored as constitutional monarchs. During the following 10 years there were experiments with democracy. Although this was a period of reform, there was great instability and social unrest. * In 1960, following a constitutional crisis, the King resumed full control, political parties were banned and a partyless Panchatayat system was introduced. The smallest political unit under the Panchayat system was the village Panchayat which consisted of a total of 4,000 to 6,000 people divided into nine wards. The elected leader of the village Panchayat was the Pradhan Pancha. Each of the 75 districts in the country had an elected district Panchayat and at the national level there was the national Panchayat (parliament). The powers of these assemblies were heavily circumscribed and the illegality of political parties did not prevent the existence of many factions, often violently opposed to each other. In 1984:, the total number of village Panchayats was increased. O Following a period of violent civil unrest early in 1990, the King agreed to abandon the Panchayat system, to allow political parties and to have his powers limited by a new constitution. The new constitution was promulgated in September 1990 and elections were held in May 1991. - 9 - schedules, appointment of Subject Matter Specialists (SMS) to provide technical support, and research-extension linkage arrangements were provided for. Loans were to be made to JTAs to purchase bicycles, while Junior Technicians (JT) and SMS were to receive motorcycles, and four districts without a four-wheel drive vehicle were to be provided with one." Research 2.4 USAID had assisted HMGN in developing its agricultural research system in a Food Grain Technology Project in the late sixties and in an Integrated Cereal Project (ICP) commenced in 1976. These projects strengthened the wheat, maize, and rice commodity programs through training and infrastructural development. The ICP was also introducing the concept of cropping systems research and demonstrating the importance of socio-economics in assessing research needs and in effective technology transfer. The AERP research component was designed to further strengtien applied and adaptive research in the Terai, especially for the eight project districts, but the research would also have some relevance for the other six districts already serviced by the T&V extension system. 2.5 Emphasis was to be given to location-specific applied and adaptive research, crop and varietal selection for the common rice-based farming systems, improved cultural practices (especially in dryland farming), and soil and water management. Trials were to be carried out on research stations, substations and farmers' fields. 2.6 Four existing research stations were designated as Regional Research Stations (RRS) and supported under the project: Nepalganj (Far Western), Bhairawa (Western), Parwanipur (Central), and Tarahara (Eastern). Two substations - Nawalparasi and Hardinath (Central) - would also be supported to provide coverage of all agroclimatic zones and research on major crops. A Research Outreach Officer (ROO) would be appointed at each of the RSS to carry out off-station trials, liaise with extension staff and have contact with farmers. 1 The DOA had a Farm Information and Communication Unit at headquarters and an Information Officer in each of the four regional headquarters to handle mass media aspects of extension. The SAR indicated this facility was "well established" and did not focus on any organizational improvement in this area. - 10 - HILL FOOD PRODUCTION PROJECT Obiectives 2.7 The main objective of the HFPP was to increase food crop production in the Hill districts of Gorkha, Lanjung, Syangja and Tanahu (Map 15049R1). This would be promoted through improved services from Agricultural Service Centers (ASC) , some critical infrastructure, and assistance with inorganic fertilizer procurement and supply: (a) constructing and equipping 20 ASC to provide offices and housing for staff of agricultural extension, livestock production/health services, ADBN and Sajhas; (b) increased staffing for the agricultural and livestock extension services, and training of extension personnel and staff of ADBN and Sajhas; (c) improved marketing through regular marketing days; (d) improvement of about 3,000 ha of small-scale irrigation schemes; (e) upgrading of trials and bridges to facilitate access; and (f) procurement of 3,200 tons of fertilizer to ensure its availability in the project area. Technical assistance" and monitoring and evaluation were also provided for. Organization 2.8 Overall responsibility for project implementation rested with the Regional Director of Agriculture (RDA) in the Western Region in which the four i8 This concept had been used under the first IDA-financed Rural Development Project (Cr. 617-NEP) implemented between 1976 and 1983. By 1979, however, it was evident that the agricultural services provided from ASCs were largely ineffective, apart from some impact on improved seed and fertilizer use in the small area of irrigation, largely due to lack of appropriate technology. See OED Report No. 6424. 19 The UNDP provided a US$0.85 m grant for the training and technical assistance component (NEP/80/019). International training (20m/m), and international consultants in agronomy (44m/m), extension training and communications (20m/m) and irrigation engineering (20m/m) were included. - 11 - districts were located. He would be assisted by a Project Coordinator and the four ADOs, one in each district. Project committees at district and regional levels were to provide cooperation amongst the project agencies. These included the above units under the DOA in MFA, as well as other Departments and units pertaining to MFA - DLDAH, ADBN, AIC- and the Sajha. DOA was responsible for project M& and was expected to use APROSC for baseline and evaluation studies. 2.9 The organization of the agricultural extension system was to be on lines similar to the T&V model described for AERP in Annex 1 (summarized in Figure 1). One difference was in having all public services in the agricultural sector (extension, credit and input supply) emanating from the ASCs (five per district) located at points of high population density and agricultural activity. Another difference was that the linkages with research were not as well defined as in AERP; there was a dependence on the Pumdi Bhundi research station for research results and the appointment of an expatriate agronomist to advise on aspec,s of extension agronomy and adaptive research programs. The importance of livestock in the Hill environment also led to specific support for a parallel livestock extension service in additional staff and trainin as well in the provision of a small veterinary dispensary at each ASC. Overall, the extension methodology was not as concretely defined in HFPP as in AERP, as it was appreciated that the standard T&V system would likely have to be adapted to the peculiar conditions in the Hill environment. CASH CROP DEVELOPMENT PROJECT Obiectives 2.10 The objective of CCDP was to promote the commercial production of a number of cash crops, mostly in the Terai, to enhance self sufficiency and reduce import requirements. The major focus was on sugarcane, tobacco and mustard, each of which had an established local processing capacity (the former two in the public sector) which was underutilized. The project also included pilot development of groundnuts (for oil) and sericulture, and provided some support to ginger development as a high-value crop. The project was in accord with the HMGN Sixth Development Plan, which aimed at enhanced self sufficiency in the production of raw materials required for agro-based industries serving domestic needs and potential export markets, and intensification into high-value crops. 2.11 The components included: 20 At the time of project preparation both crop and livestock extension, although distinct services, were under the responsibility of DOA. These functions were placed in separate departments in the MFA reorganization of 1979/80 so that livestock services moved to DLDAH. AT appraisal and in the SAR, little attention was paid to the possible adverse effects of having the livestock services being provided by a new department which was not receiving much direct support under the project. - 12 - (a) increasing yields on an existing sugarcane area of 4,660 ha (including ratoon crop) to meet the requirements of the Birganj Sugar Factory (BSF), through improved research and extension, seed cane multiplication, increased access to short-term credit, and construction and upgrading of feeder roads; (b) increasing yields and crop area (from 1,650 ha to 2,800 ha) and improvement in quality of tobacco by improved research and extension, improved tobacco redrying facilities at the Tobacco Development Company (TDC), increased access to medium- and short- term credit for curing barn modifications and production input purchase respectively, and technical assistance for training in production and curing; (c) increasing yields of mustard through further strengthening research,21 and provision of short-term credit to farmers; (d) a 350 ha groundnut production scheme supported through research, seed production and credit; (e) a sericulture pilot on 150 ha including staffing and equipping a sericulture unit and providing credit to cooperating farmers; and (f) strengthening ginger research and initiating a market study. Substantial yield increases were projected: 66% for mustard and tobacco, and 91% for sugarcane. Organization 2.12 The project covered 12 districts and the components necessitated many implementi6g agencies. The DOA was to be responsible for research and extension; the ADBN for credit; the AIC for input supply; the BSF for coordinating credit and input supply and for efficient management of cane collection, seed stock production and farmer payments; the TDC for coordinating credit and input supply and for management of the colleation, payment and redrying processes for tobacco; and the Janakpur Cigarette Factory (JCF) to maintain close linkages with extension and TDC staff to improve leaf quality. Monitoring and evaluation was the responsibility of EPAD and cash crop production and yield measurement was to be managed by DFMS, both of MFA. A Project Coordination Committee under the Secretary of Agriculture and two Sub-committees (sugarcane/tobacco, and other crops) were created for overall coordination of the implementing agencies. 21 The National Oilseeds Research Station at Navalparasi was already supported under AERP (para 2.6). - 13 - III. IMPLEMENTATION EXPERIENCE AGRICULTURAL EXTENSION AND RESEARCH PROJECT 3.1 Slow progress in construction of facilities and on the recruitment and training of the intended staff complement led to three extensions, each of one year, so that there were nine years between effectiveness and the Credit Closing Date. A significant deprecietion of the NRs against the US$ and a reduced infrastructural program led to 25% cost savings and 25% of the US$17.5 million Credit was finally canceled. Extension 3.2 Progress in construction if infrastructure to support the extension system remained slow throughout project implementation. At project closure, all housing was completed for JTs, AD0s and SMS. Despite the project extension of three years, however, only 38 of the 125 houses planned for JTAs were completed, and only limited improvements were made to the two regional training centers which were due to be upgraded. Only one of eight sub-centers (office plus training room) were built under the project, but these were largely constructed under other aid programs. Reduced and delayed budgetary provision were the main reasons for construction difficulties, but this was confounded by administrative problems in tendering and by the Indian-Nepal trade and transit crisis in 1989/90 which affected the availability and cost of construction material. Building quality suffered due to the inability t;f the Agricultural Engineering Division of MFA to cope with its supervia'.cn vork load; this, despite the increased staffing under the project (Table 2), albeit with a serious delay in the recruitment of overseers. 3.3 The procurement of vehicles, motorcycles and equipment was completed after substantial bureaucratic delays in the tendering-approval process from 1982 to 1984; however, no additional motorcycles were procured to mobilize the additional number of JTs recruited by the project (para 3.4). A recurrent problem was the frequent appropriation of vehicles by local administrative authorities for non-extension and non-agricultural duties. Loans were provided to JTAs for bicycles in most cases, but with inflation the loan amount became inadequate and towards the end of the project the maintenance allowance was suspended but subsequently reinstated. Overall, there were many deficiencies in the mobility situation, but transport was certainly much improved on the traditional extension system. 3.4 The project provided for incremental Positions in the form of 500 PLAAs, 4 ADOs, 10 SMSs, and one project coordinator for the eight districts. Following a reorganization of the Panchuyat structure in 1982, the total number of Panchayats increased from 500 to 575, and HMGN increased the recruitment of PLAAs accordingly to provide one per Panchayat. HMGN, however, not only increased the JT and JTA positions to support the additional PLAA, but intensified the intended ratio (Table 1) of JT:JTA from 1:4 to 1:3 and of JTA:PLAA from 1:3 to 1:2.5 (Table 2). This increased the operational costs of - 14 - the service and provided an average coverage of 630 households per PLAA and 1,390 households per extension staffer represented as JT or JTA. The recruitment of SMS was not increased. 3.5 HKGN did not meet its obligation to employ staff in permanent positions so that, despite continued pressure from IDA, at project closure only half of the incremental positions were permanent. This led to low morale and high turnover in ADO, SMS and JT/JTA posts due to resignations to accept permanent positions elsewhere, especially of the more experienced staff. 3.6 A persistent problem was the relatively low average quality of extension staff. The SMSs, who were largely responsible for regular training of JTs, JTAs and PLAAs, at best had a B.Sc degree, had little practical experience, and usually lacked a clear understanding of farmers' problems. The meetings with researchers were not as effective as intended (para 3.13) and recommendations developed tended to be directed at more favorable resource and high input conditions; this affected the regular training sessions of lower-level extension staff and the same bias was reflected in extension advice and in demonstration programs (para 3.10). 3.7 The performance of most JTs and JTAs was not good. Low morale (caused by housing difficulties, low travel allowances, low salaries and inadequate job security), a background that was commonly urban rather than rural, a low level of basic training21, and poor quality in their in-service training (para 3.6) resulted in a diminished capacity to analyze and respond to real production problems faced by farmers. PLAAs obtained acceptable coverage of contact farmers in their designated areas and appeared to dedicate three to four hours per day in most days of the month to their advisory duties; although not exactly "full-time" as described in the SAR, this did represent a realistic allocation of time for practicing farmers. Their practical experience as farmers helped in their communication with fellow farmers, but the type and relevance of technology recommended was limited by the assistance received from JTAs and SMS. Host PLAAs had from five to ten years of schooling. 3.8 Supervision of the overall extension program by the DADO and of PLAAs by JT/JTAs was not well developed. This was complicated by an excessive administrative burden on the ADO, a lack of incentives to carry out strict supervision, and transportation problems. Supervision further deteriorated at the end of the project in the period of political uncertainty in 1990/91. 3.9 Two studies (1984 and 1988) by APROSC report on the operational characteristics of the T&V system in the Terai. The results of these studies are reported in Annex 2 and support many of the findings of the PPAR mission; the results can be summarized as follows: (a) The coverage of both contact and non-contact farmers by PLAAs in fortnightly visits was satisfactory in most districts, but the regularity of visits was less satisfactory; 21 JTs and JTAs had a high-school diploma followed by largely theoretical training at an agricultural college. JTs had up to three years of the college training which implied a total of 12 years of education. JTAs had less training. - 15 - (b) the coverage of farmers by JT/JTAs was only about one-third that of PLAAs; (c) there was good awareness amongst all farmers of the identity of the PLAAs and a little over half of all farmors were aware of the identity of the relevant JT/JTA; (d) the ADO/AADOs, as the most senior extension staff in a district, were aware of shortcomings in the applicability of some recommendations; (e) the technical competence of about one-third of JT/JTAs was less than adequate (as measured by a knowledge test), and the JT/JTAs rated about one-third of PLAAs as being less than adequate to do their job properly; and (f) there was a generally favorable attitude amongst staff to the T&V methodology in the Terai, but not with employment conditions, so that morale was deleteriously affected. Farm Trials and Demonstrations 3.10 Farm trials and demonstrations were important in the adaptive research and extension programs.2 "Mini kits", which were used by the RRS to 22 The activities on farmers' fields were: a. adaptive research trials ("outreach" or farmers' field or pre- production verification trials) designed by researchers at the RRS and for which inputs are supplied by the RRS; site selection and supervision of implementation are shared by the ROO of the RRS and the SMS and JT/JTA of the extension service; pre-production verification trials are closer to farmer management conditions than farmers' field trials; common subjects are varietal testing and cropping patterns with a standard fertilizer input, although some sites involved fertilizer treatments. b. Mini-kits which consist only of seed supplied from the RRS; site selection and supervision are mostly done by SMS and JT/JTA; fertilizer application is in accordance with the cooperating farmers's practices; and mini-kits are largely used for widespread testing of new varieties. c. Demonstrations, which consist of seed and fertilizer packages supplied by the extension service; the JT/JTA and SMS are responsible for selection and supervision of implementation; about 90Z pertain to demonstration of new varieties; farmer inspections of plots are intended to lead to requests for seed of the demonstrated variety which are coordinated through the PLAA/JTA. d. Seed Dissemination plots, which are a limited program of the extension service in which seed of varieties which look promising in the mini- kit program is distributed in small packages to interested and progressive farmers; the extent of this program in any one year depends on the district extension budget; it provides an opportunity for wider testing and proliferation of promising varieties in a pre- release phase of the varietal selection process. - 16 - test pre-release varieties in a range of production environments, had some extension function, but the main extension thrust was through the demonstration plots financed through the district extension service budget. Unfortunately, the expenditures on demonstrations were not planned to maximize their impact on technology adoption: (a) cereal varietal demonstrations dominated the design. with high fertilizer inputs in irrigated conditions (with little attention to oilseeds and pulses, rainfed production conditions, and cultural practices); (b) selection of the cooperating farmer was usually on the basis of personal relationship between the staff and the cooperator, with little or no regard to the appropriateness of the farm or the farmer for maximizing the demonstration effect; (c) the demonstrations were largely "district-wide" packages which did not take account of the current level of technology used by farmers nor the type and level of innovation which would be most appropriate to demonstrate to the majority of farmers in the locality at the time;23 and (d) supervision of demonstrations and interpretation of the results were often deficient. Research 3.11 Construction at the research facilities was delayed as in the extension program, but was largely completed by Closing Date (Table 1). The quality of civil works, however, was poor in many cases and maintenance was deficient. Vehicles were provided as scheduled, but also suffered from appropriation by local authorities to the detriment of the on- and off-station research program. Nine of the planned 14 incremental professional staff were recruited. Unfortunately many of these staff including the ROOs were appointed on contract on a temporary basis which contributed to dissatisfaction and low morale. 3.12 The research statio,is expanded their "outreach" programs with trials on farmers' fields during the project. However, the concept of designing and implementing applied and adaptive research to address the identified needs of the many farming systems in the different agro-climatic conditions was not well developed. On many stations, the majority of resources were still devoted to applied research under conditions of more favorable resources and high input levels, emphasizing irrigated grain crops. The recommended priority for attention to rainfed areas, water management practices and the relationship of organic/inorganic soil amendments did not eventuate. 3.13 The formal structure for linkaies between extension and research with both central and district technical committees to develop and review technical recommendations and adaptive research priorities was established. So too were the seasonal workshops and bimonthly training sessions at the RRS (Annex 1 paras 11 and 12). In the absence of a well-developed farming system perspective, 23 The concentration of varietal demonstrations, of course, implies that a common district-wide demonstration would often be justified in the T4rai; the use of associated inputs and crop husbandry, however, should have been adapted to differing farmer circumstances. - 17 - however, and having ROOs and many researchers with little practical experience and limited technical depth, most of the training sessions lacked the quality needed to yield the desired results; the same characteristics also led to most recommendations being very general in nature and lacking specificity to the conditions faced by many farmers, especially in the less favorable production environments. Other components 3.14 The UNDP technical assistar,..e grant for training and consultancies was increased from $0.65 million to $0.82 million and fully utilized. Training of 42 staff was mainly in the Philippines, UK and USA (average duration of 2.7 months, but with some courses up to two years), and not in India as intended at appraisal. Staff (and Nepalese institutions) were not interested in training in India, and UNDP supported the change to other locations as India was not well organized to cater for short-term training of this type and was understood to offer theoretical rather than practical curriculum. Many staff who received overseas training left their institutions to take up permanent employment elsewhere. While this does not represent a loss to Nepal, the project institutions were in dire need of improved skills. 3.15 The consultancy funds were utilized in short term consultancies (42 person-months) covering extension management and communications and some research topics. Any sustained impact of these consultancies was difficult to identify, and UNDP considered that HMGN institutions were not geared to fully utilize the consultants and that counterpart were often not adequate. UNDP also confirmed that it was difficult to select consultants whQ could adjust to local conditions and be fruitful in short-term consultancies. 3.16 Over 26,000 tons of fertilizer were procured under the project, thus alleviating the funding constraint faced by HMGN in fertilizer purchase which required foreign exchange. This did not, however, appear to completely remove the problems of low and untimely supply of fertilizers in the project area. 3.17 APROSC carried out eight studies on behalf of project management. However, the overall quality was weak and they were not organized to provide sequential feed-back to project management nor provide substantial evaluation data. Some studies, however, were useful in contributing to an understanding of how the T&V methodology was being implemented and to measuring the impact of the services on farmer awareness of technology (Annex 2 and para 4.6). HILL FOOD PRODUCTION PROJECT 3.18 The HFPP suffered from substantial construction delays (as for AERP) and the Closing Date was extended three times so that the implementation period was eight rather than the intended five years. The total project costs in US$ were 31% less than that estimated at appraisal; this was mainly due to an incomplete infrastructural program and depreciation of NRs against the US dollar. - 18 - These factors also led to a cancellation of 33% of the US$ 8.0 million credit at project account closure. Infrastructure 3.19 The construction program for the ASCs was impeded by problems in land acquisition, tender administration and eventually in procurement of building materials (para 3.2). Only 16 of the 20 planned ASCs were completed by Credit closing, and the remaining four were completed under Cr. 1570-NEP. Due to the construction delays, extension staff (JT/JTAs) were housed in rented premises and ADBN and Sajha offices operated independently. At project closure the ASCs operated largely as a living compound for the extension service with little of the intended Integrated service functions being centered there. 3.20 The access trail program made slow progress but eventually met its targets, while only 4 of the 34 planned suspension bridges were built (Table 3). The bridge component suffered not only from the general deficiencies in the infrastructural program but also from the late recruitment of a bridge engineer and an excessive work load on the unit responsible for construction. 3.21 Although about 40% (22,500 ha) of the cultivated land in the four districts had some form of irrigation, nearly all of this was of a supplementary nature. The program to upgrade irrigation on about 3,000 ha suffered major implementation difficulties, but eventually exceeded the physical targets (Table 3). Some of the necessary specialized staff were not recruited, and poor scheme design and contractor supervision resulted in poor construction. Many of the stream diversion systems required major repairs. Only towards the end of the project were beneficiaries involved in the planning and construction process and obliged to make a small financial contribution towards construction costs and a fund for scheme maintenance; the project did not promote any effective water distribution/maintenance arrangements. 3.22 The program for provision of vehicles fared similarly to the AERP project - delays in procurement, persistent acquisition of vehicles by local administrative authorities, and scarcity *of budgetary funds for repairs, maintenance and operation. Staff Numbers and Performance 3.2? The incremental staffing plan was affected by a reorganization of the Panchayat system in 1982. The number of Panchayat was increased and it was decided to cover all parts of each district rather than to restrict coverage to the reasonably accessible areas as intended at appraisal. The number of AAs was increased but the average number of families expected to be covered by each AA was also increased from about 400-500 to 605, which was a similar ratio to that 24 4gricultural Extension Project II. SAR March 6, 1985. - 19 - adopted in many parts of the AERP in the Terai which had much easier farm access. At the same time, AAs were expected to visit only six "blocks" of farms in a fortnight, as opposed to nine "blocks" in the Terai; this meant that, in practice, household coverage must be more limited in HFPP. JTA staff losses were very high, leading to a high JT:JTA ratio of 1:1.5 (Table 3) and poor coverage of AAs in the lat.er part of the project. The relatively high SMS presence of almost three per district (designated as agronomy, plant protection and communication specialists) was attained but these staff did not normally have specialized training nor experience in these fields. 3.24 The deficiency in capability of many extension staff (SMS, JT, JTA) already noted in the Terai for AERP (paras 3.6 and 3.7) was also evident in HFPP. In the Hill zone, however, the problems of inadequate training and experience were exacerbated by the more difficult and variable production environment, very arduous travel conditions, and a scarcity of appropriate technology to improve productivity in many of the Hill production systems. The location of JT/JTAs at the centralized ASCs meant that much time was wasted in getting to AAs and farmers with a consequent reduced contact; the requirement for JT/JTAs to sign in at the ASC each day (to overcome apvnrent lax attendance to duties) further detra-ted from work time in the latter part of the project. Late provision of budgetary funds and low travel allowances inhibited the supervision function of ADOs and the training function of SMS, but the capability of many SMS to train staff in practical application of technology to address farmers' problems was questionable. The reticence for HMGN to appoint staff to permanent posts was a significant factor in low staff morale and loss of more capable staff from the HFPP extension service (para 3.5). 3.25 The livestock extension and agricultural extension did not cooperate as intended, and the DLDAH staff virtually operated a separate service which was not integrated with the cropping service. Difficulties were experienced in posting livestock JTAs to ASCs and in recruiting and training AA. The animal health program made limited progress in vaccination programs for buffalo and cattle. Fodder promotion activities were very weak, although fodder tree seedlings were distributed in the early years of the project. The HFPP Project Coordinator appeared to accept the isolation of the liveitock services and this component was dropped from the project at the time of the original Closing Date. There was no attempt to integrate or link either the agricultural or livestock services with forestry services.' 3.26 The performance of AAs was less favorable in HFPP than in AERP. The 1988 comparative study by APROSC included the HFPP district of Syangja and also the Hill district of Kavre as a district under the conventional Nepalese extension system (Table 1). The results of this study had some interesting The important role of private and community forests in farming systems in the Hill zone did not receive significant attention in the SAR nor during the project. - 20 - observations on the performance of the AAs in these environments, as well as on other characteristics of the systems: (a) Although 88% of contact farmers and about 68% of non-contact farmers in the HFPP district claimed they knew the identity of the AA, more farmers of both classes reported having been contacted by JT/JTA than by the AA (68% and 45% respectively). A similar situation was reported by farmers in the traditional district (31% and 9% having been in contact some time with JTA and AA respectively). This greater frequancy of visits by JT/JTAs than by AAs is the reverse of the SAR intention and of the situation reported in the Terai (Annex 2)26. The very low allowance given to AAs in the traditional system (NRs 50/month) could not be expected to elicit much activity, but neither, apparently, did the NRs 150/month provided under the HFPP project.27 (b) Only 15% of the JT/JTAs maintained that the regular training they received was fully relevant to field conditions. (c) The technical knowledge of about 40% of the JT/JTAs in the HFFPP district needed improvement (in accordance with a technology questionnaire), and this increased to about 70% in the case of the JT/JTAs in the conventional system in the Hills. The JT/JTAs considered over half of the AAs as being less than adequate to do their job properly in both HFPP and traditional districts. (d) The surveyed contact farmers in HFPP indicated the JTA was the primary source of technological information, followed by "other farmers" and, lastly, by the AA. Except for the smallest farmer (who acknowledged other farmers as the primary source), the non- contact farmers in the HFPP district indicated similarly. In the conventional district, the "other farmer" category was dominant as the source of technology, followed by the JTA. (e) Most JT/JTAs were satisfied with the nature of their work but objected to salary levels, working conditions, promotional prospects 26 Percentage of Surveyed Farmers who had been visited (at any time) by: JT/JTA PLAA/ T&V Terai (AERP) 45 84 T&V Hills (HFPP) 68 45 Conventional Hills 31 9 27 In practice, those AAs recruited in mid-project to cover the expanded project area were only provided the NRs 50/mth allowance. - 21 - and job security. AAs complained about their allowance and poor logistical support in both the HFPP and conventional districts. 3.27 The survey results by APROSC strongly suggest that JTAs assumed a much greater responsibility than PLAA/AAs for the face-to-face extension contact with farmers, than in the Terai. Overall coverage of farmers by the service was lower in HFPP than in AERP districts, and the conventional Hill district was even lower than in HFPP. Training of JT/JTAs was deficient in HFPP and was worse in the conventional system, which must have affected the technical guidance given to AAs and the overall performance of the service. Research-extension Linkage 3.28 The reliance on the research system, and especially the Pundi Bhumdi research station, to provide applicable technology did not function as intended. Although meetings between researchers and SMS did occur, lack of experience and technical expertise on the part of most of the researchers and the SMS, combined with a lack of appreciation of the real problems faced by farmers, severely limited the generation of relevant recommendations. 3.29 The Lumle Agricultural Center (LAC), a British Government-financed research, extension and training institution" near Pokhara in Kaski district (which is contiguous with three of four HFPP districts) had evolved an effective technology generation and transfer process based on the identification of farmers' needs. A farming system approach to research and extension had been adopted in the mid-eighties to rectify some of the deficiencies of the more narrowly focused commodity- or discipline-based programs. A Memorandum of Understanding in 1987 between the RDA, the Director of LAC and the HFPP Coordinator provided for LAC to supply relevant research for the extension service in the project area. The involvement of LAC in the linkage with ADOs and SMS resulted in a gradual improvement in the training process, but the effectiveness of this training was still constrained by the limited experience and qualifications of the SMS and the low morale problem associated with job insecurity, low travel allowances,2 delayed budget and confiscation of transport by local authorities. These factors were not constraints in the externally-funded LAC extension system. 3.30 The HFPP employed similar on-farm trial and demonstration techniques to that used in AERP and incurred the same types of problems (para 3.10). The quality of this program also improved, howeve., with the involvement of LAC, and included an expanded coverage of horticultaral crops in addition to cereals and pulses. 28 Established in 1968 as a training center under the Ex-Gurka Soldiers Resettlement Scheme. 29 District staff (ADO, AADO, SMS) received NRs 35 as per diem (TS$.83 in 1991) for up to 90 days travel per year, and subsequently NRs 10. - 22 - Other Components 3.31 APROSC was used for project evaluation activities (as for AERP) but while some baseline surveys were done in 1982 and 1984, no mid-term or final impact study was undertaken. The relevant implementation data provided by the 1988 comparative study on extension systems have already been discussed (para 3.23). 3.32 External training under the UNDP project (para 2.7) was reasonably satisfactory and exceeded targets (Table 3). However, only six of the 25 DOA staff trained in the program were still employed by HFPP at project completion. Local staff training was constrained by budgetary constraints and low traveling allowances. 3.33 The international consultancy provision was not fully utilized (Table 3) and the consultants employed did not have the expected impact. This was likely associated with a lack of previous experience by some consultants with the peculiar conditions of the Hills, and deficiencies in institutional arrangements which limited their utilization. Local consultants were reported to have performed satisfactorily. 3.34 The fertilizer procurement was timely and facilitated distribution in the first two years of the project. Distribution did suffer, however, even in this period, due partly to delays in construction of stores at ASCs (para 3.16). Subsequently, the amount and timeliness of fertilizer supply were problematical. CASH CROP DEVELOPMENT PROJECT 3.35 The project was implemented within the intended five year period and closed on schedule. Cost savings of US$920,000 (12%) were enabled by a depreciation of the NRs against the US$, which also resulted in a cancellation of US$1.4 million (24%) of the Credit at project closure. 3.36 The sugarcane component absorbed most of the project costs (52%), followed by tobacco (20%) and oilseeds (16%). Short- and medium-term credit was the single most important expenditure item, assuming 40% of project costs, followed by civil works at 35%. 3.37 The implementation of individual components are described in the following paragraphs. Throughout the project, however, a persistent problem was a lack of adequate management and coordination. The Project Coordinating Committee and its two Sub-committees (para 2.12) met infrequently and did not demonstrate the required leadership. Three successive project coordinators were appointed, none of whom had adequate powers to influence project activities. - 23 - Sugarcane 3.38 The civil works investments in sugarcane feeder road upgrading (54 km) and new construction (6 km) were close to SAR targets, and improvements at the Jitpur Research Station and Simra Seed-cane Production Farm were made (Table 4). Incremental staff were recruited for Jitpur under the project, but at the end of the project only three of the seien professional positions at the station were filled. Seed-cane selection was carried out and about half of the target area was under new varieties at project closure. Unfortunately, however, selection was made for irrigated conditions so that these varieties could only realize their potential under irrigated high-input conditions, whereas less than 10% of the cane area was irrigated. 3.39 Prior to the project, the only extension service was provided by technical staff from BSF; this, however, was mainly concerned with arranging from credit and other inputs and for operation of the cane buying stations with little attention to technical recommendations. Although the project provided for the local DOA extension service (operating under the T&V system) to take over technical extension responsibilities to farmers growing cane, this did not occur formally until three years after project initiation. The DOA extension staff, who mostly dealt with food crops, did not really get involved with extension on sugarcane until the last year of the project when a specific training program was initiated. In practice, the BSF still maintains its specialized graduate and JT/JTA level staff, which are not closely linked with the DOA system. The APROSC ex-post evaluation study" reported that in 1987/88, 88% of far. re had been contacted by BSF staff but only 1.4% by DOA extension. In addition, the research-extension linkage was generally very weak, with little on-farm trial work under rainfed conditions. 3.40 The provision of specific project funds for credit increased the consumption of production credit (in cash and in kind) through ADBN with assistance from BSF staff. The APROSC study indicated a significant increase in fertilizer usage, especially but not only on the limited irrigated area, as well as increased use of agrochemicals. The rate of fertilizer application, however, was still only about half that recommended by DOA. Tobacco 3.41 The tobacco infrastructural and equipment investments at TDC, at the Belachapi Research Station and in extension offices were delayed but largely completed (Table 4). None of the 150 large-size private curing barns were reduced in size to be more fuel wood-efficient, and ADBN credit funds for this purpose remained undisbursed. 3o APROSC - January 1989. Ex-post Evaluation of Nepal Cash Crop Development Project. - 24 - 3.42 Tobacco research did not really commence until 1986/87 when improved varieties were introduced from India (flue-cured, sun-dried and Burley varieties) for testing under local conditions; here, again, the emphasis was on irrigated conditions while most tobacco is grown under rainfed conditions. The extension service was not transferred from TDC to DOA for the initial three years of the project. Subsequently, training of extension staff was weak and, although most positions were created, they were only temporary and resulted in many vacancies. The APROSC study (para 3.39) indicated only 8.5% of tobacco farmers were contacted by DOA extension workers while 94% were contacted by TDC staff. 3.43 Although the allocated production credit was fully utilized (para 3.48), this was on a very much reduced area of tobacco (para 4.18) suggesting diversion of some *inputs to other crops. The APROSC ex-post evaluation study suggested fertilizer rates at about 70% and 43% of the recommended levels for flue-cured and sun-dried tobacco respectively; usage was higher on the irrigated than on rainfed crops. Mustard 3.44 The oilseed agronomist was only utilized for 6 months instead of five years, but the short-term credit greatly exceeded targets (paras 3.48). The APROSC study recorded an increased use of fertilizer on mustard amongst sample farmers, but also reported a scarcity of chemical fertilizers in many crop seasons. The same study indicated 15% of farmers had been contacted at some time by DOA extension staff, indicating weak extension by DOA. Some research results were available, but there was limited field work in adaptive trials and demonstrations. Other ComDonents 3.45 Although some seed was imported for testing in the pilot development program for groundnuts, no active procurement program was pursued by the Vanaspati Ghee Industry (which was to purchase the entire production). The program was abandoned in 1986. 3.46 After some delay, improvements on the main sericulture station at Khopasi and at the sub-center at Walling (Syangja District) were completed, although the cocoa dryer was not procured for Walling (Table 4). Staff were recruited as planned and all except the staff training in India was implemented; only four staff went on an observation tour in India in the last year of the project. About 100 ha (compared to a targeted 150 ha) were planted to mulberry during the course of the project, but ADBN was reluctant to provide loans to farmers. Its staff had calculated that at the HMGN-fixed farmgate price of NRe 50/kg of dry cocoons (about half that of neighboring countries), silk cocoon production was not viable. - 25 - 3.47 The small additional staff support was provided for ginger research at Pokhara, but the station failed to identify improved varieties and practices. Five solar dryers were tested under field conditions but were not adopted by farmers due to lack of technical extension support and cost considerations. APROSC carried out a marketing study but no action was taken on its recommendations. Potential growers received little support from research and extension services; this, combined with a slump in fresh ginger prices led to decreasing interest by farmers and a fall in production. Credit 3.48 The provision of credit under the project enhanced the flow of funds to production of the main cash crops in the project - sugarcane, tobacco and mustard: Production Credit: CCDP Crop Appraisal Estimate Actual Overdue ---------------NRs million -------- % 1988 Sugarcane 14.54 19.49 26.5 Tobacco 9.79 14.94 41.8 Mustard 1.92 17.68 26.1 Source: PCR and ADBN It has already been pointed out (para 3.43) that it is likely that not all credit, either in kind or in cash, was used for the purposes for which it was intended; nevertheless, most is likely to have been used for agricultural production purposes, while there would have been some deviation to debt consolidation. Repayment performance was not good. The exceptionally poor performance in tobacco borrowing can be attributed to the deteriorating economic viability of tobacco production (para 4.19), and cumbersome loan recovery procedures through TDC/JCF. - 26 - IV. PROJECT OUTCOME AGRICULTURAL EXTENSION AND RESEARCH PROJECT Production Impact of Research and Extension 4.1 As with all projects dealing with development or improvement in public sector research and extension services to the agricultural sector, it is very difficult to reliably estimate the impact of this project. The outcome of the investment in services is best measured by the incremental production and income it produces. There is great difficulty, however, in attributing causality to these output measures. They are affected by many factors other than agricultural research and extension. Macroeconomic and sectoral policies, input/output price ratios, and availability of complementary services in input supply, farm credit and market outlets are some of the other factors affecting farmer decision making and, therefore, production and income. The SAR indicated that the Evaluation and Program Analysis Division (EPAD) of MFA would analyze the production and income effects of the investments; it specified a benchmark (baseline) study in the project districts and the annual collection of yield/production data for each district. Although APROSC did some baseline studies on behalf of the project, follow-up studies were not carried out to measure impact ir. these same districts. Even if these were done, the causality issue would still have remained because of the absence of a "without" treatment to complement the "before-after" measurement. 4.2 The DEMS in MFA did collect annual production data on the major crops in each district. Although the reliability of these statistics can be questioned due to the collection procedures used, they are available for all Terai districts and thus provide an opportunity for comparison between districts. As rice and wheat are the two main crops in the Terai, yield changes in these crops should be indicative of changes in farm productivity. 4.3 Subsequent to the IDA-financed Agricultural Extension Project II becoming operational in 1986 (para 3.19), all Terai districts have been subject to some degree of the T&V extension system. For comparative purposes, however, the Terai districts can be grouped into those which have had exposure to T&V from the early eighties (1982 and 1983) and the more recent group in which T&V has been developed in the late eighties. After eliminating the non-AERP districts which had previously been subject to significant T&V extension in association with irrigation development projects and those districts with a very high proportion of cultivation under irrigation, the PPAR selected five districts to represent the T&V system under AERP31 and four districts to represent those 31 Jhapa, Dhanuska, Mohottari, Sarlahi and Kapilbastu. - 27 - recently introduced to the system32. Irrigation averaged 53.9% of the cultivated area for the former and 52.12 for the latter group of districts. 4.4 Indices of the average yields of the two groups of districts from 1972/73 to 1989/90 are presented in Figures 1 and 2. An increasing trend in wheat yields is apparent from the latter half of the seventies irrespective of extension treatment, and there is no significant yield difference between the extension treatments. In the case of paddy, there is no significant difference in the yield trend between the two groups of districts up to the early eighties when AERP was commenced. From 1982/83 to 1989/90, however, the yield trend in the AERP districts is significantly greater than the other district33. If the response to the extension treatment is lagged by two years the significance increases.3' The data set to examine a lag effect of three years is limited; the result is positive but of weaker statistical significance. If a lagged response explains what is occurring, then the yield curve for the districts more recently introduced to T&V could be expected to align with the AERP districts in the early nineties. These results suggest that factors other than the intensive T&V service were largely instrumental in the changes in wheat productivity occurring in the Terai, but that the T&V services could have contributed to yield increases in rice. In turn, this interpretation raises the possibility that adoption of some other unmeasured improved technologies may have been accelerated by the AERP service. 4.5 A separation of the respective contributions of research and extension to any production impact was not possible. The initiation (rather than full adoption) of some improvements in research procedures were reported (para 3.12), but no direct linkages to impact were identifiable. The project likely did contribute to some extent to the continued production of crop varieties, which dominated other disciplines in the Nepalese research system. Five improved varieties of rice were released for the Terai and four for the Hills in the second half of the eighties, with origins in India, Sri Lanka, Indonesia, IRRI and Nepal itself; two improved varieties of wheat for the Terai and two for the Hills were also released in the same period, with origins in India and Mexico; four maize varieties of Nepalese origin were also released, two 32 Saptari, Siraha, Bardia and Kanchanpur. 33 R-squared 0.7296; Trend T-Stat. 4.028; 2-Tail Sig. 0.007. 34 R-squared 0.8368; Trend T-Stat. 4.529; 2-Tail Sig. 0.011. 35 Supervision missions in November 1988 and February 1989 examined paddy and wheat yield data on a group of NAEP and non-NAEP districts between 1972/73 and 1986/87. These indicated no differences in the trend for either crop. The three additional years of results used in the PPAR made a difference for paddy. - 28 - each for the Terai and Hills. The Terai also benefitted from the release of two mustard varieties of Indian origin, and soybeans and groundnuts of Taiwanese and USA origins. Awareness and Adoption of Technology 4.6 In assessing the impact of extension services, it can be assumed that if farmers are made aware of the benefits of improved technology and subsequently adopt appropriate technology, this would lead to increased productivity. The previously-cited 1988 report by APROSC compared the level of awareness and adoption of specific technologies amongst seven different extension approaches .n Nepal. These included T&V in the Terai and the conventional extension system. Unfortunstely, the district selected as the conventional system in the Terai turned out to have the T&V system already functional to some degree and was dropped from the survey and not replaced; this meant that the T&V-Tere .4ample could not be validly compared with the conventional system which was reprboented only by a Hill district. Nevertheless, the survey did provide some interesting data in five T&V districts, each with samples of 40 contact farmers and 40 non- contact farmers. 4.7 In the T&V-Terai districts both contact and non-contact farmers were very aware of recommended paddy and wheat varieties (81% to 95%), but were less aware of the recommended varieties for mpizeM (20-30%), which is of lesser importance in the Terai. The adoption" of recommended varieties was also high for both contact and non-contact farmers for paddy (73% to 88Z) and wheat (85% 36 The data for maize, however, in the APROSC report can be misleading, as the report appears to report on use of the technology by all farmers rather than reporting on only farmers who grow maize; this problem does not occur with paddy and wheat because virtually all farmers in the Terai districts grew these crops. 37 -------------% Adoption of Recommendations in Terai T&V Districts ----------- Farmer Type Contact Farmer Non-contact Farmers Farm Size S' M L AVE. S M L AVE. Rice: Rec. Variety 84 82 88 74 73 84 Use of Fertilizer 60 65 80 66 43 37 73 44 Plant Protection 41 49 41 44 20 29 30 26 Wheat: Rec. Variety 92 94 96 85 91 93 Use of Fertilizer 56 69 89 66 56 66 73 59 Seed Treatment 37 38 30 35 18 23 13 18 Seed Storage 21 16 18 18 11 13 7 10 *S, M and L indicate Small-, Medium-, and Large-scale farmers, respectively. - 29 - to 96%, and was correspondingly low for maize (17% to 342). Non-contact farmers had a lower percentage awareness and adoption than contact farmers in nearly all cases for the small, medium and large farm samples; this is to be expected as the more progressive farmers would tend to be selected as contact farmers. Although these data do not allow comparison with the traditional system, they do indicate that farmers in the T&V districts were aware of and adopted this most obvious and "easiest" of technologies for their main crops, paddy and wheat. 4.8 Apart from crop varietal recommendations, the study showed that two other technology areas dominated the subject matter in farmer visits - fertilization and plant protection. Less than two-thirds of farmers applied any inorganic fertilizer at all; more contact farmers than non-contact farmers applied fertilizer; and more larger farmers applied fertilizer than smaller farmers. The average amount of fertilizer (nitrogen and phosphorous)n was about half that recommended by the extension service and did not vary significantly amongst farm sizes nor between non-contact and contact farmers. This suggests that even if the T&V extension service did have any impact on convincing farmers to use inorganic fertilizer (which is not documented), it certainly had no effect on farmers' behavior in the amount used. 4.9 As crop protection was a prominent topic in the extension program, it can be reasonably assumed that at least all contact farmers would be made aware of ecommended practices. Despite this, adoption of plant protection measures on paddy was only about 44% for contact farmers and less at about 26% for non-contact farmers. Similarly, seed treatment for wheat was only 35% for contact and about half that for non-contact farmers. Adoption of seed storage protection measures with wheat was even lower for both groups. There was no consistent trend for differences in use of crop protection measures amongst small, medium and large farmers. As a technology group, the most notable features are the non-contact farmer adoption being only a little more than half that of contact farmers, and the relatively low adoption levels overall. The low adoption of crop plant protection measures suggests that the technology was not presented in a convincing manner by the extension service as an input which was very worthwhile taking into account benefit-cost-risk considerations; 3 the marked difference between contact and non-contact farmers, however, suggests that 38 About 45 kg N/ha for paddy and 52 kg N/ha for wheat; about 17 kg P/ha for paddy and 21 kg P/ha for wheat. 39 A low benefit-cost ratio would explain and justify a low adoption score, but, in that case, it should not have been a major recommended technology. In practice, the T&V extension in Nepal paid little attention to the economics of recommendations. Sometimes, the packaging of protectants can be excessively large and discourage their purchase by small producers. However, as there was no consistent trend for more of the larger farmers to use protectants, this would suggest that this was not a strong factor in non-adoption. - 30 - adoption of this type of technology needs ready access to appropriate technical information (which could be expected for a non-traditional technique in communities with a low education level). HILL FOOD PRODUCTION PROJECT Production Impact 4.10 Despite the problems associated with the small-scale irrigation investment (para 3.21) the benefits were considerable and, although reliable production records were not available, appeared to yield a substantial economic rate of return (in excess of 25%). This component, however, represented only 18.6% of project. cots. The economic response to the investment in trails and the meager investment in suspension bridges (5.8% of costs) could not be quantified but would very likely be substantial, considering the normal benefits obtained by improved access in this region. The crop responses to the fertilizer investment (15% of project costs) as an annual input for two years should have more than covered its costs even at economic prices (para 5.5), but this represented a short-term production investment rather than a long-term development investment. 4.11 About 60% of project costs were allocated to improved agricultural extension services as the major thrust of the project. This investment did not generate the intended production response and leads the PPAR to conclude that the overall results of the project were unsatisfactory. 4.12 The DFMS district agricultural statistics did indicate a substantial increase in the production of the main cereal crops - paddy, maize and millet - and also a very big increase in wheat. Production increases also occurred with potato, pulses and oil seeds. Yields of these crops, however, with the exception of potato and pulses, did not significantly increase. The incremental production was due to an increase in cropped area and increased cropping intensity allowed by an expansion in the irrigated area in the four districts from about 35,000 ha to 128,000 ha (with only 3,200 ha financed by the project). The February 1989 supervision mission compared the DFMS yield data for paddy, maize and wheat for the HFPP districts with four other Hill districts in the centralized Western Regions and found no difference in yield trends; this suggested the services offered by the project were not generating more favorable responses than services in the other districts. The HFPP services were unable to generate the increased yields in food crops which the SAR anticipated would be possible through the adoption of low-cost technologies and strategic use of fertilizers appropriate to the situation of mostly subsistence farmers. - 31 - Awareness and Adoption of Technology 4.13 The 1988 APROSC study, in addition to measuring implementation factors, (para 3.26) attempted to examine differences in technology awareness and adoption between contact and non-contact farmers in one HFPP district and farmers in one conventional extension district in the Hills. The sample was limited to only forty respondents in each of the three farmer groups. Observations on the response data' Suggest: (a) although the awareness of recommended varieties varied amongst major crops, those farmers who indicated they were aware apparently adopted the recommended varieties (the readiness to accept varietal recommendations was also apparent in the Terai districts - para 4.8); 40 ----------------------------Technology Avareness and Adoption in Hill Districts........................ DISTRIC- HPP ..TRADITIONAL EXTENSION ELM -118 Contact Non-Contact Aver.* Non-Contact A. AWARENESS of Recommended Variety for: Maize 77 51 53 58 Wheat 91 88 88 95 Paddy 33 24 25 55 B. ADOPTION of Technology: Maize - Variety 75 46 48 42 - plant protection 2 0 12 - seed storage 4 0 0 - use of fertilizer 18 10 11 9 Wheat - Variety 87 87 87 92 - plant protection 0 0 3 - seed storage 0 0 11 - use of fertilizer 55 24 26 23 Paddy - Variety 32 23 24 42 - plant protection 13 13 34 - use of fertilizer 30 13 14 25 C. AVERAGE USE OF NUTRIENTS N Kg/ha 26 28 P kg/ha 12 14 * Average obtained by discounting contact farmer response by a factor of 12 to reflect typical situation of one contact farmer to 12 non-contact farmers. - 32 - (b) apart -rom crop varieties, the adoption of technological recommendations was generally very low in all situations; (c) there was no consistent difference between adoption rates of recommended technologies in the HFPP and the traditional district; and (d) in the HFPP district, contact farmers had a rate of adoption for many technologies which was higher than for non-contact farmers. 4.14 Few firm conclusions can be drawn from these observations because of the limitations of the data, but there appears to be no appreciable difference between the district. These findings would be consistent with the lack of yield increases observed in the HFPP districts (para 4.12). More detailed data pertaining to farm size indicated a higher proportion of larger than smaller farmers used fertilizers in both Hill districts but that there was no significant difference between the subgroups in the amounts used per ha (as for the Terai districts - para 4.7). Another general observation from the APROSC study was that both the level of fertilizer used and the percentage of farmers using fertilizer were higher in the Terai districts than in the Hill districts (para 4.4). CASH CROP DEVELOPMENT PROJECT Impact 4.15 The contribution of each component is analyzed in the following paragraphs but, overall, the project did not have the expected impact on growth of the cash crop subsector. Tobacco and mustard yield increases did not occur, tobacco area substantially decreased, and an increase in sugarcane yields only raised the average to pre-project levels. The import substitution objectives of the project, therefore, could not be met. The pilot activities in groundnut and ginger were unsuccessful and the sericulture program had very limited results. Any improvements in research and extensicn services were very marginal and far below expectations. Most of the high proportion of project costs allocated to annual production credit (nearly 40%) should have been used profitably in cash or food crop production (as farmers could be expected to make rational use of this funding resource), with the exception of some of that credit used on tobacco production which experienced significantly decreasing margins. Greater use of production inputs in specific years, however, does not represent a "permanent" investment, and did not result in the primary project objective of increased - 33 - yield of cash crope. The completed building and transport infrastructure represent permanent investments which should yield some returns. Suearcane 4.16 Although the district yield averages for sugarcane increased in the second and third years of the project, the yield statistics of DFMS for the pre- project and the project period do not allow a conclusion that the project has had a significant impact on yield trends (Table 5); this, compared to a 91% increase projected at appraisal. The APROSC study indicated a yield increase in both the plant and ratoon crops on sampled farms between 1984/85 and 1987/88. This would be consistent with the increased use of inputs reported by the same farmers (para 3.40). It is probable, therefore, that the increased use of fertilizer allowed by improved credit availability contributed to a recovery to earlier yield levels which had substantially declined in 1983/84. 4.17 The gross margins for sugarcane production during the project (as analyzed in the APROSC ex-post evaluation study) were such that it could not comn3te with a paddy-paddy-wheat land use under irrigated conditions, and, hence the area under irrigation did not expand. Under rainfed conditions, the sugarcane gross margins were better than a typical alternative of maize-oilseeds, but production outlay was higher. Consequently, there was no special incentive to continue to invest in sugarcane other than to have a guaranteed market and some preferred access to inputs.41 The latter could possibly explain the large drawdown in production input credit but the use of only about half the recommended fertilizer application rate; some inputs accessed under the sugar credit were likely diverted to other crops. Any incremental yield attributable to a change to improved cane varieties (which was limited because of its selection for irrigated conditions - para 3.38) could be considered as a medium- term investment, but production responses in the agricultural system due to increased availability and consumption are dependent, inter alia, on the continuation of favorable institutional arrangements for access to fertilizer. Tobacco 4.18 The tobacco yield statistics of DFMS for the project districts (Table 5) did not ind-icate a project impact on yields, and the area decreased from 6,900 ha is 1982/83 to 4,200 ha in 1987/88 (rather than increase by 1,000 ha as planned). The APROSC study also indicated a significant decrease in yields in rainfed tobacco between 1984/85 and 1987/88, as well as a major decrease in 41 After project closure, sugarcane prices offered by BSF increased in 1988/89 and 1989/90 which appeared to be reflected in an area expansion and in higher yields which could have been associated with higher input use. - 34 - area of tobacco planted. The proportion of local tobacco used by JCF declined from 30% to 16%, so that the import substitution objective was not met. 4.19 The prices offered by TDC as the sole purchaser of tobacco were flat from 1983/84 to 1985/86 and then increased only marginally. Differentiation between low and high grades was also inadequate, encouraging the cheaper production of lower grades. Gross margin analysis by APROSC and a comparative analysis by a 1987 supervision mission confirmed that tobacco could not compete with alternative crops (usually wheat or oilseeds), which would largely explain the demise of the tobacco crop. Mustard 4.20 National agricultural statistics for the project districts (Table 5) show that no improvement occurred in mustard yields. A major increase in production occurred, however, through a large expansion in the area planted to mustard. This was directly due to substantial price increases for the crop during the project period, which led to a very competitive gross margin against alternative crops (largely maize and wheat); it also had the advantage of a relatively low production cost outlay. Other Components 4.21 The groundnut pilot was a failure (para 3.45); the ginger component had negligible development impact (para 3.49); and the sericulture program seriously under-achieved its objectives. Although 100 ha of mulberries were planted (para 3.46), high plant mortality almost halved this area by 1987/88 and very little more was subsequently planted. The poor results in tree cultivation can be attributed to inappropriate planting material (from South India), and improper cultural method which probably reflect deficiencies in training and technical assistance. Low silk cocoon yields were associated with inadequate sanitation in feeding and cleaning, which led to disease in the silkworms; this also could reflect poor training and extension. Err-tic supply of silkworm eggs reduced efficiency. These factors, combined with the poor financial viability of even the production model under reasonable efficiency coefficients, resulted in the poor performance of the component.4 42 In 1989/90 and 1990/91 TDC offered substantially higher prices and increased the percentage price differential for high and low grades for sun-dried tobacco. This led to some increase in both area and yields in those years. 43 In 1990, grade differentials were introduced and cocoon prices raised (NRs 70 for low-quality, and NRs 100 for high-quality) from the low NRs 50/kg; this may be sufficient to stimulate moze interest in sericulture, but supporting infrastructural and services would need improvement if an efficient industry is to develop. - 35 - V. FINDINGS AND ISSUES Changes in Cropping Pattern and Productivity 5.1 The scarcity of evaluation data limits the conclusions which can be drawn from this group of three projects which attempted to provide improved services to farmers involved in food or food and cash crop production in a significant number of districts of Nepal. With few exceptions, however, the agricultural production response of farmers in the project districts appeared to be very similar to that of non-project farmers in the same production zones in Nepal: (a) The total area for each of the major crops expanded over the decade of the eighties, except for tobacco which had very unfavorable profit margins. This is in response to a high rate of population growth and increased demands for production of food and cash income. A diversity of crops has been maintained by farmers to make use of seasonal, soil type, topographical and moisture supply variations in accordance with market, profit and risk considerations. The responsiveness of farmers to economic margins is amply demonstrated by the decline in the tobacco crop because of reduced comparative profitability, and the increase in the area of mustard because of its economic benefits. At the end of the eighties, a more favorable price for sugar and expanded private sector processing led to a significant area expansion in this crop in the Terai, and there has even been a response in tobacco production in the last two years due to a more favorable pricing policy. There was also a considerable expansion of high-value fruit and vegetable crop where the availability of planting material, production conditions and accessible markets permitted sufficient financial rewards. The Nepalese farmer demonstrated his responsiveness to economic incentives in choosing crops and cropping patterns, and there is no evidence that the projects had any positive effect on altering decisions in these areas. In any case, there was negligible attention to economic considerations in research or extension programs sponsored under the projects. (b) Crop yields in the project areas reflected the general pattern in the Terai and Hills in the eighties. There was an upward trend in yields of the major food crops - paddy, wheat and maize - in the Terai, but little change in the Hills. Cash crop yields (largely in the Terai) were variable during the decade but did not exhibit an upward trend; after 1987/88, however, sugarcane started to indicate significant yield increases due to substantially improved economic benefits to farmers. The project districts did not demonstrate any yield advantages, apart from some evidence that rice yields may have been increasing at a faster rate than in non-AERP Terai districts in the latter part of the decade (para 4.4). - 36 - Changes in Production Factors 5.2 The natural production environment places limits on crop productivity. The potential productivity can be changed by productive infrastructure such as in irrigation facilities which modify the natural moisture regime. The small irrigation component in HFPP did make a contribution to such change, but the main thrust of the projects was in services to farmers to convince them to use technologies to increase the agricultural productivity of their existing production environments. Key items in this improvement were to be improved varieties and inorganic fertilizer, and appropriate husbandry measures to maximize tb2 potential of the varieties and higher levels of nutrition - plant population and spacing, plant protection, weed control, crop storage and, where appropriate, water management. 5.3 Improved Varieties: As wheat expansion in Nepal started with the introduction of exotic high-yielding cultivars, this crop has been mostly under improvad varieties (>80%) for the last 15 years, and the project has had little effect on this situation in either the Terai or Hill zone. Use of improved paddy varieties has increased nationally (to about 40% adoption) but more so in the Terai than in the Hills; while there are no data available to indicate AERP resulted in a higher adoption of improved paddy varieties, the improved coverage enabled by the service (para 3.9) and the concentration on varietal demonstrations (para 3.10) may have been instrumental in an enhanced adoption which could be associated with the higher rice yields suggested in para 4.4. Adoption of improved maize varieties (nationally, less than 30%) is less in the Terai than in the Hills where it is the primary crop in cropping systems. Hard evidence was not available to indicate any advantages of the HFPP services for this measure. The CCDP did make a contribution to the planting of improved sugarcane varieties (para 3.38) which would have contributed to the increases being obtained recently (para 5.1). The project did not significantly contribute to the production or adoption of improved varieties for other cash crops. 5.4 Inorganic Fertilizer. Fertilizer consumption steadily increased over the decade at an average of 12% per annum, reaching an average consumption level of 48.5 kg per cropped ha in 1989/90, which is low relative to most countries in the region. Consumption, however, was not uniform, with the Terai (with 78% of irrigation) and the Kathmandu Valley having more than twice and ten times the application rate per cropped area respectively of the Hills44. Another interesting phenomenon is that 58% of the 1989/90 sales were in the Central Region (which includes the Kathmandu Valley) which has the most developed road system. Consumption of Fertilizer % Kz/ha cropped Terai 64 60.3 Kathmandu Valley 11 334.0* Accessible Hills 23 28.8 Remote Hills/Mountains 3 15.0 TOTAL 100 48.0 *This efgure say be aggerated, as some private retailers to the Valley sell outside the Valley Source: Nepal Fertilizer Sector Management Study, January 1991. - 37 - 5.5 Inorganic fertilizer sales have been heavily subsidized in Nepal (para 1.14). Studies undertaken in the FAO/AIC Fertilizer and Related Inputs Program (FRIP) indicated very favorable financial returns to fertilizer application at the subsidized price." A recent study by Crown Agents on the fertilizer sector" confirmed the high returns to fertilizer application; it also found that with removal of subsidies the existing conservative rates adopted by those farmers using inorganic fertilizer in the Hills (about 65 kg/ha) would still be very attractive, but Kathmandu Valley farmers would have to reduce application rates (currently around 300 kg/ha) to maintain an attractive benefit:cost (B/C) ratio. The study concluded that national consumption has been largely determined by the supply distribution constraint rather than demand. Even amongst adopting farmers, however, the same study indicated that constraints to higher (recommended) usage included (a) suspicion about the long-term effects of inorganic fertilizer on the soil (a knowledge factor), (b) lack of irrigation to "guarantee" results (a risk factor), and (c) difficulties in accessing credit (a service factor). A further variable factor, especially in the Hills, was the usage of compost, which farmers considered (correctly) reduced the inorganic fertilizer requirements. 5.6 The APROSC studies undertaken in AERP and HFPP generally confirmed observations in other studies that (i) more wheat was fertilized than any other single crop, and (ii) more compost was used in the Hills than in the Terai, with upland maize being the main recipient of compost (although most crops in all regions also commonly received a little compost). The wheat preference is likely explained by the fact that the crop has always been based on HYV and is also a second crop for which benefits are not shared with the land owner in tenancy arrangements (so providing better B/C ratios). Maize has traditionally demonstrated good response to compost, and is commonly an upland crop which is grown on terraces in the Hills which are more accessible to the compost prepared near dwellings (where animals are housed). 5.7 An effective extension system could be expected to *-Iprove the knowledge of farmers concerning (i) long-term effects of inorganic fertilizer on the soil, (ii) the responses which could be expected to different application rates under specific conditions of moisture availability, soil conditions, temperature regimes and levels of compost applied to tu- crop and previous crop; and (iii) the economic benefits at the different resk,nse levels. Extension should also be able to liaise with complementary service institutions, at least at the district level, to assist them in improving their delivery systems to 45 FAO, Kathmandu 1991. Economic Use of Fertilizers on Rice, Wheat, Maize and Potato (4 reports cn FRIP Project). 46 Nepal Fertilizer Sector Management Study, January 1991. Crown Agents Economic Consultancy Services in association with Devtec Nepal Pvt. Ltd., for HMGN, UNDP and ADB. - 38 - farmers (credit by ADBN, and fertilizer supply by AIC, cooperatives and private dealer organizations). 5.8 Considerable data were generated by the FAO-FRIP project from 1983 to 1989 in Hill districts (which included the HFPP districts - para 5.5), and other fertilizer treatments were included in the on-farm research program and in some of the varietal demonstrations funded under the projects. Little discussion and interpretation of the results of those trials and demonstrations occurred, however, so that only general fertilizer recommendations were promulgated by the extension service. The economic relationship between composting, green manuring and inorganic fertilization (especially in the Hill zone where high transport costs dictate maximization of the use of locally produced or grown organic forms of nutrients), received scant attention by either research or extension. 5.9 There was no real evidence to suggest that AERP and HFPP had a positive influence on increasing the number of farmers using fertilizer or the amounts used per crop over and above the normal trend in the Terai and Hill districts. The CCDP project is likely to have had at least a temporary impact on greater fertilizer use, not through the provision of more appropriate research and extension services, but through the provision of the complementary production credit facilitated through the project (para 4.16). 5.10 Other Production Factors: There was no data to examine any incremental impact of the AERP and HFPP services on such factors as adoption of recommended plant populations and time of planting. The evidence on use of plant protection measures was that it was generally low in the Terai and extremely low in the Hills; comparative data on AERP and non-AERP Terai districts were lacking, but limited data on HFPP and other Hill districts showed no difference in adoption of plant protection measures. The significant difference between contact and non-contact farmers in AERP with respect to plant and seed protection supports a contention that farmers do need technical assistance to be able to use this non-traditional technology; consequently, the reputed increased farmer contact in AERP may have had advantages in making at least the contact farmers aware of application details and the benefits of adopting technology in this area (paras 3.8 and 4.9). Institutional Performance 5.11 Each of the projects incurred problems in infrastructure and equipment procurement components. Of most concern, however, is the extent to which sustainable institutional improvements were made in the research and extension services to the agricultural sector in the project areas . 5.12 Despite many operational problems, the HMGN apparently remained generally supportive of agricultural research and of the T&V approach to extension throughout the project implementation period. Delays in release of budget, confiscation of vehicles for non-project use, unwillingness to create permanent positions for many project staff, the delays in transferring extension responsibilities for sugarcane and tobacco to DOA, and poor salaries and - 39 - travelling allowances were symptomatic of generic problems and inefficiencies in the government administration rather than being specific to the projects. 5.13 HMGN and the management of the projects were obviously concerned to implement the methodology of the T&V system of extension. Targets in staff numbers at practically all levels of the hierarchy were met in accordance with theoretical requirements to provide an intensive programmed service. Visiting schedules were drawn up and visit coverage to contact farmers was reasonably good in the Terai, even if not on the intended very regular basis. In the Hills, scheduling of visits did not attain the same degree of development, but the major thrust was to have some sort of programmed extension visits, despite the difficult communication/access characteristics of the Hills. Regular fortnightly (or monthly, in some cases) training sessions of JT/JTA and PLAA/AA were carried out by SMS. Very large numbers of demonstrations and mini kits were laid out. Training sessions were held with research staff (but more regularly in the Terai than in the Hills for most of the project period). As has been demonstrated in other countries in the region, however, preoccupation with staff-client ratios and the mechanics of the T&V model is insufficient to develop an effective extension service. 5.14 In essence, despite reasonably good coverage of farmers under AERP in the Terai and enhanced coverage in HFPP, much of the technology recommended by the extension service was not adopted at all or was adopted in a substantially different way to that recommended (e.g., fertilizers). Although 100% adoption is not expected for any technology, Section 4 of the PPAR concludes that, apart from use of improved varieties, adoption of recommendations was generally low, and there was negligible evidence to indicate that the extension services in the projects had positive effects on adoption with the possible exception of some technologies (particularly those which impacted on paddy) in the Terai. As farmers are generally responsive to opportunities to increase production and profit after considering any risks involved (para 5.1), it must be assumed that many of the recommended technologies were not perceived by farmers as appropriate to their circumstances. 5.15 The recommended technical messages of the extension service were mostly of a general district-wide (or more extensive) nature and tended to be directed at higher-potential production conditions which would justify high use of inputs. In addition, the messages were delivered by agents who were generally not adequately trained in the responses which could be expected to a particular technology under different condition. When it is considered (a) that only 45% of the cultivated land in the Terai and only about 20% in the Hills have any form of irrigation, (b) that production environments are particularly diverse, especially in the Hills, (c) that most farmers are at or near subsistence production levels, and (d) only about 602 of farmers have access to credit (with a bias to larger farmers) and only one-fifth of these use institutional credit (para 1.12), it is obvious that recommendations have to be adjusted to the particular resource circumstances if they are to be seen as relevant. 5.16 The T&V model is intended to have an efficient feed-back system from the front-line extension workers to the applied researchers and formulators of - 40 - recommendations so that relevant technologies are proposed. Unfortunately, a largely top-down procedure for developing recommendations evolved, with a high cost in lose of relevance and responsiveness to the circumstances of farmers in the widely-varied production environments." Similarly, the large numbers of demonstrations were not designed to maximize their impact and cater for the agroecological and socioeconomic production environments (pares 3.10 and 3.30); they were geared to better resource circumstances, represented general recommendations, and did not adequately consider personal influence factors in the selection of cooperators. Negligible attention was paid to farm economics in promoting technologies.8 5.17 The traditional practice of a centralized bureaucracy in setting implementation targets in the form of number of meetings, hectares of specific crops, demonstration plots, etc. acts, unfortunately, against the development of a truly responsive service. The T&V system in Nepal, despite its objectives of meeting farmers' needs, was affected by these traditional practices. Apart from the general bureaucratic deficiency and a preoccupation with the mechanics rather the quality of the methodology, a major factor directly influencing the reduced relevance of recommendations was the inadequacy of in-service training. 5.18 The training deficiency can be linked to the low quality of the SMS staff and the low effectiveness of the linkage with the research system. SMS generally were not specialized or experienced in a particular area and were usually young graduates appointed to fill a SMS post regardless of background or experience (paras 3.6, 3.7 and 3.24). The ROOs at RRS assigned to liaise with the extension staff had little experience and were temporary contract appointments; nor were they usually fully integrated into the research units at the stations. The research-extension meetings tended to be a forum for discussion of research results and planning trials on farmers' fields rather than as a truly interactive process for discussing field conditions and adapting recommendations accordingly, and for influencing the applied research program. 5.19 In the latter two years of HFPP, the involvement of the Lumle Agricultural Center (para 3.29) resulted from a recognition of the relative 47 The preoccupation with numbers and organization rather than quality and performance was also exemplified in the decision in HFPP in 1983 to cover all Panchayats in the four districts; this led to more staff appointments to cover all areas "on paper", but with an expanded farmer-extensionist ratio and no increase in supporting SMS. 48 Knowledge of farm economic data on fertilizer applications originating from the PAO-FRIP project (para 5.5 and F.N. #45) appeared to be confined to some experts in the Soils Division of the research complex, and the data were virtually unknown outside this enclave. These series of trials over five years covered paddy, wheat, maize and potato in 14 to 22 Hill districts, depending on the crop; unfortunately, in the case of paddy and wheat, these trials and demonstrations mainly related to irrigated conditions. - 41 - ineffectiveness of the project research and research/extension linkage arrangements. This led to some improvement in the training process and an improving relevance in recommendations, but the weakness of the SMS cadre, low morale, and the deficiencies in the ASC concept (para 3.24) limited the results which could be obtained. 5.20 Despite the very important role women played in decision-making on the farm and in performing agricultural functions, negligible attention was paid to this group, either as sole farm decision-makers or joint decision-makers with spouses. The virtual absence of women in the staffing of the extension service in district offices, in the contracting of PLAA/AA, and in the selection of contact farmers reflected the male bias. Potential cultural problems in acceptance of female extension staff and of men dealing with women farmers would have had to be carefully analyzed and catered for in the Terai; surveys carried out in three Hill districts, however, indicated that male farmers were strongly in favor of more attention to the role of women as farmers, and in one district male farmers expressed a preference for female extension agents. The same Hill farmers also indicated that extension representatives should be experienced and knowledgeable about farming, be mature and have a reasonable education, implying that they considered women could meet these criteria. 5.21 In Nepalese farming systems, livestock make demands on the farm labor resource, especially in collection of fodder and bedding material, but can make significant contributions to cash income, organic manuring of cropland as manure and compost, and draft power. Baseline surveys by APROSC50 demonstrated the 49 SERED Report #15, May 1988. "Women Participation in Agricultural Extension Activities. A Case Study of Myagdi, Baglung and Parbat Districts." NARSC. The male farmers indicated a preference for female extension representatives at 25%, 61% and 21% respectively in the three districts. 50 - APROSC, December 1984. Nepal. The Hill Food Production Project. Household Baseline Survey - APROSC, June 1986. Household Baseline Survey. Agricultural Extension Project (Bardiya, Kailali and Kanchanpur Districts) ZONE-LIVESTOCK FARM SIZE VS* S M L TERAI - Buffalo Female n.a. 0.7 0.9 1.0 - Cow n.a 1.9 2.5 5.0 - Bullock n.a. 2.0 3.6 6.1 HILLS - Buffalo Female 2.1 2.1 2.8 4.2 - Cow 1.5 1.6 2.4 4.1 - Bullock 1.0 1.3 1.4 2.1 * VS - very small; S - small; M - medium; L - large farm. - 42 - importance of the livestock resource in both the Hills and Terai. The Hills had more female buffaloes (for milk), while the Terai farms tended to have more bullocks (for plowing), with larger farms having more large ruminant livestock resources; three to five smallstock (sheep/goats) were also usually kept. In AERP, no attention was paid to the role of livestock in the SAR nor during project implementation, apart from recommendations on improved composting methods which were included in the extension messages. Coordination with the separate livestock extension service received negligible attention. In HFPP, the importance of livestock as an integral part of the farming system appeared to have been more appreciated and the separate livestock service was supported in the project design; during implementation, however, the livestock service operated in isolation of the cropping service (para. 3.25) 5.22 Forests in the vicinity of villages provided significant amounts of fodder (grass and tree foliage) and most of the bedding for livestock (as well as firewood, poles and timber for household requirements). The Forest Department which was responsible for supervising the use of the forest resources acted without any real interaction with either the crop or livestock services; nor was any effort apparently made at appraisal of HFPP or AERP to provide for any such linkage.51 This meant that the potential for agro-forestry to contribute to sustainable agriculture was ignored, despite an understanding that soil erosion was an increasing problem, especially in the Hills (para 1.6). Overall, the services provided in the three projects did not appear to pay any significant attention to environmental concerns. 5.23 The research system which received some support under AERP and CCDP continued to make a contribution to agricultural productivity (paras. 3.38 and 4.5), but fell far short of the projects' institutional development goals. The creation of the ROO positions and function was inadequate to develop a research orientation which gave priority to understanding the requirements of the common farming systems and to directing research to address the problems and 51 In the Lumle project (para. 3.29), three separate services were provided for crops/horticulture, livestock and forestry, but they were strongly integrated. 52 Integrated Pest Management (IPM) was not addressed as an issue in any of the projects. However, due to the diversity in cropping systcias, the scarcity or lack of knowledge of the dynamics of the pest-predator relationship under a range of production conditions, and the scarcity of trained and specialized staff, an IPM program would not have been a practical proposition in these projects. Lack of attention to the high and unbalanced levels of inorganic fertilizer application in the Kathmandu valley (which was not located in any of the three project areas) was also indicative of very little real action on environmental issues by the Nepalese agricultural services. - 43 - potentials of those systems. The procedure of having ROOs responsible for off- station research and most dealings with farmers and extension staff enabled researchers who were less interested in being responsive to farmers' needs to continue with their traditional behavior. The extent to which individual researchers were obliged to participate actively in field trials (with ROOs) and the respect, statue and support given to the ROOs varied amongst RRS and appeared to depend largely on the quality of leadership at each station. 5.24 In most cases, research continued to be largely directed at better resource environments (especially irrigated conditions) and assumed farmers' willingness and ability to provide high levels of production inputs. Varietal development and selection, although always very important, tended to excessively dominate both the on- and off-station research programs to the disadvantage of crop husbandry investigation. Farm constraints were not given adequate consideration and farm economics were given negligible attention. 5.25 USAID followed its ICP research project (para 2.4) with an Agricultural Research and Production Project (ARPP) which introduced two new divisions (Farming Systems Division and Socioeconomic Division), a National Agricultural Research Service Center (NARSC) at Khumaltar, and a rejuvenated Research Coordination Committee. The Farming System and the Socioeconomic Divisions introduced the concepts of having research more oriented towards farmers' socioeconomic conditions and of testing technologies in an integrated farm setting. The farming systems activities, however, have been limited to five Hill sites (one in each region) and had only weak linkages with research stations and the extension service. The Divisions were not fully integrated into the research system and appointments to the units were on a temporary contract basis. The Socioeconomic Division developed some very useful techniques of rapid rural survey using key informants. One technique, called "Samuhik Bhraman"53 used a multidisciplinary group visit of about five working days to interact with key informants and subsequently with the farmers in groups to discuss and further develop initial interpretations of socioeconomic and technical conditions in the survey site. This valuable experience was limited, however, by funding and per diem constraints and scarcity of trained personnel, and remained peripheral to the main research and extension functions. It should be noted that activities of both of these new Divisions were much more designed to assist the research complex in directing applied and adaptive research to meet farmers' needs rather than to assist in the extension process. The programs were not pervasive enough to have a substantial institutional impact on the RRS which were supported under the IDA credits, although all researchers were at least made aware of these concepts. ss Mathema, S.B. and Galt, D.L. (1989) Samuhik Bhraman: A Multidisciplinary Group Activity to Approach Farmers. SERED, Khumaltar, Nepal. - 44 - Sustainability 5.26 Systems for providing research and extension services are likely to be sustained if they can attract the continued support of budgetary authorities and if satisfactory professionalism and morale are maintained by staff in the services. 5.27 Difficulty was experienced in ascertaining the budget situation with respect to the extension system due to the way budget items are grouped. An analysis of budgets allocated for agriculture, livestock and horticulture services to each agricultural district for the years 1988/89 to 1991/92, however, can be assumed to include most extension allocations (Table 6).4 ThiL indicates that the total budget has declined slightly even in nominal terms, whereas the CPI has risen by approximately 48% in the same period. Budgets for regular extension activities declined significantly in nominal terms for the Terai districts as a whole, but increased for the other districts; this appears to be largely due to the emphasis given to horticulture in the Hill zone. The total district budget as a percentage of agricultural GDP (Table 7) has also declined from 0.78% in 1988/89 to 0.51% in 1990/91. 5.28 Budgets allocated to the Nepal Agricultural Research Council (NARC) over the last few years (Table 8), indicate a slight decline in nominal terms in the annual budgets and a substantial decline in real terms. As budgetary constraints are imposed in most national research systems, increasing proportions of the budget are typically allocated to salaries; this did not occur, however, in the case of Nepal. Researcher salaries in Nepal are very low by international standards and are low even by local regional standards. There was apparently some rationalization of staffing in 1990/91 which permitted a reduction in the total budget for salaries. The research budget as a percentage of agricultural GDP (Table 9) has declined in recent years. 5.29 Overall, it is apparent that the responsible officials in the Ministry of Finance appreciate the importance of the agricultural sector and of the potential benefits of research and extension services; considerable budgetary support has been maintained despite fiscal constraints. The amount in real terms, however, has been declining in recent years, as has its proportion of S4 It can be assumed that actual expenditures in all years were significantly below budget awards. This occurred in each year for the budgets allocated under AERP, and amounted to only 67% and 45% of sanctioned amounts for research an-4 extension respectively. This can be attributed to delayed release of funds, bureaucratic delays and management inefficiencies. It should also be noted that all the described budgetary allocations have occurred while Nepal has continued to enjoy very substantial financial assistance in the sector, not only from multilateral agencies such as the World Bank, ADB and UNDP, but also under major bilateral programs including from Japan, USA, Germany, Canada, UK, Switzerland and many other countries. Total commitments showed an upward trend during the eighties (reaching about US$300 million per year), with about 55 percent of this directed at agriculture, forestry and rural development. - 45 - agricultural GDP. This likely reflects some dissatisfaction by the central ministries with the output and effectiveness of the research and extension services. Ministry of Finance officials expressed concern to the PPAR mission about (a) the research system not being sufficiently responsive to the real needs of the majority of farmers, especially those with less resources; (b) poor linkages between researchers, extension agents and farmers; and (c) the high annual operational costs of the extension service and doubts about the appropriateness of the T&V methodology for many productions environments in Nepal. This implies that continued and expanded support for the research and extension systems will only be likely if the services are made more effective and if this effectiveness is clearly demonstrated to budgetary officials. 5.30 The APROSC surveys in the mid-eighties reported that most field level staff were relatively satisfied with the extension system but were dissatisfied with their conditions of service (paras 3.9 and 3.26). Queries have been raised, however, on many aspects of the extension methodology by DADOs and senior officials in the DOA. In 1989, the DOA assigned an expert group of its own staff to review extension and develop a sustainable and efficient approach to provide extension services. This recommended a number of important modifications, including greater emphasis on defining farmers' real technological needs, improving the competence of staff not only in technical but also in social and economic fields, enhanced technical support to field staff, a group approach rather than using individual contact farmers, and an increased role of the private sector in providing complementary services. The DOA leadership, however, did not fully agree with the proposed changes and the report was not acted on. Nevertheless, concern continued and ADO staff and other officials were encouraged to express their views on how to improve the extension system in a national symposium held in Kathmandu in September 1991. Significant changes were subsequently introduced and are elaborated below (paras 5.31-5.33). In essence, the extension methodology adopted under the IDA projects had a number of deficiencies which became apparent to the more observative and analytical extension professionals, so that the original version was not sustained. The AERP and HFPP projects did appear to raise the level of professional self-esteem amongst lower level staff. The concern for their conditions of employment, however, continued to detract from this positive element and has resulted in a relatively low level of morale. Substantial improvements in salaries are unlikely in the foreseeable future. Under the circumstances, it is obvious that professional satisfaction will have to play an important role and be further enhanced if improved performance of the service (which is necessary to attract further budgetary support) is to be obtained. 5.31 The research investments under AERP promoted the installation of ROOs in an attempt to improve the relevance of research and increase the support for extension. This concept was not fully developed under the project and would not likely be sustainable without improvements in its effectiveness and more integration into the research systems (para. 5.23). - 46 - Recent Institutional Developments 5.32 During the period of political uncertainly in 1990/91 after the King agreed to allow political parties (see F.N. #16) and before a new elected government was installed, the PLAA/AAs were organized to go on strike to seek classification as civil servants (rather than annual contract employees). In November 17, 1991 the newly elected government abolished all PLAA/AA posts. As these posts were a critical element of the T&V system supported not only under the three projects but also under the on-going Agricultural Extension Project II, this naturally disrupted field services" and obligated the development of alternative arrangements. A new model was discussed by senior DOA and MFA officers and regional and district staff and was finaliy approved on December 9, 1991. 5.33 The major modifications in the system were to discontinue the contact farmer and the PLAA/AA service concepts. These were replaced by JTAe assuming responsibility for direct communication with farmers who would be contacted through farmer groups. However, not all farmers would be in groups; about three groups were assumed to be developed in each Village Development Council (VDC - which replaced the Panchayat unit), each with about 30 members and with one group leader elected by the members who would came together voluntarily to form the group; the group leader would visit members twice per month (probably in subgroups) and would arrange for monthly meetings of all members in which the JTA would attend. JTAs and group leaders would be trained monthly, and it was "proposed" that NRa 50 be given to the group leaders to cover "expenses" at each monthly training session at the Agricultural Service Centers (9 per district). JTAs would spend about one-third of their time attending to non-group farmers. 5.34 The brief description of the new program provided to the Bank makes reference to the concept of assessing farmers real needs before organizing training sessions, and that the groups would be voluntary and, therefore, cohesive and able to work together. It is apparent from these changes that DOA does appreciate the advantages of working with groups and of the need to make technical recommendations more relevant. The PPAR considers, however, that much more analysis, planning and training will be needed if these concepts are to be successfully introduced and used to the best advantage in the extension service (paras. 5.37 to 5.45). Of special concern is the lack of definition of what constitutes a "group", of the means for its identification, and of the management of interaction between the JTA and group members. The new program also appears to have grossly underestimated the number of non-group members (as a residual in each VDC) in its numerical example. Finally, it is worth noting that the proposal appears to have accepted that annual operating costs will not be reduced; the previous AA allowances of NRs 50, NRs 150 or NRs 300/mth are replaced by a proposed training "allowance" per group leader of NRs 50 per session, which would be for three times the number of AAs or the equivalent of 35 During the 1990/91 period, the political/administrative uncertainty had also caused some breakdown in the operational routine of the T&V program where this had previously been functioning. - 47 - NRs 150 per month for the area previously covered by one AA; nor were chages in any other cost items proposed. 5.35 The agricultural research system has also recently undergone important institutional developments. In 1987, NARSC (para 5.25) was gi-en the status of an agency within MFA and became responsible for research activities previously carried out by both DOA and DLDAH (para. 1.10) covering agriculture and livestock.5 It was renamed National Agricultural Research Center and, subsequently, Council (NARC) by an Act in May 1991; in November 1991, the Act was supported by bylaws. This effectively created a structure in which all crop, horticulture, livestock and fish/aquaculture research and all related research stations and farms would be under the single NARC entity reporting to a National Agriculture Research Board chaired by the Secretary of MPA. 5.36 In 1987 and 1988, HMGN undertook the preparation of a Master Plan for the Forestry Sector. This resulted in the Forest Policy of April 1990 which departed from the old emphasis on industrial forestry and reforestation and indicated that all accessible Hill forest would be handed over the local communities for their (sustainable) management. Legislation .had provided fcr management of forest by village users since the Decentralization Act of 1982, but extremely little forest had been effectively transferred to user responsibility under community forestry schemes. The new policy emphasized the application of a "bottom-up" approach and appreciated that its widespread adoption would necessitate retraining of the entire staff of the Ministry of Forestry and Soil Conservation as advisors and extensionists; the traditional attitudes of the forestry profession as promoters of industrial timber forestry and guardians of the forest resources had been largely responsible for the slow progress in community foresty. In the late eighties, a number of externally a%,sisted projects had undartaken pilot schemes which resulted in the transfer of blocks of forest/scrub to user responsibility within an agreed management and forest use plan. A notable project was the Nepal-Australia Forestry Project" which has demonstrated how national staff can be trained to implement a "bottom-up" approach; existing groups and subgroups of users (farmers) are identified by staff who subsequently facilitate (not dictate) the development and implementation of an equitable and technically sound management plan by forest users. These principles have relevance to improvement in the agricultural extension service with which the forest service should closely coordinate, as the use of forest products is an irportant part of most farming systems." 36 Agroforestry research is a new discipline in Nepal and was initiated within the NARC structure in ARPP's hill districts. However, some agroforestry research is also carried out by the Forestry Research Division of the Department of Forestry and Plant Research of the Ministry of Forestry and Soil Conservation, and at the Institute of Agriculture and Animal Husbandry (IAAS) and Institute of Forestry. 37 Discussed by Gilmour, D.A. and Fisher, R.J., in "Villagers, Forests and Foresters" June 1991. Sahayogi Press, Kathmandu, Nepal. U Due to a wide interest by donors and the early promising (although variable) results with community forestry, a major Hill Community Forestry Program was adopted. This is supported by a number of donors in specific areas so that most of the Hill districts are covered; the Bank approved a project as part of this program in May, 1989 (Cr. 2028-NEP). - 48 - Concepts for Improved Institutional Performance 5.37 Responsiveness to farmers' Conditions. It is apparent that agricultural research and extension services in Nepal have major deficiencies. The experience from the three projects which are the subject of this PPAR, however, and other experience from inside Nepal and from other countries in similar circumstances, indicate areas in which improvements have to be and can be made. From a purely technical viewpoint, if the agricultural services are able to identify agroecological production zones, determine products and the associated technologies which will utilize the potential of these zones, and then make farmers aware of these technologies, it could be argued that the publicly-funded research and extension functions have been fulfilled. Such an approach, however, will have limited impact when dealing with a farming population with a generally low level of education and within which the majority have scarce resources and limited access to further resources. Such is the case in Nepal. The potential of any agroecological zone must be tempered by the resource circumstances of the households who derive a living from farming in the zone; technologies need to be developed which are relevant to these circumstances; and technologies and opportunities must be presented so that they are perceived as being relevant to farmers' situations and able to help them reach their farming objectives. If designed to address these issues, complementary services in research and extension can be effective in increasing productivity and farmer's welfare and make good use of public sector funds. The farming system work initiated under ARPP (para 5.25) provided a valuable introduction of these principles into the technology generation and transfer systems in Nepal. 5.38 Farmina System Perspective. Ideally, the farming system research would be based at each major research station and be greatly expanded as part of the NARC's outreach activities, having a major impact on both research and extension programs. This would allow a much greater coverage of production circumstances or "recommendation domains". NARC does intend to expand its farming system research along those lines, but financial and manpower constraints limit the number of sites at which the full treatment of farming situation aialysis, design of an improved farming system, field testing of comp3nent technology and of the new system and agroeconomic monitoring can be undertaken. These-constraints, however, need not apply to the widespread adoption of the farming situation analysis or diagnostic phase of the approach. This would allow the adoption of a farming system perspective (FSP) in the development of all applied technology and in the formulation of technical recommendations; it would provide the basis for a responsive approach by the services 5,. 5.39 Reference Groups. The Lumle Agricultural Center (para 3.29) has been prominent from the mid eighties in highlighting the relevance of socioeconomic differences within communities to the applicability of agricultural research and extension. It was evident from its farming system work that distinct subgroups with different technological requirements exited within communitiee and within so These are a number of overlapping functions in the Farming System and in the Socioeconomic Divisions of NaRC. The latter has a broader mandate in socioeconomic surveys and impact evaluation, but has played a critical role in the diagnostic phase of the farming system research. Should these units be combined or rearranged as NARC develops, the farming system diagnostic and associated training function should be preserved and receive priority. - 49 - general recommendation domains.60 Initially this was verified through detailed surveys which categorized socl )economic groups in a community and ascertained the reactions of individuals within each group to specific technologies. It should be noted that techniques such as Samuhik Bhraman (para 5.25) practiced by NARC's Socioeconomic Division were able to indicate the existence of groups of different socioeconomic circumstances in a community; the Lumle work took this process further by placing individual households in specific groups. Members in each group had a mix of resources and characteristics which generated similar requirements in technology and responsiveness to technological innovations. These groups can be classed as "reference groups"", with members identifying themselves as part of a group of households in similar circumstances. 5.40 Reference groups are not created; they normally exist in a community. The characteristics of a reference group make it a more predictable entity than the broader "recommendation domain" and make it the ideal medium to develop effective interaction between agricultural extension services and farmer clients. The new proposals for revamping the extension service (para.5.33 and 5.34) refer to the benefits of using groups ri the focus for extension contacts, but it is not apparent that the major benefits of using existing reference groups of farmers is appreciated. On the other hand, the user-group approach employed in the Nepal-Australia Forestry Project (para. 5.36) divides the forest users into distinct sub-groups according to their forest requirements and interests, and then ensures that each sub-group is catered for (See F.N. #57); this is a similar principle to the use of defined reference groups as extension contact points. 5.41 The use of detailed surveys to determine reference groups, however, is impractical as an operational tool due to the staff time and funds required. A much more rapid method of resource assessment and classification (which had been previously used in Africa) was adopted. It was a form of rapid rural appraisal which involved selected members of a local community identifying the relative wealth of households in their community (termed "wealth ranking"), and has now been successfully tested and used at Lumle.6 The technique can be carried out in six to eight hours of field work by a single trained operator and is briefly described in Annex 3. Other techniques have been used to rapidly 60 Bell, K.L. (1986). Farming Systems Research Strategy for the Hills of Nepal. Technical Paper 86/9. Lumle Agricultural Research Center Nepal. Factors which could be involved in an individual identifying himself with others in a similar situation include land ownership, type of land, amount of household labor, off-farm employment and its security, and non-farm income, as well as influences which impinge on socio-economic conditions such as caste or class. In sociological terms, members of a reference group are usually described as having similar values, attitudes and behavioral expectations, so that there is a significant degree of "reference" to each other. 62 Seeley, J. (1989). Wealth Ranking in Magar communities: examples from Luale Agricultural Center Extension Command Area. Working Paper 89/2. 0 Kiff, E. (1991). Effectiveness of Lumle's Extension Services in Reaching Both Women and Men and Farmers from All Resource Bases. (a) Bajung Survey. Technical Paper 8/91. (b) Durlung Surveys. Technical Paper 9/91. (c) Summary of Results of Three Villages. Seminar Paper 9/91. * Gradin, B.E. (1988). Wealth Ranking in Smallholder Communities: A Field Manual. Intermediate Technology Publications Ltd., London. - 50 - identify reference groups (e.g., in Thailand), and are usually similarly based on interviews of informed persons; the wealth ranking technique has already been demonstrated as applicable to communities of Nepal, at least in the Hill zone, which supports its more widespread use. 5.42 The technique would be very complementary to the farming system diagnostic work to be undertaken by MARC, involving both R00s and extension sta&f. If carried out prior to the major informed person surveys or Samuhik Bhraman interviews, it could improve the quality of data collection by selecting the key informants from each defined reference group. It would also provide the basic reference groups membership data for subsequent use by the extension staff. 5.43 The wealth ranking technique could also be used as a general tool throughout the extension service (i.e., without relying on formal farming system diagnostic surveys) to define existing reference groups as a basis for extension. This would be possible with some training,63 and would overcome the main concern expressed in para 5.34 in which the PPAR doubted that the planned change to group extension was designed to take full advantage of the potential for technology transfer using reference groups. 5.44 The benefits of using a reference group as the contact entity with extension staff are substantial: (a) Members are at ease in interacting with each other and are therefore willing to express their concerns and opinions more freely; this avoids the frequently observed dominance by traditional or official leaders of larger groups or communities (to the detriment of less assertive members and subgroups), and also encourages the participation of women as household heads or active members of the agricultural households; (b) As members are generally in similar socioeconomic circumstances, farming constraints and potentials are largely common to all members so that a particular mix of technologies is relevant to most group members; discussion of the applicability of a technology by the group, with intervention/demonstration by the extension agent as appropriate, allows any necessary adaptation and a common conclusion on its relevance; and (c) the group is a naturally occurring entity and tends to be cohesive, so that it can reliably enter into agreements to undertake specific activities which are mutually benefici., including commitments with their extension agent. 63 The wealth ranking done with the involvement of the Socioeconomic Discussion of NARC in the farming system diagnostic work could provide useful case study training material for the extension staff. This implies an expansion of the role of this unit in NARC in giving more attention to the extension aspects of its farming system work. - 51 - It may not be practical to have each defined reference group as a contact point for extension when there is a very small number of members in a group. More than one reference group could be combined into an extension group for practical reasons, as long as the existence and membership of each are acknowledged and catered for in discussions with the group and in the presentation of technical solutions and management of test plots, etc. 5.45 Problem Census. The potential of the reference group for effective technology transfer is maximized if interaction with the extension agent is initiated through a problem census (usually on an annual basis), and continued in an agreed action program for the group members and the agent which addresses the constraints as determined by the group in the problem census and in subsequent meetings. Successful ex erience and methodology in conducting problem census meetings are documented, but their introduction into Nepal may initially need external assistance in practical aspects of the technique. Essentially, the extension agent acts as a facilitator in the problem census procedure; the group splits into subgroups of five to ten persons, each of which discusses and then prioritizes problems; this is followed by further interaction between the small groups to arrive at a carefully considered analysis of the serious constraints they face in their farming system." The extension service subsequently (a) discusses and agrees with the farmers on the factors involved in their constraints, and on those measures which can alleviate them or which need to be further tested in the local situation, and (b) refers those aspects which are beyond its domain to other relevant agencies (and maximizes cooperation and follow-up with these agencies). This process is the basis for an agreed action plan and programmed visit/assistance schedule by the extension agent. It is a completely "farmer-centered" approach in that any technology or intervention is analyzed from a benefit-resource cost viewpoint, the amount of risk involved in using additional resources (especially in production response and market price), and how a technology can be interpreted as meeting the expressed concerns of the farmers in the group. The process is applicable to a programmed extension system as long as the required leadership and training are provided." The role of the problem census technique and its interaction with farming system diagnosis and the use of reference groups in the recommended 64 Crouch, B. (1984) Problem Census: Farmer-centered Problem Identification. Training for Agriculture and Rural Development, Rome, FAO. 65 Farmers are encouraged not to restrict their analysis to cropping issues but to cover any factor they consider to be limiting their productivity, and may include subsistence or cash cropping concerne, horticulture, livestock, use of forest resources, transport access, input supply or marketing problems. 66 See also: " OED PPAR Report #8808, June 1990. India: West Bengal Agricultural Extension and Research Project (and four other extension projects). Annex 4. o OED PPAR Report #10101, November 1991. Thailand National Extension II Project. Paras 4.4 and 4.5. - 52 - approach to the technology generation and transfer process are presented diagrammatically in Figure 3. 5.46 Research-extension Linkage. The FSP/reference group/problem census concepts together vould provide an excellent basis for closer interaction between research and extension services in each district. The research staff would become more aware of the technological and adaptive research requirements in a greater number of more specific recommendation domains; and the extension staff would have a reliable basis for defining actual farmer constraints in their interaction with researchers to facilitate adaptation of recommendations to the different circumstances of farmer reference groups within these domains. 5.47 Demonstrations. Large numbers of relatively uniform demonstration plots were utilized in the projects. Their effectiveness was limited, however, by not ensuring that each demonstration was designed to ensure that in matched the circumstances and requirements of the farmers in the area of the demonstration; they tended to be "top-down" demonstrations (paras 3.10 and 3.30). Having JTAs working directly with reference groups and agreeing a program of interaction with members (para 5.45) would facilitate (a) small demonstrations or test plots being designed to suit the local requirements and to address issues important to the farmers, (b) any demonstration becoming part of the action plan agreed with the group, and (c) the group deciding on where the plots should be (within any required technical guidelines given by the JTA), and assuming "ownership" of the plots, with the output being identified as "their results". This approach requires more flexibility and greater responsibility at the district level, but would yield a much higher return on the current considerable investment in on- farm demonstration. 5.48 Mass Media. The recommended concepts also enable the development of a much improved mass media program for extension. The two most important factors influencing the effectiveness of mass media are coverage of the clientele and the perception by the clientele of the material as relevant to their Qircumstances. Although improvement mass media communication was not addressed in any of the three projects (see F.N. #17), this offers a very cost-effective means of complementing the face-to-face extension services. The adoption of the farmer- centered emphasis in technology development and transfer and the programmed interaction with all types of farmers provide the information necessary to design mass media programs which are relevant to the situation of the clientele. Material does vary in its suitability for effective maas communication, but the potential impact of programs is increased as the media become more localized. Relevance is also increased by using mechanisms to enhance identification of the audience with the program, such as in the Nepalese radio program "the Old Lady and the JTA". Although the use of the recommended extension principles provides relevant material, farmer literacy and the coverage of households by radio are low in many areas of Nepal. Nevertheless, extension staff can be organized to collect this information for analysis and use by experts in charge of designing programs. The programs need not be carried out by the public sector, as the private sector and some NG0s are usually much more adept in this function. - 53 - Institutional Implications and Requirements 5.49 Extension. The process of revision of the extension service at end-1991 and in 1992 provides the opportunity to incorporate extension principles which can gradually improve the effectiveness of the service. The adoption of the FSP- reference group - problem census concepts is feasible, but will require a firm commitment by management to their implementation and a well conceived training and monitoring program to support implementation. 5.50 One possibility is to expand the role of the Socioeconomic Division of NARC into training of extension staff in these concepts which are, essentially, based on established sociological and economic principles. The unit should be initially supported, however, by experts with practical experience in this use of the concepts in extension programs. The adoption of the reference group/problem census system could only be implemented at the rate staff could be trained. The early examples of problem census with groups, and of the analysis with them of factors involved in constraints and possible solutions, would provide appropriate case study material for further JT and JTA training. The extension service would also have to appoint specialists who have received more intensive instruction to continue the training of extension staff in the methodologies and to monitor and provide assistance in their implementation; this would likely be at the level of district SMS. 5.51 Practical training of SMS in technical and economic aspects of production are of equal importance with the training referred to above; without competent SMS to provide the technical support to JTAs interacting with farmer reference groups, the effectiveness of any system will be seriously constrained. 5.52 The principle of programming extension visits would be maintained. The subject of each visit, however, ..ld not be determined on a zonal or even district basis. It would be planned in accordance with a program developed with the group for each season after considering which technologies or actions may be appropriate to resolve constraints. In practice, there would be common problems across smilar gfoups, so that the appropriate time for attention to specific problems or issues would be the same, even though the means of addressing the problem may be different in accordance with circumstances in a particular group. The monthly training sessions for JTAs would lose any flavor of top-down programming and would represent analysis and discussion of how available technologies can be adapted to meet the requirements of each group. 5.53 The monthly training at each ASC of group leader selected by each extension group (in addition to JTAs), as proposed by DOA, is an unknown quantity in that the extent to which such a representative will volunteer his or her time free-of-charge is questionable. Nevertheless, the advantage of having some form of elected group representative to assist the JTA in organizing meetings and coordinating programs is obvious and should be encouraged. 5.54 The use of the modified approach to demonstrations (para 5.47) would mean a change in DOA strategy and also training of district staff in the physical and financial management of the more responsive demonstration policy. - 54 - 5.55 The monitoring/supervision functions would be retained, but the most important criterion for effectiveness would be the extent to which group members indicated the extension service was servicing their requirements. JTAs would be made fully aware of this form of assessment, as would the group members. This would increase the accountability of the JTA to his farmer clients, and also of the higher levels in the district service whose assessment would depend on JTA performance. Monitoring would also cover the farmers' opinions of factors other than technology which are constraints to their improving their productivity, income and welfare. This not only contributes to an understanding of progress or lack of progress in technology adoption, but also reinforces the feedback regularly channelled through JTAs and the extension service to district and higher authorities responsible for complementary services (para 5.45). 5.56 The modifications recommended above make greater demands on the extension service in being expected to identify and respond to real requirements of groups in the farming community, rather than to work within the target-driven, top-down, less complicated system they have been used to. The modified monitoring system may provide an incentive for use of the improved practices and for improved performance, but the current relatively low morale and output are only likely to be improved if: (a) staff are properly trained in the new techniques so that they understand how they can substantially increase their own effectiveness; (b) the technical back-up that the lower-level staff receive is able to support them in the "responsive approach"; (c) the operational support for mid-level staff in per diems and transport facilities is adequate; and (d) promotion and higher-education training policies are instituted to make the extension service an attractive career. These measures would, in themselves, reflect a commitment by the DOA management to a much improved service with a more appropriate modus operandi. In view of the common perception that promotion and progression in the government bureaucracy has historically been more dependent on patronage and connections rather than ability, initiative and effectiveness, it is essential that extension staff perceive a change for the better under any extension strategy in the new era of a democratically-elected government. 5.57 The recommended approach to extension services could reduce costs by avoiding expenditures in the more intensive and costly services now provided under special schemes, but its main impact would be in making the services much - 55 - more cost effective.67 The adoption of an extension strategy based on agreed principles would allow adaptations to different zones where socio-economic and production conditions are markedly different. The specific strategy, however, should oblige external donors to fit in with it rather than to initiate other systems with their associated counterpart funding requirement, "draining" of scarce management expertise, and the usual creation of temporary positions with temporary personal benefits which detract from the development of a sound extension service with career prospects. Much greater collaboration between the services of agriculture, livestock and forestry would certainly be more cost- effective for HMGN overall, and could be organized to reduce costs. As the reference group deals with existing groups of farmers (rather than "crested" groups), the use of this concept and of the group problem census have application to all three services. Ideally, the annual problem census by a group should be supported jointly by the three services. This would require an explicit commitment at the most senior levels of MFA and Ministry of Forestry and Environment, as the managements of individual services are unlikely to want to release anp degree of autonomy (as was evident even within MFA in the use of ASCs in HFPP). Other obvious ways to reduce costs are to maximize the coo eration with and use of NGOs to assist in implementing the extension strategy, and to 67 It is also possible that staff numbers and associated costs could be reduced if a policy is adopted to provide services only to the extent that they are likely to be significantly beneficial to farmers, rather than to have a uniform service to all farmers in all areas. This certainly does not suggest restricting services to those with resources which will alluw them to easily respond to use of high levels of inputs. Rather, it implies that, after a problem censVa it may be obvious that what the extension service can offer does not justify a monthly visit regime, but perhaps less frequent but timely visits to help the group on specific items on which it is agreed they need assistance; in other situations it may be more appropriate to place more reliance on mass media. 68 It is unlikely that any institutional amalgamation of field extension services would be possible at this stage, but this would offer further cost savings. If such amalgamation should occur, it is probable that a separate animal health service should be maintained because of the nature of these activities, and also a special cadre of forest rangers or other specialized and trained personnel in each district to progressively cover communities (and their reference groups - para 5.36) in developing forest use plans. 9 The more participatory/responsive approach recommended in this PPAR is almost universally accepted by NGOs working in the agricultural sector. The use of NG0s to implement mass media programs based on relevant information generated by the research and extension services could be considered. - 56 - use private sector firms (especially those in agroprocessing and input supply industries) to provide detailed technical services to specialist crop producers. 5.58 The PPAR has emphasized the need for adoption of more demand-driven applied research which is reiponsive to the needs of farmers. Without appropriate technology being generated, the capacity of the extension service to assist farmers must be severely limited. The use of the recommended concepts in improving the research-extension linkage in bi-monthly and semestral meetings and collaboration in on-farm trials has been described in para 5.46; this is not enough, however, to ensure the required services. 5.59 Research. Recent reviews of the research system have made firm recommendations for improvement. The Bank, at the time of writing this PPAR, was considering support for a research project, and will undoubtedly address a whole range of issues designed to complement the activities of other donors in strengthening NARC. Based on the experience of the research components under the IDA projects, the PPAR considers that four aspects should receive special attention if a sustainable service in technology development is to be provided - the outreach program from research stations; the responsiveness of the discipline and commodity divisions; personnel policies and professional support; and measures to address the research funding constraints. 5.60 The outreach (off-station) adaptive research program should receive increased prominence in the research strategy. The current dichotomy between ROOs and "pure researchers" needs to bf 1roken down to have all reseaxchers on stations having greater participation in off-station activities. The Farming System and the Socioeconomic Divisions should assist each of the RRS in developing outreach activities to include farming system work (and aspecially diagnostics - para 5.39), and the overlapping responsibilities of these two divisions should be rationalized. Greater prominence needs to be given in this work to social-economic-risk considerations and to non-crop activities (livestock, forest use, and agroforestry), apart from the current emphasis on irtegrated cropping systems. 5.61 The discipline and commodity divisions should be more responsive to the farming system/recommendation domain data generated by the outreach program. Because of the impact of soil fertility on production levels, this especially applies to the soil science discipline; consideration must be given to not only inherent fertility but also the relationship of organic and inorganic fertilization, scarce financial resources to procure fertilizer, and the possible economic use of agroforestry and specific green manure crops (such as Sesbania rostrata) to contribute to nutrient supply. Recognition of the needs of real farming systems should lead to greater attention to crop performance in less favorable production environments and at lower input levels, to improved crop husbandry measures which are not cash-intensive, and to commodities which have not received adequate attention to date (especially in horticulture). 5.62 Personnel policies need to be substantially altered if staff quality and output are to be improved and sustained. Little investment has been made since the mid-eighties to upgrade the technical qualifications of professional - 57 - staff, so that about 60% of staff have only B.Sc degrees despite an average experience of seven years. Of equal concern is the high proportion of temporary posts (18% of professional technical posts), which has especially occurred in eternally-assisted projects such as AERP and ARPP; this has been detrimental to institutionalizing the new functions supported under these projects. The policy of a single, low entry level regardless of qualifications and experience inhibits recruitment of more qualified professionals, and promotion which is based on the usual public sector bureaucratic system discourages scientific innovation and output. The recent granting of a semi-autonomous status to NARC provides the opportunity to develop transparent promotion systems based on merit, and also to address to some degree the very unsatisfactory salary scales for researchers. 5.63 Despite the recommendations for NARC to be more demand-driven in its research program, to cover research areas which have received inadequate attention in the past, to expand the outreach program and to improve its salary scales, constraints on public sector funding are likely to continue. This means that NARC will have to (a) have a clearly defined research strategy based on assessed pr*.rities (which need not detract from the demand-driven principle); (b) rely on substantial contributions from international centers for research and research institutions for similar environments outside Nepal; (c) ensure efficient use of the national research resource by coordination of agricultural research amongst national institutions; and (d) develop a research evaluation capacity to be able to demonstrate the returns on its research to those responsible for funding decisions. 5.64 Environment. Resource degradation in Nepal is continuous and complex. A large part of erosion is due to geologic factors; however, the use of fragile soils for cropping, the use of sloping land without adequate soil conservation measures, the stress placed on forest resources by increasing population and associated livestock requirements, and the environmental neglect forcec by conditions of poverty are all factors which can be influenced by agricultural research and extension services. There have been a number of government initiatives which show concern for environmental degradation, such as in reforestation projects, but these have been piecemeal and there has not been a common strategy for resource management to affect all development programs. Environmental concerns appear to have had little priority in either agricultural research or extension, and it beholds these services to address this concern if the benefits to be derived from agricultural technology are to be used in sustainable systems which contribute to real economic development and rural household welfare. Bank Performance 5.65 Design. There were some weaknesses in the design of the projects, especially with respect to the assessment of risks. The T&V methodology was assumed to be able to provide improved extension services and result in faster adoption of improved technology and substantial yield increases. This was based on "encouraging results" obtained in irrigation development projects in the Terai. However, although yield increases had been recorded for the concerned - 58 - districts, no evaluation studies had been made to indicate the extent to which any incremental benefits were due to improved extension, or irrigation, or input supply or other factors. The possibility of the methodology not being as effective in production systems which were dominantly rainfed (due to a scarcity of relevant technology for the lower resource - higher risk conditions) was not considered. The SARa for AERP and CCDP stated that "suitable technology for different conditions will not be a constraint as there is a backlog of technology as yet to be adopted". This is a frequent T&V design error. Apparently, no attempt was made to ascertain why much of this available technology was not adopted. An obvious question would be the extent of its relevance to the production circumstances of farmers. The SAR for HFPP did not raise the availability of appropriate technology as a risk issue. 5.66 A further design weakness was the acceptance that the T&V methodology would be implemented by the DOA in the expanded area in a relatively ideal manner. This would provide effective feedback of farmers' constraints to those responsible for formulation of recommendations and to researchers in charge of the adaptive research program. A Nepalese public sector institution has traditionally been bound by hierarchies based on caste, position, education level and age factors, and is affected by other social customs which are r.ot conducive to frank expression of opinions and communication upwards in the bureaucracy. Under these circumstances, it is to be expected that a top-down communication system will prevail unless major measures are taken to counteract it; in practice, no special precautions were taken and the top-down system did predominate. 5.67 In HFPP, the SAR did appreciate that some adjustments to T&V would likely be necessary in the Hill zone, but did not pay enough attention to the process of generation of applicable technology (which was eventually addressed to some degree by the involvement of LAC in 1987-para 3.29). It expressed concern about extension staff being willing to live at ASCs, but did not address the more practical concern of the inefficiencies in having JTAs located centrally at the ASC9 which involved excessively long treks to their farmer clients. 5.68 The absence of any project components to enhance the impact of mass media in technology transfer probably reflected the common emphasis given to the T&V methodology in extension projects supported by the Bank at the time, without serious consideration of alternative or complementary activities for particular situations. 5.69 The CCDP suffered from a design problem which is inherent in many projects which are directed at specific annual crops in mixed crop production systems. The ultimate impleentor is the farmer who will make decisions on use of his resources for particular crops and on the level of inputs used on each crop in accordance with his assessmen, of benefits and costs; this is not always consistent with the project's objectives for specific crops. The project also was excessively complex and the projected crop yields over-optimistic. 5.70 Very little attention was paid to environmental concerns in any of the projects. Again, this was probably symptomatic of the Bank's approach to - 59 - development in this period, in that it assumed environmental issues would be dealt with properly without special emphasis. Although the Bank was concerned about the environmental impact of development in Nepal, there was no overall resource management strategy which might have influenced the emphasis given to this topic in projects such as those reviewed in this PPAR.70 5.71 Supervision. The supervision experience was mixed. Some missions on AERP and HFPP appeared to concentrate on implementation mechanics of the SAR's extension methodology, while others showed legitimate concern about the relevance of recommendations and the need to develop a more responsive approach. Towards the end of the projects, supervision missions were suggesting the introduction of reference groups and problem census techniques (as recommended in this PPAR) to address these concerns. 5.72 Successful Bank interventions in the latter part of HFPP resulted in the involvement of LAC which was very advantageous, but cooperation between the agricultural, livestock and forestry/soil conservation services was not given its deserved attention. Insufficient attention was given in AERP and HFPP to ensuring the completion of impact evaluation studies which could have provided more useful quantitative data to assist in improving the services. 5.73 The input of the projects into the research system was limited, so that they could not be expected to have had a major impact on the institutional aspects of that system (which evolved into NARC towards the end of the implementation period of the projects - para 5.35). However, supervision missions wera generally attentive to the main issue of research relevance raised in this PPAR (para 3.11-3.42). 5.74 In CCDP, the supervision missions attempted to address the many problematical issues as they arose. However, the continuous rating of the project from 12/85 to 6/88 as a project with major problems obviously implies that it should have been subject to a major review and redesign or other action in the third year of implementation. 7o OED, 1992. Natural Resource Management in Nepal: 25 Years of Experience. Report # 10330, February 11, 1992. Table I NEPAL: District-level Orianization and Operation of Asricultural Extension Systems TRAOfflONAL SYSTEM PLANNED AERP T&V SYSTEM (TERA PLANNED HFPP T&V SYSTEM (HILL ZONE) Oranization Unit Post No. Ratio Post No. Ratio Post No. Iatio. District DADO 1 1 DADO 1 1 DADO 1 1 AADO 1 DADO XI AAOO 1 0A00 X 1 AADO 1 DADO X I SMS 0-1 SMS 2 DADO X 2 SMS 3 DADO X 3 Subdistrict JT 5 DADO X 5 Jr 5 DADOX 5 ASC 5 DADO X 5 (4-7) JTA 14 JTX3 JTA 14 JT X 3 JT 5 ASCX1 JTA (AG) 10 ASC X 2 JTA (L'S) 5 ASCX1 Panchayl AA 40 JrA X 3 PLAA 56 JTA X 4 AA 50 (AG) JA (AG) X 5 (2590)1 (Part-time. (Ful-time, on (Past-time, on contracl) 25 (U) JTA (L'S) X 5 on cotract)' contract) - - - CF 4.000 PLAA X 70 Agr. Ext. Service: Y Farmer Groups 400 AAX8 CF 1.400 Group X 3.4 IAAX28 - Farmer 45.000 - Farmer 45,000 PLAA X 550 to 1,300 Farmer 20.000y AA X 400 CF X 1I CFX 14 Tra*ning No regular In-service training for a Fostnightly In-service training of JT. JrA and *Monthly training of Jr. JTA at district HQ. Jr. JTA and AAs PLAA by SMS Fortnightly training at ASC for AA by JUJTAISMS Minimal pre-service training for AAs 0 Pre-service PLAA training Periodic DADO/SMSfresearcher workshop rregular and infrequent visis to a Preseason DADOISMS/researcher workshop B1-monthly DADO/Sresearcher training session research stations by DADOIJT/JrA 0 Preseason JT. JTA and PLAA training sta session a 1-monthly DADO/SMS/researcher training sessions at RRS * JT/JTA conduct extension with help * PLAAs to visit all contact farmers at set time In * PLAAs to visit all contact farmers at set time in of AAs each fortnight and give recommendations in each fortnight and give recommendations in a No fixed farmer vislt program accordance with previous training session accordance with previous training ssion " Limited, part-time extension aciMVty JTA to assist PLAAs in extension and JrA to PLAAe In extension and expected of AA demonstrations, and visit each PLAA on a demonstrations, and visit each PLAA on a fixed o Neo ble supervision of JT/JTAs fixed day of each fortnight day of each fortnight or oor o d JT to supervise JTA and PLAA Jr t supervise JrA and A o Very poor nobilly and technical * DADO/AADO to supervise Jr and SMS DADO/ADO to supervise Jr and SMS AB MTM:1n the I rorte, Is a greater numbor ot Panchayats (Village Deveopment Committees) per district due to sand distances between hamlets and villages t households per Panchayats are therefore, ASC. Agricultural Serve C r lower In the Hil zone at abut 300-750, compared to 700-1.100 In Teral. CF: Contact Farmer At NRo saImth allowance. AADO* District Agricultural Development Officer R/ NR. 15/ith Initially (raised to NR. 3001mth In 195 for the Teral only). J- Junior Technician Y The parallel livestock service was to be coordinated with the agricutural extension service but the JrA Junior Technical Asistent arrangements and organization were not specific In the SAM PLAAAA* Panchayat Agricutural AnistanAgricutural Asist 21 The HFPP design did not attempt to cover all Panchgarts, but only about 52% which were realAstically accessible. ..: Sub5 h amar specialist - 62 - Table 2 Page 1 of 2 AERP - PERFORMANCE INDICATORS etan Uni M Revid Astu Teiget Turet AaNevmentU A. SatN nequirmenes 1. AgnUtal Extension .KA no. 600 674 " 674 -JTA no. 126 186 i§ -.J no. 43 74 74 • M8 nlo. 10 "1.• I1 • AADO no. t. 10 •ADO no. 0 n.r. 6 •TO no. 2 n.. 2 •PC no. "ir• • ut1 ~taN no. ..132 2. Avr~tu Rmach Cim l sta no. 4 n.r. 4 Cua n no. 13 n.. 13 Cla~s in st1rt no. 09 ni. 04 JT 4Tainn) no. 4 np. 4 JT {~ s no. 7 .• 7 JTA no. 7 fi. 7 RO0 no. 4 ni. 4 k~par StMI no. •n.i. 20 3. Agrural r ing Avriadnra Wngin no. 1 1 Ansiat agil. nmkum no. 1 2 2 Oveser no. 9 ni. I. cv W~d 1. Trining F~ities (Extemimn) Hl builig 2 f. Faming unit 2 ni. 2 E~ri utst aupphy unit 2 n.. 2 2. Hmun (£Ntenn JTA unit 126 66 3911 JT utm 36 32 32 AADO unit 1 0 0 ADO unit 6 4 4 oms unt 17 17 S rt staff unit 24 23 23 3.O.h.ontere ' elt 0 1 4.Lb is 0s1sr0i Off~m ~oty " nit 1 -. leidaorry iunit 2 r. 2 nes hewe unit 3 2 2 Pum he no. 4 rt. TrFs dhd no. n.. M TmNn. 7 ni. 7 gme no. 2 nM. 2 Denir no. 1 0 0 。―〕…， 州、’呈鑒，豐寫認豐0，露呈一“-.，豐．00娌煙： B，久留徑 暹彥”&&&&&&&&&&&&&&i邑 iI 。透＇廈 廈要藝藝 呈居。，::,:,：賽，&，·：：二。．發姿露雙要盒認 法一”!11{ 〕【〕．-&! 鑒豐寫煙要常憂 ,,,j,，。。。。。。，1,,,,,,,,I〕｝l!l}.1 涌薔豐藝棗呈廈1曆 屆，藝賽豐‘整t,, !11!I!!i!}: 。｝:;:!：〕｝-!!!!!!！〔｛!{}!l!!!，。｝〕｝-----!―〔―〕 ‘蕣’&j삐 i&I&l &i&&&&!&&&&!&&&&&&&i&&&&&&l‘啣 & ,,,,,，· ,,,, ,,,m, ,,,,,,,,，··············…… [ . In· · Ch , & . 1 ，「 &-”·4 && &’紗··’ &&”·h&--&&&&-&.--”。”--．섶‘ l & : & 喚『 & & ‘· 『『『” ·…‘·。’。 辦；’『’ －··…’－뻐－·’뢴 蕣 ··… 。，。： :.:.．엄： .&.－·낙．&.，版；..．·．&...--．。．-,, ]; Table 3 - 65 Page 2 of 2 §mo unit no. ni. L ComMumv 1. E~nem e 4) Awammy mim 44 24 26 bl E~cn trad a "VM 20 a CO~c~ 63 Irsoffimn wwr~ mun 20 20 Told M/M 84 60 49 2. Lmd cw~ 9) Momw« w4 eva~ mim ni. 7 bl A~ mim 20 la cl tu~ coffamunk~ "VM 14 3 T~ mim 6 40 29 Tt~ . et sk Ac~ d en~ fft/M 4 ni. 2516 bl PA~ May t~ M/M ni. a d At~ ý~ M/M ni. 0 41 A9~ mim ni. 41b plann~~*~ 2. ~ Tt~ 41) ADsm nthn 110 ni. njL Ale ~ wim 35 ni. n4. 6~ etO M/M ni. M^ d) 8~ ffa~ gr~ na. n^ el A@~ ~on DADOffi~ffims mim 29 ni. n^ SM* MIdm 22 ni. njL -JT mim 10 ni. n^ JWJTA fft/M 36 OJ. n44 Tt~ v~ nö. 13 ni. 0 U~Må Exter~ . JTMTA mim in ni. n^ .A^ nvm 50 ni. rh& - Trå~ vkics ni. MOTMOTO, 11 am~ Md ~ mucecd ta bo @ca d by Wd c# pr~ v ni. - n*~ to~ w At am~ 4 wo w~ ~ C~d w~« w~ ha or~ te 4 d~ ud 194 p~M 0~ 9~ km i pii*tm pom a~ m 24* P~= k~ in 9~ 4 d~ 9*~ m~ #ha t9 ~~ d k~ ~~ frun or~ m~inb~c~ h~ el^ ~ OMSIO-NEP. 6 Twe mm ~^ tw w~ 6~~ hd u~ 4~ ta 0 -1 m cm ~ cwm~ w~ pr~ er-, $~1p- 7# Inr~ et om ~ ad am w*~ twe 4.ND dm~ c& 0~ - 0 k~ u~ UNW udv~ 4~ w~ w mim - man~ ta 12 mamw~ en ~ u~ m~m~ Db marm~de en w~ 4 numn- mede m 4~ OM Oredu~ Pk~ Source: RM PCR - 66 - Table 4 Page 1 of 2 CCDP - Performance Indicators Pg Un SAR AUGGI Target Achivmet A. SUARCANE DELVLOPMNdnT CMVI WORKS Road D~vicrem EaM flin & gavelng of .ining roex a km. 59 55 Conusuonof new rods km. 5 6 Jitpur Romas Station unk 3 3 Subject MauerSpii. unit å Seed Producio Frm. Sim Fencing of Sued Production Arm a 4,600 4,600 Improving of Irrigation Syom una i VERMCES Pick.up Van no. Motortycie no. 1 2 RECRUlTMENT Jitpur Resmarch Station Stff saffly&sM 15 15 Subject Mansr S*cialist 5saNy.auu 5 5 TECHNICAL PERPORMANCE nwn 72,700 n/a B. TOBACCO DEVE~ MEN CMVI WORKS RodPinP% unt l Redyng Pan Staf uni 8 k Resu~rekSte uni* 7 10 Exana staff Uni 2 2 EOUIP9MEN AND FUNIIUR Redry.Ag Pha (g r no. 2 1 Raehå R*~gggg (kbgiuip~(li Enainn (office aquipam& &anine) s 2 1 VEMICLES Minibu no. i i Pickup Van Moomycis no. 4 1 RECRUrrMEN ai Re~dying Pa Staf afffyear 15 15 Upgding TDC StaN mNyma 20 20 " a Raa~e SWiceStaff mawy«ar 90 90 Subject Mair Sesii m~a&y.ar 10 10 Tob pafo~yaO r 5 Cureno. 90 20 Prnuw d 1,600 negtive a - no availabf at Data nol avoilabia for areaa undermsgancan. See Table 5 -67.- Table 4 Page 2 of 2 Unt SAR Auag Targ Ach C. SERICLTURE- DEVLOPMENT CIVL WORKS Kpas. Sriculu Station building 10 9 Waing Sub-cenar building 3 4 EOIPMENT AND FURNfrUR Kp 5 s' Waing Subcmffr set 13 12 VEHICLES Pick-up Van 4WD no. 1 Mogmcycis no. 1 2 Station a~ycar 25 22 Walling Sub-emr ffryar 20 :0 Tminin Fannar Traiing peson 0 467 Staff In-mvice Training prsons 65 Staff Traing india peron 10 4 Y TECHNICAL PERFRMANCE New Mulb~rry Area Doveloped ha. 150 55 Cocons Sold by Farmrs m.o 115 10 D. OILS D DEE N RECUff MENT Oil~d Agvon i sa yar 5 0.5 TECHNICAL PERFORMANCE Ru nd m~san Incremal Poducion m.cn 960 9.600 1' hn~as in Ama ha. 350 aM inrte~l Production m. 245 n/a E. GINGER DEVELOPMENT EQUEPMENT AND FURNMrRE Dr Uni no. 5 12 RECRLUffMENT Giqge Azoon s =aWysar 5 3 1' Source: CCDP PCML S' 1Ma cocoo dtryr for the Syangja di~uit was no pmourd. Itk was pland that 10 ofileers would undergo ightmonths of fblopwhip training. Du. to adminimatv. problems, only four ofEcor wern on a brif obrvion ur. 9 Du. to ama inas which canna be enried to the projct. o Tha gingur agronomic who was working uader the provisinoof CCDP rfeg~ed in un 19g6, bu an Atinart Vegeable Developan Ollcer was eo-mui limitd asiss work. - 68 - TABLE 5 Crop Yields (mt/ha) in CCDP Districts 1980/81 1981/82 1982/83 1983/84 1984/85 1985/86 1986/87 1987/88 Sugarcane 25.5 28.0 27.6 - with project 22.0 23.8 31.3 27.0 26.2 Tobacco 0.59 0.64 0.66 - with project 0.67 0.87 0.49 0.63 0.53 Oilseeds 0.62 0.78 0.78 - with project 0.dl 0.69 0.51 0.53 0.65 Source: PCR and DFMS - 69 - TABLE 6 NEPAL: Budgets to District Agricultural Offices!/ 1988/89 - 1991/92 (NRs '000) 1988/89 1989/90 1990/91 1991/92 Terai Districts: 1. Regular Extension Programs 109,578 103,142 83,371 79,265 2. Development Projectsl 11,387 14,971 14,471 18,633 Hills/Mountain Districts: 1. Regular Extension Programs 129,543 136,811 123,443 139,248 2. Development Projects 76,472 77,545 67,190 76,391 Totals: Terai and Hills Mountain Extension Programs 239,121 239,953 206,814 218,513 Development Projects 87,859 92,516 81,661 95,024 Grand Total 326,980 332,469 288,475 313,537 Includes agricultural, livestock and horticultural services. Includes extension services as well as other development expenditures. TABLE 7 NEPAL: Airicultural District Budizets and Agricultural GDP 1988/89 - 1990/91 (NRs million) 1988/89 1989/90 1990/91 1. Total District Budgets 327 332 288 2. Agricultural GDP 42,148 50,032 56,346 Ratio 1/2 (M) 0.78 0.66 0.51 Source: District Budgets Supplied by Ministry of Agricilture. - 70 - TABLE 8 NEPAL: Agricultural Research Budgets 1988/89 - 1991/92 (IRe '000) Administration Salaries Operations & Maintenance Investment Total 1/(1+2+3) (1) (2) (3) (4) (5) 1988/89 37,748 31,517 20,806 29,930 120,003 0.42 1989/90 48,442 36,250 20,921 21,000 126,613 0.46 1990/91 36,813 39,142 20,392 12,316 108,663 0.38 1991/92 36,714 39,032 13,260 25,489 114,495 0.32 TABLE 9 NEPAL: Agricultural Research Budgets and Agricultural GDP 1988/89 - 1990/91 (NRs 'million) 1988/89 1989/90 1990191 1. Total Agricultural Research Budget 120 126 108 2. Agricultural GDP 42,148 50,032 56,346 Ratio 1/2 (%) 0.29 0.25 0.19 Source: NARC. WHEAT YIELD INDICES: 1972/73-1989/90 5 AERP & 4 OTHER DISTRICTS 1.5 1.4 1.3 1.2 1 0.9 0.8 0.7 07 72/73I 74/75 75/77 78/79 80/81 82/83 84/85 85/87 a889 ' 73/741 75/76 77/78 79/80 81/82 83/84 85/86 87/88 89/190 YFARM D AERP Districts + Other Districts Index 1.0 = Yield average for 1977/78 to 1979/80 RICE YIELD INDICES: 1972/73-1989/90 5 AERP & 4 OTHER DISTRICTS 1.5 1.4 1.3 1.2 a 1.1 z 1- 0.9 0.8 0.7 72/73 775 777779 8 182 83 84 85 887 89 73/74 75/76 77/78 79/80 81/82 83/84 85/86 87/88 89/9G0 YIEAR to 0 AERP Districts + Other Districts Index 1.0 = Yield average for 1977/78 to 1979/80 ReePonsive Atricultural Research - Extension - Farmer Interactions FARMERS National Secroral RESEARCH Development Agricultural Policies Research Strategy Identification by the Extension Farminx System Diainosis Service of Subgroups vIth X sindlar circumstances (L-ference groups) in a community Discipline Identifi a of Constraints and Research Coumodity Adaptive Research Factors Research - for Recomendation Affecting Situation Further Lbcal testit e of Technology Testing of Improved Farming System Applicable ( Adfnce of Measures TecInolodies to overcome constraints and Realize Potentials n h Increased Production and FaFmer Welfare - 75 - ANNEX 1 Page 1 of 3 Description of the T&V Extension Service Designed for AERP - The Plan (Largely Extracted from the SAR) 1. The project would provide farmers on a regular and systematic basis with demonstrations and up-to-date advice on farming practices best suited to their specific conditions and capable of having an immediate impact on production and income. This would be achieved by programmed visits from regularly trained field staff, supported by professional advice from within the Department of Agriculture. 2. At the village level, the Panchayat-Level' Agricultural Assistant (PLAA) would be the extension worker and would have completed about eight years of schooling and come from a farming background. PLAAs would be residents of the Panchayat in which they would work and would be mainly selected from among existing Agricultural Assistants (AA)2 who have the requisite background and qualification. Unlike AAs, they would be full-time extension workers and would earn Re. 150 pm compared with Rs. 50 pm earned by AAs. In the project districts, there are 375 Panchayats and about 363,000 farm families and the number of farm families per Panchayat ranges between about 530 and 1,330. Initially one PLAA would be postwd to each Panchayat, but in some Panchayats where population is sparse or in irrigated areas, more than one PLAA would be required. In the first case because of long distances to be travelled and because of the need for more intensive coverage in the second case. 3. The Extension Methodology. Under the proposed agricultural extension system, an intensive scheduled program of farm visits would be introduced. Each ward would be visited by the PLAA, on a fixed day of the week, once every two weeks. Thus, the PLAA would cover five wards in one week and four wards in the next week. In some Panchayats, where additional PLAAs are provided, the number of farm families covered by one PLAA would not exceed 700. During daily visits, PLAA would spend the full day in the scheduled ward, would visit fields during morning hours, and in the afternoon, either hold meetings in the village at a pre-determined time or hold discussions with farmers or meet farmers in their homes. One of the three remaining working days each fortnight would be devoted to in-service training, where PLAA would discuss and learn recommended practices to be given to farmers during the following fortnight and also bring farmer's problems to the attention of trainers. Thus, every two weeks the PLAA would spend nine days in visiting each of his nine wards and one day in training. During the two remaining days, PLAA would visit field trials, arrange special I The Panchayat system had both political and development connotations, and had village, district and national levels. A village Panchayat comprised nine administrative wards, usually grouped into two or three villages. 2 AAs were selected farmers who were given minimal training and were expected to give technical advice to other farmers in the vicinity of their homes on a part-time basis. ANNEX 1 - 76 - Page 2 of 3 extension activities, make up visits missed because of illness or holidays, and do occasional but limited office work. All PLAAs would maintain a chart showing days for visiting each ward as well as time and place of group meetings. 4. It would not be possible for the PLAA to reach all farmers in the wards during field visits. Thus, under the guidance of supervisors and in consultation with the Panchayat and farmers themselves, PLAAwoul1 select contact farmArs from each ward. Contact farmers could be from all levels of village society, and would be selected for their potential influence and willingness to collaborate with the extension service in following recommendations. PLAAs, during their field visits, would make special efforts to reach at least all contact farmers' fields, and they, in turn, would share the responsibility of organizing and attending afternoon meetings or any other village group activity. Within their area of operation, PLAAs would concentrate on the most important crops, focussing on practices which bring the best economic results. 5. Organization and Technical Support. PLAAs would receive technical support from Junior Technical Assistants (JTAs) and one JTA would guide and supervise the work of 3-5 PLAAs working in three Panchayats. One Junior Technician (JT) would guide and supervise the work of three JTAs. Each JTA would spend at least nine days in the fortnight visiting PLAAs in the field. He would also help the PLAA in carrying out demonstrations and simple trials. He would attend meetings organize by PLAAs and meet each of his PLAAs on a fixed day of the fortnight. This close ongoing relationship between PLAA and JTA would relieve the former of the need for formal reporting. PLAA and JTA would, however, maintain a daily diary of activities. The JT would supervise the work of PLAAs and JTAs and ensure effectiveness of extension operations. He would spend at least four days in the week meeting his PLAA and JTA in the field. During his visits, he would be able to find out whether visits are being made as planned, inquire from farmers about visit days in their villages, messages learned, etc. 6. At the district level, there would be an Agricultural Development Officer (ADO) to supervise all :xtension activities. To free the ADO from routine activities and to allow frequent field visits so necessary to review extension operations, one Assistant Agricultural Development Officer (AADO) would be provided. He would take over responsibility for reporting, attending coordination meetings at district level, and input supply arrangements. 7. To ensure a continuous flow of timely, locally-proven and adapted farm recommendations, and adequate back-stopping and training of PLAA, JTA and JT, it would be necessary to strengthen the technical support staff at district Level. Two Subject Matter Specialists (SMS) would be provided in each district (one each for Agronomy and Crop Protection). SMS would spend one-third of their time in training groups of PLAA, JTA and JT, one-third of their time in field visits providing technical support and guidance to PLAA and JTA, and one-third of their time in building their own stock of knowledge through a regular dialogue with research workers, carrying out simple field trials in farmers' fields, and attending short training couias. 8. At the national level, project activities would be coordinated by a full time Project Coordinator. ANNEX 1 - 77 Page 3 of 3 9. Training. In each project district, a number of extension sub-centers would be eatablished and which would be the headquarters of the JT. Each sub- center would cover about nine Panchayats and would be used for fortnightly training of extension staff. There would be 43 sub-centers in the project diotricts. 10. The project would strengthen training programs for extension workers and other administrative staff. The keynote would be the continuous fortnightly training of PLAA, JTA and JT by SMS wto are conversant with the latest research results, active in the identification of local problems, and who participate in local verification trials. Training would be practical and concentrate on the few specific recommendations relevant to farming operations in the coming fortnight. Other in-service training courses Aor PLAA, JTA and JT would include a pre-season workshop/training session for each major season and one or two special short courses per year on a specific technical topic. 11. Also before each season there would be workshop at the Regionai Research Station attended by ADO, SMS, Research Outreach Officer (ROO) and research workers. During the season, research workers and extension staff would review the previous season's research results, specify recommendations and plan the coming season's research and field trial programs. 12. Improving the knowledge of SMS is crucial to the success of the project. For this, there would be a training session held once in two months at the Regional Research Station of each region. Ths training would be attended by all SMS in the region, ADO, ROO and other research staff. The main objective would be to review results achieved, modify recommendations as necessary to suit prevailing field conditions, review findings and progress of field trials and develop suggestions for future research topics and field trials. - 78 - ANNE 2 Page 1 of 2 Implementation of the T&V Methodology in the Terai - APROSC Reports 1. The T&V methodology of fortnightly visits to CFs was reasonably carried out by PLAAs. This was evidenced in two monitoring studies contracted by the AERP Unit in 1984 and 1988. 1 In the earlier 1984 study, the following observations were recorded: Responses in the Eiaht Proiect Districts by Contact Farmers by Non-Contact Farmers " Knowledge by farmer of 742 to 972 602 to 892 the PLAA 882 average 732 average " Visit by PLAA at 742 to 982 362 to 902 least once in last month " Regularity of visit 212 to 812 202 to 832 by PLAA on a specific day 502 average 402 average o Non-contact farmer knowledge - 252 to 912 of contact farmer 522 average 2 of sample with holdings 22 33 below 1.0 ha X of holdings in Terai 58 districts below 1.0 ha2 2 of sample farmers with 42 35 >502 of holding irrigated 2. Those results indicate some variation amongst districts in the operational mechanics of the programmed visit system, but the coverage was acceptable even though the regularity of visits was low. The non-contact farmer knowledge of the contact farmer to whom he/she pertains was surprisingly low in some districts. The data suggest some bias in the selection of contact farmers towards larger farmers and a very small bias towards those with more irrigation. (The APROSC survey also appears to have been skewed away from the smallest landholders in both contact and non-contact farmer samples). About half of the non-contact farmers and 80% of the contact farmers considered the PLAA APROSC, July, 1984. Monitoring Survey of the Visit of Panchayat Level Agricultural Assistants: AERP. APROSC, May, 1988. Suitability of Training and Visit System of Agricultural Extension Approach in Nepal (included 5 Terai districts with T&V, four of which were in NAEP). I IDS. The Land Tenure System of Nepal 1986. - 79- ANNEX 2 Page 2 of 2 advice useful; the scope for this response, however, was toi wide to be a real indicator of the potential impact of the advice. 3. The later report (APROSC, 1988) also examined the extension awareness and visit frequency amongst farmers. The survey revealed a high awareness amongst farmers of the PLAA (90% and 802 amongst contact and non-contact farmers respectively) and a relatively high awareness of the JTA responsible for their area (632 and 512 respectively). A visit frequency of at least twice in the previous two months by the PLAA was reported by 902 of contact farmers and 572 of non-contact farmers; the corresponding figures for JTA visits were 37% and 20% respectively. 4. The 1988 study reported the opinions of the DADO/AAD0s concerning the relevance of various groups of technical recommendations. They all considered varietal and seeding rate recommendations were relevant (apart from fruit varieties - only 402 considered these relevant); 702 considered fertilizer recommendations were relevant, and 80% considered plant protection measures relevant (again, apart from fruit culture - only 252 considered these relevant). This indicates a concern about the relevance of advice in some aspects of technology even at the most senior levels in each district. The SMS were reported by both JT/JTAs and PLAAs as their major source of technical information, as was intended in the project. In a test on knowledge of a specific set of technical recommendations, 68% of the JT/JTAs were 100% correct, while 94Z were at least 752 correct; the corresponding figures for PLAAs were 39% and 792 respectively. The JT/JTAs rated about one third of PLAAs as not being fully capable with respect to carrying out the visit schedule, disseminating messages or assisting in demonstrations. 5. Most PLAAs indicated they were satisfied with their jobs except that their allowance was far too low. JT/JTAs mostly reported satisfaction with the nature of their work, but were dissatisfiad with their salary, facilities, promotional prospects, security of employment and departmental policies. This suggests a general satisfaction with the T&V methodology but not with employment conditions. Responses on desirable frequency of farm visits also give some support to this observation: 562 of JT/JTA considered weekly or fortnightly farm visits were most desirable (31% favoring "as necessary"), while 68% of PLAA favored weekly or fortnightly visits (only 13% favoring "as necessary"). - 80 - ANNEX 3 Page 1 of 2 The Wealth Ranking Technique and its Use in Extension 1. Wealth ranking is a rapid rural appraisal technique which involves members of a local community identifying the relative wealth of households in their own community. It relies on the detailed local knowledge of selected informants in the community. In Nepal's Hill zone, community members expressed enthusiasm for the concept and readily accepted that farmers of different "wealth" have different problems and needs which affect the relevance of specific technologies and the capacity of an individual to adopt technologies. One distinct advantage of the technique is that individual households retain privacy over details of their situation, while their resource base which is common knowledge to community members is disclosed. 2. The technique is based on a card-sorting method, in which the name of each household head in the community being studied is written on a small card and key informants are asked separately to place the cards in piles according to the wealth of each household. Procedures can be summarized as follows: (a) The surveyor elucidates the purpose of the survey with village leaders and interested community members. (b) Key informants are selected based on their local knowledge and on representation of a cross section of the community. There is usually a tendency for informants to have more dettiled knowledge of the resource base for those households which have resources not too dissimilar to their own, with grouping together of households with much greater or lesser resources than themselves; hence the necessity to have the informants from high, medium and low resource groups. (d) The size of the community being examined must be no larger than a realistic "knowledge area" for the informants. In the Hill zone, this would imply from 100 to 150 households, but would vary. (e) The number of key informants for a community would normally be at least three. (f) The surveyor discusses the concepts of wealth with the informants to ensure that outside resources of income from employment and assets are considered, in addition to land holding, land quality, house quality and possessions, and the number of producers and consumers in a household. The informants, however, are encouraged to make their own assessment of wealth ranking based on what they consider as indicative of a particular status. In the Lumle experience, informants often defined the categories in terms of a household's ability to feed itself (e.g., number of months for which there is self-sufficiency in basic foods without seeking casual outside employment, or surplus grain and income available). - 81 - ANNEX 3 Page 2 of 2 (g) Each informant is allowed to place the households in as many categories as he/she considers as discernible; this would usually be between four and eight. After sorting of cards, a discussion is held with each informant about the basis for their categorization and notes of record are made by the surveyor. Should an individual's rankings differ greatly from other informants, the individual is encouraged to recheck rankings and to discuss the basis with the surveyor. Each session of explanation, card sorting and discussion usually takes about two hours. 3. The concept of using a reference group (PPAR text paras 5.39-5.44) as an extension group target is based on the premise that members of that group have a set of characteristics which are conducive to their having similar attitudes to technological change and risk, and similar technological requirements. The wealth ranking techni-ue provides a practical way of grouping households according to their resource base, which is likely to be the most important variable affecting the response to a particular technology; each ranked group is a form of reference group, with members being able to identify themselves as having circumstances similar to others in the group and, therefore, able to relate to each others opinions and experiences. 4. Depending on the size and number of the reference groups defined by the wealth ranking technique, it may not be practical to provide programmed extension services to each group. More than one reference group could be combined into an extension group for practical reasons, as long as the existence and membership of the different reference groups are acknowledged and catered for in discussions with the group and in the presentation of technical solutions and management of test plots, etc. 5. The F.N. #61 in the text refers to the Lumle reports which elaborate on use of the wealth ranking technique. - 83 - ATTACHMENT Page 1 of 4 COMMENTS ON THE PPAR COMPONENTS REFERRING TO CR. 1100-NEP, CR. 1101-NEP AND CR. 1339-NEP BY THE JOINT SECRETARY, MINISTRY OF FINANCE - APRIL 7, 1992 (Retyped Version for Clarity) His Majesty's Goverxment Ministry of Finance Kathmandu Nepal Date: April 7, 1992 Mr. Graham Donaldson, Chief, Agriculture and Human Development Division, Operations Evaluation Department, Washington, D. C. Dear Mr. Donaldson, Re: NEPAL - AERP, HFPP, CCDP and Draft Project Performance Audit Report I wish to refer your letter dated February 18, 1992 addressed to the Finance Secretary, regarding the above subject. We understand that other related line agencies have already sent their comments on the said report and I request you to consider the attached observation along with the other comments. With best regards, Sincerely Yours, (signed) R. B. Bhattarai Joint Secretary - 84 - ATrACHMENf Page 2 of 4 Comments on Project Performance Audit Report for AERP, HFPP & CCDP Operation Evaluation Division of the World Bank has prepared a comprehensive Project performance audit report for the above projects. We are grateful for the preparation of the report which has covered not only the area of the already completed projects but also highlighted recommendations for future undertakings. We would also like to make a few remarks on the report as following: Slow progress in the project implementation is one of the critical issues, which the government is facing today. Reasons for the delay could be many, including delay in budget release, conservative financial rules and regulation and low motivation on the part of the government employees. Without making significant changes in the above areas, it is not possible to realise the substantial progress in near future. Recognising these factors, HMGN is seriously pursuing to improve the institutional set up of the administrative organisation, and also in the process of modifying the budget release procedure. To this end, HMGN has already set up a high level Administrative Reform Commission constituted under the chairmanship of Right Honourable Prime Minister. The Commissioa is expected to submit its report in the near future regarding the overall improvement of the civil service structure including the benefits and other facilities. In the same manner, HMGN has also recently changed the financial rules and regulations. Under the new regulation, adequate authority has been provided to the manager in the site. Moreover, HMGN is also reviewing the budget release procedures. The basic idea behind this review is to make available budget to the project manager on a timely basis. With these changes, it is hoped that the project implementation in future will improve significantly. - 85 - ATACHMENT Page 3 of 4 COMMENTS ON THE PPAR COMPONENTS REFERRING TO CR. 1100-NEP AND CR. 1101-NEP BY THE PROJECT COORDINATOR FOR THE NEPAL AGRICULTURAL EXTENSION PROJECT - MARCH 30, 1992 (Retyped Version for Clarity) The OED of the World Bank has prepared a comprehensive Project Performance Audit Report (PPAR). The report has nicely reviewed and recommended the future measures to be taken by the Bank based on the lessons learnt in the above projects. The following are a few comments on the report. 1. HFPP was rated as unsatisfactory as compared to AERP. This was partly because HFPP had activities not directly related to extension service such as trails, bridges and irrigational structures which reduced the attention to agricultural extension. 2. Both the projects were unable to use the allocated funds due to the devaluation of Nepali currency, which especially affected the civil works loan allocations. 3. The T&V system of extension of extension introduced in both projects was more a top- down approach than a bottom-up, demand-driven, farmers' needs-based approach. In addition, the routine visits to contact farmers were very difficult due to transportation problems (especially in the Hills). 4. Several missions from World Bank supervised the projects regularly. However, the supervision missions were more concerned with the amount spent as per appraisal for civil works and procurement, rather than the proper utilization of funds and the impact on the beneficiaries. 5. Some impact monitoring studies were done by APROSC for the projects; however, the studies are vague and doubtful with regard to adoption of technology and sustained increase in production. 6. Since agricultural extension is not solely responsible for increased production and there are other agencies/factors which influence an increase or decrease in production, the project alone should not be considered as influencing productior. Technologies conveyed or demonstrated by extension agents have a potential to be very appropriate to specific areas, but many farmers would not adopt them due to their poor socio-economic conditions and/or the lack or untimely supply of required inputs. - 86 - ATTACHMENT Page 4 of 4 7. The PPAR states that the project had no apparent impact on wheat yields; in irrigated areas, farmers are growing other crops such as potato, winter vegetables and mustard instead of wheat in the winter season. Due to irrigation facilities, the croppir patterns have been changed and farmers income have been increased. Because of this fact, wheat production might have decreased but total crop production has not decreased. Therefore, in evaluating the projects, increase in farmers incomes rather than the production of individual crops should be considered. 8. Maize yields in the Hills has been decreasing every year due to lack of available technology for different agro-climatic conditions, lack of inputs, lack of farmer purchasing power and loss of fertile surface soil through soil erosion. Only in accessible areas of hills where available technology and inputs exist and the farmers can afford to buy the inputs, maize yield has been increased. Therefore, it is difficult to infer that the maize yield of project beneficiaries has been decreased without authentic study. 9. Regarding research-extension linkage, there was poorer linkage in HFPP districts than in AERP. Outreach trials conducted in AERP districts were not farmers'-problem based. They were based on the perception of researchers on the farmers problems and they were similar to the trials conducted inside the research farm. The Agricultural Extension Project has addressed this issue, and a joint group trek of researchers and extensionists in farmers' fields is being used to design the trials based on farmers' problems. 10. Sustainability of agricultural extension and research services are not only related to continued support from budgetary authorities, satisfactory professionalism and morale as stated in PPAR, but also with the extent to which they address the target groups' problems. Both should be demand-driven rather than being activities to fulfill the numerical targets of extension and research. 11. The PPAR recommendation to use problem census technique in programming extension activities seems to be quite relevant. However, the carrying out of these functions should be within the mandate of the extension service. シぐ 82- BRI) 23595 NEPAL ECOLOGICAL ZONES AND "d -l CASH CROP DEVELOPMENT PROJECT INDIA -30P HUI ly MOLINTAIN 1COLOGICAI. 1.11E 'I";' CA E A. HILL ECOLQGICAL ZON15 TOSAC 0 SERICULTUK TERA ECOLOGICAL ZONE OILSEEDS KATH NDU VALLEY L-JURA "IN ROADS CROP RESEARCH STATION.' GINGER EXISTING TOBACCC -k- UNDER CONSTRUCTION SERIOLILTURE L"RA 11CONDARI ROADI OILSEED DOTEE ..HAM JUMLA RAILROADS NATIONAL CAPITAL TE ýM iD BROAD GAUGE METER GAUGE DEVELOPMENT REGION BOUNDARIES ADMINISTRATIVE ZONE BOUNDARIES KAILALI DAILEKH ~IRKOT M~ DISTRICT BOLINDARIES INTERNATIONAL 80UNDARIES RUKUM RIVERS SURKHET W MVAGDI KILO.ETERS "1 12 SALYAN KA o A K iý 6,0 TO 0Q BAGLU NG MILES ULMI p LI, M, ADIN E-1 DANC~I.TRI --OK H«HANC141 WAKOT UMBI, Kt-11 NG kl(WAN UR 4 HAP INDIA OKHAL HU ý 15,1,10H Lir "0 NýBHOJPUR A~ -v J1ERHAýT PANCHTHýR ILKLAYPUR SýN5AIII 11 1 MORANG! SAPTARI JHAPA BAN L'A D Es> T.A.m El, MARCH 1992 NEPAL HILL FOOD PRODUCTION PROJECT CHINA PROJECT AREA AND FACILITIES M A - A N G- PROJECT HEADOUARTERS ( DISTRICT TOWNSICENTERS A PROPOSED AGRICULTURAL SERVICE TRUNK ROADS CENTERS AANNAXURNA --- ROADS UNDER CONSTRUCTIOoN /y -- - PROPOSED TRAIL IMPROVEMENTS RDIS UNDARIO 7937 -281W** DISTRICT 80UNDARIES 2w PROJECT DISTRICTS D TRAINING CENTERS INTERNATIONAL BOUNDARIEH AIC STORAGE DEPOTS RIVERS Lm O PROJECT RELATED VILLAGES MUTISNEEAIN DISTRICTS WEANDEG A N D A K I I Mok S PaIl ngrtw 1 N G RK- 3 Kb=dokE X A 4 Twrgr Makaing GKdi Bararu 2 Sera Saudi. LA.JUNG 3 Aarikos / 4 Sundler Bawa 5 TinDipale A Rr A Pokhara AWW 1 Cm.guro. w BAGMATI 2 Darsing Dahatur SYAVIA 3 Blirwa Amchale 4 Gaungi 5 Daring Maadkot sativowis ? NANU 2d3 Ranl Pokhti 2 O METERS .rt s Y Af A w0dw KILOMETERS sG. 2~r ==WtEcqonc.o B23~~~N,n GA*N~hIRdU