Document of The World Bank Report No: ICR00003422 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-40400 TF-98827) ON A CREDIT IN THE AMOUNT OF US$12.00 MILLION EQUIVALENT TO THE KINGDOM OF BHUTAN FOR A DECENTRALIZED RURAL DEVELOPMENT PROJECT June 23, 2015 Global Food and Agriculture Practice Bhutan Country Management Unit South Asia Region CURRENCY EQUIVALENTS (Exchange Rate Effective December 2014) Currency Unit = Ngultrums Ngultrums 63 = US$1 US$ 1.45 = SDR 1 FISCAL YEAR January 1 – December 31 ABBREVIATIONS AND ACRONYMS ACIAR Australian Centre for International ICRR Implementation Completion and Results Agricultural Research Report AF Additional financing IEG Independent Evaluation Group AFD Administration and Finance Division IRR Internal Rate of Return AMC Agriculture Machinery Centre ISRs Implementation Status Reports BAS Budget and Accounting System MoAF Ministry of Agriculture and Forests BTN Bhutan Ngultrum M&E Monitoring & Evaluation CAS Country Assistance Strategy MOU Memorandum of Understanding Chiog Village level with Tshogpa (elected head of MTR Midterm Review village) NBC National Biodiversity Center DA Designated Account NHPC National Post Harvest Centre DAOs District Agriculture Officers NOC No Objection Certificate DzongkhagDistrict administrative level NPP National Potato Program Dzongda District Administrator NPV Net Present Value DRDP Decentralized Rural Development Project NSC National Seed Centre EA Extension Agent O&M Operations & Maintenance EFRC Environmentally Friendly Road Construction PAD Project Appraisal Document EIRR Economic Internal Rate of Return PDO Project Development Objectives EMP Environmental Management Plan PMT Project Management Team FIRR Financial Internal Rate of Return PMU Project Monitoring Unit FM Financial Management PTTs Power Tiller Track FYP Five Year Plan QAG Quality Assurance Group GAFSP Global Agricultural and Food Security Program QEA Quality at Entry Assessment GDP Gross Domestic Product RDC Research Development Centre Geog Block (lowest level of government RGoB Royal Government of Bhutan administrative system) RNR Renewable Natural Resources GCC Geog Community Centre RUAs Road Users Associations GYT Geog Yargey Tshogdu (Block Development SIL Sector Investment Loan Committee) SIMF Social Impact Management Framework HHs House Holds WUAs Water Users Associations Vice President: Annette Dixon Country Director: Johannes Zutt Senior Global Practice Director: Juergen Voegele Practice Manager: Shobha Shetty Project Team Leader: Winston Dawes ICR Team Leader: Imtiaz Alvi KINGDOM OF BHUTAN DECENTRALIZED RURAL DEVELOPMENT PROJECT CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph 1. Project Context, Development Objectives and Design ............................................... 1 2. Key Factors Affecting Implementation and Outcomes .............................................. 4 3. Assessment of Outcomes .......................................................................................... 11 4. Assessment of Risk to Development Outcome......................................................... 23 5. Assessment of Bank and Borrower Performance ..................................................... 24 6. Lessons Learned ....................................................................................................... 27 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners .......... 29 Annex 1. Project Costs and Financing .......................................................................... 30 Annex 2. Outputs by Component ................................................................................. 31 Annex 3. Economic and Financial Analysis ................................................................. 44 Annex 4. Bank Lending and Implementation Support/Supervision Processes ............ 56 Annex 5. Beneficiary Survey Results ........................................................................... 57 Annex 6. Stakeholder Workshop Report and Results................................................... 61 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ..................... 62 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ....................... 74 Annex 9. List of Supporting Documents ...................................................................... 75 MAP A. Basic Information Decentralized Rural Country: Bhutan Project Name: Development Project Project ID: P087150 L/C/TF Number(s): IDA-40400,TF-98827 ICR Date: 06/14/2015 ICR Type: Core ICR Lending Instrument: SIL Borrower: Kingdom Of Bhutan Original Total USD 12.00 M Disbursed Amount: USD 12.00 M Commitment: Revised Amount: USD 12.00 M Environmental Category: B Implementing Agencies: Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 01/22/2004 Effectiveness: 04/15/2005 04/15/2005 03/2011 Appraisal: 11/29/2004 Restructuring(s): 12/2013 Approval: 03/01/2005 Mid-term Review: 5/2008 Closing: 09/30/2009 12/30/2014 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Satisfactory Risk to Development Outcome: Moderate Bank Performance: Moderately Satisfactory Borrower Performance: Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Moderately Satisfactory Government: Satisfactory Satisfactory Implementing Moderately Satisfactory Quality of Supervision: Agency/Agencies: Overall Bank Moderately Satisfactory Overall Borrower Moderately Satisfactory Performance: Performance: i C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments (if Indicators Rating Performance any) Potential Problem Project at any time Yes Quality at Entry (QEA): Satisfactory (Yes/No): Problem Project at any Quality of Supervision No None time (Yes/No): (QSA): DO rating before Satisfactory Closing/Inactive status: D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Agricultural extension and research 12 12 Irrigation and drainage 18 18 Roads and highways 64 64 Sub-national government administration 6 6 Theme Code (as % of total Bank financing) Decentralization 20 20 Other rural development 20 20 Rural services and infrastructure 40 40 Water resource management 20 20 E. Bank Staff Positions At ICR At Approval Vice President: Annette Dixon Praful Patel Country Director: Johannes Zutt Alastair J. McKechnie Senior Global Practice Juergen Voegele Director: Practice Manager: Shobha Shetty Adolfo Brizzi Project Team Leader: Winston Dawes Daniel Sellen ICR Team Leader: Imtiaz Alvi ICR Primary Author: Sati Achath & Imtiaz Alvi ii F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The objective of the project was to improve market access and increase agricultural output for rural communities in selected areas of Bhutan. Revised Project Development Objectives (as approved by original approving authority) The objective was not revised (a) PDO Indicator(s) Original Target Formally Actual Value Achieved at Values (from Revised Indicator Baseline Value Completion or Target approval Target Years documents) Values Indicator 1 : Average time taken to walk to motorable road • The walking time to nearest road point1 is within ½ an hour for 78.2% (572) HHs; between ½ to 1 hour for 14.5% (106) HHs; between 1-3 hours for 4.4% (32) HHs and between 4-6 hours for 2.9% (21) HHs. Value 1.5 hours on quantitative or Average time = 6.5 hours average • 71% HHs were living Qualitative) within 1 hour walking time to motorable road2 as compared to 43% in year 2000. • The walking time to nearest motorable road has reduced to 1.3 hours in year 2013 from 3 hours in year 2000. Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % Exceeded the target by a significant margin achievement) 1 Source: Field survey: DRDP Impact Assessment Study(Oct-Nov- 2014) 2 Source: Agriculture Statistics 2000, 2011 and 2013 iii Indicator 23 : Increase in food crop production 15% Increase 2013 over 2009 increase in production (and yield): above the Paddy: + 18.2% (+44%) Value 20% increase in varies by crop (aggregate baseline of (average of 11 quantitative or food crop production index = 100) 2009 (in Dzongkhags) Qualitative) reference to Maize: + 21.3% (+49%) paddy and (average of 14 maize) Dzongkhags) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments The target was fully achieved for paddy and maize (incl. % Since GLS affected maize growing areas were found beyond the original 11 project achievement) districts, DRDP extended support to three additional districts. Indicator 3 : Increase in cash crop production 15% Increase 2013 over 2009 increase in production (and yield); Value above the (average of all 11 30% increase in cash quantitative or baseline of Dzongkhags): crop production Qualitative) 2009 (in Potato: + 10.7% (-6.7%) reference to potato) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 71% achievement. The reason for less than100% achievement is that the major project Comments intervention of increased improved and disease free potato seed production under (incl. % aeroponic culture has been successfully launched and will start showing results from achievement) 2017 onwards. Indicator 4 : Direct project beneficiaries Value quantitative or 15,200 41,360 Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % Exceeded the target by 270%. Indirect beneficiaries are estimated to be about 374,240. achievement) (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Indicator 1 : Farm roads constructed (km) • 138.27 km of new farm road Value constructed (145% (quantitative 95.5 achieved) or Qualitative) • 69.35 km of farm roads improved (during DRDP-AF): 3 For PDO Indicators 2 and 3, all data source is the same, namely: Agriculture Statistics 2009; Agriculture Statistics 2008; Ministry of Agriculture and Forests, Department of Agriculture. Based on these data comparisons of 2013 with 2009 were made for the calculation of averages. iv (462% achieved) • 128.99 km of power tiller tracks constructed (113% achieved) • 42.50 kms of new irrigation channel constructed (96% achieved) • 5 bridges constructed (83% achieved) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments As a whole, Overachieved the targets by145%. (incl. % achievement) Indicator 2 : New irrigation channels constructed (km) • 42.50 km of new Value irrigation channel (quantitative 44.3 constructed (96% or Qualitative) achieved) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 96% achievement. Shortfall of 4% is due to the fact that project communities preferred Comments rehabilitation of existing channels (achievements level 133%) over the construction of (incl. % new channels. This notion is further strengthened by beneficiaries’ perception (survey)1 achievement) against the relevance of activities, wherein renovated irrigation was considered most relevant for 24.7% of total respondents as against 5.5% for new channels. Indicator 3 : Irrigation channels rehabilitated (km) • 521.24 km of irrigation channels renovated (311.16 km during DRDP- Value AF) (133% (quantitative 450.60 achieved) or Qualitative) • 1 pipe and 1 storage irrigation schemes were implemented as pilot. Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % Exceeded the target by 133% achievement) Indicator 4 : Timely submission of accurate FM accounts Value • Timely submission (quantitative 100% of progress and or Qualitative) financial reports Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % Fully achieved achievement) v Indicator 5 : Improved roads (km) • 138.27 km of new farm road constructed (145% achieved) • 69.35 km of farm Value roads improved (quantitative 15 (during DRDP-AF): or Qualitative) (462% achieved) • 128.99 km of power tiller tracks constructed (113% achieved) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % Exceeded the target from 113% to 462% achievement) Indicator 6 : 25% of households have adopted high yielding rice varieties, output increased by 20% 25.00% adoption rate HH adoption rate Value (March 2011) was 65%; average (quantitative n/a n/a yield increase was or Qualitative) 20% increase in 18.2% production (March 2011) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments 260% achievement for household adoption rates in the main rice growing areas of (incl. % Wangdue-Punakha; 91% achievement in terms of yield increase as average for all project achievement) areas. Indicator 7 : Replacement of maize seeds in Grey Leaf Spot (GLS) affected areas 2,899 acres of GLS affected area Value 80.00% 3,616 acres of GLS affected (including three non- (quantitative n/a (2,893 acres) area project areas) or Qualitative) (March 2011) received quality seed replacement Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 100% achievement. No baseline was recorded in the project documents, however the Comments 3,616 acres GLS affected areas including 4 project districts (Dagana, Trasigang, Tsirang, (incl. % Zhemgang) and 3 non-project districts (Lhuentse, Pemagatshel and Yangtse) was achievement) communicated by MoAF (Maize Coordinator). Indicator 8 : Production of potato basic seeds 875 tons of basic Value seed potatoes were 700 mt (quantitative 350 mt (2011) n/a (March 2011) produced and or Qualitative) marketed. Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments Exceeded the target by 125%. While there was no baseline for this indicator in the PAD, (incl. % the 350 mt baseline was introduced through the TF Grant Agreement in March 2011. achievement) Indicator 9 : Construction of Power Tiller Tracks Value 114 km 129 km vi quantitative or Qualitative) Date achieved 30-Apr-2014 31-Dec-2014 Comments 113% achievement against the target (incl. % achievement) Indicator 10 : Construction of suspension and motorable bridges Value quantitative or 6 5 Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 83% Achievement. The project constructed 6 bridges as per the target at appraisal. Two Comments bridges were washed away during the 2012 cyclone that caused huge losses in Bhutan. (incl. % The project rebuilt one of the damaged bridges with the additional support of JICA. achievement) The overall output against this subcomponent remained 5 bridges against a target of 6. Indicator 11 : Construction of RNR Centers Value quantitative or 6 6 Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % Fully achieved achievement) Indicator 12 : Farmers trained and adopting demonstrated technologies Value (quantitative 0 1200 n/a 1200 or Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments In the “Project Paper on a proposed additional grant and restructuring in the amount of (incl. % US$5 Million under the Food Crisis Response Trust Fund” (Dec. 21, 2010, p.7) the achievement) indicator was dropped as it had been already achievement. Indicator 13 : Farmers benefitted by extension services (of which 40% female farmers) Value (quantitative 0 n/a 8500 3311 or Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 39% achievement with approx 50% female participants. This indicator appears in the “Project Paper on a Proposed Additional Grant and Restructuring” (Dec. 2010) and in Comments Annex1 of the “Restructuring Proposal for DRDP Additional Financing (May 2013), (incl. % but not in the Trust Fund Agreement (March 2011). The ICRR mission was also told achievement) that in any event extension services are regularly delivered and financed by RGoB, which may explain that exact data on this indicator were not available and the achievement rates given here are based on the PCR report of the borrower. Indicator 14 : Formation of “Growers Groups” Value (quantitative 0 n/a 5 No data or Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments There is anecdotal evidence that farmers groups have been involved (rice grower groups, (incl. % vegetable groups, etc), but no hard data on the names of groups, numbers of membership, achievement) or trainings completed, have been reported. vii Community based seed production groups (CBSP) are formed and linked with the Indicator 15 : National Seed Center Value (quantitative 0 n/a At least 2 CBSP 9 or Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % The target overachieved by 450% achievement) Capacity of 14 agricultural field staff increased through training in production and post Indicator 16 : harvesting technologies Value (quantitative 0 10 (2011) 14 (2013) 252 or Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % Exceeded the target by 1800% achievement) Indicator 17 : Capacity of Research and Development Staff enhanced Value (quantitative 0 4 (2011) 6 (2013) 8 or Qualitative) Date achieved 06-May-2005 30-Apr-2014 31-Dec-2014 Comments (incl. % Exceeded the target by 133% achievement) G. Ratings of Project Performance in Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 06/06/2005 Satisfactory Satisfactory 0.50 2 11/24/2005 Satisfactory Satisfactory 0.56 3 06/02/2006 Satisfactory Moderately Satisfactory 0.68 4 12/16/2006 Satisfactory Moderately Satisfactory 0.95 5 05/23/2007 Satisfactory Moderately Satisfactory 1.78 6 12/25/2007 Moderately Satisfactory Moderately Satisfactory 2.90 7 06/28/2008 Satisfactory Moderately Satisfactory 3.91 8 12/30/2008 Satisfactory Moderately Satisfactory 4.58 9 05/27/2009 Satisfactory Moderately Satisfactory 4.92 10 11/28/2009 Satisfactory Satisfactory 6.11 11 05/28/2010 Satisfactory Satisfactory 6.31 12 12/11/2010 Highly Satisfactory Satisfactory 6.64 13 06/05/2011 Highly Satisfactory Satisfactory 6.80 14 10/18/2012 Highly Satisfactory Satisfactory 7.11 15 06/12/2013 Satisfactory Satisfactory 7.11 16 12/06/2013 Satisfactory Satisfactory 7.11 17 06/04/2014 Satisfactory Satisfactory 7.11 18 12/02/2014 Satisfactory Satisfactory 7.02 viii H. Restructuring (if any) The project underwent the following five restructurings: (a)First amendment of the Credit was in the definition of the Incremental Operating Costs; (b) Second restructuring was to extend the Closing Date of the project by 15 months to December 31, 2010 to complete road contracts which required between 2-15 months for completion after the September 30, 2009 closing date; (c) Third Restructuring took place to extend the closing date by three months (up to March 31, 2011) to provide sufficient time for review and internal processing of a new FPCR TF Grant; (d) Fourth restructuring was to provide additional financing of US$5 million through the FPCR TF Grant, including a further extension of project closing date to December 30, 2013. Changes were also made to the project component as indicated above, and introduced a new set of performance indicators to measure Outcomes; and (e) Fifth restructuring extended project closing date by 12 months (up to December 31, 2014) to utilize the savings resulting from depreciation of the local currency versus US dollar. I. Disbursement Profile ix 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal. The 2000 Poverty Assessment had found that absence of roads was a major contributing factor to poverty in Bhutan. The Transport Sector Note called accessibility “the defining development issue in Bhutan, be it access to opportunity, enterprise, markets, or services”. Many Bhutanese communities lacked access to road network, and depended on animal and head-load transport. One third of Geogs (blocks- lowest level of government) had effectively no motorized access, and only 40 percent of Bhutan’s population lived within one hour’s walking distance from a road. Access to education and health services was limited. The efforts to diversify into cash crops had a limited impact as in most cases agriculture markets were beyond the reach of farmers. Besides contributing one third of GDP, the agriculture sector provided income, employment and food security to most Bhutanese, especially the poor. The path to Bhutan’s vision of Gross National Happiness, therefore, required concerted efforts by the Ministry of Agriculture and Forests (MoAF) to focus on strengthening crop and livestock production, forestry, farm roads and natural resource management. The Renewable Natural Resources (RNR) Strategy under the 2002 9th Five Year Plan (FYP) aimed at reducing rural poverty by achieving higher rural incomes and employment, improved food security and sustainable natural resource management. The path to Bhutan’s vision of Gross National Happiness, therefore, requires concerted effort by the Ministry of Agriculture (MoA) in the Renewable Natural Resources (RNR) sector, which comprises crop and livestock production, forestry, farm roads and natural resource management. Rationale for Bank involvement. The Royal Government of Bhutan (RGOB) had prepared development plans for 201 Geogs that ably reflected the expectations of rural people in the 20 Dzongkhags (ddistricts) and had launched central programs to support realization of the Geog Plans. As RGOB financial resources and donor provided funds were inadequate to fully support these plans, RGOB sought the World Bank support for financing the resource gap. The Bank had the ability to mobilize adequate financial resources to meet priority needs for the 9th FYP and complement the RGOB and donor programs. As a knowledge based organization, the Bank was able to bring in the global knowledge, experience and best practice in the domains of rural infrastructure, fiscal decentralization and community-driven development. 1.2 Original Project Development Objectives (PDO) and Key Indicators The objective of the Decentralized Rural Development Project (DRDP) was to improve market access and increase agricultural output for rural communities in selected areas of Bhutan. Key indicators: • Average time taken to walk to motorable road (reduced from 3 hours to 1.5 hours on average); • Increase in food crop production (by 27% with 2004 as baseline); and • Increase in cash crop production (by 37% with 2004 as baseline). 1 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification While PDO remained unchanged, key PDO indicators (and a number of intermediate outcome indicators) were revised with the provision of Additional Financing of US$5 million through a Food Price Crisis Response (FPCR) TF Grant in March 2011: • Average time taken to walk to motorable road (reduced from 3 hours to 1.5 hours on average); • Increase in food crop production (by 15% with 2009 as baseline); and • Increase in cash crop production (by 15% with 2009 as baseline). 1.4 Main Beneficiaries At the time of project preparation, expected beneficiaries of the project included: • Farmers who would benefit from improved rural access to markets as well as from new or rehabilitated irrigation facilities; • Government Staff of Geogs and Dzongkhags who would benefit from training programs; and • Geogs, Dzongkhags, and MoAF would be strengthened through the institutional strengthening activities. At project completion, the total number of beneficiaries was about 415,600 comprising 41,360 direct beneficiaries and 374,240 indirect beneficiaries. The female beneficiaries were 202,445 out of which about 91,640 (45.25%) were direct beneficiaries and remaining about 110,800 (54.73%) were indirect beneficiaries. Besides the farmers, government officials also benefited from the project primarily through the capacity building activities. The operation strengthened existing technologies and facilities of the implementation partners such as National Seed Centre (NSC), National Plant Protection Centre (NPPC), National Post Harvest Centre (NHPC) and Renewable Natural Resources Research and Development Centers (RNR-RDCs). 1.5 Original Components The project comprised the following three components: Component 1: Rural Infrastructure (USD 5.10 + US$0.80 million) (a) Rural access: This subcomponent was to focus mainly on farm roads and upgrading of mule tracks to permit passage of power tillers. A target of 89 kilometers (km) of farm roads, and upgrading mule tracks to 114 km of power tiller roads were to be screened and implemented in accordance with the MoAF’s Farm Road Guidelines, and the “Environment Friendly Road Construction” (EFRC) rules. Six small suspension bridges, with the capability for power tillers and motorcycles and one motorable bridge to connect a farm road, were to be constructed. 2 (b) Irrigation: A total of 45 km of new irrigation channels and 231 km of rehabilitated channels were proposed. Component 2: Renewable Natural Resources (RNR) Centers (USD 0.75 million) (a) RNR Center Construction: The subcomponent was to finance at least six RNR Centres for which prototype designs already existed. In addition to this, the Geogs requested an addition of offices for the “Gup” (Geog Head) and clerk. (b) RNR Capacity Building: Provision of training and carrying out of demonstrations of high value crops, livestock-raising practices and new agricultural technologies, and farmer-to-farmer field visits in select Geogs. Component 3: Institutional Strengthening (USD 0.35 million) Institutional strengthening included capacity building of finance and procurement officials of Geogs, Dzongkhags and the Administration and Finance Division (AFD) of MoAF. Capacity building was to include Budget and Accounting System (BAS), planning, administration, community mobilization and progress reporting. Another key area was capacity building in social/environmental screening and assessment of sub- projects, especially farm roads, irrigation works and RNR centers. 1.6 Revised Components Following changes were made to the project components through Additional Financing in March 2011, and FPCR TF Grant (US$2.42 million, 1.92 million and 0.66 million for Component 1, 2 and 3 respectively). Subsequent restructuring in December 2013 allocated the project savings, due to exchange rate gains, to those components which had the capacity to utilize additional funds. Component 1: Rural Infrastructure (a) Rural access: Additional Financing (AF) increased the target for farm road construction from 89 km to 95 km, and added a target of rehabilitation of 15 km (increased further to 53 km in December 2013) of farm roads. (b) Irrigation: AF raised the target for the rehabilitation of irrigation channels from 231 km to 393 km. It was further increased to 448 km in December 2013. Component 2: Renewable Natural Resources (RNR) Centers and Support for Improved Technologies for Marketing of Rice and Maize (a) RNR Construction: This sub-component remained unchanged. (b) RNR Capacity Building: It comprised provision of training for research and extension staff and farmers, and carrying out of demonstrations of high value crops and food crops (potato, maize, rice), livestock-raising practices and support for new agricultural technologies including seeds and post-harvest technologies in select Geogs. 3 Component 3: Institutional Strengthening This component remained unchanged. 1.7 Other significant changes The project experienced the following key changes during implementation: (a) First amendment of the Credit in the definition of the Incremental Operating Costs; (b) Second Restructuring to extend the project closing date by 15 months to December 31, 2010 to complete roads contracts which required between 12-15 months for completion after the then closing date of September 30, 2009; (c) Third Restructuring to extend the closing date by 3 months (up to March 31, 2011) to provide adequate time for review and internal processing of a new FPCR TF Grant; (d) Fourth restructuring to provide additional financing of US$5 million through the FPCR TF Grant, including a further extension of project closing date to December 30, 2013. Changes were also made in the project component as indicated above, and introduced a new set of performance indicators to measure Outcomes; (e) Fifth restructuring in December 2013 extended the closing date of FPCR TF Grant by 12 months to utilize the savings resulting from depreciation of the Bhutanese Ngultrum versus the US dollar. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry Quality at Entry. The project’s quality at entry was moderately satisfactory. • The project design was simple, straight forward, realistic and well structured. The design addressed Bhutan’s highest priority for improving rural access by building rural infrastructure and strengthening investment planning and implementation capacity at the district and local levels. It was based on a Bank instrument that was a winning combination of programmatic and Sector Investment Loan (SIL) operations and used the national systems effectively for implementation.The project was focused and proposed investments in six districts, since other districts were already getting support from the government and other donor resources. • In view of the fact that Bhutanese economy was predominantly agrarian and major source of income and employment for a majority of rural population, and that several studies had clearly indicated lack of access as a fundamental constraint to increasing agricultural production, the project judiciously focused on infrastructure and agriculture-related investments. The project appropriately endorsed concrete roads with higher upfront costs with almost no maintenance costs over the first 10 years of the infrastructure creation. • It relied on the decentralized rural development model that had already been tried and tested in a number of projects including the UNCDF pilots which had shown substantial impacts. • Because of the constraints experienced in terms of remoteness of locations and availability of local population for limited time periods, the project opted to utilize the services of the private contractors for infrastructure activities. • The results framework was robust and the causal links were solid and clearly articulated. 4 On the other hand, the design had some drawbacks as given below: • While risks were suitably identified, the mitigation measures were not well defined in the project document. • Economic and financial analysis was not very rigorous. For example, cost-effective analysis was limited to roads infrastructure, and irrigation investments were not included. Furthermore the analysis lacked projected FRR of the benefits of the increased agricultural production; relying on broad studies to show benefits by comparing villages with access and without access, was not a very robust ERR/FRR approach. • The PDO indicated only increase in agricultural production rather than yield increase (as a critical indicator of efficient water use). • As mentioned in Section 2.3, M&E design had shortcomings such as absence of baseline data and weak M&E capacity at Geog level. Soundness of background analysis. As part of project preparation, sector background was studied, key sectoral issues were analyzed in depth, and the government strategies to deal with the issues were duly considered. For instance, investment priorities had already identified through a major participatory exercise for all 200 Geogs wherein the rural population had accorded high priority to rural roads, irrigation and other infrastructure. Lessons of earlier operations taken into account. The following key lessons were taken into consideration: (i) Decentralization is a long-term process that requires commitment, predictable flow of funds, institutional strengthening and capacity building on regular basis, (ii) simple and effective systems and procedures so that funds and subprojects can be managed adequately at Dzongkhag and Geog level; and (iii) Dzongkhag officials are “on-board” by incorporating project responsibilities into their regular mandate, as opposed to being seen as additional burden. The project design was based on the ongoing decentralization process in Bhutan and it devolved most resources and decision-making powers at the Dzongkhag and Geog levels. Risk Mitigation Measures. Although the risks were correctly identified, corresponding mitigation measures were not well defined in the project document. For example, the risk related to non-adoption of demonstrated technologies should have been mitigated by recommending technologies that were successfully demonstrated in other parts of Bhutan. There was little in the PAD about how this was to be achieved. The risk that farmers would be unable to find markets was assumed to be mitigated by reduced transport costs and improved access to motorable road. This was unlikely to be the case for many remote locations where improved access alone would not have mitigated this risk. Similarly, the risk related to rural population not utilizing infrastructure could have been mitigated through awareness raising, but these measures were not included in the project. Another risk was associated with the fact that there was no separate PMU proposed for DRDP. The project was housed in MOAF along with its other routine operational responsibilities and this could have easily resulted in less than the required focus on expediting implementation. A second key risk that was not identified related to 5 procurement – given that the locations were really remote and may have ended up facing getting the bids or timely completion of contracts. Adequacy of Government Commitment. At the time of project preparation, RGOB demonstrated a high level of commitment to the project objectives. For example, the government was committed to decentralization by empowering local governments, strengthening fiscal management at the local level, and undertaking capacity building of government officials accordingly. 2.2 Implementation The Bank conducted a Midterm Review (MTR) in May 2008, to assess the project performance, identify any implementation issues and recommend actions to ensure a successful completion of the project. The MTR endorsed the MoAF revised target of 107.66 km length for eleven power tiller tracks (PTTs) against the original plan of 114 km of PPTs. MTDF on Food Crisis Response provided a grant for US$5 million as additional financing (AF), effective March 21, 2011, to mitigate the impact of rising food prices in Bhutan due to low productivity, crop disease and global food price trends. The project decided to focus on rehabilitation of irrigation canals and enhanced production of rice, potato and maize. It resulted into an increase in the number of: (i) project Dzogkhags from six to eleven, (b) targets for irrigation rehabilitation increased from 231 km to 393 km; and (c) additional indicators for specific crops. Factors affecting implementation. The following factors had affected project implementation: (i) A lack of familiarity with the Bank’s procurement procedures and absence of a dedicated procurement specialist led to delays, especially in the initial years of project implementation. For example, a number of contracts had to be re-tendered due to this reason. (ii) In the early stages of DRDP, there were delays in the execution of farm roads and irrigation schemes. It was primarily due to remote and inaccessible areas, non-availability of building materials, unavailability of skilled labor, and limited technical capacity of Dzongkhags’ engineers, especially in design, estimate preparation and supervision of construction of farm roads. (iii) Delays were experienced in funds release and timely reporting of expenditure statements in the early stages of implementation, as Dzongkhag accounts staff were not familiar with the project’s financial system and disbursement arrangements. Initially, the project funds were not released and utilized on time, which led to the funds being carried forward at the end of fiscal year. It led to problems in reconciling the funds, especially when the project finance team could not provide adequate and timely oversight. The fact that Dzongkhag officials were inadequately trained and experienced in project’s procedures proved to be a bottleneck for project management. 6 (iv) Construction work on four roads (Etonyesa-Simnorbu, Rakshidrangra-Chanutre PTT, Pakshikha Zomgang, and Jangbee) was delayed due to a lack of response from the bidders, exceptionally high bid rates or non-responsive bids leading to re-tendering. (v) A high turnover of the Bank’s task team leaders (TTLs) made it challenging for the government to maintain sustained communication with the Bank. As there was no TTL for a protracted period of time the project, it lead to implementation delays. (vii) On the positive side, despite several constraints and inaccessible project areas, an increasing number of project Dzongkhags were proactive and efficient in implementing the project activities. Judicious use of funds and efficient implementation made it possible for the Dzongkhags to exceed their respective targets. It was made possible by the effective leadership provided by the dzongdas (district administrators) and the district agriculture officers (DAOs). 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization M&E Design. Outcome Indicators and Results Indicators were developed to monitor the project performance. The indicators included: (i) Average time taken to walk to motorable road, (ii) Increase in food crop production, and (iii) Increase in cash crop production. Although MoAF had a system in place to annually collect, compile and publicize agricultural statistics for Geog and Dzongkhag level, initially the project was missing baseline data. The project relied on the government system for data collection and no separate arrangements were made to regularly gather and compare productivity/yield data for project supported irrigation-schemes. M&E Implementation. While focal persons (Agriculture Extension Officers) from Dzongkhag engineering section, collected the necessary data for each intervention (cost, beneficiaries, acreage, Cheogs, Geogs, start and completion dates etc.), its consolidation at the central level into a comprehensive database remained weak. Consequently, there were limitations to carry out any detailed analytic work during implementation or at project completion. Nevertheless, ‘Commodity Coordinators’ made good efforts for collection and reporting of data on rice, maize and potato. M&E Utilization. Data collected from Geog was regularly analyzed and used for informed decision-making. For example, data collected by the respective “Commodity Coordinators” for rice, maize and potato were used to assess the seed replacement rate and the progress made with farmers’ training and capacity building measures. Considering the shortcomings such as lack of project-specific M&E system, relying completely on the government system of data collection and reporting, weak institutional arrangement for M&E, and lack of regular reporting, performance of M&E system is rated as moderately satisfactory. 7 2.4 Safeguard and Fiduciary Compliance Safeguard Compliance. Environmental Safeguards. DRDP has resulted in net environmental benefits to the project communities. Amongst the key achievements are: (i) improved access to market through the construction of farm roads, PPTs and bridges, (ii) ensuring dependable irrigation supplies by constructing irrigation channels, (iii) judicious use of limited available water for irrigation, (iv) improved on-farm irrigation efficiencies, (v) substantial increases in crop yield per unit of water, (vi) low carbon foot print due to the reduction in time to access market, and (vii) decrease in post-harvest losses. The project supported community driven small-scale interventions such as farm roads, PTT, irrigation schemes, and agricultural production enhancement through improved seed supply, training, exposure visits, etc. While no significant adverse environmental impacts were expected during implementation, the project developed an environmental management framework, and each activity was screened to avoid and/or mitigate any specific environmental concerns. The screening did not indicate any significant environmental issues. No farm roads, irrigation schemes and other infrastructure supported by DRDP are located in protected areas or in a known critical natural habitat. None of the irrigation schemes’ source stream was known as important fish habitat or migration route for protected/endangered/rare aquatic life/fishes. Minor and site specific environmental issues encountered for some interventions (particularly farm road, PTT and irrigation channel) were related to landslides and erosion, management of spoil/debris, water management/ drain-outlet, felling of trees/ vegetation and workers health and safety. These were locally managed. Amongst the good environmental practices adopted by the project were: (i) obtaining No Objection Certificate (NOC) from the Forest Office for works in the forest area, (ii) disposal of spoil/debris in safe places, (iii) slope retaining and water management structures, and (iv) bio-engineering/plantation for controlling/improving existing landslides/erosions. Release of any new crop varieties was coordinated with the National Biodiversity Center (NBC) which ensures conservation of local crop variety. The project did not support any purchases of chemical pesticides and fertilizer. Instead, green manure, such as use of sesbania (Dhaincha) as bio-fertilizer plant, was promoted. Regular monitoring of environmental mitigation measures of small-scale interventions spread over dispersed and difficult to access locations was challenging as the District Environmental Officers responsible for environmental monitoring were overworked. In the absence of adequate training and third party monitoring, documentation and reporting remained sub-optimal. Requirements for the implementation of Environmental Management Framework were not included in the project legal agreements. The overall rating is satisfactory. Social Safeguards. The project triggered OP4.12 (Involuntary Resettlement) to address potential social risks associated with acquiring additional land for project interventions. 8 In order to assess social impacts and necessary mitigation measures for managing any adverse social impacts, the project developed a Social Impact Management Framework (SIMF) including resettlement framework. The primary instrument that the project used to ensure due diligence in safeguards compliance was the social screening for new infrastructure planning. Social screening was carried out for new farm roads and irrigation schemes. In the case of additional financing, since all construction activities were confined to the renovation of small community-managed irrigation channels and improvement of farm roads, it did not require any social screening. The procedure employed for voluntary land donation i.e. MoU were viewed by the community members as satisfactory and working well. No major negative impacts were reported or complaints lodged by the beneficiaries on loss of large parcels of land or assets, loss of livelihoods or relocation of households due to project interventions. In some cases (15%), land was donated on voluntary basis. The overall environmental and social safeguards rating is rated as satisfactory. Fiduciary Compliance Financial Management. The FM performance of the project was unsatisfactory during the initial years of implementation due to various issues such as quality and timeliness of the FMRs, funds flow related issues, cash management and disbursement of funds, limited capacity and understanding of the finance staff about the Bank’s fiduciary requirements, Although some good progress was noted during the mid-term mission, even at that time, the issues of disruption in contractor payment schedule were noted due to: (i) delays in the release of funds to Dzongkhag/Geog, (ii) “outages” in Designated Account, and (iii) reconciliation issues in FMR. The efforts jointly made by the MOAF and the Bank including continuous handholding and training of staff, streamlining of cash flow issues, regulated payments to the contractors the FM performance improved considerably in the later years of project implementation. It have had a positive impact on the implementation speed and project outcomes. Procurement: The project in general complied with the fiduciary covenants during implementation. Initially, there were serious capacity issue as the officers at the Dzongkhag level did not have any experience in handling procurements funded by the World Bank. Although internal control arrangements were in place, and adequate system for financial management, procurement records, and disbursement monitoring were maintained and improved considerably over time, an absence of any in-house or hired (consultants) procurement expertise led to considerable procurement delays during the early years of implementation. In several cases, it resulted in rebidding. Nevertheless, the Bank’s post-reviews did not identify any noticeable procedural violations, collusion or F&C issues. In fact appropriate recordkeeping was identified as an area requiring significant improvements. Combining all relevant factors, fiduciary compliance is rated as moderately satisfactory. 9 2.5 Post-completion Operation/Next Phase (a) Transition arrangements. The transition arrangements for continuing future operations appear to be adequate, as demonstrated by the following measures adopted by MoAF. Institutional: Project activities and processes (M&E, social and environment safeguards practices etc.) are already well integrated within the RGOB’s institutional framework. Enhanced staff capacity resulting from the project, are well streamlines and internalized, especially at the Ministry, Dzongkhag and Geog level. Furthermore, Institutional capacity of implementing partners have been strengthened. For example, research units have better facilities for research by having irrigation, necessary field equipment, training and exposure to staff, improved inputs, and with land terracing. The facilities for seed production have considerably improved with the installation of seed separators, fencing of seed production areas, construction of irrigation system, installation of processing units and aeroponics facility. The e-pest surveillance system is ready to capture all pest organisms and establish a database for necessary pest control and prevention, to minimize crop losses. The national database on irrigation is now available for use for development and improved O&M of the irrigation infrastructure in the 11th Five Year Plan (2013- 2018). Technical: Activities implemented by the project have been incorporated into the government’s 11th FYP. Some of the programs outlined in the 11th FYP that would ensure continuity of the project activities include the following: • Rural Infrastructures: (i) construction of new irrigation schemes for rice- based farming system, (ii) improving existing irrigation infrastructures, (iii) promoting and strengthening of effective functioning of water user associations (WUAs), (iv) developing irrigation water allocation system in the country, (v) pilot implementation of lift/pump irrigation including water harvesting, and (vi) construction and improvement of farm roads. • RNR and Extension Activities: (i) promoting the use of high yielding varieties (HYV), (ii) introducing, evaluating, assessing and producing high yielding varieties including biotic and abiotic stress tolerant varieties and characterization of local germplasm, (iii) promoting and demonstrating integrated plant nutrient management and soil fertility management including green manuring, integrated pest management technology including disease diagnosis and surveillance, (iv) promoting mechanization services through hiring services, mobile clinics, supply of machineries including privatization, and (v) enhancing the capacity of farmers and extension agents on operation and maintenance of farm machineries. • MoAF is working out institutional arrangements and budgetary allocation for sustainable use of the DRDP-supported roads. Road User Groups (RUGs) have been formed for regular maintenance works including minor repairs. However, some roads, especially those stretching over a long distance, would require higher level of inputs for maintenance than what RUGs can voluntarily contribute in the form of labor. Major repairs requiring machinery and skilled labor will be beyond their capacity, and it would require financial and technical support from Dzongkhag and Geog. The 10 PMT is preparing a plan for putting in place suitable institutional arrangements to finance the maintenance work for each farm road in consultation with the users, as well as the Dzongkhag and Geog authorities. • As per the government guidelines, no MoUs or agreements are to be signed with any communities which are benefiting from the project. Instead, handing over a document is signed for O&M after the completion of work. As stipulated in the farm road guidelines and national irrigation policy, communities are responsible for routine and minor maintenance and upkeep of the infrastructure. Budgeting: In the annual budget, the RGOB has allocated funds for operation, maintenance, renovation and rehabilitation of DRDP supported interventions including road maintenance, and for institutions such as National Seeds Center to sustain their activities. Further, Geog administration and communities are committed to maintain and sustain the project interventions. Nevertheless, should a natural calamity hit a Geog, govt. allocated budget may not be sufficient to meet all the emergency needs for rehabilitation and reconstruction. Staffing: Capacity and technical knowledge of MoAF, Dzongkhag and Geog officials have been considerably strengthened through several training programs, study tours and international exposure visits organized by the project. For example, extension agents now have better knowledge on paddy, maize and potato production and farmers’ group formation. Similarly, many engineers now have improved knowledge on irrigation survey, design, cost estimation, construction technology, water management and O&M that guided implementation effectively. (b) Follow-on project. While an IDA funded follow-on project is not under preparation as such, Bhutan has been awarded grant funds from the Global Agricultural and Food Security Program (GAFSP). Accordingly, MoAF and the Bank are preparing a Food Security and Accelerated Poverty Reduction Project which, to the extent feasible, will build up on the lessons and success of DRDP. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Relevance of Objective. The project objective remains highly relevant to Bhutan's current development priorities. DRDP activities are in line with the government policy to enhance self sufficiency in food crops that was challenged by the remoteness of villages without road connectivity and irrigations facilities. For example, access roads and irrigation for agriculture production with a focus on extension, seed production and post- harvest technology are exceedingly relevant for Bhutan. While several donors are engaged in this sector, there are opportunities for the Bank to continue supporting this sector. Relevance of design and implementation: Although the core design and implementation arrangements remain extremely relevant, valuable lessons learned from the project should 11 be taken into account while designing future projects. Likewise, while project implementation process turned out to be relevant, efficient and effective for achieving the project objective, additional capacity building measures on quality management, local institution building, fiduciary and M&E are recommended. The implementation model has considerable relevance, especially due to decentralized implementation at Dzongkhag and Geog level, participatory identification of investment priorities, O&M of assets with the help of user associations, Geog financing for O&M maintenance, and districts supporting O&M in emergency situations. Rating: Considering these factors, the relevance of objectives, design, and implementation is rated substantial. 3.2 Achievement of Project Development Objectives As per the 9th Fifth Year Plan, over 800 km of existing schemes were to be rehabilitated and over 300 km of new irrigation channels were to be constructed4 in Bhutan. DRDP contributed 65% and 14% to the national targets by supporting construction of 42.5 km of new channels and rehabilitation of another 522 km. Satisfactory. The project was successful in achieving its objective. The following achievements demonstrate the direct correlation between the project outputs and the resultant outcomes. For example: (i) access to the nearest road point has greatly improved as now it is within ½ an hour reach for 78.2% HHs, (ii) with the construction of farm roads and PPTs, farmers are able to transport cash crops and produce, such as cardamom, oranges, rice, maize, potatoes, vegetables directly to agriculture markets, and are earning more by improving their net margins, and (iii) following the construction of farm roads, several service facilities such as basic health units, construction of new buildings for primary schools, RNR centers, and community centers are now available at closer distance within the Geogs. The project has made valuable contribution in enhancing the productivity of major crops such as rice, maize, potato through double cropping, provision of quality seeds, improved farm management practices, improved extension services, rehabilitation of irrigation schemes and post-harvest support. Overall, the project has made a tremendous social and economic impact on livelihoods of the beneficiaries. The details of the project’s major outcomes and achievements are as follows: (i) Improving market access for rural communities • In terms of walking distance to the nearest road point, an evaluation in 2011 found that only 50.6% HHs were within walking distance of 1 hour to the road point. The project has greatly improved the accessibility of rural households to the nearest road 4 DRDP Project Appraisal Document (2005) 12 as 78.2% HHs are within a walking distance of ½ an hour, 14.5% HHs are between ½ to 1 hour of walking distance, 4.4% HHs are between 1-3 hours, and only 2.9% HHs are between 4-6 hours of walking distance. • Prior to the construction of farm roads, farmers were spending significant amounts of money for the transportation of cardamom, oranges and other cash crops to the market. The project-built farm roads have enabled the farmers to transport these crops directly to the agriculture markets. As a result, their net margins have increased and household incomes have gone up. An increasing number of farmers are now planting cash crops including cardamom and oranges. • Following the building of farm roads, several services and facilities are now available at closer distance within Geogs. For example, basic health unit, construction of new buildings for primary schools, RNR centers and community centers have opened up across the project areas. People are able to access extension services at closer distance, and extension workers are able to reach the rural communities with relative ease, unlike in the past when the workers had to walk long distances on one hand and farmers had to leave their farms and gather at a common point on the other hand. • Introduction of public transport service, purchase of private vehicles by local residents and improvement of existing houses and construction of new housing units have provided significant economic stimulus at the local level and have created new economic opportunities at the local level. • As a multiplier effect, formation of agricultural groups (mainly for vegetables) and establishment of livestock farms have picked up. This created significant impacts on income and cash at hand to women as most of the vegetable group members are women. • Access to credit has also improved with road construction. Mobile banking facilities and commercial banks have opened up branches in several Geogs. About 21% of the households availed loans primarily for the following purposes: (i) purchase of seedlings such as cardamom and orange, (ii) acquisition of improved cattle, (iii) house construction, (iv) buying of power tillers, power chain, equipment to minimize post-harvest losses, (v) establishment of agricultural farms (mainly livestock related), and (vi) commercial enterprises. By and large, all Geogs and Dzongkhags connected with farm roads and PTTs, have experienced increased economic activities in the form of new shops, enterprises and increased access to essential edibles and other items. Local vendors are regularly collecting vegetables and dairy products from the farmers and selling them in Thimphu and other towns. Farmers’ transportation costs for taking oranges, ginger and vegetables to agriculture markets have been reduced significantly. There is enough anecdotal evidence to suggest that the benefits enjoyed by the project communities include: • reduction in post-harvest losses as the yield reaches the market relatively quickly, and at less cost; • increased access to the regional and international markets; 13 • middlemen/buyers are able to reach the farmers easily and buy produces at farm gate; • new business and job opportunities for local people such as transport and allied businesses including tyre repair, auto workshops, oil and filter change, auto mechanics, repair of agricultural implements, agri inputs supply, sale of auto parts, grocery stores, food and teas stalls. (ii) Increasing agricultural output for rural communities The project has contributed significantly in enhancing productivity of major crops such as rice, maize, potato through double cropping, provision of quality seeds, extension services, rehabilitation of irrigation schemes and post-harvest support. The achievement of PDO indicator 2, “15% increase in food crop production (rice and maize) from baseline (2009) for targeted Geogs” and PDO indicator 3, “15% increase in cash crops production (potato and mandarin) from baseline (2009) for targeted Geogs” has to be assessed by considering the following factors: • Due to the lack of DRDP-specific baseline to assess any increase or decrease in production or yield, it was agreed to use the national Agriculture Statistics Data 2009 as baseline. • Crop production output was dependent on a variety of external influences beyond the project’s control, in particular the loss of agricultural land to other non-agricultural uses, increased rate of agricultural land being left fallow due to labor shortage and exposure to wildlife damage. Thus, any productivity enhancement in terms of increased yields could be offset by the loss of land and could potentially not result in an increase of production per Geog or Dzongkhag. • Since DRDP did not work on all the irrigation schemes in a Geog, production gains in project-schemes could be offset by the production losses in non-project schemes within the same Geog. This begs the question whether the result indicator “production per Geog” was appropriate, or if measurement of productivity in terms of “yield per acre per project irrigation scheme” would have been a better choice. This, for example in the case of paddy, is supported by case-studies of very successful “project-farmers” in selected schemes of Thedtsho (Wangdue) and Sampheling (Chukha). • A comparison of agricultural production from one year to another generally runs the risk of being influenced by annual fluctuations due to extreme weather conditions or occurrences of pest and disease attacks. Accordingly, the results can change dramatically by choosing different reference years. In order to assess general production or productivity trends it is more useful to compare the averages based on 2-3 years. The below graphs exemplify the variability between years for the paddy production in Bhutan. 14 A comparison of project Geogs or Dzongkhags to non-project Geogs or Dzongkhags is challenging as DRDP benefitted non-project areas as well. An example is the distribution of GLS-resistant maize varieties to GLS affected project and non-project Geogs and Dzongkhags (e.g. 3 non-project districts Luenthse, Trashiyangtse, and Pemagatshel received 4,460 kg of seed covering 297acres), or distribution of 9.7 mt of potato seed in Haa and Mongar districts. The use of “control-Geog” or “control- Dzongkhag”, therefore is of limited use, but nevertheless may serve as an orientation. Geog level data were only available for up to 2012, whereas Dzongkhag level data was available up to 2013. Therefore, the production or productivity increase in 2013 or in 2014 could not be reflected in the ICR. It is especially relevant for potato, as the aeroponic seed production unit set up by the project at Paro research station will only release its first batch of 3G potatoes in 2016 for further multiplication by farmers and from there on for commercial production in 2017. Hence, the impact of the adoption of improved potato seeds cannot be measured at this point in time. In consideration of the above, the ICR mission decided to use Dzongkhag data to report on increase or decrease of production area, production volume and yield for rice, maize and potato. “Total project” includes Chukha, Dagana, Trongsa, Zhemgang, Tsirang and Wangdue from DRDP Phase 1, and Punakha, Sarpang, Mongar Trashigang and Samtse from DRDP Phase 2. “Total control” includes 2-4 non project districts. 15 Table 1: Paddy Production % increase PADDY - DZONGKHAG 2013 over 2009 TOTAL PROJECT Area (acres) -18.3% Production (mt) +18.2% Yield kg/acre) +44.6% TOTAL CONTROL (Samdrupjonkhar, Lhuntse, Haa, Paro) Area (acres) -19.2% Production (mt) +1.5% Yield kg/acre) +25.6% The project supported Dzongkhags achieved the target by increasing paddy production by 18.2%, despite heavy land losses of -18.3%, and on average, the yield increased by 44.6%. In contrast control Dzongkhags experienced an increase of production by 1.5% and yields by 25.6%. Table 2: Maize Production % increase MAIZE – DZONGKHAG 2013 over 2009 TOTAL PROJECT (11 project districts) Area (acres) -23.7% Production (mt) +12.9% Yield kg/acre) +48.0% TOTAL ADDITIONAL (3 non-project districts, but receiving project support) Area (acres) +16.0% Production (mt) +69.1% Yield kg/acre) +45.8% TOTAL DISTRICTS RECEIVING DRDP SUPPORT (14 districts) Area (acres) -18.8% Production (mt) +21.3% Yield kg/acre) +49.4% TOTAL CONTROL (Haa & Paro) Area (acres) +18.4% Production (mt) +25.2% Yield kg/acre) +5.8% DRDP fully achieved the maize production target of +12.9% in the 11 project districts despite heavy land losses of 23.7%. Furthermore, the DRDP distribution of GLS resistant maize varieties to three non-project districts (Pemagatshel, Lhuntse and Trashiyangtse) in 2011- 2013 resulted in impressive production and yield increases of +69.1% and +45.8% respectively. As a whole, DRDP support led to an average maize production and yield increase of +21.3% and +49.4% respectively. 16 Table 3: Potato Production % increase POTATO - DZONGKHAG 2013 over 2009 TOTAL PROJECT Area (acres) +18.7% Production (mt) +10.7% Yield kg/acre) -6.7% TOTAL CONTROL (Haa, Paro, Lhuntse & Samdrupjonkhar) Area (acres) +3.9% Production (mt) +1.2% Yield kg/acre) -2.7% DRDP fully achieved the potato target by increasing production by 10.7%, in comparison with 1.2% production increase in the control districts. Nevertheless, an important contributing factor to this growth could be the increase in acreage, as in several cases farmers were confronted with deteriorating seed quality. It is reflected in an average yield decrease of 6.7% during the last years when comparing averaged yield data from the period 2011-2013 with yield data from the period 2009. DRDP implemented counteracting measures by: (1) replacing seed potatoes by moving seed potatoes from higher altitude farmers to lower altitude farmers, and (2) producing disease free improved potato seeds through aeroponic culture. It is important to note that the full project impact is expected in 2017 when the first improved potato seed produced through the aeroponic unit will be released to the farmers. A more detailed analysis shows considerable success in increased production in 4 of 11 project Dzongkhags, namely Chuka (+72.1%), Trongsa (+99.6%), Mongar (+36.7%) and Trashigang (+73.6%). In these districts DRDP fully achieved its target of at least 15% increase in production. In summary, the overall achievement is as follows: Indicator Target Actual Value Increase in food 15% increase above the Paddy: + 18.2% (+44%) crop production Baseline of 2009 (average of 11 Dzongkhags) Maize: + 21.3% (+49%) (average of 14 Dzongkhags) Increase in cash 15% increase above the Potato: + 10.7% (-6.7%) crop production Baseline of 2009 (average of 11 Dzongkhags) 3.3 Efficiency Pre-Investment Economic and Financial Analysis No overall rates of return were provided in the financial and economic analysis annex of the February 2005 PAD and no discussion was presented of irrigation investment benefits. However, calculations were presented as to expected improvements in travel times due to investment in road access in the six Dzonkhags (districts) covered by the project under Component 1. Secondary data was also provided on the experience of changes in agricultural production following road construction, the variance in household incomes 17 between those with and without road access, and educational enrolment rates with and without road access. No estimates of the impact of irrigation infrastructure provision (also under Component1) were given. Under Component 2 (RNR Centres) and Component 3 (Institutional Strengthening) information was given only on construction targets (6 new RNR Centres) and training (5 course types), respectively. Post-Investment Economic and Financial Analysis No estimates of direct benefits to road construction under Component 1 have been made as there is insufficient data available. The principal impact of the irrigation investment (also in Component 1) was to permit the expansion of wetland paddy production, and this was critical for rice production. The benefits arising from the investment in irrigation are thus subsumed within the crop income calculations. However, the beneficiary survey undertaken by the borrower shows that improved road and power tiller track access has important benefits beyond the key crops of paddy, maize and potatoes. Nearly half of all households benefitting from improved access reported increases in vegetable production, while smaller percentages increased dairy and citrus output. Commercial activity (trading, processing etc.) also increased. These benefits are not captured by improved productivity of the target crops. For Component 2, the completion survey determined that 12-18 percent of potential households in the 11 project Dzongkhags were beneficiaries of support to rice, maize and potatoes. The survey also identified percentage output increases for each crop. As there is no breakdown in investment costs for individual crops, it is not possible to calculate returns for each crop. No attempt was made to quantify return to institutional strengthening activities (Component 3). Whole Project Analysis In all, the project provided benefits across 7,609 hectares and may have affected as many as 35,000 households. This latter figure must be treated with care, however, as some households may be active in more than one of the crops, leading to double counting. In total, the crop-related benefits associated with the project reach over USD 1 million per annum, although there is no longitudinal information to determine the year in which such benefits were first generated. This constraint has been addressed by making conservative assumptions as to the build-up of benefits over time. No crop benefits at all are assumed for the first three years of the project and full returns are only achieved at Year 6 of implementation. Financial returns to crop productivity improvements under the project were not only strongly positive, but robust. Even when including all irrigation investment costs, returns from crop productivity improvements reach almost 30 percent and remain above 20 percent even when the results were tested against a 20 percent decline in crop benefits. A whole project analysis based entirely upon crop-related benefits yields negative financial and economic rates of return as approximately 60 percent of total project expenditures were directed at road and power tiller track improvements and construction. In order to assess the financial returns to total project investment therefore, and on the 18 basis of results from comparable projects in Bhutan and elsewhere in Asia, calculations were undertaken based on assumed increases in incomes for the estimated 6,000 households directly benefitting from new and improved farm roads and power tiller tracks. The results are shown in the table below. It can be seen that even under the assumption that access to rural roads or power tiller tracks increases beneficiary household income by only 10 percent (considerably lower than results from comparable projects), a FIRR of over 15 percent is achieved, while if household incomes increase by 15 or 20 percent, the return to the project increases to approximately 20 and 25 percent. Two main factors were considered for the economic analysis; labour costs and exchange rates. Although labour costs are significantly higher in Bhutan than in other regional countries, so is productivity and a 2010 World Bank study considered labour rates to be competitive. No adjustments were therefore made in this factor. A 2014 IMF review stated that there was a significant over-valuation of the Ngultrum (estimated at 16 percent), but as imported items comprise such a small portion of crop production costs, the adjustment is minor. EIRRs for the scenarios considered increased only marginally, to 15.5, 20.5 and 25.1 percent. On the basis of this analysis, therefore, and even in the absence of considerable key performance data, the project it is considered likely to have been highly satisfactory in both financial and economic terms if the above assumptions of moderate increases in beneficiary household incomes is accepted. It is worth noting that over 35 percent of respondents interviewed in the beneficiary survey noted that they had increased vegetable production as a result of road access, while 72 percent stated that market access had improved. In addition, other important non-quantified benefits for which little data is available include expanded dairy production (rendered feasible by improved road access), small enterprise development and improved access to educational, health and employment opportunities. 3.4 Justification of Overall Outcome Rating Rating: Satisfactory Relevance: As explained in Section 3.1, the relevance of objectives, design, and implementation is rated as substantial. 19 Achievement of PDOs: As explained in Section 3.2, the project has many satisfactory achievements to its credit. Thus, based on the extent of outcome of the project, the achievement of PDOs is rated as satisfactory. Efficiency. Various examples shown in Section 3.3 and Annex 3 have demonstrated that there were operational and administrative efficiencies which came as a result of design choices and the actual implementation approaches brought under the project. Based on the above factors and the discussion in sections 3.1–3.3, the overall outcome is rated as satisfactory. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development Poverty Impacts: (i) Availability of access roads, irrigation channels, improved knowledge and skills and increased income from agricultural products have helped villagers to purchase household assets including television, rice cookers, refrigerators, furniture, livestock. Many households reported to have purchased vehicles (cars, pick-ups and trucks) including power tiller following the construction of road and PTT. For example, people from Barshong Geog (Tsirang) reported that 80% of the households have repaired their houses or constructed a new unit as the road enabled them to get construction materials quickly at lower rates. Another multiplier effect is the economic transformation at the local level by increased mobility to and from the project villages, increased access to jobs and employment, and opening of new small businesses in the villages, on road junctions and in the nearby towns. The project provided assets to the project partners such as aeroponics unit, field equipment and more importantly, e-pest surveillance system. (ii) DRDP had a substantial impact on enhancing livelihood of rural people and in reducing poverty, especially in remote and far flung communities. For example, as per the borrower’s Impact Assessment and Project Completion report, in 2014 the annual income per HHs was above Nu. 30,000 for 43.5% HHs, which is an increase by 354.3% as compared to HHs with this level of income in the past six years. (iii) With the availability of road/transport and ease in marketing, intensification of vegetables, with new varieties including potato and maize, and increases in paddy production, it has been reported that there is improvement in nutrition at households. Gender aspects: (i) Women comprised 45% of the direct beneficiaries of the project. Due to increased mobility, several producer groups (mainly women vegetable groups) have established 20 their own farms and marketing systems. It has caused significant positive impacts on women’s social and economic status including increased income and cash in hand. (ii) Reduction in travel time has significantly reduced multitasking burden of women as well as the opportunity cost of their time and dependence on men. It increased their productivity and income, enhanced their mobility, and improved their access to health and education services. With improved transportation facilities resulting from new roads, women are able to travel to nearby towns, sell their produce, and earn higher incomes. Women’s increased incomes have reduced their vulnerability. Many women are reported to have savings and have opened bank accounts. As new schools and hostels for students have been constructed in the project areas, more girls are now attending schools. (iii) Agricultural mechanization has helped to reduce workload for women. For example, construction of power tiller tracks have helped to reduce their workload and working hours. Females are using power tillers not only to plough fields but also for several non- farm purposes. Women do not have to carry loads of essential items themselves as in the past; now they use power tillers in groups to bring in essential food items, agricultural inputs, fish, vegetables, medicinal plants and dairy products. Social development (i) The project has made significant impact in improving social harmony, community networks and overall social cohesion amongst rural population. The farm roads and bridges in critical locations have improved the access and linkages between and amongst: remote villages, rural townships, state service delivery, and marketing of local produce and commodities. Better transportation facilities have improved communication amongst rural communities and enabled them to mingle with each other more frequently. In addition, improved social interactions have resulted in uplifting the confidence level of people living in remote areas. It has been reported that ambulance services are now accessible to remote locations, making it easier for emergency medical treatment at the nearest basic health unit and hospital. (ii) A greater sense of community ownership has evolved due to the fact that, communities were directly responsible for identification and implementation of infrastructure such as irrigation channel rehabilitation, PTT and farm road. Local institutions such as water user groups and farmer producer and marketing groups also enhanced the trust and confidence (social capital) among farm producers and women. (b) Institutional Change/Strengthening (i) Capacity of farmers have been strengthened through focused training programs, skills enhancement and exposure visits related to crop production, demonstrations and trails on improved variety of seeds, seed production and selection, integrated pest management and demonstration on farm machinery, amongst others. The project provided valuable training on hill irrigation and various orientations on environmental and social safeguards. 21 (ii) DRDP’s capacity building efforts have had good impacts on project staff and rural population. MoAF staff, and extension agents now have adequate knowledge, understanding and skills for farmer group formation, paddy, maize and potato production, improved farm management, pest control, livestock management, etc. (iii) Government engineers have acquired greater knowledge and are better skilled for irrigation survey, design, construction technology, water management, operation and management which guided and facilitated effective project implementation. Significant contribution has been made in the form of capacity building on aeroponics for potato seed production. Officials and farmers have better understanding and skills in the areas of post-harvest and value addition. Another significant impact is the development of e-pest surveillance system for scaling up. (iv) Dzongkhag officials have acquired a deeper experiences and understanding of environmental and social issues and are well trained in undertaking social and environmental screening and ensuring compliance. (v) MoAF and government departments at Dzongkhag and Geog have greater technical, financial management, procurement, implementation and O&M capacity to adequately design, budget, implement, operate and manage government programs, annual plans and funds. (c) Other Unintended Outcomes and Impacts (positive or negative) • The use of pipes for irrigation has not only minimized losses through infiltration, but also, provided useful exposure to the engineering staff. The engineers now have adequate knowledge and experience in irrigation survey, design and water management. It helped to introduce lift irrigation and water harvesting interventions. • DRDP enabled the National Potato Program (NPP) and the National Seed Centre (NSC) to produce potato seed tubers using aeroponic facility at the NSC complex since April 2012. It is the first of its kind in the country for seed potato production. The facility was built using one of the greenhouses that already existed at the NSC. • The project was instrumental in the promotion and spread of biotic and abiotic stress tolerant crop varieties (drought tolerant rice varieties and disease resistant maize varieties). The positive selection process to raise awareness about the importance of quality seed to produce quality seeds and to impart knowledge on safe storage of seed for next season were important innovations of the project. The farmers were imparted with the knowledge of quality household seed production to sustain easy seed access within the community. Farmers were trained in site selection, variety selection, rouging at different crop stages in the standing crop, processing, labeling, packing and storage, etc. • The e-pest surveillance system is another important innovation of the project that is designed to capture information of all pest organisms (insect pests, disease organisms, weeds) that causes or have the potential to cause damage and loss to crops over time and space.y 22 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops Highlights of the borrower commissioned Impact Assessment and Project Completion Report are as follows: • The project has made good achievements in successfully implementing activities as planned: (i) 101% achievement for new farm roads construction, (ii) 102% for farm road improvement, (iii) 100% for power tiller track construction, (iv) 100% for construction of new irrigation channels, (v) 99.48% for rehabilitation of irrigation channels, (vi) 100% for construction of RNR/Geog Centers, (vii) 83% for construction of bridges, and (viii) 276% for farmers’ trainings/demonstrations and trails. • Productivity (yield) of paddy increased by 36.4% and production by 6.4% in 2012 as compared to 2006 and for potato there was 31% increase in production in 2013 as compared to 2006. In case of maize, there was 30% increase in yield and 11.5% increase in production in 2013 as compared to 2000. • Some of other key outcomes are: 19% higher yield than local variety for Bajo Maap 2 and Khangma Maap (paddy), 49% higher yield than local variety for IR-64 (paddy), 17345 minitubers of potato harvested in 2013 from aeroponics, and an increase from earlier 5 tons production to 60 tons potato seed from Phobjikha National Seed Centre (NSC) in 2012. • 18.5%, 18.1% and 12.2% of respondents respectively experienced increased paddy maize and potato production. 12.7% households witnessed increased cash income, 12.6% benefitted from improved marketing, 12.5% households’ walking distance was reduced significantly. Other benefits included availability of transportation vehicles, saving labor for agriculture and service facilities being available at closer distance. The marketing of agricultural products have improved in the past six years for 71.9% of respondents, and for 2.78% it remained the same as before. • In terms of walking distance to the nearest road point, it is within ½ an hour for 78.2% HHs, between ½ to 1 hour for 14.5% HHs, between 1-3 hours for 4.4% HHs, and between 4-6 hours for 2.9% HHs. It shows a substantial improvement in access to market, as in 2007 the nearest road was within 1 hour walking distance for 45.7% HHs, which improved to 50.6% in 2011 and now to 78.2% HHs having nearest road within ½ an hour walking distance. 4. Assessment of Risk to Development Outcome Rating: Moderate The ICR team considers that the risk to development outcome would be moderate in light of the following factors: • RGOB has adopted and incorporated the DRDP approach, learning and best practice into the 11th FYP covering five year period of 2013–2018. It shows that to the extent feasible, DRDP experience and best practice has been mainstreamed at the appropriate levels in the MOAF, Dzongkhags and Geogs. 23 • In the annual budget, the RGOB has allocated funds for operation, maintenance, renovation and rehabilitation of infrastructure include DRDP interventions. Besides coordinating with the farmers’ user groups for O&M, the government’s financial and human resources are available in the event of natural calamity/emergency such as landslides, floods, torrential rains and any major damages to infrastructure. • Dzongkhags and Geogs have the ownership of the DRDP delivered interventions and are responsible for O&M. For this, they have dedicated: (i) expertise and staff, (ii) equipment and machinery, and (iii) budgetary allocation of NU 2.0 million annually is available to each Geog for O&M, are good indications of post-project sustainability. Likewise, Geogs are committed to maintain and sustain the project interventions. For example, Geog Yargey Tshogdus (GYT- Block Development Committees) have been mobilizing community labor to maintain the assets. Communities are also proactively interested and involved in O&M. • As mentioned in Section 2.5 for infrastructure investments, Road User Groups (RUGs) have been formed for regular maintenance works including minor repairs. As per the government regulations MoUs are not required to be signed with any benefiting communities, instead a handing over document is signed for O&M after the completion of works. • As stipulated in the farm road guidelines and national irrigation policy, beneficiary communities are responsible for minor O&M activities (drain cleaning, spot filling, small debris removal from form roads and PTT, vegetation clearing, blockage removal, intake and outlet structures maintenance for irrigation channels, checking pins and joints for bridges etc.) • Enhanced staff capacity resulting from the project has been well internalized, especially at the ministry, Dzongkhag and Geog level. For instance, procedures and processes such as M&E, gender, social and environment safeguards practices etc. are already well integrated within the RGOB’s institutional framework. • In some cases, DRDP initiatives need further strengthening of technical know-how and up-scaling for sustenance. For example, NSC which has taken over the aeroponics for seed production would require continued support for technology improvement and up-scaling. DRDP initiated system for poly-crops and e-pest surveillance needs to be scaled up and would require capacity building of users on a regular basis. Similarly, adequate research funds are required to further improve the seed farms, quality of seeds and for promotion and replication of improved varieties of seeds. The government is fully aware of this situation and is committed to meet the capacity building and funding needs to these important project interventions. 5 Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Satisfactory A Quality at Entry Assessment (QEA) of the project was carried out by the Quality Assurance Group (QAG) in July 2005, and it received a satisfactory rating. It comprised highly satisfactory rating for strategic relevance and approach, satisfactory rating for risk 24 assessment, fiduciary, environmental, policy and institutional aspects, and moderately satisfactory for poverty, gender and social development, and implementation arrangements. QAG remarked that: • Focus of the project is on the binding constraint to Bhutan’s development (viz. rural access) and its relevance is, therefore, self-evident; • The project is realistic in scope, time frame, complexity, and within the implementing institution's capacity; • The DOs are simple, clear and realistic; • The combination of SIL and a programmatic approach, the selection of the areas of focus, the decisions on which areas to exclude from the project were all appropriate and high-quality; • The project was kept simple by excluding all but the essential elements for increasing rural access. Most experiences and lessons learnt from previous and similar operations were captured in the project approach; • Though much AAA had not been done by the Bank on agriculture and decentralization, the team diligently sought out experience of other donors, studied conditions on ground, and applied lessons learnt; • The economic rationale was well presented and the analysis was kept simple and sensible: Conditionality was well focused and kept simple, mainly on fiduciary aspects. The financing plan was adequate and well constructed; • The early adoption of country systems and simplified interim financial reports is commendable; • The PAD is an exemplary document, clear, candid, simple project design, practical results framework, and well balanced in its coverage of different elements. The task team deserves to be applauded for this. During preparation and appraisal, the Bank took into account the adequacy of project design and all major relevant aspects including technical, financial, economic and institutional, including procurement and financial management. Several alternatives were considered for the project design. In addition, major risk factors and lessons learned from the earlier projects in the rural development sector were incorporated into the project design. Nevertheless, as mentioned in Section 2.1, the project design had some drawbacks. The overall quality at entry is, therefore, rated as moderately satisfactory. (b) Quality of Supervision Rating: Satisfactory. The Bank's performance during project implementation was overall satisfactory. During the initial phase, high turnover of TTLs and lack of a designated TTL for some period, affected project implementation. Subsequently and especially in the last few years, the Bank task team has provided high quality implementation support to the government. Sufficient budget and staff resources were allocated, and the project was adequately and intensively supervised. The task team prepared Aide-Memoires regularly and alerted the government and PMT to any issues and challenges the project was facing and facilitated remedies in a timely manner, in conformity with the Bank procedures. 25 The Implementation Status Reports (ISRs) realistically rated the performance of the project both in terms of achievement of development objectives and project implementation. The task team also monitored safeguard and fiduciary compliances. The Bank conducted a MTR in May 2008, and assessed progress on all project components, implementation issues and the actions to be taken to ensure a successful completion of the project. (c) Justification of Rating for Overall Bank Performance Rating: Moderately satisfactory ith a moderately satisfactory rating for quality at entry and a satisfactory rating for quality of supervision, the overall Bank performance is rated as moderately satisfactory in accordance with the IEG’s harmonized rating criteria. 5.2 Borrower Performance (a) Government Performance Rating: Satisfactory As mentioned in Section 2.1, RGOB had shown commitment to the objective of the project from conception to completion. The Minister, Secretary, and Project Director of MoAF consistently maintained their commitment throughout the implementation period. The government officials worked closely with the Bank’s task team on a regular basis, and cooperated fully with the Bank staff. Appropriate levels of review and approval were usually in place. Financial accountability and follow-up was observed, and expenditures were duly authorized before they were incurred, and documentation was maintained properly for periodic review. The project did not suffer from any counterpart funding problems, as RGOB took timely corrective measures and made appropriate budget provisions. (b) Implementing Agency or Agencies Performance Rating: Moderately satisfactory The MoAF and the Project Management Team (PMT), in particular, were fully committed to the project. PMT was led by the same Project Director throughout the life of the project, which helped to ensure consistency in project leadership and institutional memory. The project was an integral part of the RGOB’s annual program and budget. As a result, the project enjoyed a high level of ownership both amongst the government officials and communities. The project director and other government officials regularly visited project communities and were able to streamline and strengthen the implementation arrangements. The field staff spent significant amounts of time with the communities to provide the much needed guidance and support for identification, design, and implementation of project interventions. The local political leaders, especially Geog administration heads were fully engaged in project implementation. Financial Management. Overall the FM performance is rated as moderately satisfactory. For details see Section 2.4 26 Procurement: The procurement performance is rated as moderately satisfactory. For details see Section 2.4 M&E: As mentioned in Section 2.3, Dzongkhags and other collaborating partners collected information from Geogs on needs basis. Geog officials regularly collected and submitted data to Dzongkhag. The data collected from Geog was analyzed and used for informed decision-making. Nevertheless, a dedicated project specific data collection and analysis system was not established. Consequently, in-depth analysis and reporting of project implementation progress remained weak. The performance of M&E system is rated as Moderately Satisfactory. (c) Justification of Rating for Overall Borrower Performance Rating: Moderately satisfactory With a satisfactory rating for government performance and a moderately satisfactory rating for implementing agency’s performance, overall borrower performance is rated as moderately satisfactory in accordance with the IEG’s harmonized rating criteria. 6. Lessons Learned The lessons emerging from the DRDP are as follows: Project Design: • For a decentralized project like DRDP, in a country with considerable capacity constraints, institutional strengthening, streamlining of procedures and capacity building of staff on a regular basis have to be an integral part of the project design. It calls for a proper capacity appraisal of implementing/executing agencies. The assessment should include institutional strength, skill and knowledge, past experience in designing and implementation of similar projects, financial management and procurement system capacity, etc. In line with the capacity needs assessment; a comprehensive capacity building component needs to be made a part of the overall project design. • For a small-sized country like Bhutan, it is pertinent to have an integrated and simple project combining various sectors such as infrastructure and agriculture for sustainable impacts on ground, and greater ownership of the project from the government and rural communities. Implementation: • When an umbrella Ministry is the same, projects could benefit from sharing learning and experience of staff engaged in various past and present Bank-funded projects. While the Bank staff would be available for support, getting hands-on help and support from the experienced colleagues is a valuable source to be utilized fully. • Community ownership is critical for the sustainability of project infrastructure. DRDP successfully revamped the exiting informal mechanisms of water channel maintenance by forming Water Users Associations (WUAs) and Road Users Associations (RUAs) for roads. It helped to ensure ownership and long term 27 sustainability of project interventions. WUAs and RUAs have worked well for operation and maintenance of irrigation channels and roads. The formal registration envisaged by the project and even by law (land act) did not prove to be successful, as WUAs/RUAs largely preferred to work informally. Future projects should continue forming such associations with a greater focus on orientation and training about O&M and raising funds for O&M, rehabilitation and replacement. Registration under relevant law/act; MOU with Geog administration; and compliance of maintenance plan (as given, for instance, in “Farm Road Guidelines by MOAF”) would be important for long terms sustainability of the project. • RGOB’s prevalent institutional arrangements largely worked well. Contrary to creating a separate project implementation unit with dedicated project staff, DRDP relied on the existing government system and staff. This cost effective implementation arrangement helped strengthen the devolved service delivery mechanism, ensured bottom-up grassroots participation, as all interventions were identified, planned and budgeted at the Geog level and approved at the Dzongkhag/central level, and created project ownership at all levels of government and communities. Monitoring: • DRDP has fully achieved or exceeded all project targets. The project did not develop a dedicated monitoring and reporting system; instead, need-based reporting took place. Data collection and reporting remained weak. For future projects, it is important to create an effective monitoring and reporting system supported by a dedicated tracking system and baseline information. • Integration of DRDP approach and interventions into regular government programs can be seen as best practice, and could benefit considerably by improved M&E either through dedicated project M&E staff or by strengthening the M&E capacity of MoAF, Dzongkhag and Geog. Fiduciary: • The use of country fiduciary system has worked well. As the country’s Public Financial Management system is being strengthened (either through the Bank’s support or otherwise), the FM capacity at the project level is also improving. Capacity building of fiduciary staff and their familiarity with the Bank’s fiduciary requirements at the initial phases of project implementation is important. Sector-specific: • DRDP experience shows that for community managed small-scale interventions, it would be beneficial to include a third party monitoring of project interventions on a regular basis. • Orientation and capacity building of project staff and community members in ESMF implementation and compliance should be built into the project design. It should include handling and management of construction spoils, plantation/bio- 28 engineering works, health & safety at construction, preparation of screening, site- EMP, and hill irrigation. • In several developing countries including Bhutan, rural access is the binding constraint to poverty reduction and needs to be addressed upfront. DRDP is exemplary in this regard. Yet, once this constraint abates, the resultant growth may not respond to meet the expectations of rural population unless an early start is made to identify both traditional and new sources of growth in rural areas. • Future projects will need to pay greater attention to agriculture value chains, agri- business, seed production and storage and marketing facilities. • There is a need to be careful in agriculture diversification. The Bhutanese experience during the Indian ban on food grains from 2007-2009, led RGOB to prioritize food self-sufficiency and introduce carefully planned proportional diversification. Arguably, cash crop is important but it should not be at the cost of food crops. • Project interventions in infrastructure development (irrigation schemes, farm roads, market places, etc.) should be aligned with the capacity building needs in the same Geog/beneficiary groups, such as formation or strengthening of producer groups, marketing groups, improved production technologies, improved seeds, appropriate extension services, etc. • In order to increase food security/self-sufficiency, promotion of key commodities through infrastructure and capacity building should be aligned to possible economic clusters to fully exploit the available agro-ecological potential. • Restriction on project intervention areas (only certain Geogs/Dzongkhags) may restrict the impact if a project works on select commodities (area approach versus value chain approach). In fact, DRDP had to work outside the designated project Dzongkhags in order to effectively address the Grey Leaf Spot disease in maize, or to enhance seed potato production in climatically favorable Dzongkhags such as Haa, which was not part of DRDP. Important factors for the success of a project: • Continuity of project staff, led by an experienced project director, capacity building and training, community ownership and proactive participation in project design and implementation are important both for the Bank and the Borrower. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies Ministry of Agriculture and Forests, Royal Government of Bhutan, reviewed the ICR document and did not offer any comments. (b) Cofinanciers N/A (c) Other partners and stakeholders (e.g. NGOs/private sector/civil society) N/A 29 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Appraisal Percentage Additional Actual/Latest Estimate of Appraisal Components Financing Estimate (USD (USD millions) (USD millions) millions) Component 1: Rural Infrastructure 5.10 2.42 7.81 153 Component 2: Renewable 0.75 1.92 2.97 396 Natural Resources Centers Component 3: Institutional 0.35 0.66 1.09 311 Strengthening Total Baseline Cost 6.20 5.00 11.87 Physical Contingencies 0.28 Price Contingencies 0.52 Total Project Costs 7.00 5.00 11.87 191 Front-end fee PPF Front-end fee IBRD Total Financing Required 7.00 5.00 11.87 191 (b) Financing Appraisal Actual/Lates Type of Estimate t Estimate Percentage of Source of Funds Cofinancing (USD (USD Appraisal millions) millions) Borrower 0.00 0.00 0 International Bank for Reconstruction 7.00 7.00 100 and Development Trust Fund (Additional Financing) 5.00 4.87 97 30 Annex 2. Outputs by Component Component-1: Rural Infrastructure Farm Roads The project overachieved the target by 162% and 144% respectively by constructing 138.27 km of new roads exceeding the target of 89 km at appraisal and 95.5 km at DRDP-AF. Against an output target of maintaining 53 km of road, the project maintained 69.35 km of road, which shows an impressive achievements of 131% against the DRDF-AF target. While appreciating this achievement, the Impact Assessment and Project Completion report 5 noted that the farm road improvements were done in certain places and not for the entire length of the road. Based on the field observations and witnessing the pliability of sampled improved roads, the ICR mission is in agreement with the finding of the Completion Report. Figure1 illustrates the achievement of outputs for new roads construction against the target. Power Tiller Tracks: Power tiller track is the first level of improvement to the existing pedestrian way/mule track to permit passage of power tillers while leveling, grading and widening (up to 3 meter) of earthen path. The project exceeded the target by 113% and constructed 129 km of PTT against the target of 114 km at appraisal. Irrigation Channel: The project achieved 96% of the target and constructed 42.5 km of new irrigation channels against a target of 44.3 km at appraisal/DRDP-AF. The output with regards to improvement/rehabilitation of irrigation channels was 521.98 km, exceeded the target of 448 km set at DRDP-AF. Bridges: The project constructed 6 bridges as per the target at appraisal. Two bridges were washed away during the 2012 cyclone that caused huge losses in Bhutan. The project rebuilt one of the destroyed bridges with the additional support of JICA. The overall output achievement level was 83% and 5 bridges were constructed against the target of 6. Summary Achievements (%) Of the 6 infrastructure subtypes, 4 types exceeded the PAD/DRDP AF targets; while the remaining two fell slightly short of the target values as explained above. Fig: 2 reflects these % achievements by infrastructure type. 5 The Impact Assessment and Project Completion Report initiated by Department of Agriculture and Forest 31 Geographical Outreach and Beneficiaries Farm Roads: New and rehabilitated farm roads were implemented in 6 Dzongkhags as planned at appraisal. Twenty three new road schemes - with a total length of 138 km - benefited 2,520 households with an average of 110 HHs per scheme and 18 HHs per km. Similarly, 16 rehabilitated farm roads - with a total length of 69 km - benefited 1,575 HHs with an average of 98 HHs per scheme and 23 HHs per km. Table:1 Farm Road (Improvement/Rehabilitation) by Dzongkhag No. of S.N0. Dzongkhag Length (km) Beneficiary Households Schemes 1 Chukha 6 13.2 902 2 Dagana 3 9.0 167 3 Trongsa 3 15.3 185 4 Tsirang 2 11.3 50 5 Wangdue 1 9.0 207 6 Zhemgang 1 11.6 64 Total 16 69.4 1,575 Table:2 Farm Road (New) by Dzongkhag No. of Beneficiary S.N0. Dzongkhag Length (km) Schemes Households 1 Trongsa 6 31.8 377 2 Wangdue 6 30.9 238 3 Chukha 5 15.1 494 4 Zhemgang 3 34.2 538 5 Tsirang 2 12.0 677 6 Dagana 1 14.3 196 Total 23 138.3 2,520 32 Table:3 Farm Roads by Year No. of Year Length (km) Beneficiary Households Schemes New Construction 2008 5 31.1 320 2009 10 66.3 949 2010 8 40.9 1,251 Total 23 138.3 2,520 Improvement/Rehabilitation 2011 2 11.3 50 2012 3 12.6 498 2013 10 33.9 963 2014 1 11.6 64 Total 16 69.4 1,575 Irrigation: A total of 164 irrigation channel rehabilitation schemes were implemented in 9 Dzongkhags which benefited 4,607 HHs and 11,839 acres of farmland. These schemes included one piped irrigation scheme in Samdrup Jongkhar and one water reservoir in Punakha Dzongkhag as pilot projects. Eleven new irrigation channels benefited 418 HHs and about 678 acres of farmland in 6 Dzongkhags. Table:4 Irrigation Rehabilitation Schemes by Dzongkhag No. of Beneficiary S. No. Dzongkhag Length (km) Schemes Households 1 Wangdue 37 145.5 979 2 Dagana 30 81.85 567 3 Chukha 27 56.68 641 4 Tsirang 25 84.81 532 5 Zhemgang 21 60.9 559 6 Trongsa 12 56.2 498 7 Punakha 10 33.3 824 8 Mongar 1 2 7 9 Samdrup Jongkhar* 1 - - Total 164 521.24 4,607 Table:5 Irrigation Rehabilitation Schemes by Year 2008- Grand Year 2008 2009-10 2010-11 2011-12 2012-13 2013-14 09 Total No. of Schemes 40 10 28 34 14 31 7 164 Length (km) 105.15 17.83 101.6 103.02 56.3 111.34 26 521 Beneficiary HHs 823 321 741 706 548 1229 239 4,607 33 Achievements of Outcome To assess progress against the project development objective of “Improved market access” one simple indicator was suggested at appraisal: “Average time taken to walk to motorable road decreased by one half (from 3 hours to 1.5 hours) for targeted Geogs”. The PAD suggested to use the Ministry of Agriculture’s 10 percent sample survey of all households, as published in “Renewable Natural Resources (RNR) statistics for measuring the outcome indicator of reduced walk time. The published RNR statistics somehow did not consistently reflect such information. A field survey conducted for the project completion report in October-November 2014, covering 731 respondents of targeted Geogs, showed promising results towards the achievements of outcomes. The survey showed that walking time to the nearest road point is within ½ an hour for 78.2% (572) HHs, between ½ to 1 hour for 14.5% (106) HHs, between 1-3 hours for 4.4 % (32) HHs and between 4-6 hours for merely 2.9% (21) HHs: These results show that the walking time to motorable road is within 1 hour (against a target of 1.5 hr at appraisal) for nearly 93% of surveyed Geogs/households. While the Project Result Framework suggests one indicator (as noted above) for improved access to market, the impact assessment survey also captured access time to other facilities. Fig:3 illustrates the percent of households benefitting various services and facilities within a walking distance of less than 1 hour. The data extracted from the Annual Agriculture statistics for various years show an upward trend for the households having improved access to motorable roads in terms of walking time. For instance, DRDP PAD referred to agriculture statistics of year 2000, with 43% of HHs within one hour of walking time to motorable road. Whereas in year 2011, 69% and in year 2013 more than 71% of HHs were living within 1 hour walking time to motorable road. Further analysis of data reveals that with improved connectively through farm roads, PPT and bridges, the walking time has reduced to 1.3 hours in year 2013 from 3 hours in year 2000. Such analysis provides sufficient evidence that the project has successfully achieved its outcome objectives with regards to road infrastructure. Table:1 below shows proportion of households by walking time against various time intervals with an average estimated time to reach motorable road. 34 Table:6 Proportion of HHs by walking time to motorable Road Proportion of households by walking time taken to reach nearest motor-roads points Year Average Less than 4 to 6 >8 Hour to 1 to 3 hours (+) 6 hours 1 hour hours hour Motorable Roads Estimated Average 0.33 2 5 7.5 12.5 - in interval (hours) 2000 42.7 31.1 10.8 5.3 10.2 3 2011 69 15 6 10 - 1.6* 2013 71 18 5 6 - 1.3* Source: Agriculture Statics 2000, 2011 and 2013 * ICR Team calculations based on estimated average in interval (hours) as used in DRDP PAD-Page 22 Project Contribution in overall Irrigation Development: As per the 9th Fifth Year Plan, over 800 km of existing schemes were to be rehabilitated and over 300 km of new irrigation channels were to be constructed 6 in all twenty Dzongkhags in Bhutan. By supporting construction of 42.5 km of new channels and rehabilitation of another 522 km, DRDP contributed about 14% and an impressive 65% respectively for new and rehabilitation targets set for the overall irrigation development in 9th FYP. Table:7 DRDP Contribution in Irrigation Development for 9th FYP Targets Irrigation Channels Irrigation Channels Description (New) (Rehabilitation) 9th FYP Targets (km) 300 800 DRDP Contribution (km) 42.5 522 DRDP Contribution (% of FYP) 14.2 65.3 Key Findings Quality of Implemented Infrastructure: In view of the remote and inaccessible locations and limited capacity, ensuring quality of infrastructure was challenging. Not surprisingly, implementation largely focused on achieving the annual/project targets. For instance, in case of farm roads, the MoAF provided guidelines/specifications for base course thickness, appropriateness of gravel-earth mix, required compactions were not always followed. Consequently, ruts and pot-holes were witnessed by the ICR mission at places. In case of irrigation channels, in some cases, the inlet structures were found missing/washed away due to design flaws or poor Implementation quality. At certain 6 DRDP Project Appraisal Document (2005) 35 locations, irrigation pipes were sagging, reflecting use of lower quality pipes. Some of the contributing factors are as follows: • Inadequate monitoring by engineers, • Engineers’ limited knowledge and experience (especially during the initial years). It improved considerably through the project provided quality training to the engineers, • Capacity constraints at Dzongkhag level – limited no. of engineers, inadequate logistics, excessive workload, • Challenging tterrain and remoteness of project sites, and • Lower bid/award costs by contractors than the engineering estimates. Sustainability: Despite the shortcomings, the promising aspect is the sustained continuity of benefits generated by the DRDP supported infrastructure. An in-place, constrained, yet satisfactory and institutionalized O&M mechanism explains the likelihood of infrastructures sustainability. The key characteristics include the following; • Water/Road User Associations have been formed, which are active and performing routine maintenance. During initial 2-3 years of project implementation these associations were formally registered with department but later preferred to work informally. For achieving, lasting sustainability of infrastructure subprojects, it would be useful to strengthen these associations through O&M training and awareness raising. This would help sensitizing them towards the importance of timely maintenance and would contribute to fully complying with the MoAF guidelines that require registration under Cooperative Act, proper composition7, signing of formal MoUs8 with Geog Administration and fully compliance of maintenance plan9 as embedded in the guidelines. • Beneficiary community is responsible for minor O&M activities (drain cleaning, spot filling, small debris removal for form roads and PTT, vegetation clearing, blockage removal, intake and outlet structures maintenance for irrigation channels, checking pins and joints for bridges, etc.) • Dzongkhags are responsible for financing and provision of machinery for major O&M in case of floods and land sliding. As per the letter signed between RUAs and Geogs, the representative of RUAs would inform the administration in case of any major repair required. 7 Includes Chairperson, Secretary, Treasurer and Road Care Taker 8 Annex F: MoU Format between Geogs Administration and Road User Group; Guidelines for Farm Road Development Revision 2013; Department of Agriculture, Ministry of Agriculture and Forests. 9 Routine maintenance plan includes; (i) Cleaning of obstruction from road, (ii) Cleaning of side drains, (iii) Cleaning of drainage structure, (iv) Repair of scour checks and drain lining, (v) repair of pot holes, (vi) repair of structures, (vii) cutting of vegetation, and (viii) checking and repairing bio-engineering. 36 • NU 2.0 m annually is available at each Geog for possible utilizing and prioritizing for O&M, • Additional Funds are made available to each Geog in case of emergency needs. Impact: The infrastructure interventions have provided significant economic and social benefits to the project communities. The road projects (farm roads, PPT and bridges) yielded huge benefits. Approaching remote villages via newly/rehabilitated farm roads and interacting with the beneficiary communities, the ICR mission came across some impressive success stories. For instance, the farmers of Geog Rakshay-Rangthangling, Tsirang Dzongkhag used to sell oranges at Nu10,000 per acre. With the DRDP built PTT, the earnings rose to Nu 30,000 per acre. The PTT was upgraded to a farm road and the income increased to Nu 150,000 per acre. Of the 30 households in the Geogs (2 more Geogs with 60 HHs are also beneficiaries), 10 households have bought new cars, 3 households purchased mini trucks to start transport business. More than 50% HHs have renovated (cemented from traditional wood construction) their houses. Furthermore, the beneficiaries of irrigation channels (Tsirnag, Bajo), are growing a second crop with significant financial gains. The reported benefit streams generally include the following; • Reduced walking time to motor-able roads, • Reduced Transportation Cost, • Improved access to resources/services (health, education, market, and farm inputs etc.), • Improved income of farmers, particularly for cash crops/fruit, • Increased irrigated area, • Improved and reliable availability of irrigation water, • Improved crop yields, • Second Crop cultivation, • Improved access to Power Tillers, • Transformation from traditional means to farm mechanization, • Reduced travel time between interconnecting communities (Geogs/cheogs) through bridge construction, and • More social cohesion among communities/cheogs/Geogs, Component 2: Renewable Natural Resources Centers (RNR) Component 2 comprised two subcomponents, namely (a) RNR construction and (b) RNR Capacity Building. 1. RNR Construction The subcomponent financed six RNR centers. The project used existing prototype designs. Typically a RNR-Centre would comprise an office building housing three MOAF Extension 37 Agents (EA), one each for agriculture, livestock and forestry, as well as a multipurpose meeting hall, residential quarters for the three EAs, and the Geogs Office for the “Gup” (Geogs Head) and the Geogs clerk. Sometimes fencing and landscaping, water supply and drainage were added on needs basis. RNR Centers attracted additional funds from other donors, such as Geogs Community Centers (GCC) which offered telephone, fax or email services. Key outputs of RNR Construction Results-Framework Indicator: “6 new RNR Centers constructed” Status: fully achieved DRDP achieved the target of constructing 6 RNR centers in the following Geogs (Dzongkhags): • Drakten (Trongsa): RNR Office, staff quarter, multi-purpose hall, drainage and water supply; • Gosaling (Tsirang): RNR Office, staff quarter, multi-purpose hall, fencing and landscaping; • Ngangla (Zhemgang): RNR Office, staff quarter, multi-purpose hall, Gup Office; • Phangyul (Wangdue): RNR Office, staff quarter, multi-purpose hall, Gup Office, fencing, water supply; • Drala (Chukha): RNR Office, additional staff quarter, multi-purpose hall, fencing; • Goshi (Dagana): RNR Office, staff quarter, fencing. 2. RNR Capacity Building With the aim of increasing production of food and cash crops, the subcomponent focused on: (i) capacity building of farmers and extension staff, (ii) dissemination and adoption of appropriate technologies, (iii) improved seed varieties, and (iv) use of agricultural mechanization where feasible. Key outputs of RNR Capacity Building a) Rice: Results-Framework Indicator: “Adoption of new rice varieties by at least 25% of farming HHs” Status: partly achieved Improved rice variety adoption rate is 65% in terms of household adoption and 56% in area coverage. However, it is only valid for the Wangdue-Punakha valley as support in other Dzongkhags was not as extensive and data are not available. Other key outputs include: • As per the Impact Assessment and Project Completion Report, 64.7% households reported cumulative increase in crop production for rice, maize and potato; • Promotion of 26 tons of high yielding variety of paddy seeds in 6 Dzongkhags benefited 3,406 farmers in 6 Dzongkhags; • Double cropping of rice introduced. In Thedthso Geog, 48 households made substantial gains from double cropping of rice on 22 acres of land; 38 • Mechanized farming adopted on 5000 acres of wetland including tray nursery technology on 25 acres; Land terracing done for extended research within the RNR-RDC sub-centers. b) Maize: Results-Framework Indicator: “80% maize seeds of the Grey Leaf Spot (GLS) affected farmers are replaced” Status: fully achieved Seed replacement target rate of 80% for GLS affected farmers with GLS resistant varieties was fulfilled by either direct seed supply through the project or indirect supply through the National Seed Centre (NSC). Direct supply to 7 dzonkhags (4 project-, and 3 non-project Dzongkhags) was as per the table below. Year 1 Year 2 Year 3 Grand Total Dzongkhag acres HH Acres HH acres HH acres HH Dagana 200 300 280 340 315 370 795 1010 Lhuentse 0 0 120 100 135 147 255 247 Pemagatshel 133 160 130 184 145 205 408 549 Trashigang 200 230 205 253 235 306 640 789 Tsirang 130 188 217 287 320 317 667 792 Yangtse 0 0 0 0 64 53 64 53 Zhemgang 0 0 0 0 70 60 70 60 Total 663 878 952 1164 1284 1458 2899 3500 The disease reached epidemic scale in 2007 and affected 4,821 acres in 12 districts. Of which 75% was in the 7 Dzongkhags, listed above, leading to a baseline of affected area of 3,616 acres. With direct seed replacement for 2,899 acres, DRDP fully achieved the target. Furthermore, indirect supply was possible through DRDP-supported 9 CBSPs with a total of 85 members. These groups produced 88,000 kg of quality maize seeds over a period of 3 years under a contract with NSC. The NSC purchased seeds from the groups and supplied it to GLS hit farmers across the affected Dzongkhags. Considering a seed rate of 15kg/acre, additional 58,886 acres have been covered under this seed replacement drive. Other key outputs included: • 582 farmers were trained in various topics such as maize seed selection and management of new maize varieties, • Support to on-station research for basic seed production and variety maintenance, • Initiation of spring maize program and hybrid maize introduction as part of the cropping system intensification initiative, • Publication of brochure on “knowing maize varieties”. 39 c) Potato: Results-Framework Indicator: “100% increased production of potato basic seeds from 350 mt to 700 mt” Status: fully achieved A total of 875 tons of basic seed potatoes were produced and marketed, thus the project exceeded the target of increasing seed production from 350 mt to 700 mt by 121%; • With the project support, aeroponics facility for potato seed production at the National Seed Centre (NSC) is operational since April 2012. Its harvest of minitubers was 11,616 in 2012 and 17,345 in 2013; • Three Generation Seed Strategy (3G) for revamping the seed potato production system is at an advanced stage. About 8.8 tons of G2 potato seed that was harvested in 2014 will be planted in 2015 as 3rd generation on NSC fields. The next cycle will be multiplied by contract farmers; • Publication of 500 booklets for aeroponic production of potato mini-tubers; • Publication of training manual on positive selection to improve quality of farm saved potato seeds; • Distribution of seed potatoes to 800 households (500 DRDP, 300 RGoB) who had not replaced their seeds for a long time. According to some estimates the seed replacement rate increased from previous 10 to 18%. d) Farmers’ Group Formation Results-Framework Indicator: “At least 5 functional “growers group” formed with necessary skills on marketing and post harvest technologies” Status: partly achieved • There is anecdotal evidence to suggest that farmer groups (rice grower groups, vegetable groups, etc) formed by the farmers are proactive, but data such as names of groups, numbers of membership, or trainings completed, was not regularly collected and compiles. Results-Framework Indicator: “At least 9 community based seed production groups are formed and linked with the national Seed center“ Status: fully achieved • DRDP facilitated establishment and training of 9 Community Based Seed Producer Groups (CBSPs) with a total of 85 members. e) Training of Farmers: Results-Framework Indicator: “Number of farmers benefitted by extension services (maize, potato, rice) at least 8500, of which female beneficiaries constitute 40%” Status: partly achieved 40 Farmers were trained through various means including field days and demonstrations, classroom training, study tours. The Impact Assessment and Project Completion Report stated that both males and females equally benefitted from training. Summary of the Trainings and Demonstrations to Farmers (2005 - 2014) Number of Type of trainings farmers trained Demonstration / Field days 966 Trainings 2295 Study tours 50 Total 3311 f) Training of MoAF Staff Results-Framework Indicator: “Capacity of 14 agricultural field staff increased through training in production and post harvesting technologies” Status: fully achieved DRDP exceeded the target on training of staff by a significant margin Summary of the Trainings to officials (2005 - 2014) Number of Type of trainings staff trained Trainings related to Commodity (Maize, Potato and Paddy) including research, 145 pest surveillances (in-country + outside) Training related to finance, administration and ICT 18 Study tours/ seminars and workshops 85 Trainings related to Irrigation technologies and techniques 12 Results-Framework Indicator: “Capacity of 6 research staff enhanced through training” Status: fully achieved Among the 145 staff trained (see table above), at least 8 researchers were trained in the following areas: • 1 staff attended wheat improvement and pathology course for two weeks at CIMMYT-South Asia, Katmandu, Nepal; • 1 researcher attended a rice breeding course of 4 weeks at International Rice Research Institute, IRRI-Philippines; • 1 research assistant working on maize attended 3 weeks on-the-job training on maize breeding, agronomy and seed production in Thailand; • 5 Researchers from the Renewable Natural Resource-Research and Development Centres (RNR-RDCs), National Post Harvest Centre (NPHC), and Agriculture Machinery Centre (AMC) attended training on maize products diversification. 41 g) Other Achievements DRDP successfully achieved additional targets, which will significantly contribute for the growth of the agricultural output. Some of those major achievements are listed below: • Institutional support to the NSC and its farms (Bondey, Jephu, Paro, Bajo, Phubjikha) through construction of irrigation facilities and renovation of pumps, fencing the premises, upscaling the Aeroponics Unit, and road maintenance; • Strengthening of RNR-RDCs Bajo, Bhur and Wenggkhag through supply of rice mills, nursery trays and rice planters as well as power weeders; • Strengthening of agricultural machinery centre by supplying 21 power threshers for hiring services; • Banana tissue culture started at RNR-RDC Paro; • E-pest surveillance system installed and ready for use; • 110 sets of deep fat frying machines supplied to farmers for chips and cornflake making; • 30 on farm potato stores and 40 on farm maize stores constructed, and around 200 other seed storage containers distributed; • Setting post harvest machineries (rice mill, flour mill and corn flake machine) and supplied to 9 CBSP groups; • Supplying seeds and demonstrations on using dhaincha (sesbania aculeata) as green manure for rice. 3. Key Findings a) Project Development Objective – Indicators • Two of the three PDO indicators state a “15% increase for food and cash crop production per targeted Geogs” as a target. The choice of these indicators led to substantial problems when it came to impact assessment. For instance production in terms of metric tons per Geog is a function of area under cultivation and yield per area, and DRDP was not always working in all irrigation schemes of targeted Geogs and, therefore, gains in some schemes could be offset by losses in other schemes within the same Geog. Accordingly production gains or losses per Geog have to be seen in the overall context, including factors beyond the project’s control, such as land loss or increasing shortage of farm labor. • Considering the above, project achievements may appear lower than they actually were. A more appropriate indicator could have been to measure increase in productivity (yield increase per acre) per renovated or newly built irrigation scheme. 42 b) Lack of Baseline Data • After attempts to establish a project specific baseline through surveys it was finally agreed to use 2009 statistical data as baseline. Although this serves the purpose to monitor production data on Geogs level, it does not allow to measure productivity increases in renovated or newly built irrigation schemes. c) Data Collection & Processing / Monitoring / Record Keeping • In line with the RGoB policy, implementation of DRDP activities was decentralized and the responsibility for implementation rested with the respective Dzongkhags and eventually the Geogs. There was no separate Project Management Unit (PMU) as such, but a Project Director who, besides DRDP also had to take care of his usual duties. While this approach was useful in integrating DRDP into regular MoAF activities, some key functions required dedicated project staff and were not provided for in the project. For example, there was no dedicated M&E system in place to ensure regular project specific data collection, collation and analysis. • Records (e.g. training sessions for farmer groups, curricula/topics of training sessions, etc) were often not available at the central level. However, some of them were available at Dzongkhag or Geog level. Crop specific data (rice, maize, potato) were mainly collected and stored by the commodity coordinators. d) Alignment of infrastructure development with capacity building • In the first phase (2005-2010), the project focused primarily on infrastructure. From 2011 onwards there was greater emphasis on capacity building and the provision of quality extension services. e) According to available data, Geog or farming households benefitting from new or renovated irrigation schemes did not necessarily receive training in improved agricultural technologies (other than the existing extension services). f) Farmers’ Training / Producer Group Establishment / Marketing • Farmers’ training could have been more extensive and should have included more sessions on group formation and strengthening as well as on establishing common input supply and marketing mechanisms and group owned machinery services. 43 Annex 3. Economic and Financial Analysis A. Pre Investment Analysis Pre Investment Economic and Financial Analysis No overall rates of return were provided in the financial and economic analysis annex of the February, 2005 PAD. However, calculations were presented as to expected road access improvements in the six Dzonkhags (districts) covered by the project under Component 1 (Rural Infrastructure). Road investment was projected to affect 5,611 households, averaging almost 38 percent of target area households. Data was also provided from secondary sources on previous experience of changes in agricultural production following road construction, the variance in household incomes between those with and without road access, and educational enrolment rates with and without road access. Unfortunately, the PAD provided no analysis or estimates of the impact of irrigation infrastructure provision (also under Component1). Under Component 2 (RNR Centres) and Component 3 (Institutional Strengthening) information was given only on construction targets (6 new RNR Centres) and training (5 course types), respectively. B. Post Investment Economic and Financial Analysis: Sub Components Analytical Methods Form of the analysis. The analysis of the project is able to be more specific about the costs and benefits than was the case at appraisal because sub-projects, investments and activities have been identified and implemented throughout the project period. However, no baseline survey was conducted, nor was any dedicated monitoring and evaluation (M&E) system established, for the project. With the exception of cost of production data for rice and maize, data has therefore been largely obtained from other sources, including an independent survey of over 1,400 producers and producer groups undertaken as part of the borrower ICRR. Primary data collected during the ICRR mission was also used. Component and Sub Component Analysis Component 1: Rural Infrastructure (USD 7.86 million). This component comprised two sub-components: (a) Rural Access, and; (b) Irrigation. The main beneficiaries comprised isolated rural communities with little, if any, road access to the national transport system. Achieved rural access investments are shown in Table 1 below, but do not include the five bridges (of a total of six targeted). New farm roads, rehabilitated farm roads, power tiller tracks and rehabilitated irrigation channels all exceeded original targets, while the construction of new irrigation channels and bridges was slightly less than targeted. As no baseline data was collected on productivity or other economic activities for communities targeted for improved infrastructure, it is not possible to estimate returns to project investments in these areas. However, enough data does exist for the average cost per kilometre of construction and per beneficiary to be calculated, and these costs compared with standard estimates and guidelines. 44 Table 1: Achievements and Beneficiary Households from Rural Access Investment Overall, the rural access component incurred total costs of Nu.153.4 billion and benefitted 10,938 households, or approximately 50,000 persons. This represents almost 27 percent of all households within the 11 target Dzonkhags and equates to a cost of Nu. 14,000 (USD 219) per household, or Nu. 3,066 (USD 48.1 per person), not including the additional beneficiaries arising from the five bridges which were constructed. In Table 2, the expenditures for each infrastructure category are presented in per unit terms. As no differentiation is made in the project expenditure data between rehabilitated and new farm roads, the costs shown represent an average across the two categories. Table 2: DRDP Expenditures on Infrastructure per Unit Roads The majority of road investment (138.3 km of a total of 207.6 km) was for the construction of new roads – the remainder was for rehabilitation. Detailed cost data per kilometre of road constructed was only available for a number of samples in a single Dzonkhag (Wangdue). These are shown in Table 3 below. Given the relatively small number of communities and households served by the roads[1], the cost per kilometre and beneficiary is high; Nu. 869,000/km and USD 463/person. In the PAD, cost guidelines [1] The performance assessment of the IFAD-financed Agriculture, Marketing and Enterprise Promotion Programme (AMEPP) in the West of Bhutan, noted a beneficiary per kilometer ration of 10 households per kilometer and deemed it ‘sparse’. AMEPP Project Performance Assessment. Report No. 3300-BT, IFAD, August 2014. 45 issued by the MOA are given, which indicate that farm road costs should not exceed USD 425/beneficiary and the PAD itself anticipates project road costs at USD 385/person. No updated version of the MOA costs could be identified, but adjusting these costs for US dollar inflation since the guidelines were published suggests that the equivalent cost in 2009 would be USD 459 – very close to the average of the six roads for which there is data available, although still higher than the predicted costs in the PAD, adjusted for inflation (USD 416/person). Table 3: Rural Road Costs per Beneficiary, Wangdue District It should be stressed, however, that significant benefits are subsumed within increased yields and household incomes for the key project crops as a result of irrigation and road investments. In particular, the borrower ICRR (project completion report) participant survey determined that expansion in maize and potato production were closely linked to improved road access and, therefore, act as a proxy for road benefits in this analysis. It is worth noting that an amount of Nu.2 million (USD 31,348) is set aside in each district annual budget for road maintenance and repair, although this may not be sufficient for all road works required annually within each district. Power Tiller Tracks The data in Table 1 and 2 show the number, length and number of beneficiaries from the power tiller track investments, and an average cost of almost USD 5,500/km. This represents almost one quarter of the per kilometre cost of farm roads constructed under the project but it should be noted that the roads served more than twice as many beneficiaries per kilometre as the tiller tracks. Irrigation No assessment of irrigation impacts or benefits was provided in the PAD for this project, and this analysis assumes that irrigation benefits are subsumed within expanded crop productivity within the project target area. However, the beneficiary ICR survey determined that the principal impact of the irrigation investment was to permit the expansion of wetland paddy production, and respondents indicated that this was more important for them than road access in the case of rice. The benefits arising from the investment in irrigation are therefore subsumed within the crop income calculations presented under Component 2. 46 Irrigation activities included the renovation of 521 km of channels in 58 Geogs, benefitting 4,594 households and with command area of 4,733 ha, as well as 42.5 km of new channels built in 11 Geogs benefitting 418 households with a command area of 276 hectares. Achieved renovation work therefore exceeded considerably the ex-ante estimates in the PAD and AF Project Paper (393 Km.) but were marginally below the projected new construction (44.3 Km). The average costs per kilometre of canals and associated works are shown in Table 2 as USD 5,362 for new construction and USD 3,182 for rehabilitation of existing systems. Actual investment costs for irrigation works as assumed in the PAD and as incurred at implementation, are shown in Table 4. It can be seen that actual costs were approximately 12 percent lower than assumed at design for new construction, but almost 33 percent higher for rehabilitation. Table 4: Projected and Incurred Costs for Irrigation Infrastructure Works According to data drawn from the relevant Five Year Plans, the 11 Dzonkhags included in DRDP (five of which were included only later at the time of the AF) contained 521 schemes with 1,617 Km of canals. By the close of the project these numbers were 1,220 schemes and 3,060.2 Km of canals. In total, the project thus accounted for approximately 32 percent of the work carried out in these Dzonkhags over this time period[2]. Bridges A total of 5 bridges were constructed under the project, at an average cost of almost USD 32,500 per bridge. This amount may over-estimate average costs as one bridge was damaged by heavy flooding after the bridge was completed, rendering it necessary for the bridge to be rebuilt. However, some additional funding was received by RGOB from JICA towards this reconstruction cost. Component 2: RNR Centres and Support for Improved Technologies and Marketing of Rice, Maize and Potato (US$2.28 million). This component constructed 6 RNR centres benefiting 2,430 HHs and provided support for capacity building among RNR staff. In addition, under the revised project structure in December 2010, additional funds were made available to include direct support for improved crop production technology, including improved seed, mechanization and cultivation practices. It should be noted, however, that not all crop based benefits derived from these additional funds, as the [2] This calculation assumes that schemes which were renovated were not counted in the original numbers. 47 provision of improved road access was, as noted above, important for the improvement in maize and potato production, while the training offered through the new RNR centres also contributed to benefits in this area. In addition to the rehabilitation and construction of irrigation channels, other crop-related improvements provided through the project included: (i) hybrid and high yielding varieties, as well as seed with improved genetic resistance to grey leaf spot and other diseases; (ii) the establishment of aeroponic production facilities for potatoes; (iii) upgrading of research and testing facilities on MOAI research stations; (iv) the development of an e-pest surveillance system; (v) the installation of a number of polynet houses for both rice seedlings and vegetable production, and; (vi) improvements in mechanization equipment and distribution. The borrower’s completion survey identified that from 12-18 percent of potential households in the 11 project Dzongkhags were beneficiaries of support to rice, maize and potatoes[3]. The survey also identified output increases for each crop, as well as income increases for maize only. Table 3, below, summarizes the agronomic results of the project investments, through a comparison of harvested area, production and yield within the project area with national statistics over the period 2009-2013. Although the starting point (2009) is some four years after the project commenced, there was little direct support to crop productivity during the initial years of the project. In fact, in the case of potato, a major expansion of the aeroponic seed production facility has yet to be completed, implying a far greater impact from improved seed in the future. While the use of national statistics for comparison with project area results is not wholly satisfactory, as the project area accounts for from 62 to 90 percent of total national production of the crops, the identification of a suitable control area gives rise to even greater problems, due to the need to identify a significant production area not affected by the project. Given the important role of project area production in national output, however, the subsequent calculation of returns is based on the changes within the project area. It can be seen gains of over 20 percent in total production were achieved for paddy and maize over the period 2009-2013. As the harvested area of both crops declined significantly, these increases are due entirely to yield increases – up by almost half for [3] Although mandarin was originally included in the list of target crops, this was later dropped and no project activities directly targeted at mandarins were undertaken. 48 both crops. By contrast, potato balanced a small decline in yields with a major increase in harvested area to generate a 10 percent increase in output. Table 5: Changes in Output of Key Crops – Project Area and National Source: Project Crop Programmes and RNR Annual Surveys Using the increased production levels from Table 6, together with data generated by the various project-supported Crop Programmes on producer costs and output prices, it is possible to calculate the incremental benefits generated since 2009 by investment in paddy, maize and potato – the principal crops supported by the project. Production increases in all three crops generate estimated annual net benefits of USD 1.02 million per annum. Table 6: Annual Net Benefit from Incremental Project Area Changes in Production The project only indirectly financed crop productivity activities in the period 2005-2010 through irrigation construction and rehabilitation and some capacity building of RNR staff, but direct productivity support investments were introduced into the project through the 2010 Additional Financing, with USD 1.92 million allocated to Component 2 activities at that time. By 2010, the RNR Centres had already been completed; as a result this additional amount went almost entirely to crop support measures. Using this data, it is therefore possible to calculate rates of return to these investments (Table 7). Benefits are assumed to be only partial in early years, deriving from the irrigation investment, and reach full level only in 2011. To account for these initial indirect benefits, the costs of the irrigation investment (USD 1.1 million) are also assigned to this component. 49 The estimated FIRR, of 28.9 percent, is high, but all road infrastructure and management costs incurred in Components 1 and 3 are excluded, as well as those Component 2 expenditures (principally on the RNR Centres) prior to the additional financing. Table 7: Financial Rates of Return to Crop Productivity Investments These results are robust, even with the inclusion of all irrigation investment costs. Lowering the rate of crop productivity benefits by 10 percent reduces the FIRR by less than three percent, to 26 percent. Even with a 20 percent reduction in benefits the FIRR remains at 23 percent. Component 3: Institutional Strengthening (USD 1.32 million). This component included project management and training of government staff from MOAI at central, Dzongkhag and Gewog level. Training comprised a series of short courses covering such areas as financial management, capacity building, environmental and social safeguards and monitoring & evaluation. No attempt has been made to assign quantifiable benefits to these activities. C. Post Investment Economic and Financial Analysis: Whole Project The project had undoubted non-quantifiable benefits that cannot be measured in this case due to the lack of adequate baseline or M&E data. The most important example is that of non-crop production benefits arising from the investment in road construction and rehabilitation, although other significant benefits are expected to arise from improved access to education, health services and employment. For crops normally grown outside of irrigation schemes and livestock, road access is critical, reducing input costs and providing access to markets through lower transport costs as well as the expanded presence of traders once vehicular access is feasible. Most transport projects undertaken in Bhutan in the last decade have focused heavily on larger roads than the farm and community access roads addressed by this project, and none included power tiller tracks. Nevertheless, a review of World Bank and other IFI experience in this sector shows the magnitude of benefits which are possible. Table 8 below shows that considerable increases in business activities and remittances, as well as declines in transport costs can be expected. 50 Table 8: Results of the World Bank Bhutan Rural Access Project, 2006 Source: Rural Access ICCR, World Bank, 2006 A 2008 study of an IFAD-funded transport project in Bangladesh calculated an average 39 percent decrease in transport costs, in addition to an estimated a 19 percent increase in yields, although no mention was made of changes in cropping patterns[4], while an older World Bank report estimated a 15-20 percent reduction in the cost of marketing agricultural products in the Philippines[5]. Other unquantified benefits arising from the project include the development of micro- enterprises for the processing and marketing of maize, vegetables and even livestock products, which only became financially attractive once improved road access was available. Increased social capital from the formation of a number of community-based maize seed production groups may also have generated financial benefits. In fact, according to data collected by the beneficiary survey conducted as part of the borrower ICRR, the largest impact on the production of marketable surpluses was not in the three primary crops targeted, but in vegetables (35 percent of households), dairy (15 percent) and citrus (14 percent). No data is available, however, on the quantities or net incomes arising from these activities. Financial Analysis According to the borrower ICRR, the project provided direct benefits to 28,850 households and indirect benefits to a further 42,994 households. However, improved crop productivity under Component 2 comprises the only element of the project for which returns can be directly calculated, as there is insufficient M&E data or baseline information to permit the assessment of quantitative benefits to infrastructure activities. As the crop-based benefits relate to only approximately 30 percent of total investment costs, they are not in themselves sufficiently large to provide positive returns to the entire project. As a result, overall project financial returns have been calculated on the basis of a [4] Agriculture and Rural Transport: Joint IFAD/IFTRD/ILO Workshop. June 2008. Reported by the International Forum for Road Transport and Development (IFRTD) at: iftrd.org/new/issues/agriculture.php [5] A Strategy to Fight Poverty. World Bank, Philippines. 1996 51 series of alternate assumptions concerning benefits arising from the provision of improved access. These benefits have been identified in the borrower ICRR as resulting from the increased production and sale of vegetables, dairy and citrus products, as well as expanded commercial activity in non-agricultural areas. The studies cited above suggest that reductions in transport costs alone are a significant benefit, ranging from 15 to 40 percent. As a result, the impact of increases of 10, 15 and 20 percent in average household incomes among beneficiary households have been calculated using average household expenditure data for project Dzongkhags as given in the 2012 Bhutan Living Standards Survey as a proxy for income, and the result used in calculating returns to the project. These calculations are shown in Appendix 1 to this annex and summarized in Table 9 below. Under the assumption of phased benefit accrual as civil works are completed, starting in the third year of implementation, an increase of only 10 percent in direct beneficiary household incomes, the financial rate of return to the whole project switches from negative to over 15 percent. This rises to an FIRR of over 20 percent for a 15 percent increase in household incomes and almost 25 percent for a 20 percent increase in household incomes. Table 9: Total Project Financial Returns under Different Assumptions as to Benefits of Rural Access Infrastructure Thus a very conservative estimate of an increase of only 10 percent in household incomes among beneficiary households served by rural access infrastructure under the project would render the project financially attractive. Economic Analysis The economic analysis attempts to modify market prices for costs and outputs to reflect the real net benefit to the economy of the project. Two factors normally considered for adjustment under economic analysis are labour and foreign exchange costs. Labour costs are particularly important to this analysis as such a major portion of crop production costs in Bhutan are for labour. The Bhutan Investment Climate Assessment Report of the World Bank, published in 2010, states that average labour costs in Bhutan are 45 percent higher than in India, a country to which Bhutan is very closely linked through trade, a tied exchange rate, and substantial inbound investment to Bhutan. Nevertheless, the report also points out that labour productivity is much higher in Bhutan than in other regional countries and, as a result, per unit costs are lower than the regional average. The study concludes that labour costs are, in fact, competitive. As a result, no adjustment has been made in labour costs for the economic analysis. 52 Nominal exchange rates (those at market prices) frequently do not adequately reflect the real cost of imported items or the value of exported ones, due to distortions arising from currency controls, speculation, trade imbalances or a number of other factors. In the IMF Article IV Consultation Report, published in 2014, the authors specifically examine the Ngultrum to determine if the Real Effective Exchange Rate (REER) is significantly different from the nominal one. They conclude that the Bhutanese currency is overvalued in the range of 8.5-21 percent and propose an average valuation of 16 percent. This means that all imported items used in production have a real value to the Bhutanese economy of less than the nominal exchange rate suggests. In the EIRR and ENPV, therefore, the cost of imported items used in the production of the three target crops has been reduced. However, almost all production costs for rice, maize and potatoes are labour-based, with only fertilizers and perhaps other agrochemicals being imported[6]. No more than five percent of total production costs – according to the data provided by the project crop coordinators – is accounted for by such costs, reducing the impact of the adjustment to little more than one percent of total costs. The results of these adjustments are shown in Table 10 in relation to the financial returns calculated in the Appendix. Table 10: Economic NPV and IRR Estimate – Initial Disbursement As can be seen, as a result of the very small reduction in costs the calculated EIRRs increase only marginally, from 15.3 percent (financial) to 15.5 percent (economic) under the assumption of a 10 percent increase in household incomes for direct beneficiaries of project rural access investment. Conclusions Due to lack of data, a key set of assumptions has been applied in the above analysis, relating to both crops not targeted by the project and to increases in non-crop activities and earnings. The benefits from increased crop productivity are sufficient to cover all costs associated with this activity as well as all investment made by the project in irrigation infrastructure. Even when estimated crop benefits are reduced by 20 percent, the FIRR to this activity is still 23 percent. [6] In fact, even the fertilizer may well have been manufactured in Bhutan, but the feedstock for the plant will be imported. 53 Using comparative data from other projects, both in Bhutan and elsewhere in Asia, the analysis has shown that only minor (10 percent) increases in the household incomes of those directly benefitting from access to new and rehabilitated rural roads and power tiller tracks are sufficient to generate acceptable financial and economic rates of return. If household benefits are higher than the minimum 10 percent assumed, the project is strongly positive. On the basis of this analysis, therefore, and even in the absence of considerable key performance data, the project it is considered likely to have been highly satisfactory in both financial and economic terms if the above assumptions of moderate increases in beneficiary household incomes is accepted. It is worth noting that over 35 percent of respondents interviewed in the beneficiary survey noted that they had increased vegetable production as a result of road access, while 72 percent stated that market access had improved, rendering these assumptions credible. Given that previous studies cited in this annex have shown much higher impacts than this, and the robustness of the quantified benefits arising from increased crop productivity, it is believed that the project will have generated a strongly favourable impact on beneficiary household incomes and welfare. 54 Annex 3 - Appendix 1: Total Project Returns under Different Assumptions as to Rural Access Benefits Assumptions: The following table shows increased household and total beneficiary incomes across all households directly benefitting from transport investments under increases of 10, 15 and 20 percent. When these increased household income benefits are introduced into the prior estimation of financial returns at the levels of 10, 15 and 20 percent, the following results are obtained (Tables 1-3). 55 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Madhavan Balachandran Sr Financial Management Specialist GSUCA Debabrata Chakraborti Senior Procurement Specialist GGODR Grahame Beaumont Richard Dixie Adviser GFADR Samantha L. Forusz Program Manager HRDPR Drona Raj Ghimire Environmental Specialist GENDR Malcolm A. B. Jansen Consultant GENDR Surendra Govinda Joshi Consultant GTIDR Zarafshan H. Khawaja Lead Social Development Specialist GSURR Yuka Makino Sr Natural Resources Mgmt. Specialist GSURR Manvinder Mamak Sr Financial Management Specialist GGODR Yoshiko Masuyama Program Assistant SASDO - HIS Shyam Sundar Ranjitkar Consultant GSURR Binyam Reja Lead Transport Specialist GTIDR Kumaraswamy Sankaravadivelu Consultant GGODR Suran KC Shrestha Program Assistant GFMDR (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY04 15.00 117.93 FY05 30.28 159.84 Total: 45.28 277.77 Supervision/ICR FY05 6.22 24.05 FY06 16.45 51.78 FY07 8.34 33.54 FY08 23.61 59.00 FY09 13.13 53.51 FY10 5.47 31.10 FY11 5.70 29.45 FY12 10.30 49.29 FY13 10.33 58.52 FY14 17.53 84.89 FY15 20.93 117.13 Total: 138.01 592.26 56 Annex 5. Beneficiary Survey Results Background The primary objective of impact assessment and preparation of project completion report was “to ascertain and record major achievements of the project, success stories and critically reflecting on the implementation performance of all project partners, and assess the extent to which the project succeeded in reaching its stated objectives and goals.” Approach and Methodology Field Research was conducted to collect both qualitative as well quantitative data from six sampled Dzongkhags and 22 Geogs. Consultations were carried out with the collaborating and implementing partners (RNR-RDCs, National Plant Protection Centre, NSC, Dzongkhag and Geog Agriculture officers, and Dzongkhag Engineers) and farmers. Survey Design Within these Dzongkhags, Geogs were further sampled using sampling criteria as follows: • Geogs where majority of the activities were implemented • Geogs (villages) with and without road access • Selecting a minimum of 40% of the Geogs within each selected Dzongkhag. From DRDP coverage of 78 Geogs in eleven Dzongkhags, 22 Geogs from six Dzongkhags (28.2%) was visited for the field work. The number of targeted beneficiaries for discussion and interview were based on the number of activities in each Geog. In total, 731 farmers (56.2% males: 43.8% females) were interviewed and additionally 682 numbers of same farmers (55.1% males: 44.9% females) were also consulted through focus group discussion. Limitations of the Study Though 2009 was identified as baseline, a system to record production and yield data for beneficiaries’ households during project period did not exist. The Geogs and Dzongkhags collect Geog’s production and yield data annually which is reflected in agricultural statistics that is based on sample survey and represents the entire Geog. Therefore, measuring the changes brought about by the project for beneficiaries households (HHs) alone using secondary data was not possible. However, direct impact on beneficiaries’ households is supplemented by farmers own reflection on the yield and production as collected during field survey. 57 Overall Assessment (i) In comparison with targets in PDO, the project has made good achievements in successfully implementing activities as planned: 101% for new farm roads construction; 102% for farm road improvement; 100% for power tiller track construction; 100% for construction of new irrigation channels; 99.48% for rehabilitation of irrigation channels; 100% for construction of RNR/Geog centers; 83% for construction of bridges and 276% for farmers’ trainings/ demonstrations and trails. (ii) Analysis on productivity and production revealed that on an average the productivity (yield) of paddy increased by 36.4% in 2012 as compared to 2006; and production by 6.4%. For potato there was 31% increase in production in the year 2013 as compared to 2006. For maize there was 11.5% increase in production in the year 2013 as compared to 2000; and 31% increase in yield. (iii) Some of other key outcomes are: 19% higher yield than local variety for Bajo Maap 2 and Khangma Maap (paddy); 49% higher yield than local variety for IR-64 (paddy); 17345 minitubers of potato harvested in 2013 from aeroponics; increase from earlier 5 tons production to 60 tons potato seeds from Phobjikha NSC in 2012; 7 tons of G2 potato seed of Desiree variety harvested in 2014; 875 tons seed potatoes produced and marketed; 9 community maize seed production group formed that produced 27MT of seeds in 2012 with 37% increased income as compared to 2011; 15% seed replacement (in 2011), 45% replacement (in 2012) and another 15% replacement (in 2013) done for maize (for Grey Leaf Spot: GLS disease affected areas). (iv) Impacts based on livelihood domains were diverse and many. With regards to physical assets, availability of farm roads and increased income from agricultural products have helped farmers to purchase many assets. Out of 1148 multiple responses from respondents, followings were reported with respective percentage: purchased TV/ refrigerators (28%); constructed new house (13.2%); constructed new toilets (7.9%); undertook CGI roofing (7.2%); purchased agricultural tools and machineries (6.5%); purchased improved cattle (6.3%); and purchased vehicle/power tillers (3.4%). The comparison of the rural infrastructures (mainly roads and irrigation schemes) at the start of the project and closure shows the followings: irrigation schemes (521 in 2007 vs 1220 in 2013); length of irrigation channels (1616.93 km vs 3060.205 km now); and length of farm roads (473.02 km vs 2392.87 km now).As compared to food last in HHs in past six years, presently food lasts for 1–3 months has increased for 10.3% HHs. While food lasts for 4–6 months and 7–11 months decreased for 55.2% and 30.6% HHs respectively, the sufficiency for 12 months increased for 123.8% HHs10. (v) The project interventions towards environment and common resource base were related mainly with farm roads constructions. With earlier knowledge and experiences on bio-engineering, environmental assessment and mitigation measures, environmental friendly survey and design were incorporated as well during implementation. 10 PCR and Impact Assessment Survey, Bhutan Consulting Associates, Oct-Nov 2014, 58 (vi) Farmers mentioned having acquired better knowledge and skills for: pest management; planting potato on ridges for higher yield; cultivate improved variety of paddy and maize (draught tolerant and disease resistant respectively); proper harvest and seed selection for maize; and up-land rice cultivation techniques. Extension agents now have added proper knowledge on paddy production and farmers group formation. Similarly, a few engineers now have proper knowledge on irrigation survey, design, construction technology and water management that guided implementation in proper ways. Significant contribution has been the capacity building on aeroponics with which now the unit is functional for potato seed production, to be continued by NSC. Some officials also have added knowledge in the field of post harvest, value addition, and e- pest surveillance. (vii) The survey revealed: increased paddy production (for 18.5% of total respondents); increased maize production (18.1%); increased cash income (12.7%); eased marketing (12.6%); reduced walking distance (12.5%); increased potato production (12.2%); and smaller percentages for other such as availability of transportation vehicles, saving labor for agriculture, and service facilities being available at closer distances. (viii)The marketing situations for agricultural products have improved over past six years for 71.9% of respondents, have remained the same as before for 27.8%, and has become more difficult for 0.3% of them (only from Zhemgang Dzonkhag where markets and consumers are not available along national highways despite having farm road connectivity). (ix) In terms of walking distance to nearest road point, it is within ½ an hour for 78.2% 572 HHs; between ½ to 1 hour for 14.5% (106) HHs; between 1-3 hours for 4.4% (32) HHs and between 4-6 hours for 2.9% (21) HHs. This is a substantial improvement in access to markets, as in 2007 the nearest road was within 1 hour walking distance for 45.7% HHs, which improved to 50.6% in 2011 and now to 78.2% HHs having nearest road within ½ an hour of walking distance. This shows 191% increase over May 2008 MTR results on HHs having reach to road within ½ an hour of walking distance. (x) The main source of income for rural population was agriculture for 57.7% of respondents. These were from marketing of orange, vegetables, rice, ginger, potato, chilli, and maize. Livestock products remained another source of income for 17%, followed by wage income (14.5%), remittances (7.4%), business (3.2%), and forest resources (0.2%). As compared to HHs income in the past six years (pre and post situation), the annual income per HHs is above Nu. 30,000 for 43.5% HHs, which is an increase of 354.3% as compared to HHs with this level of income in past six years. Presently, remaining 24.4% HHs have income between Nu. 5001–Nu. 15000 per year; 16.4% have between Nu. 15001–Nu. 30,000 per year; and 15.7% still have less than Nu.5000 per year. (xi) With regards to sustainability, the activities initiated and implemented by DRDP are picked up and programs are further incorporated into the government’s 11th FYP, the post project period (2013 – 2018). However, adequate resources for these and other unplanned activities could still be an issue for sustenance. (xii) There were some innovations of the project. The use of pipes for irrigation was new that not only minimized losses through infiltration but also provided experiences to the 59 engineers. The aeroponics facility for seed production is the first of its kind in the country and also probably first in the South Asian region especially in terms of its usage in potato seed production. The e-pest surveillance system is ready for use. The positive selection process to raise awareness on the quality seed, i.e to produce quality seed and to impart knowledge on the safe storage of seed for the next season was an innovation of the project. Another innovation was in providing irrigation facilities for vegetables cultivation with pipes for people from Orong Geog where farmers suffered without proper irrigation facilities. 60 Annex 6. Stakeholder Workshop Report and Results N/A 61 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Project Development Objective and Components: The objective of the project was to “improve market access and increase agricultural output for rural communities in selected areas of Bhutan to contribute to RNR strategy of 9th FYP”. Project Components: The project activities were distributed amongst three components: (a) Rural Infrastructures (improvement of farm roads and irrigation canals); (b) Renewable Natural Resources Centers and Support for Improved Technologies and Marketing of Rice, Maize and Potato; and (c) Institutional Strengthening, Project Management and Monitoring. Main Beneficiaries: At project appraisal 62 Geogs were targeted and by closure, the project impacted (direct 11 ) beneficiaries were from 78 Geogs with 28,860 households (population of 189563) which is 40.76% households from a total of eleven Dzongkhags. At project completion, the total HHs (for 11 Dzongkhags) were 71804 HHs (32520 HHs for six Dzongkhags and 39284 HHs for additional 5 Dzongkhags) with a total population of 415583 numbers 12. The women beneficiaries are 202445 numbers out of which 91643 numbers are direct beneficiaries (45.25% of total female population) and remaining 110802 numbers (54.73% of total) are indirect beneficiaries. Apart from farmers, government officials were also the beneficiaries (through capacity building). The implementation partners’ (government units and centers) such as NSC, National Plant Protection Centre (NPPC), National Post Harvest Centre (NHPC), and Renewable Natural Resources Research and Development Centers (RNR-RDCs)’s existing technologies and facilities were strengthened by DRDP-AF. Monitoring and Evaluation: A social and environmental screening criterion was used during planning period and before implementation of the activities in case of rural infrastructures (for farm roads, PTT and irrigation channels). The responsibility for regular monitoring during the process of implementation remained with Geog officials (agriculture and the Gup). Dzongkhag Engineers and Agriculture officers were also involved in the monitoring. However, the heavy work load for Dzongkhag engineers and merger of irrigation section with Dzongkhag engineering section diluted focus on irrigation and farm roads and full time monitoring by the engineers were not in place. Though project trained a few engineers on survey and design of irrigation, inadequate technical expertise with engineers on irrigation and farm roads and also on the environmental engineering parameters were mentioned by most of the Dzongkhag engineers. The execution and supervision was mostly based on the in-house experiences gained by the engineers in similar structures. The officials from PCU and collaborating partners (RNR-RDCs, NPPC and NSC) were also involved in monitoring activities specific to them. 11 Direct beneficiaries: Households receiving project services directly; Indirect Beneficiaries are remaining households 12 10th and 11th Five Year Plans, Gross National Happiness Commission, Royal Government of Bhutan 62 Key Factors Effecting Implementation and Outcomes: (i) Initially being not oriented to WB procurement guidelines, it took much time for PCU and Dzongkhag officials to get acquainted with it. Being a strict process for selection of contractors, many got disqualified on a few occasions and required re-tendering at times. On the aspects of free labor contribution by the beneficiaries for renovation, rural households facing shortage of productive manpower with rural-urban migration faced difficulty in providing timely labor. (ii) All DRDP Farm Roads and PTTs were screened and environmental clearances were issued by the District Environmental Office. A factor contributing to improved environmental knowledge of engineers and construction of environmental friendly roads was the technical support provided by Netherlands Development Organisation (SNV- Bhutan) in the initial years of DRDP. With SNV support to Dzongkhang in preparation of site specific environmental management plans (EMPs), the district capacity to prepare these plans improved. All DRDP farm roads since then had EMPs prepared. These EMPs were presented in simple tabular form and in general, identified potential issues and risks. (iii) In general the co-management involving PCU, Dzongkhags, Geogs and relevant partner organisations was another key issue that lead to smooth implementation. Flexibility on utilising savings from an activity provided opportunity to invest in the same or other required but associated activities that was not funded by the government. Project Costs and Financing: The consolidated (reconciliation) for the period 2005 to June 2013 shows that a total of US$ 11.01 million was received against expenditure of US$ 10.71 million. The total fund received for loan and grant by the year are presented in Table 1. Table 1: Fund received from World Bank as credit and grant Loan Fund received from World Bank Year 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12 2012-13 Nu. 23,015,000 53,760,000 85,155,734 49,714,669 45,290,500 33,337,000 7,366,239 10,822,000 Nu. in million 23.0 53.8 85.2 49.7 45.3 33.3 7.4 10.8 Nu. in UD$ in Nu. million million Total 308,461,142 308 7 Grant Received from World Bank Year 2011-12 2012-13 Total Nu. Received (Nu.) 60,682,000 126,299,000 186,981,000 Nu. In Million 61 126 187 US$ in million 4.16 63 Overall Assessment Project Achievements and Outputs: The project’s achievements were the following: renovated irrigation channel (58 Geogs); new irrigation (11 Geogs); new farm roads (16 Geogs); farm road improvement (12 Geogs); Power tiller tracks (11 Geogs); RNR centres (6 Geogs); Maize activities (9 Geogs); Paddy activities (7 Geogs) and Potato activities (9 Geogs). The seeds promotion and distribution as supported through RNR-RDC and Potato Program is across all Geogs of the Dzongkhags, wherein if accounted for coverage, would be all Geogs (135 numbers from 11 Dzongkhags). The project has made good achievements in successfully implementing activities as planned till December 2014. The consolidated physical progress for the period 2005– 2014 reflects the following achievements: 101% for new farm roads; 102% for farm road improvement; 100% for power tiller track construction; 100% for construction of new irrigation channels; 99.48% for renovated irrigation channels; 100% for construction of RNR/Geog centres; 83% for construction of bridges and 276% for farmers’ trainings/ demonstrations and trails .These % achievements (for road, irrigation and PTT) are however compared to the revised targets as reflected in progress reports and not as per PAD or DRDP-AF targets, which if measured would be as follows: • 138.27 Kms of new farm road constructed against the DRDP-AF target of 95.5 Kms (145% achieved) benefitting 2520 households. • 69.35 Kms of farm roads improved against the DRDP-AF target of 15 Kms (462% achieved) benefitting 1575 households. • 128.99 Kms of power tiller tracks constructed against DRDP PAD target of 114 Kms (113% achieved) benefitting 1831 households. • 42.50 Kms of new irrigation channel constructed against the DRDP-AF target of 44.3 Kms (96% achieved) benefitting 418 households with a command area of 687acres. • 521.24 Kms of irrigation channels renovated against the DRDP-AF target of 393 Kms (133% achieved) benefitting 4594 households with command area of 11784.56 acres. • 6 RNR centres constructed against the same target of DRDP PAD benefiting 2430 HHs. • 5 bridges constructed against the DRDP PAD target of 6 (83% achieved). • 3311 numbers of farmers trained against the DRDP PAD target of 1200 (276% achieved) Apart from rural infrastructures, as capacity building and extension activities related to paddy, maize and potato were also the focus of the project during DRDP-AF, summary of the outputs reported by various commodity coordinators (rice, potato and maize) are as follows: 64 • promoted 26 tons of high yielding variety of paddy seeds in 6 Dzongkhags, • spring rice cultivation undertaken (for 22 acres) with double cropping benefitting 48 HHs (Thedtsho Geog), • land terracing done for extended research within the RNR-RDC sub-centres, • contributed to 5054 acres of wetland being brought under mechanisation till August 2013 owing to farm mechanisation demonstrations, recording 1553 acres of rice paddies under mechanisation from Punakha and Wandue Dzongkhags alone (a eight times more than previous year), • advancement done in research of rice germplasm, • double cropping of rice initiated at Lamoyzinkha (Dagana Dzongkhag) on 6 acres land, • poly houses in Baap Geog (Punakha) also used for vegetables nursery apart from paddy alone, • tray nursery training undertaken leading to 25 acres land under paddy mechanisation, • publication undertaken for “knowing maize varieties,” • aeroponics facility for potato seed production operational since April 2012, • 60 acres land brought under potato seed production in Phobjikha, NSC, • 3G seed strategy for revamping the seed potato production system is in its advanced stage, • 17345 minitubers for Potato harvested unit in 2013 from aeroponics, • 7 tons of G2 potato seed of Desiree variety harvested in 2014, • 875 tons seed potatoes produced and marketed, • increased from earlier 5 tons production to 60 tons potato seeds from Phobjikha NSC in 2012, • published booklet for aeroponic production of potato mini-tubers and 500 copies made, • published manual on positive selection to improve quality of farm saved potato seeds and 1000 copies made, • published poster on 3G strategy and 16 copies made, • 9 community maize seed production group are functional having total of 85 members, • 1250 kgs of hybrid maize seeds imported and planted in 100 acres, • spring maize introduced (249 acres cultivated covering 324 HHs), • Grey Leaf Spot - GLS tolerant foundation maize seeds were produced (3.2 MT), • report on maize products diversification is ready for up-scaling, • Banana tissue culture instituted, • E-pest surveillance system installed and ready for use. Furthermore, other significant achievements are development of national irrigation data base along with maps at Geogs and Dzongkhag levels as until recent the data on irrigation was old and incomplete as well unreliable; and the NSC facilities for seed production were enhanced with support for water reservoir, micro irrigation, installation of 2 units of seed processing units, construction of GI chain link fence and also supply of four power tillers. 65 Relevancy: The relevancy of the activities from beneficiaries’ (farmers) perception reflects that most relevant were: farm roads (for 28.8 % of total 731 respondents); renovated irrigation channels (24.7%); power tiller tracks (12%); seed distribution (10.3%); Trainings / demonstration (5.5%); new irrigation channels (5.1%); Geog / RNR centres (4.8%); bridges (3.5%); and smaller percentages for other such as nursery houses, post harvest machineries / equipments, trainings and support for group formation (mainly road user and water user groups). Linking relevancy to value addition for the households (HHs) and for improved access to markets, the survey revealed: increased paddy production (for 18.5% of total respondents); increased maize production (18.1%); increased cash income (12.7%); eased marketing (12.6%); reduced walking distance (12.5%); increased potato production (12.2%); and smaller percentages reported for others such as availability of transportation vehicles, saving labour for agriculture, and service facilities being available at closer distances. All of these show that activities were much relevant to the needs and situations of the communities. DRDP activities were in line with the government policy to enhance self sufficiency in food crops that was challenged with remoteness of the villages without road connectivity and irrigations facilities. The construction of power tiller tracks opened space for widening some of them to farm roads. Having road connectivity and irrigation facilities presently enabled Dzongkhags to undertake agreement with the present government on increasing production for self sufficiency. Apart from that, the focus remained on activities that lagged without fund such as promotion and spread of improved seeds, acquiring more knowledge and exposure to technologies for implementers, overcoming the challenge faced with maize (devastating GLS disease) with seed replacement program, and timely support for potato seed production using aeroponic facility, which were thus much relevant. Efficiency: The survey illustrated that the quality of infrastructures were good (for 64.8% respondents), very good (for 21.3%), and not so good (only 2.3%), while remaining 11.5% remained non-beneficiaries of rural infrastructures. Similarly the quality of improved seeds promoted and distributed (maize, potato and paddy) were found good (by 46.8% respondents), very good (for 11.1%), not good (for 0.5%), and remaining 41.5% remained to be non-recipients of seeds. The quality of demonstrations/trainings in terms of delivery and methods, and understanding by the farmers showed that majority did not avail any trainings supported by DRDP (55.7% respondents) as usual agricultural trainings to farmers is a regular mandate of the Geogs extension supported by RGoB fund. Out of remaining 44.3% respondents that attended trainings, it was very good (for 23.7%) and good (for 20.7%), confirming that the quality and efficiency of demonstrations and trainings to farmers Effectiveness: As analyzed in earlier sections, the activities have led to achievements of targets. In terms of achieving the goals and objectives as project outcomes, the overall increase in production and yield of paddy for 37 Geogs supported for irrigation were 66 assessed using available agricultural statistics. The analysis shows sharp drop in harvested area owing to factors outside the control of the project (i.e, rural-urban migration, shortage of agricultural manpower in rural households, land being left fallow, and migration of people from villages for better income opportunities).Paddy productivity (yield) has increased to average of 1076/acre/kg in 2012 as compared 984 in the year 2010; but had overall decrease in production owing to decreased harvested area. A separate analysis on potato for targeted Geogs showed an average increased productivity from 4947 /acre/kg in the year 2010 to 5374 in the year 2012; and increased production from 9333360 kgs (2010) to 11352540 (2012). A similar analysis for maize showed increased productivity from 1025 to 1034; and increased production from 4256530 kgs to 4461293 kgs. Looking at Dzongkhag data for paddy, it showed that for DRDP first phase 6 Dzongkhags, there have been increase in production from 20289 MT to 23719 MT; and increased yield from initial 823 to 1440 in the year 2012 as compared to 2006. For all eleven Dzongkhags, there was increase in production (from initial 49623 MT to 56310 MT) and increase in yield (from initial 1159 to 1446) in the year 2012 as compared to 2010. Note: Compiled data for Dzongkhags for paddy for the year 2013 is not available yet. Similarly Dzongkhag’s data for potato showed increased yield (from initial 3643 to 19864); and increased production (from initial 211.52MT to 20619 MT) in the year 2013 as compared to 2006 for first 6 Dzongkhags. For all eleven Dzongkhags, there has been increased production (from initial 24870 to 32576) despite decrease in yield in the year 2013 as compared to 2010. Dzongkhag data for maize showed that there have been increased yield (from initial 848 to 1111) and increased production (from initial 20199 MT to 22514 MT) in the year 2013 as compared to 2006 for first phase 6 Dzongkhags. Similarly, there was increased production (from initial 43233 MT to 54315MT) and increased yield (from initial 932 to 1177) in 2013 over 2010 for all Dzongkhags. In terms of vegetables (reported having intensified after having road), Dzongkhag data revealed increase in production and harvested area for major vegetables like beans, cabbage, cauliflower and carrot. In case of mandarin (the main cash crop apart from potato), despite decrease in numbers of bearing trees and in production, there was increase in yield (from initial 38 to 42) in the year 2012 over 2006 for first 6 Dzongkhags. In the year 2012 compared over 2010, there was decrease in numbers of bearing trees; decrease in production and increase in yield (from initial 33 to 44) for all Dzongkhags. During field survey, in areas where irrigation channel were constructed or renovated, farmers have reported increased paddy production owing to availability of water during season and for the period required. The food lasts from own cultivated paddy was also 67 reported to have increased. The case studies collected during field survey shows substantial outcomes out of irrigation. Impacts: Impacts based on livelihood domains were diverse and many. Collected through FDGs and interview with beneficiaries and Dzongkhags and Geogs officials, with regards to physical assets, availability of farm roads and increased income from agricultural products have helped farmers to purchase household assets such as television, rice cookers, and refrigerators and to bring in good furniture for few. Sergithang farmers (Tsirang) reported that after road and electricity, now 60% households have television. Many reported to have purchased vehicles (cars, pick-ups and trucks) including power tiller after having road and PTT. For example people from Barshong Geog (Tsirang) reported that about 80% of the households have either repaired their houses or made a new one after they could transport needed materials at cheaper costs after having farm road. Likewise, apart from supply of post harvest machineries on subsidy from the project or the government, few in every Geogs visited reported to have purchased on their own after having road and also irrigation. Better shed for cattle, purchase of improved cattle and establishment of livestock farms (poultry) were also reported in few occasions. Rangthangling Geog farmers (Tsirang) mentioned that after road, every household now have a trained driver, even if not owning a vehicle yet but soon would have capacity to purchase vehicles with income from cardamom. Rukha village farmers (Athang: Wangdue) despite having suspension bridge at entry point to power tiller track from national highway (Tsirang – Wangdue), three people have power tillers that is used on hire, not only for ploughing but also for transporting (bringing in essential items to villages and marketing their traditional dried fishes). The comparison of the rural infrastructures (mainly roads and irrigation schemes) at the start of the project and closure shows the followings: irrigation schemes (521 nos in 2007 vs 1220 nos in 2013); length of irrigation channels (1616.93 Kms vs 3060.205 Kms now); and length of farm roads (473.02 Kms vs 2392.87 Kms now). With regards to food security, increases in paddy production were reported by farmers wherever irrigation was supported. With roads and ease in marketing, intensification of vegetables and with varieties including potato was reported that contributed towards improvement in nutrition at households. For majority of the households rice is the staple food while maize is also consumed by some. As few examples, Bongo (Chukha) farmers mentioned that after renovation of irrigation channel, they have rice sufficiency for at least 6-7 months for all HHs. Dremtse (Mongar) farmers mentioned that even if they consume maize alone, they will not only have sufficiency for 12 months but will also have some marketable surplus. Impacts on institution creation and strengthening were more evident in areas where RNR centres and farm roads were constructed. With RNR centres, people now have 68 extension facilities at nearer distances. With road, extension service providers reach their village unlike in past requiring them to walk long distance and gather at one point. There has been increased workload for rural women after having irrigation and roads (except in very few places) as in many places earlier fallow land were cultivated for paddy and vegetables production were intensified, as reported by women respondents. The increase in workload is equally for their male partners too. As gender concerns, lives however have improved for women with comfortable households assets (rice cooker saving time for cooking; time saved from fetching firewood, contributing to better health without smoke in kitchens, and lighting facility with electricity). Decision making is mostly joint between male and female, and most of the trainings or meetings were usually participated by women. The access to credits has also improved after having road with mobile banking facilities and banks being opened up in several Geogs now. About 21% of the respondents reported having availed loans by their HHs mainly for the following purposes: purchase of seedlings (cardamom and orange); purchase of improved cattle; house construction; purchase power tillers, power chain, post harvest machineries; establishment of agricultural farms (mainly livestock related); and even for commercial business enterprises. With regards to financial resources and income, the main source of income for rural HHs was agriculture for 57.7% of respondents. These as mentioned were from marketing of orange, vegetables, rice, ginger, potato, chilli, and maize. Livestock products remained another source of income for 17%, followed by wage income (14.5%), remittances (7.4%), business (3.2%), and forest resources (0.2%). As compared to HHs income past six years (pre and post situation), the annual income per HHs is above Nu. 30,000 for 43.5% HHs, which is an increase by 354.3% (manifolds) as compared to HHs with this level of income in past six years. Presently, remaining 24.4% HHs have income between Nu. 5001–Nu.15000 per year; 16.4% have between Nu. 15001–Nu.30,000 per year; and 15.7% still have less than Nu.5000 per year. Sustainability: For continuity and sustainability of programme initiated and implemented by DRDP, resources both in terms of human and fund have to be made available for all activities (under three components supported by DRDP) by the government to maintain and upscale the momentum gained thus far. To this effect, the activities thus initiated and implemented by DRDP are actually picked up and programs are further incorporated into the government’s 11th FYP, the post project period (2013– 2018). Innovations: There were some innovations of the project. The use of pipes for irrigation was new that not only minimized losses through infiltration but also provided experiences to the engineers. Engineers now have added knowledge on irrigation survey, design and water management, through which lift irrigation and water harvesting techniques were added as initiatives of the project. 69 • DRDP enabled the National Potato Program (NPP) and the National Seed Centre (NSC) to produce potato seed tubers using aeroponic facility at the NSC complex at Paro since April 2012. It is the first of its kind in the country and also probably first in the South Asian region especially in terms of its usage in seed potato production. • Other innovation was promotion and spread of biotic and abiotic stress tolerant crop varieties (drought tolerant rice varieties and disease resistant maize varieties). The positive selection process to raise awareness on the quality seed: to produce quality seed and to impart knowledge on the safe storage of seed for the next season was an innovation of the project.. • Another significant innovation is providing irrigation to farmers from Orong Geog (Samdrupjongkhar) for vegetables cultivation. Lack of adequate irrigation water was a limiting factor for agriculture farming in the Geog and local school children suffered from malnutrition owing to lack of adequate nutritious vegetables in the school’s mess menu. While formation of vegetable group was supported by the government to make link for vegetables supply to the school, the project funded irrigation scheme (long piped irrigation water supply scheme). With a flow of 5.3 litres per second the scheme has now alleviated the problem of water shortage for vegetable production in the area. • The e-pest surveillance system is another innovation of the project that is designed to capture information of all pest organisms (insect pests, disease organisms, weeds) that causes or has potential to cause damage and loss to crops of importance (economically and food security) over time and space . Performance of Partners: On the whole the implementation support and guidance provided by the Bank to the project team was good and satisfactory. However, at the early stage of project implementation and following the transfer of the original Task Team Leader (TTL) there was a complete break in communication between the Bank and project team. In absence of a replacement and without an interim TTL the project team could not communicate with Bank on any issues. Added to it, the inexperience of the project team in planning and project management collectively contributed to delayed implementation of project activities on the ground. The Engineering Division under Department of Agriculture (DoA), Ministry of Agriculture and Forests (MoAF) was the Project Coordinating Unit (PCU). Instead of a separate management unit, the project implementation was instituted within the existing institutional set up of MoAF. However, without a separate project management unit with adequate personals, the PCU suffered a lot, as evident from inadequate documentation of the progress and also inadequate monitoring. With regards to procurement procedures, for each contract to be financed by the Credit, the different procurement methods or consultant selection methods, the need for prequalification, estimated costs, prior review requirements, and time frame needed to be agreed between the Borrower and the Bank project team in the Procurement Plan were some criteria. In the beginning, the project team was not conversant with the Bank’s 70 procurement guidelines and system and a lot of difficulties were faced then. Though an assessment of the procurement capacity of implementing agencies were done in the beginning of the project, it was also identified that government procurement differed from Bank’s guideline mainly for rejection of bids, dispute resolution and negotiation with bidders. Lack of familiarisation on Bank’s procurement procedures with the Dzongkhags were also identified in the very beginning. Therefore many tenders had to be re-tendered in the beginning as one or another procurement requirement were not understood and met, which led to delays in implementation of project activities. However, towards the later part of original DRDP phase and after the appointment of a new TTL and with few changes in the Bank’s team members many of the issues faced earlier were addressed. Since then project implementation in general became smooth and was on track. The project management was able to receive timely support and assistance from the Bank’s team. The support of SNV during the original DRDP played a crucial role towards Institutional Strengthening and Capacity Building component. Given that the project coordination unit consisted of only the Project Coordinator and finance personal, SNV’s support was much needed and beneficial. Delay in fund release and timely reporting of expenditure statements was an issue during the early stages due to the Dzongkhag accounts staff not being familiar with the project’s financial system and fund disbursement arrangement. Often funds were not utilised on time, which led to funds getting “spilled over” at the end of every fiscal year. This caused problems in reconciling the funds, especially when the project accounts team did not provide adequate and timely oversight. Here again, inadequate numbers of Dzongkhag staffed who were trained and experienced in project’s procedures became a problem for the project management. On the positive side, some project Dzongkhags were more proactive and efficient in implementing their activities. Judicious use of funds and efficient implementation made it possible for these administrations to surpass their project targets. This was possible mainly due to the strong leadership provided by the Dzongda (district administrator) and the district agriculture officer (DAO). The Department of Public Accounts & Department of National Budget provided necessary support to the project without major issues. The delays in fund release that occurred were mainly due to the weakness of PCU and implementing agencies. The collaborating partners in DRDP-AF period (RNR-RDCs, NPPC, NSC and AMC) were all proactive and implementation were taken as priority as the project supported those key areas that was lagging without fund and many supported activities were initiatives and innovations that added much value not only to the end beneficiaries (farmers) but also to the collaborating partners. 71 7. Lessons Learned • The importance of a strong project management team for a successful project implementation cannot be overemphasized. The team needs to be also staffed adequately. In this regard, DRDP with its inexperienced project coordinator and finance staff suffered during its early stages. In the absence of focal persons the project coordinator had to perform the duties of procurement officer, environment and social focal persons and M&E officer in addition to other regular mandates within the division. Needless to mention, the project coordinator could not fulfill all the needs. • The importance of M&E officer to monitor, collect and document project data, information, impacts and experiences is amply demonstrated by the existence of minimal project reports. • Often most donor funded projects have differing fund disbursement and procurement systems.than the RGoB’s. This creates confusion among the implementing agencies, especially when there are many stakeholders and involves local governments. This leads to delay in fund release, replenishment, reconciliation and re-bidding of tenders. It is therefore deemed crucial to conduct a training workshop for all the key stakeholders on the project’s budgeting, fund release and procurement systems in the first few months of project. Such an exercise once annually for the next two years would have been much appropriate. • The DAO as a focal person needs to nurture and maintain a close working relationship with the Dzongkhag Engineering Sector for civil works. Most of the farm roads under DRDP were constructed with budget allocation far below the prescribed amount of Nu 3.0 million per Km. Consequently, their quality was below the minimum standard required for a farm road. This came about because planning and budgeting were left to the local governments as required by the decentralization policy without any control from the Centre (Project Coordination Office). To ensure quality and make farm roads pliable throughout the year future projects should make it mandatory for new farm roads to meet the minimum technical standards, and a strict monitoring can be instituted only by having M&E personal under a separate project management unit. • The CBSP approach used under DRDP to grow and multiply quality maize seeds has been successful to a great extent. This could be replicated in similar projects in the future. • The lengthy process (engineering studies, social, environmental clearances, bidding) involved in getting start farm roads and irrigation construction activities and the delays it causes indicate that these activities should be undertaken and completed in the first year for all the sub-projects. This will leave the remaining years for the actual construction. The current practice of planning and implementing these activities is time consuming and thereby causes delays. 72 • New farm road designs should be done only by qualified and experienced civil/road engineers. Prior assessment needs to be carried out whether this capacity exists with the implementing agencies. Where it is not available this task needs to be outsourced to local consultants. This strategy is critical in creating and maintaining good and environmental friendly roads. • Institutional capacity building is one of the key factors in sustainability of any project interventions. Few good examples to cite are: the training of NSC staffs on tissue culture of potato and banana has led to propagation of high quality and disease free basic seeds. Over the past year the quantity of these seeds propagated and multiplied for mass distribution has increased manifold. Same is the experience with maize seed. • Despite devolution of authority for planning and implementation to the local government, often activities had to be implemented and monitored only by Geog agriculture officers at Geog level. There are many times weak ownership of the infrastructures with the local government as after completion of the construction, it usually lies again with Geog agriculture office for further monitoring (mainly for irrigation channels and roads). • Though Dzongkhags and Geogs agriculture officers implemented the activities, they had inadequate knowledge and skills with regards to roads and irrigation technologies. This also applies for the engineers as all are not trained on road or irrigation, Though measures were taken to expose and train few engineers, it would have been good to orient all agriculture officers and engineers on basics (technologies) of roads / PTTs and irrigation schemes before actual implementation of such a big infrastructure development activities. 73 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders N/A 74 Annex 9. List of Supporting Documents • Project Implementation Plan • Project Appraisal Document for Bhutan: Decentralized Rural Development Project (DRDP) dated February 2, 2005 (Report No: 30208) • Aide Memoires, Back-to-Office Reports, and Implementation Status Reports. • Project Progress Reports. • Borrower's Evaluation Report dated December 2014 • Agriculture and Rural Transport: Joint IFAD/IFTRD/ILO Workshop. June 2008. Reported by the International Forum for Road Transport and Development (IFRTD) at: iftrd.org/new/issues/agriculture.php • A Strategy to Fight Poverty. World Bank, Philippines. 1996 • The performance assessment of the IFAD-financed Agriculture, Marketing and Enterprise Promotion Programme (AMEPP) in the West of Bhutan, noted a beneficiary per kilometer ration of 10 households per kilometer and deemed it ‘sparse’. AMEPP Project Performance Assessment. Report No. 3300-BT, IFAD, August 2014. *including electronic files 75 Map IBRD 33672 76 IBRD 33672 90 ° 00' 90 ° 30' 91 ° 00' 91 ° 30' 92 ° 00' ELEVATION IN METERS: 5000 BHUTAN 28 ° 30' 28 ° 30' I G H H I M A L A Y A 3000 1000 DECENTRALIZED RURAL DEVELOPMENT PROJECT H H I M A L A Y A E R N I N PROJECT DZONGHAGS (DISTRICTS) H I M A L AYA FOOTHILLS RIVERS AIRPORT PAVED ROADS 89 ° 00' 89 ° 30' UNPAVED ROADS NATIONAL CAPITAL GASA DISTRICT CAPITALS 28 ° 00' OTHER TOWNS 28 ° 00' hu oC C DZONGKHAG (DISTRICT) BOUNDARIES Ph Gasa ha INTERNATIONAL BOUNDARIES m ka Ch BHUTAN u Mo C h u LHUENTSE Tashithang BUMTHANG TRASHI WA N G D U E Minjey ' YA N G T S E 0 10 20 30 PUNAKHA Tang THIMPHU Bumdeling Shengana Lhuentse KILOMETERS Tango PHODRANG Kurjey Tangmachhu Trashi T 'Yangtse rashi 'Yangtse Pa Chari ro PA R O Samtengang K u lo n Punakha Ch Lobesa Trongsa u Jakar K u r u C hu Drukgye Changkhala g 27 ° 30' 27 ° 30' THIMPHU Ch Wangdue Yangnyer Bu Haa Shari u Paro m Chu Bartsham th Damthang an g Khasadrapchu Phubjikha Phongme Haa T R O N G S A San Drametse ko Tsaphel Trashigang Lingmithang Mongar sh Yadi Yadi HAA as an TRASHIGANG M Zhemgang Kilikhar Chaskhar ZHEMGANG MONGAR Wamrong Torsa Yebilaptsha Tendru DAGANA Thrimshing Chhukha Daga SAMCHI TSIRANG Gonpasingma W CHHUKHA Gomphu an Damphu S A R PA N G PEMA g Sipsu Dorokha Dungna SANDRUP 27 ° 00' Norbuling Pemagatshel 27 ° 00' Chu G AT S H E L JONGKHAR To n gsa Samrang Daifam Mirchim Dagapela Pelrithang Dalim Panbang Ch Samtse Gelephu as Man u Sarpang Pagli Phuentsholing Nganglam Samdrup Bhangtar Singhi Lalai Jongkha This map was produced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information Raidak Phibseo shown on this map do not imply, on the part of The World Bank Kalikhola Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 89 ° 00' 89 ° 30' 90 ° 00' 90 ° 30' 91 ° 00' 91 ° 30' 92 ° 00' NOVEMBER 2004