Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00004665 IMPLEMENTATION COMPLETION AND RESULTS REPORT 47771-IN ON A CREDIT IN THE AMOUNT OF SDR 127.3 MILLION (US$197 MILLION EQUIVALENT) TO THE REPUBLIC OF INDIA FOR THE Uttar Pradesh Sodic Lands Reclamation III Project June 18, 2019 Agriculture Global Practice South Asia Region CURRENCY EQUIVALENTS (Exchange Rate Effective May 31, 2019) Currency Unit = Indian Rupee (Rs) 69.88 = US$1 US$1.38 = SDR1 FISCAL YEAR April 1 – March 31 Regional Vice President: Hartwig Schafer Country Director: Junaid Kamal Ahmad Senior Global Practice Director: Juergen Voegele Practice Manager: Mary Kathryn Hollifield Task Team Leader(s): Abel Lufafa ICR Main Contributor: Kunduz Masylkanova ABBREVIATIONS AND ACRONYMS BCR Borrower’s Completion Report CAS Country Assistance Strategy CPF Country Partnership Framework ENPV Economic Net Present Value ERR Economic Rate of Return ESMF Environmental and Social Management Framework FM Financial Management FPO Farmer Producer Organization GoI Government of India GoUP Government of Uttar Pradesh GP Gram Panchayat (Village Council) IDA International Development Association ID Irrigation Department IA Implementation Assistant IUFR Interim Unaudited Financial Report KVK Krishi Vigyan Kendra M&E Monitoring and Evaluation MTR Mid-Term Review OFD On-Farm Development O&M Operation and Maintenance NGO Non-Governmental Organization PDO Project Development Objective PG Producer Group PIP Project Implementation Plan RH Rural Haat (Sodic Haat) R&R Resettlement and Rehabilitation SIC Site Implementation Committee SHG Self-Help Group ToC Theory of Change UP Uttar Pradesh UPBSN Uttar Pradesh Bhumi Sudhar Nigam (UP Land Development Corporation) UPSLRP Uttar Pradesh Sodic Lands Reclamation Project WB World Bank WSHG Women Self-Help Group WUG Water User Group TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 6 A. CONTEXT AT APPRAISAL .........................................................................................................6 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION ................................................................9 II. OUTCOME .................................................................................................................... 11 A. RELEVANCE OF PDOs ............................................................................................................ 11 B. ACHIEVEMENT OF PDOs (EFFICACY) ...................................................................................... 11 C. EFFICIENCY ........................................................................................................................... 13 D. JUSTIFICATION OF OVERALL OUTCOME RATING .................................................................... 14 E. OTHER OUTCOMES AND IMPACTS (IF ANY) ............................................................................ 14 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 15 A. KEY FACTORS DURING PREPARATION ................................................................................... 15 B. KEY FACTORS DURING IMPLEMENTATION ............................................................................. 16 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 18 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 18 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ..................................................... 19 C. BANK PERFORMANCE ........................................................................................................... 20 D. RISK TO DEVELOPMENT OUTCOME ....................................................................................... 20 V. LESSONS AND RECOMMENDATIONS ............................................................................. 21 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 23 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 31 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 33 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 34 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ... 38 ANNEX 6. SUPPORTING DOCUMENTS .................................................................................. 43 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name P112033 Uttar Pradesh Sodic Lands Reclamation III Project Country Financing Instrument India Investment Project Financing Original EA Category Revised EA Category Partial Assessment (B) Partial Assessment (B) Organizations Borrower Implementing Agency Department of Economic Affairs Uttar Pradesh Bhumi Sudhar Nigam Project Development Objective (PDO) Original PDO The project development objective (PDO) is to increase agricultural productivity of degraded lands in selected areas of UP. The objective would be achieved by reversing water-induced land degradation, enhancing soil fertility, and improving the provision of agriculture support services. PDO as stated in the legal agreement To increase agricultural productivity on degraded lands in selected areas of Uttar Pradesh by: (i) reversing water- induced land degradation; (ii) enhancing soil fertility; and (iii) improving the provision of agricultural support services. Page 1 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 197,000,000 197,000,000 174,629,819 IDA-46400 200,000 200,000 200,000 TF-98748 Total 197,200,000 197,200,000 174,829,819 Non-World Bank Financing 0 0 0 Borrower/Recipient 49,200,000 4,920,000 4,920,000 Local Communities 25,800,000 22,300,000 22,300,000 Total 75,000,000 27,220,000 27,220,000 Total Project Cost 272,200,000 224,420,000 202,049,819 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 30-Jun-2009 18-Sep-2009 29-Apr-2013 31-Dec-2015 29-Dec-2018 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 28-Aug-2014 88.24 Change in Loan Closing Date(s) 03-Aug-2017 147.24 Change in Loan Closing Date(s) KEY RATINGS Outcome Bank Performance M&E Quality Satisfactory Moderately Satisfactory Modest Page 2 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 30-Nov-2009 Satisfactory Satisfactory 0 02 26-May-2010 Satisfactory Satisfactory 3.18 03 12-Dec-2010 Satisfactory Satisfactory 5.82 04 09-Jun-2011 Satisfactory Satisfactory 22.01 05 17-Feb-2012 Satisfactory Satisfactory 35.73 06 02-Sep-2012 Satisfactory Moderately Satisfactory 40.92 07 24-Dec-2012 Satisfactory Moderately Satisfactory 52.39 08 29-Jun-2013 Moderately Satisfactory Moderately Satisfactory 63.49 09 29-Dec-2013 Moderately Satisfactory Moderately Satisfactory 69.04 10 06-Jun-2014 Moderately Satisfactory Moderately Satisfactory 88.24 11 31-Aug-2014 Moderately Satisfactory Moderately Satisfactory 88.24 12 16-Jun-2015 Moderately Satisfactory Moderately Satisfactory 115.06 13 29-Oct-2015 Satisfactory Moderately Satisfactory 115.06 14 09-Jun-2016 Satisfactory Satisfactory 134.96 15 15-Dec-2016 Satisfactory Satisfactory 134.96 16 19-Jan-2017 Satisfactory Satisfactory 147.24 17 20-May-2017 Satisfactory Satisfactory 147.24 18 07-Dec-2017 Satisfactory Satisfactory 153.40 19 06-Jun-2018 Satisfactory Satisfactory 174.14 20 03-Dec-2018 Satisfactory Satisfactory 174.83 Page 3 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) SECTORS AND THEMES Sectors Major Sector/Sector (%) Agriculture, Fishing and Forestry 90 Crops 50 Irrigation and Drainage 30 Public Administration - Agriculture, Fishing & Forestry 10 Industry, Trade and Services 10 Agricultural markets, commercialization and agri- 10 business Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Social Development and Protection 0 Social Inclusion 20 Participation and Civic Engagement 20 Urban and Rural Development 0 Rural Development 67 Rural Markets 10 Rural Infrastructure and service delivery 10 Land Administration and Management 47 Environment and Natural Resource Management 0 Renewable Natural Resources Asset Management 14 Biodiversity 7 Landscape Management 7 ADM STAFF Role At Approval At ICR Regional Vice President: Isabel M. Guerrero Hartwig Schafer Page 4 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Country Director: N. Roberto Zagha Junaid Kamal Ahmad Senior Global Practice Director: John Henry Stein Juergen Voegele Practice Manager: Gajanand Pathmanathan Mary Kathryn Hollifield Task Team Leader(s): Animesh Shrivastava Abel Lufafa ICR Contributing Author: Abel Lufafa Page 5 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Context 1. At appraisal Uttar Pradesh (UP) was experiencing slow growth and persistent poverty. For most of the years preceding appraisal, economic growth in the state was comparatively lower than for most of India (i.e. 5.3 percent, compared to the national growth rate of 7.7 percent), and so was its per capita income – standing at less than 50 percent of the national average. Between 35 and 50 million of the state’s 142 million rural inhabitants subsisted below the poverty line. 2. Regarding agriculture in particular, although growth of sector in the state eclipsed the national average (i.e. 2.1 percent compared to 1.1 percent between 2002-2007), this performance was still suboptimal. This was due to a confluence of structural and institutional issues, including: low productivity, land degradation, and weaknesses in dissemination of agricultural technology, inter alia. The low sector performance was a cause for concern especially as the share of agriculture (then a source of livelihood for two-thirds of UP’s population) in GDP had been steadily declining - from about 33 percent in 1999-2000 to 26 percent in 2007-2008. 3. The Government of Uttar Pradesh (GoUP) therefore sought to accelerate agricultural growth (targeting an initial annual growth rate of 4 percent in 2007, which was expected to reach 5.1 percent in 2012) and had undertaken several initiatives to address some of the constraints to sector growth through a number of schemes, programs and missions1. However, there still remained areas where intervention was critical if the target agriculture growth rates were to be achieved. These included: (i) physical reclamation of large tracts of land (estimated at 5,140,000 hectares - ha - at the time) facing a variety of degradation problems ; (ii) provision of support services to predominantly small and marginal farmers who cultivated about 60 percent of arable land; (iii) strengthening of farmer organizations, especially producer groups (PGs) and water user groups (WUGs); (iv) improving physical facilities for marketing of agricultural produce; and (v) and linking farmers to remunerative value chains that provide opportunities for post-harvest value addition. 4. Intervention in these areas required significant resources, beyond the medium-term state budget outlay for agriculture and therefore, the GoUP had sought World Bank (WB) financing to support some of these investments. Hitherto, the Bank had over the last 15 intervening years, partnered with the GoUP to reclaim over 258,000 ha of degraded (sodic) lands through two separate phases (Phase 1 - 1993 to 2001; and Phase 2 - 1997 to 2007) of the Uttar Pradesh Sodic Lands Reclamation Project (UPSLRP) and was uniquely positioned to continue supporting the GoUP’s land reclamation efforts - by scaling out the UPSLRP interventions to other degraded areas - as well as the related activities including: strengthening community participation, agricultural produce marketing, and investing in community institutions for improved agricultural livelihoods. 1 Including: improving soil health (with a budget of US$74 million between 2007-2012); support to program for extension services(US$73 million between 2007-2012); Kisan Mitra (Farmer’s friend) scheme (US$18 million between 2007-2012); technology mission on oilseeds, pulses and maize (US$45 million between 2007-2012); macro-management of agriculture (US$168 million between 2007-2012); drains maintenance; enhancing Rabi pulses production(US$46 million between 2007- 2012). Page 6 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) 5. It is against this backdrop that the WB accepted GoUP’s funding request for Phase 3 of the UPSLRP, which was approved by the Board in 2009. The expectation was that, improving agricultural productivity on degraded lands (predominantly cultivated by small and marginal farmers and low-income groups), would raise incomes of the poor, contribute to addressing income inequality - a key thrust of the then prevailing Country Assistance Strategy (CAS) - and foster inclusive growth, in line with the Government of India (GoI) priorities as had been laid out in the Eleventh Five-Year Plan (2007-2012). The support was consistent with the CAS’ pillar of “ensuring sustainable development” under which the Bank sought to support India to reduce the burden of land degradation on the population. Theory of Change 6. The project was designed to increase agricultural productivity on degraded lands. This was envisaged to be accomplished through: (i) land treatment to reclaim degraded lands; (ii) provision of agriculture support services; and (iii) strengthening and building capacities of institutions for market access. Increased crop productivity, it was assumed, would lead to increased rural income thus contributing to narrowing the income inequality in the long-term. Figure 1 below summarizes the project’s theory of change (ToC). Figure 1. Project Theory of Change 7. The ToC was based on several implicit assumptions, including: (i) timely availability and supply of key inputs (e.g. gypsum and press mud) necessary for effective land reclamation of sodic lands; (ii) interest and willingness of farmers to adopt new technology; and (iii) willingness of farmers to organize for cooperative action, both in production and output marketing. Page 7 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Project Development Objectives (PDO) 8. The project development objective (PDO) was to increase agricultural productivity on degraded lands in selected areas of Uttar Pradesh by: (i) reversing water-induced land degradation; (ii) enhancing soil fertility; and (iii) improving the provision of agricultural support services. Expected Outcomes and Outcome Indicators 9. The key expected outcome was increased agricultural productivity on degraded lands. This outcome would be measured by the following indicators: (i) increase in productivity of wheat and rice (tons/year); (ii) increase in cropping intensity (percent); (iii) increase in crop income per unit land (percent); and (iv) increase in land value (percent). Components 10. Component A: On-Farm Development and Land Treatment (appraisal cost2 US$170.7 million; actual cost US$184.3 million). This component would support sustainable reversal of water-induced land degradation on about 130,000 ha of sodic lands3,4 and pilot the reclamation of ravine lands on 5,000 ha. Key activities included: (i) mobilization of village communities; (ii) mapping and classification of sodic lands; (iii) formation of WUGs; (iv) on-farm development (OFD) through land-leveling, bunding, and linking field drains to link and main drains; (v) provision of shallow tube-wells for reclamation operations and irrigation; (vi) application of chemical/organic amendments and plant nutrients to the soil; and (vii) cultivation of rice-wheat-green manure crop cycle. 11. Component B: Improvement of Drainage Systems (appraisal cost US$39.2 million; actual cost US$42.3 million). Support under this component would improve the drainage systems to remove/leach effluents, excess rain, and irrigation water from reclaimed land and adjoining areas. Key activities included: (i) rehabilitation and maintenance of main drains and (ii) training and capacity building of Uttar Pradesh Bhumi Sundar Nigam (UPBSN), Irrigation Department (ID) staff, Site Implementation Committees (SIC)5 on technical and management aspects, and WUG6 for appropriate operation and maintenance (O&M) of drainage networks. 12. Component C: Agriculture Support Services (appraisal cost US$9.4 million; actual cost US$10.2 million). Project support under this component would fund increased agricultural productivity by promoting improved technology, better agronomic practices, and more effective provision of key support services. Main activities included: (i) training farmers in effective land and water management practices; (ii) dissemination of improved agricultural technology and production practices through on-farm demonstrations; (iii) support for livestock production; (iv) exposure visits, farmer fairs, animal health camps; and (v) training and capacity building of line department staff and other relevant support service providers. 2 Appraisal base costs included physical, but not price contingencies. 3 For project purposes, classified as: “C class land” (the most severe case of sodicity - mostly barren); “Class B land” (single- cropped sodic land - with very low productivity); and “Class B+ land” (double-cropped land - with low productivity). 4 Among others, targeted based on the following eligibility and selection criteria: not located in UPSLRP Phase 1 or Phase 2 villages; existing in patches at least 40 ha in size and where 65 percent of the patch was “C class land”; and exclusion of absentee farmers. 5 A village level body to coordinate project activities and distribute gypsum and other inputs. 6 WUG are informal groups of around 10-15 farmers operating around 4-5 ha of land and formed for collective use of irrigation from one pump set, of maintenance to field drains, implementation of OFD works, and to serve as a field operational unit for distribution of gypsum and agricultural inputs among other activities. Page 8 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) 13. Component D: Institutional Strengthening and Capacity Building for Market Access (appraisal cost US$9.7 million; actual cost US$10.5 million). This component would improve farm profitability and enhance livelihoods of the poor through improved marketing, value addition and enhancement of skills. Key activities included: (i) mobilization and capacity building of community-based institutions like Women Self-Help groups (WSHGs) and PGs; (ii) value chain analysis of selected commodities; (iii) support to PGs for productive assets; (iv) investment support for productive infrastructure; (v) improving rural market infrastructure; (vi) exposure visits and buyer-seller meetings; and (vii) organization of innovation forums. 14. Component E: Project Management (appraisal cost US$22.9 million; actual cost US$24.7 million). Under this component, project financing was made available to ensure smooth implementation of all project activities, monitoring of project progress, results and outcomes, learning from project experience, and dissemination of learning to the wider development community. B. SIGNIFICANT CHANGES DURING IMPLEMENTATION Revised PDOs and Outcome Targets 15. No changes were made to the PDO and outcome targets. Revised PDO Indicators 16. There were no revisions to the substance of the PDO indicators. However, as presented in Table 1, at mid-term review (MTR) in April 2013, the indicator “increase in productivity of wheat and rice” was revised to “increase in productivity of wheat and paddy”, replacing “rice” with “paddy” for a more apt representation of the rice crop yield. In addition, the indicator “percentage increase in land value” was more narrowly defined to refer to “Class C land”. The corresponding unit of measure for this indicator was also changed from “percent” to “Rupees per ha”. Table 1. Appraisal and revised PDO indicators and targets Original PDO Indicators Revised PDO Indicators Original Target Revised Target Increase in productivity of wheat and Increase in productivity of wheat 6.4 tons/year - rice and paddy Increase in cropping intensity - Baseline+200% - Percentage increase in crop income per - Baseline+71% - unit land7 Increase in Class C land value Percentage increase in land value8 Baseline+200% 650,0009 (Rs./ha) Revised Components 17. There were no revisions to project components. Other Changes 18. Changes to sodic land selection and eligibility criteria: At MTR, sodic land selection criteria were relaxed to allow consideration of patches that were smaller (at least 30 ha) than the initial 40 ha size. The selection criteria were further relaxed in September 2015 to reduce the share of “Class C land” from 65 7 Including all crops produced on a given land. 8 Land value was based on farmers’ perception. 9 This target figure is still “baseline + 200%” but in 2009 prices. Page 9 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) percent to 50 percent and increase the share of “Class B land” from 25 percent to 40 percent. At this time, the earlier eligibility criterion, precluding reclamation in Phase 1 and Phase 2 villages, was also dropped. In 2017, the target for sodic land reclamation through press mud treatment (compost obtained from sugarcane waste) was reduced from 6,000 ha to 1,226 ha. 19. Revisions to irrigation and main drain rehabilitation targets: When the eligibility criteria was relaxed to include sodic lands in Phase 1 and Phase 2 villages, actual requirements for irrigation development were reduced from 98,000 ha to 68,496 ha and those for rehabilitation of main drains reduced from 5,700 km to 3,256 km10. 20. Scaling up ravine land reclamation: Based on success of the 5,000-ha ravine land reclamation pilot and availability of foreign exchange rate savings, an additional 18,943 ha of ravine land were reclaimed bringing the total area of ravine land reclaimed to 23,943 ha. 21. Discontinuation of support to seed related activities: At MTR, seed production and dissemination activities were discontinued. 22. Closing date extension: The project closing date was changed twice - for a cumulative extension of 36 months. The first extension - for 24 months - was approved on August 28, 2014, moving the closing date from December 31, 2015 to December 29, 2017. In the second extension, approved on August 3, 2017, the project closing date was extended by an additional 12 months, from December 29, 2017 to December 29, 2018. Rationale for Changes and Their Implication on the Original Theory of Change 23. The changes to sodic land selection and eligibility criteria at MTR were occasioned by a combination of inadequate implementation capacity at UPBSN, limited availability and supply of gypsum to meet the technically recommended application rates for reclamation, and the diminishing availability of a critical mass of sodic land patches that would meet the project’s technical selection and eligibility criteria. Further relaxation of the sodic land selection and eligibility criteria in September 2015, was done in recognition of the restrictiveness of the criteria and the impact this had on the available area of sodic land sufficient to meet the project’s reclamation targets. Reductions to the target area of sodic land to be reclaimed through treatment with press mud were due to constrained availability and high cost of the press mud as well as limited acceptance of the technology by farmers due to environmental and safety concerns (refer to para. 48). The downward revision of the area under assured irrigation and the length of main drains that were to be rehabilitated was occasioned by the reduced requirements as a result of moving into Phase 1 and Phase 2 villages where some of these investments had earlier been made. Seed production and dissemination activities were discontinued as there was a new parallel seed program through which beneficiaries’ seed needs were being adequately met. The first closing date extension was required to address quality issues that had arisen in the OFD elements of the sodic land reclamation process11. The second extension of the closing date was justified on the basis of the need to scale up ravine reclamation, provide technical support to newly established FPOs, and consolidate achievements of the three UPSLRP phases by revisiting and improving rehabilitated drainage systems and re-treating earlier reclaimed patches where, for a number of reasons, sodicity had recurred. 10 Because Phase 1 and Phase 2 had invested in irrigation development and rehabilitation of main drains that would serve some of the areas reclaimed in Phase 3. 11 With pressure to meet the annual targets, some field staff were allowing slippages in the implementation timeliness and thoroughness of the OFD process, which posed a threat to sustainability of reclaimed lands. Page 10 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) 24. In sum, the above changes were mainly undertaken in the spirit of meeting project targets, promoting efficient use of project resources, satisfactorily completing project interventions, ensuring sustainability of project investments, and achieving the PDO. These changes did not affect the original ToC. II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating 25. The project objective remains valid today. The objective is consistent with the Bank’s 2018 -2022 Country Partnership Framework (CPF), which among others, posits that to sustain its growth trajectory and spread more equitably the gains of economic growth, India needs to pivot to a more resource- efficient, inclusive, and diversified growth in the rural sector - including by increasing productivity while improving efficiency in the use of land resources in agriculture. In addition, the objective is still consistent with the Government of India (GoI) priority to double farmers’ incomes by 2022, partly through enhancing agricultural productivity. The relevance of the PDO is therefore rated High. B. ACHIEVEMENT OF PDOs (EFFICACY) Assessment of Achievement of Each Objective/Outcome 26. Increase agricultural productivity on degraded lands(PDO)12: As indicated earlier, achievement of the project objective was to be assessed by four PDO indicators (see Table 2 below). A combination of reclamation and improved provision of agriculture support services increased: (i) annual cereal (wheat plus paddy) productivity from 4.09 t/ha to 6.02 t/ha (94 percent achievement of the target of 6.4 t/ha); cropping intensity from 45 percent to 211 percent (surpassing the target of 200 percent); (iii) crop income per unit area of land by 76 percent against the target of 71 percent; and (iv) the market value of previously “C class land” from Rs. 200,000 to Rs. 648,279 per ha – an almost 100 percent achievement of the target. In sum, targets for two of the four indicators were fully achieved, while achievement for the other two indicators (“increase in productivity of wheat and paddy” and “increase in C class land value”) was substantial - standing at 94 percent and 99.7 percent of the project targets, respectively. The project therefore significantly achieved its objective as measured by the indicators. Table 2. Achievement of PDO outcome indicators PDO indicators End of Project Target Achievement Increase in productivity of wheat and paddy(ton/year) 6.40 6.02 Increase in cropping intensity (percent) 200 211 Percentage increase in crop income per unit land (percent) 71 76 Increase in C class land value (Rs./ha)13 650,000 648,279 12 The level of achievement of PDO indicators slightly diverges from that in the borrower completion report because of the different approaches to the assessment. Annex 6 provides the methodology used for the ICRR. 13 Relevant PDO outcome indicators are assessed in or adjusted to 2009 prices. Page 11 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) 27. Improved soil quality: A total of 142,677 ha (above the target of 130,000 ha) of sodic land, spread across 32 districts, was reclaimed by the project. The project also reclaimed 23,943 ha of ravine land (against the target of 18,300 ha) in 158 watersheds14. Assured irrigation was provided to 68,612 ha of reclaimed land against the target of 98,000 ha, reflecting lower than projected needs for irrigation development (refer to para. 18). About 3,235 km of main drains were rehabilitated, against the target of 5,700 km, also reflecting lower than projected actual needs (refer to para. 18). Additionally, the project facilitated the improved maintenance of rehabilitated main drains by strengthening capacities of responsible bodies15 and influencing the GoUP to allocate resources for maintenance through the state budget16. A total of 34,883 WUGs were formed and trained in maintenance of field drains. Baseline and post-intervention soil sample tests confirmed improvements in overall soil quality of reclaimed sodic lands, with restoration of pH and Electrical Conductivity (EC) levels to within ranges appropriate for crop production17. Main drains rehabilitation and maintenance reduced waterlogging in the project areas by 95 percent, further improving soil quality for crop production purposes. A sustainability assessment18 conducted at project closing, shows that 88 percent of reclaimed sodic lands were still being used for agricultural purposes. While this proportion of utilized reclaimed land compares favorably well with levels observed at completion in other land reclamation projects funded by the Bank (e.g. 78 percent in the China – Heilongjiang Land Reclamation Project; 82 percent in the Rwanda Land Husbandry and Water Harvesting and Hillside Irrigation Project; and the 85 percent in the China Loess Plateau II Project), it is still low compared to the 93 percent level observed for Phase 2 of the UPSLRP. Given the low average landholding in the state and as observed in Bank supervision reports, most of the unutilized reclaimed areas seem to belong to absentee landlords. Building on the success of pilot of 5,000 ha, the project successfully reclaimed 23,954 ha of ravine lands in 158 watersheds. 28. Increased adoption of improved agricultural technology: The project supported dissemination and transfer of technology mainly under the rubric of improved farming practices and crop diversification. In total, 19,996 demonstrations (148 percent of the target) on Integrated Plant Nutrient Management (IPNM), 7,537 on Nadep compost (80 percent of the target), 7,625 on vermi-compost (90 percent of the target), 1,656 on Integrated Pest Management (IPM) - 92 percent of the target, 13,888 on crop diversification (163 percent of the target), and 137 on Sodic Kisan Mela (137 percent of the target) demonstrations were funded by the project. As a result of these technology transfer activities for example, IPM practices were adopted on a total of 5,284 ha (against the target of 4,000 ha). In reclaimed sodic lands, previously used for cereal production only, shares of oil seeds and pulses (a proxy of diversification) reached 8.8 percent (against the target of 7.5 percent) while the shares of vegetables reached 2.8 percent (against the target of 2.5 percent). Average milk productivity increased to 8.55 liters per animal per year for buffalos (against an end of project target of 7 liters per animal per year) and 6.33 liters per animal per year for cows (against an end of project target of 5 liters per animal per year). 29. Improved access to markets: With project support, 120 FPOs (against the target of 100) with total membership of 112,590 farmers were formed and strengthened. At closing, 82 FPOs were qualified to 14 Based on the success of the ravine reclamation pilot, GoUP had requested for additional financing at MTR to scale up the pilot. The WB turned down the request it also included development if groundwater irrigation. 15 Maintenance responsivities lay with UPBSN and ID for main drains, SIC for link drains, and WUG for field drains. 16 Over the course of the project, GoUP released Rs. 892.1 million for maintenance of 10,608 km of main drains. 17 Soil pH declined from 10.4 to 8.4; EC from 2.9 dS/m to 0.3 dS/m - against a target of below 4.0 dS/m; and organic carbon content increased from 0 to 0.31 percent compared to a target of 0.30 percent. 18 Conducted by Remote Sensing Application Center (RSAC), the study monitored utilization of reclaimed sodic lands in 27 project villages located in 7 project districts. Page 12 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) receive project resources as planned under the Community Investment Fund (CIF) for working capital and basic infrastructure. Twenty FPOs secured statutory licenses enabling them to operate in regulated wholesale markets for agricultural produce. An additional 69 FPOs registered with e-NAM, a pan-India online trading platform for agricultural commodities, to enable them access markets in different states. Another 57 FPOs were linked to the Small Farmers’ Agribusiness Consortium (SFAC) from where they accessed grant and credit financing for their business needs. Through “buyer-seller” meets, 15 FPOs entered into binding supply contracts with several flour mills and had delivered over 46,000 tons of paddy and wheat before project closing. At closing, cumulative FPO business transactions were estimated at Rs.1.2 billion, 60 percent of which accrued from agricultural produce marketing, 38 percent from agriculture inputs trade, and the rest from provision of agro-services. Agriculture input and output marketing services are reported to have reduced crop production cost by 10 percent, and to have increased price realization by 10-15 percent. Around 37 percent of paddy and 32 percent of wheat producing farmers supported by the project sold their produce through the FPOs. Additionally, the project improved market infrastructure in 102 rural markets - often referred to as rural haats (RH) - against the target of 125. Anecdotal evidence as presented in the Borrower’s Completion Report (BCR) points to substantial increases in price realization (ranging from 9 percent for gram to 109 percent for bottle gourds) as a result of infrastructure improvements in the RHs. 30. Landless families mobilized for livelihood activities: In total, 73,257 women, organized in 6,555 Women Self Help Groups (WSHG), were trained in income generating skills and supported by the project - through the Livelihood Support Fund (LSF) grants - to engage in multiple livelihood activities including, but not limited to beekeeping, goat rearing, jewelry making, tailoring, and packaging. The impact assessment survey found that through these income-generating activities, individual WHSG members were on average, able to generate up to Rs.19,150 in annual profit. Justification of Overall Efficacy Rating 31. Overall efficacy rating is Substantial as the project almost fully achieved its objective as measured by the outcome indicators (see Table 2). C. EFFICIENCY Assessment of Efficiency and Rating 32. Economic efficiency: An economic rate of return (ERR) of 25.1 percent and economic net present value (ENPV) of Rs.9.5 billion were estimated at project appraisal. These returns were assessed for a 25- year period (including five years of implementation), at 2009 prices and a 12 percent discount rate. Estimated benefits were to derive from improved crop productivity, intensification and diversification on reclaimed sodic lands and non-sodic lands benefitting from improved drainage, agriculture support services and market access, and improved livestock productivity resulting from improved extension and animal breeding services. The appraisal returns were overestimated as they were based on an optimistic implementation timeframe and correspondingly overestimated benefit accumulation timelines and incremental returns. The project returns were re-estimated following the methodology adopted at appraisal, using actual and revised future costs and benefits. The returns were assessed for 29 years, including nine years of implementation, at 2009 prices and a 12 percent discount rate. The re-estimated ERR is 19.7 percent and the ENPV is Rs.4.0 billion. The divergence between the appraisal and completion returns was due to: (i) overestimation of benefits at appraisal based on unrealistic assumption of the Page 13 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) project timeframe; (ii) delayed benefit realization by three years; (iii) lower than projected areas of “Class C land”; and (iv) lower than expected utilization of reclaimed lands. 33. Administrative efficiency: Mainstreaming elements of project implementation into GoUP structures (e.g. through ID and at the district level) to reduce administrative overheads (in addition to strengthening ownership and ensuring sustainability) was a positive efficiency attribute. However, staff turnover, especially at the highest levels (Managing Director), gaps in staffing for prolonged periods, and weaknesses in implementation capacity at UPBSN affected the timeliness of implementation. 34. Efficiency rating: The cost efficiency per ha of reclaimed land remained within the projected ranges. Re-estimated economic returns at completion are within the recognized and acceptable range for similar investments elsewhere and are also comparable to those reported for Phase 2 of UPSLRP. To a certain extent, delays in implementation, tempered down overall project efficiency, which in this case, is rated Substantial. D. JUSTIFICATION OF OVERALL OUTCOME RATING 35. The relevance of the project’s objectives is high as the operations’ objectives were substantially achieved. In addition, the project demonstrated some good elements of efficiency, although some of the lustre was stained by delays in implementation and concerns about quality of implementation. In view of the above, the overall outcome rating is assessed as Satisfactory. E. OTHER OUTCOMES AND IMPACTS (IF ANY) Gender 36. Project support to land reclamation was gender blind as it was targeted to areas meeting specific eligibility and selection criteria. Yet, in recognition of the dismal level of land holding among women in the project area, a gender strategy had been developed as part of the Environmental and Social Management Framework (ESMF) and specific support tailored to women was organized in the form of livelihood activities under project Component D. As discussed earlier, 6,555 Women Self Help Groups (WSHG), were trained in income generating skills and supported by the project - through the Livelihood Support Fund (LSF) grants - to engage in multiple livelihood activities including, but not limited to beekeeping, goat rearing, jewelry making, tailoring, and packaging. Institutional Strengthening 37. The principal focus of the project’s institutional strengthening efforts was UPBN staff and line departments. This mainly entailed human capacity building especially in the areas of participatory reclamation, O&M of drainage systems, land and water management, and remote sensing technology. In addition, the project trained and strengthened the capacity of progressive farmers to serve as change agents by facilitating knowledge transfer to their peers. FPOs were also trained on provision of extension and marketing services to member farmers; proactive youth were trained on maintenance of irrigation and drainage systems and crop and livestock farming; and SICs were trained on maintenance of link drains. As part of the project’s exit strategy, these paraprofessionals are expected to be linked to line departments to continue provision of extension support. The status of these proposed links is unknown but if they were to materialize, they should contribute to strengthening the extension system in UP, which has been chronically weak, partly because of manpower deficiencies. Page 14 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Mobilizing Private Sector Financing 38. Contribution in cash by smallholders (beneficiaries) is estimated at around US$22.3 million. Poverty Reduction and Shared Prosperity 39. Project beneficiaries of land reclamation support were small and marginal farmers, owning less than 1 ha of land, and generally among UP’s rural poor. Field missions did not detect cases of elite capture and therefore, it is generally understood that project benefits from land reclamation largely accrued to this population segment. In addition, most of the livelihood support activities funded by the project were of the kind that could be carried out with limited means and were generally supported for rural residents with little resource endowments (landless poor, women, scheduled caste, scheduled tribes, and other backward castes). It is also acknowledged that very vulnerable households (e.g. the elderly, the disabled, etc.) that could not meet or afford the labor requirements imposed as the beneficiaries’ contribution to reclamation activities, were bypassed by the project. Other Unintended Outcomes and Impacts 40. No unintended outcomes and impacts were documented. III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 41. Objectives and activities: The project objective was generally clear, simple and realistic, as were the associated activities in support of land reclamation and improved access to agriculture support services. While project activities in support of improved market access were pertinent and justified given the context, their direct contribution to the objective as was defined is very debatable, and at best, tenuous19. In this case then, the PDO could have been better defined to explicitly capture improved market access as an additional objective. 42. Results framework: While most of the outcome indicators cohered well with the project activities and were measurable, the specific outcome indicator on “percent increase in land value” was more complex to measure and accurately apportion project contribution to overall changes in its value. This indicator could easily have been dropped without compromising ability to track attainment of the PDO. The results framework could also have included better measures of improved market access e.g. improved price realization by farmers, or volumes of market transactions in RH or through FPO marketing services. 43. Design: Project design built on the positive experiences from Phase 1 and Phase 2 and sought to pursue innovative approaches to land treatment and management, as well as new interventions such as crop diversification and improved marketing to promote sustainability of project outcomes. The project’s components were mutually reinforcing: land reclamation and agriculture services to unlock agricultural production on degraded lands, and marketing support to deal with second generation challenges of produce marketing that accrue from increased production. The government’s proposal to reclaim 60,000 ha of ravine lands was suitably addressed by testing out ravine reclamation on a smaller 5,000-ha pilot, which could then be scaled up (and was indeed scaled up), after being proven successful. To the extent that timely implementation greatly hinged on ready availability of requisite amounts of gypsum, yet this 19 Mostly promoting sustainability of the PDO Page 15 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) was based on unrealistic assumptions, the project timeline was ambitious, and this appears to have been a design shortcoming (see section on Key Factors for Implementation). 44. Risk mitigation: At appraisal, risks to attainment of the PDO were identified as: (i) placement of greater emphasis on physical aspects of land reclamation than on enhancing agricultural productivity, linking producers to markets and strengthening agricultural livelihoods; (ii) weak coordination between line departments for effective implementation; (iii) elite capture and political interference; (iv) poor quality of OFD, which would affect success of land reclamation; (v) frequent changes and high turnover of staff in implementing agencies; (vi) weaknesses in aggregated institutions-WUGs and WSHGs. Of all the above risks, frequent changes, high turnover, and shortage in staff posed the most daunting challenge during implementation. Throughout implementation, UPBSN was unable to attract and retain the right skills and number of staff due to low wages and allowances, and delayed staff appointment approvals. This affected both the timeliness and quality of implementation. The risk of poor quality of OFD and the anticipated impact on success of land reclamation, materialized in various stages of implementation, but this was identified through field missions and mitigated partly by allowing for more implementation time. 45. Readiness for implementation: By approval, key staff at UPBSN were onboard and District Project Units (DPUs), responsible for project implementation at the district level had been adequately staffed. A Project Implementation Plan (PIP), procurement plan, procurement manual, financial management manual, and safeguards documents had been finalized. In addition, bids for works and goods for the first cohort of land reclamation activities (planned in five out of the 32 districts) had been issued and recruitment for fiduciary and safeguards specialists had been initiated. By approval, the area of sodic land available for reclamation as per design eligibility and selection criteria remained unknown and the implementation arrangements for the component D on market access were expected to be detailed out during implementation. This was responsible for some of the delays in implementation. B. KEY FACTORS DURING IMPLEMENTATION 46. Strong GoUP interest in the project: The GoUP ensured that key implementation arrangements and mechanisms were in place at all levels (state, district, sub-district and village) as per the project design. With a few exceptions, the GoUP also ensured timely transfer of funds to UPBSN for implementation. 47. Implementation capacity: UPBSN seemed unable to attract and retain qualified staff both in adequate numbers and skills due to low renumeration and field allowances, delayed approvals of staff appointments, and fee adjustments by GoUP. The shortage in Implementation Assistants (IAs), who played a critical role at the field level, was particularly marked - with around 30 percent gap between project year 1 and 5. An additional 54 established staff positions at UPBSN remained vacant throughout implementation. Furthermore, approvals for revisions to emoluments and field allowances by the GoUP were delayed until project year 5 and 7. Staff turnover was also substantial which undermined timely provision of training for technical staff, especially for IAs. Capacity weaknesses in M&E, were also reportedly responsible for the consistent disconnect - as voiced in supervision reports - between progress data and the actual situation on the ground. These capacity weaknesses negatively affected the quality and timeliness of implementation. 48. Availability of cost-effective soil amendments for land reclamation: At appraisal, annual targets for reclamation had been set at 20,000 ha in the second year, 35,000 ha in the third year, 40,000 ha in the fourth year, and 35,000 ha in the fifth year. Out these, 2,000 ha each, in the second, third and fourth year were to be reclaimed through press mud treatment. However, annual supply of gypsum was sufficient for Page 16 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) only 20,000 ha and recourse to imports from other countries was economically infeasible. Similarly, timely procurement of requisite amounts of press mud at financially feasible cost, turned out to be a challenge as it was no longer available for free, its bulky nature rendered transportation costly, while its drying, decomposition and application was labor intensive. Additionally, farmers were reluctant to accept this treatment option due to environmental (e.g., foul odor, attraction of insects) and safety (e.g., catching fire) concerns. Constrained availability of these soil amendments mediated a scaling down of annual reclamation targets and was part of the justification for closing date extension. 49. Sodic patch selection criteria: During implementation, it became increasingly difficult to locate sodic patches conforming to the eligibility and selection criteria proposed at appraisal as such patches were becoming more dispersed and unavailable in critical amounts within a given area. To address this issue, the eligibility and selection criteria were relaxed to allow taking up for support, patches in Phase 1 and Phase 2 villages, and those with a lower share of “Class C lands” (from 65 percent to 50 percent ) and higher proportion of “Class B land” (from 25 percent to 40 percent). Even with relaxation of the eligibility and selection criteria, towards the end of the project, potential patches for reclamation were more scattered, rendering implementation costlier and more difficult. There were cases when the time pressure to meet the physical targets led to misclassification of sodic lands, and also, since continued operations of individual DPUs depended on the existence of sufficient amounts of patches eligible for support in a particular district, and the renumeration for IAs had been tied to the proportion of “C class lands, there were cases where project teams tended to overestimate “C class land” area. This affected the quality of implementation and the efficient use of project resources. The decision to take up patches in Phase 1 and Phase 2 villages for reclamation reduced the overall project requirements for irrigation development and main drain rehabilitation as some of the would-be requirements had been partially met in Phase 1 and Phase 2. Resources released from the drainage and irrigation interventions were used to scale up ravine reclamation. 50. FPO formation and strengthening: Formation and strengthening of FPOs for improving market access was uncharted territory for the UPBSN and therefore, as indicated above, detailed implementation arrangements for this activity were expected to be finalized during implementation. Formation of an adequately staffed Marketing Cell, within UPBSN was delayed20 and this in turn affected timely recruitment of a service provider to help with formation and strengthening of FPOs and preparation of guidelines for provision of project support through the CIF to qualifying FPOs. As a result, it was not until the fourth year of implementation that the first FPOs (about 20) were formed, and the eighth year, that initial disbursements from the CIF to FPOs were made. Inadequate experience at UPBSN in implementation of this activity also delayed disbursement of the CIF. Indeed, the first extension of the project closing date was requested partly to allow satisfactory completion of FPO strengthening activities to ensure their sustainability. 51. Bank supervision: Timely and systematic technical guidance by the Bank in areas of sodic land reclamation, drainage, and agriculture services considerably contributed to satisfactory implementation. Observations based on extensive site visits, consistent emphasis on quality and sustainability aspects, identification of opportunities for convergence with other government programs by Bank staff are particularly notable. However, implementation support and backstopping in other areas such as improving market access and M&E could have been more intentional, regular and thorough. 20The reasons behind the delay are unknown. However, since these were posts expected to be filled by civil servants through deputation, it is probable that qualified staff were unavailable. Page 17 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 52. Overall, the project had good M&E design elements. The design provided for: (i) internal monitoring by implementing line departments and UPBSN; (ii) participatory monitoring by beneficiaries and beneficiary organizations; and (iii) external agency for building M&E capacity of UPBSN, progress monitoring to enhance the effectiveness of project implementation, and project M&E. A baseline for key parameters was available and specific assessments were planned to measure progress and impact at midpoint and end-of- project. The ToC was generally clear, although as discussed under the section on Key Factors During Preparation, the direct contribution of project activities in support of improved market access to the PDO was tenuous, and the outcome indicator on “percent increase in land value” was not easy to measure - as this required monitoring of external market factors - and its linkage to the PDO was hard to establish. In addition, indicators to measure success with respect to improving market access could have been considered. M&E Implementation 53. As indicated above, a project baseline was available. All indicators were systematically monitored and reported, and midterm and end-of-project surveys were conducted in a timely manner. This notwithstanding, the lack of appropriate controls made it impossible to compare project outcomes with the counterfactual to capture general trends that are not attributable to the project, and the surveys designed to assess achievement of the outcome indicators had weaknesses (e.g., the project’s attribution and impacts of external factors were not accurately captured). As explained in Annex 6, these weaknesses are responsible for the differences in the BCR and ICRR estimates of the level of achievement of PDO indicators. Similarly, the project achievements in improving market access were inadequately monitored (e.g., incremental price realization and marketing transactions were not captured). Moreover, the quality and reliability of some of the progress data was a key concern during implementation and thus undermined the integrity of the monitoring system. M&E Utilization 54. Project progress data was the basis for a number of decisions and midstream course-corrections. This included: relaxing eligibility and selection criteria for sodic patches to be taken up for reclamation and reducing annual targets for land reclamation in light of reduced supplies of necessary soil amendments. There were also cases where, for unknown reasons, corrective measures were never taken even when field findings by the Bank supervision provided justification for undertaking such remedial measures. This, for instance, included the overestimation of the proportion of “C class land” in some district, which does not seem to have triggered sufficient management action. Similarly, sustainability assessment findings, which suggested that not all reclaimed sodic lands were being utilized, did not trigger corrective actions. Justification of Overall Rating of Quality of M&E 55. Overall rating for M&E quality is Modest. The project had a clear objective, corresponding ToC, and mostly measurable outcome indicators. Appropriate baselines existed and performance indicators, as Page 18 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) presented in the results framework were reported in ISRs. Progress data also influenced decision making in some aspects of project implementation. As reported above, there were shortcomings related to usefulness of some of the indicators in assessing the PDO, lack of better indicators to assess progress with respect to improving market access, candidness of reporting, and proper assessment of outcome and results indicators. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 56. Environmental safeguards: The project was classified as category B in accordance with the WB’s environmental safeguard policy classification, and the following policies were triggered: Environmental Assessment (OP 4.01); Natural Habitats (OP 4.04); and Pest Management (OP 4.09). Against this backdrop, the government had prepared an ESMF providing appropriate mitigation actions. Under the ESMF, a specific Wetland Management Plan was implemented to prevent draining of natural wetlands. An Environmental Manager was in place throughout the project and regularly monitored and reported on safeguards compliance. Environmental safeguards compliance was satisfactory overall, and no significant adverse impacts were observed. 57. Social safeguards: The project triggered OP/BP 4.12 on Involuntary Resettlement and an appropriate resettlement and rehabilitation framework had been prepared by appraisal. In addition, a strategy for gender and vulnerable group development was available. Although a Social Development Specialist was not recruited for most of implementation, there were no reports of adverse social impacts as a result of project activities. RH infrastructure was built on land owned by Gram Pachayats and was free from encumbrances and field drains were constructed in beneficiaries’ own fields and didn’t involve land acquisition. 58. Financial management: The fund flow arrangements functioned well. Budget allocation and release of funds by the GoUP were timely, except in FY 2012/13, 2013/2014 and 2018/19 with attendant negative effects on implementation. For most of implementation, the staffing arrangements were adequate. At UPBSN, Finance Controller, a deputed official from the State Financial Services, provided regular oversight. The accounting books were maintained in an ‘off the shelf’ accounting system at UPBSN (Headquarters and DPUs), and in manualized form in other accounting locations. Submission of Interim Unaudited Financial Reports (IUFRs) was timely. Effective internal and external audit mechanisms were in place leading to timely preparation and submission of project audit reports. There were some audit observations related to: (i) delays in settlement of advances; (ii) liabilities appearing in the books pending settlement; and (ii) poor accounting, bookkeeping and inventory management at DPUs, WUG, WSHG), and GP. The Sodic Project Audit Committee meetings reviewed the audit observations to ensure appropriate resolution of issues. NGOs were tasked to support the user groups. The project accounting staff were regularly trained leading to improved accounting and internal control. The statutory audit of UPBSN Limited (Entity Audit) that was supposed to be undertaken for each financial year as agreed to was significantly delayed (due to delay in the appointment of entity auditor) and had not been completed by MTR. Currently, UPBSN is working on closing outstanding issues and submission of final disbursement claims by the grace period deadline on June 29, 2019.21 UPBSN has agreed to submit the last year (FY 2018-19) project audit report to the Bank by December 31, 2019. 59. Procurement: The project procured a mix of small works, goods, non-consultancy and consultancy services. Some challenges were faced in procurement of gypsum due to limited number of suppliers and 21 Extended from April 29, 2019 to June 29, 2019 because of election related delays. Page 19 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) high transportation costs. This resulted in limited number of bidders. Procurement of fertilizers and seeds was also constrained as these are regulated items in India. e-Procurement was introduced towards the project closure for the procurement of civil works by ID. UPBSN took appropriate follow-up actions on the findings of procurement post review. To minimize the procedural violations by other implementing agencies, UPBSN procurement team conducted review of the procurement handled by these agencies. Extensions of project closing dates required multiple extension of many contracts. The project did well on the contract management front. From a fiduciary perspective, there were no incidents of malfeasance reported or discovered in the course of the project’s implementation. C. BANK PERFORMANCE Quality and Entry 60. The project design was well aligned with the development priorities of the GoUP, GoI and the Bank and most of the project activities drew lessons from the earlier phases of UPSLRP as well as from other Bank-financed projects. In addition, most of implementation risks were adequately identified and appropriate mitigation measures proposed. However, GoUP’s own processes for project staff recruitment and approvals of improvements and changes to staff emoluments and fee adjustments should have been studied to assess their implication for the project, which was dogged by implementation capacity issues. M&E systems should have received greater attention during project formulation and more assessments could have been done to gauge the extent to which the sodic patch selection criteria was realistic, for example by assessing remaining sodic lands following the earlier phases’ reclamation. Availability of requisite supply of gypsum and press mud could also have been studied as this was key to timely implementation. Quality at entry, therefore, is rated Moderately Satisfactory. Quality of Supervision 61. Several positive attributes to implementation support performance include: (i) regular and timely supervision missions with adequate composition of expertise22; (ii) candid reporting and adequate recommendations by missions through most of the project’s implementation; (iii) effecting appropriate changes at MTR; and (iv) responding positively to government’s requests for extension of the project closing date - to address quality issues, satisfactory completion of activities, and scale-up implementation - as well as the end of disbursement closing date. The Bank could have done more in providing support to market access activities and to M&E as these were identified early on as areas where UPBSN and line departments had limited experience and capacity. It seems though, that Bank support in these areas is also recognized as weak across most of India’s agricultural portfolio and might merit a broader response. The quality of supervision is rated Moderately Satisfactory. Justification of Overall Rating of Bank Performance 62. The Bank’ strong performance on the project design front as well with respect to regular missions during most of implementation is modulated by shortcomings in ensuring adequate M&E, providing adequate support to market access as well as livelihood activities, and ensuring adequate supervision between 2015-2017. Overall Bank performance is therefore rated Moderately Satisfactory. D. RISK TO DEVELOPMENT OUTCOME 22Except for some instances between 2015-2017 when field missions were irregular and some missions too short (e.g. 2 days only). Page 20 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) 63. The project sought to increase agricultural productivity on degraded lands. The sustainability of these productivity gains hinged on ensuring: (i) adequate drainage of the reclaimed sodic land - since waterlogging was a key driver of degradation; (ii) continued cultivation of the degraded land following a well-defined cropping cycle; (iii) continued use of good land and water management practices by farmers; and (iv) continued effective functioning of producer collectives. The ID is responsible for maintenance of main drains, and adequate resources are available for this purpose through the state budget. Additionally, the maintenance of the field drains is vested with the WUG and these obligations are generally within the WUG’s technical competence and financial reach and should be easy to maintain. The WUGs will need to be regularly supervised and guided by the ID. Increasing agricultural land scarcity in UP should generally guarantee the continued cultivation of most of the reclaimed land and support through other government programs e.g. the National Mission for Sustainable Agriculture23 which provides soil and fertilizer testing services and test-based tailored extension services among others should meet some of the beneficiaries’ needs for knowledge on land and water management. However, other state agencies e.g. Krishi Vigyan Kendras (KVK), and the Department of Agriculture will have to fulfill their responsibilities too, especially the provision of advisory services to farmers on the sustainable use of the reclaimed lands. Finally, all the project FPOs have been linked to the Small Farmers Agribusiness Consortium (SFAC) for continued support and strengthening for the purposes of establishing forward and backward linkages in their preferred value chains. Continued technical support to these FPOs is required, especially to those 38 FPOs that did not qualify for the CIF support. Taken together, these arrangements should ensure the sustainability of project outcomes. V. LESSONS AND RECOMMENDATIONS 64. Criteria meant to guide identification and selection of project targets should be based on a proper understanding of the universe from which the selection is to be made: The largest share of project investments was meant to reclaim sodic land patches identified and selected based on specific technical selection criteria. During implementation, it became evident that patches in the target districts, meeting the criteria were limited. An assessment of the available patches during preparation would have led to more pragmatic identification and selection criteria and enhanced the timeliness and quality of implementation. 65. Formation of fully functional FPOs is a long and arduous process that is hard to accomplish within typical project timeframes. Even with the hiring of specialist service providers, to help UPBSN with the formation and strengthening of FPOs, it was only in year 5 of implementation that the initial set of FPOs were deemed ready to undertake their activities and by project closing, many FPOs, (around 38 FPOs) were still considered weak and in need of strengthening if they were to serve their roles. In this case, while there are merits to FPOs, projects seeking to improve market access through such farmer collectives should consider working with/strengthening already existing FPOs or anticipate the impacts of the typically long FPO gestation period in the design. 66. Project designs with clear and agreed to strategies for maintenance of project supported infrastructure guarantee sustainability of outcomes: Infrastructure development in many countries has 23This national program was established in 2014. Under the Soil Health Management (SHM) component, the program promotes adoption of Integrated Nutrient Management, provides soil and fertilizer testing services and tailored recommendations, provides extension and demonstration services to farmers, and controls quality of fertilizers. Page 21 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) often followed the build-neglect-rebuild paradigm mainly because of poor asset maintenance. Project design anticipated challenges with OM and made provisions for maintenance of the rehabilitated drains, with clear responsibilities for both government (ID) and beneficiaries. These maintenance arrangements were triggered upon completion of rehabilitation of any drains and by project closing they were still functional and should help sustain project outcomes. . Page 22 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: increase agricultural productivity on degraded lands in selected areas of Uttar Pradesh Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Increase in productivity of Metric ton 4.09 6.40 6.02 wheat and rice (tons/year) 18-Sep-2009 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Increase in crop intensity Percentage 45.00 200.00 211.00 18-Sep-2009 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Page 23 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Percentage increase in crop Percentage 0.00 71.00 76.00 incomes per year 18-Sep-2009 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Increase in "C" class land Number 204000.00 650000.00 648279.00 value (Rs./ha) 09-Feb-2010 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): A.2 Intermediate Results Indicators Component: On-Farm Development and Land Treatment Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Area reclaimed Hectare(Ha) 0.00 130000.00 142677.00 18-Sep-2009 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): Page 24 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Component: Improvement of Drainage Systems Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Area provided with irrigation Hectare(Ha) 0.00 98000.00 68612.00 68612.00 and drainage services (ha) 11-Dec-2016 30-Nov-2016 12-Aug-2013 29-Dec-2018 Area provided with Hectare(Ha) 0.00 98000.00 69496.00 69496.00 irrigation and drainage services - Improved (ha) 18-Sep-2009 30-Nov-2016 12-Aug-2013 29-Dec-2018 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion length of main drains Kilometers 0.00 5700.00 3235.00 3256.00 maintained 18-Sep-2009 30-Nov-2016 12-Aug-2013 09-May-2018 Comments (achievements against targets): Component: Agriculture Support Services Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Area under Integrated Pest Hectare(Ha) 0.00 4000.00 5284.00 Page 25 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Management (IPM) (ha) 18-Dec-2009 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion % increase in milk Hectare(Ha) 3.41 6.00 7.00 productivity (litres/lactating cow/year) 18-Sep-2009 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): The milk yields are an average of cow and buffalo milk production Component: Institutional Strengthening and Capacity Building for Market Access Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of producer groups Number 0.00 100.00 120.00 formed 18-Sep-2009 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Page 26 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) No. of SHGs formed Number 0.00 5500.00 6555.00 18-Sep-2009 30-Nov-2016 29-Dec-2018 Comments (achievements against targets): B. KEY OUTPUTS BY COMPONENT Objective/Outcome 1: Increase agricultural productivity of degraded land 1. Increase in productivity of wheat and paddy 2. Increase in cropping intensity Outcome Indicators 3. Percentage increase in crop income per unit of land 4. Percentage increase in land value 1. Area reclaimed (ha) 2. Area provided with irrigation and drainage services (ha) 3. Soil quality improved (pH, EC, and organic carbon) Intermediate Results Indicators 4. Length of main drains maintained (km) 5. Area under Integrated Pest Management (IPM) (ha) 6. Percentage of gross cropped area (GCA) under oilseeds and pulses (%) 7. Percentage of gross cropped area under vegetables (%) Component A. 1. 142,677 ha of sodic lands reclaimed Key Outputs by Component 2. 69,496 ha are under assured irrigation Component B. 3. 3,256 km of main drains rehabilitated Page 27 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) 4. 29,436 km of main drains are maintained Component C. 5. 5,284 ha of arable lands are under IPM 6. 8.8 percent of sodic GCA are under oilseeds and pulses 7. 2.8 percent of sodic GCA are under vegetables Page 28 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) C. STATUS OF ACHIEVING OBJECTIVES/OUTCOMES No. Indicators Baseline 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 Project Achievement Data Frequency of Data end % status source data collection PDO Indicators collection method 1 Indicator 1. Increase in productivity of wheat and paddy (tons/year) Weighted average for utilized Target 5.49 5.78 6.20 6.40 6.40 6.40 82.5% External Pre & post- Beneficiary largely achieved reclaimed lands Actual 4.14 2.41 2.46 5.92 5.92 5.92 6.15 6.13 5.28 M&E agency reclamation interviews Indicator 2. Increase in cropping intensity (%) 2 Weighted average for utilized Target 45 200 200 200 200 200 92.6% External Pre & post- Beneficiary largely achieved reclaimed lands Actual 26 26 200 200 200 205 209 210 212 185 M&E agency reclamation interviews 3 Indicator 3. Increase in crop income per unit land (Rs/ha) Weighted average for utilized Target BL BL BL+43% BL+52% BL+65% BL+71% BL+71% 102.6% External Pre & post- Beneficiary largely achieved reclaimed lands Actual 7,113 0 12,486 M&E agency reclamation interviews 4 Indicator 4. Increase in C class land value (Rs/ha) Target BL 650,000 99.7% External Pre & post- Beneficiary achieved Actual 204,000 648,279 M&E agency reclamation interviews Intermediate Results Indicators 5 Soil quality Target BL-1 BL-1.3 BL-1.6 BL-2 8.40 109.1% Bi- annual Field visits UPBSN/line a. pH level exceeded Actual 10.4 9.5 9.2 9 7.7 7.70 and reports departments Target ds/m <4 <4 <4 <4 <4 100.0% achieved b. EC level Actual 2.9 1.9 1.1 0.96 0.3 0.30 Target % 0 0.2 0.25 0.28 0.3 0.30 103.3% c. Organic carbon exceeded Actual 0 0.15 0.16 0.21 0.31 0.31 Area reclaimed (ha) 6 Target 20,000 35,000 40,000 35,000 130,000 109.8% Bi- annual Field visits UPBSN/line exceeded Actual 21,160 26,043 20,765 25,023 15,074 12,758 10,121 11,734 142,677 and reports departments Area reclaimed through press-mud treatment (ha) 7 Target 2,000 2,000 2,000 6,000 20.8% Bi- annual Field visits UPBSN/line did not achieve. Actual 85 599 565 1,248 and reports departments Area provided with irrigation and drainage services (ha) 8 Target 15,500 26,250 30,000 26,250 98,000 70.9% achieved. The actual Bi- annual Field visits UPBSN/line needs were met and reports departments Actual 12,776 14,008 10,408 11,180 6,672 5,524 4,244 4,684 69,496 Page 29 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) No. Indicators Baseline 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 Project Achievement Data Frequency of Data end % status source data collection Area under IPM (ha) collection method 10 Target 800 800 1,200 1,200 4,000 165.6% exceeded Actual 624 1,600 104 1,220 1,736 640 432 268 6,624 11 Increase in milk productivity (lt/day/lactating animal a. Buffalo Target 2.5 2.5 6.0 8.0 92.5% External Pre & post- Beneficiary largely achieved Actual 6.2 6.2 6.7 6.9 7.2 7.3 7.3 7.4 7.4 7.4 M&E agency reclamation interviews Cow Target 2.5 2.5 6.0 8.0 60.0% surveys did not achieve Actual 3.7 3.7 4.1 4.3 4.4 4.5 4.6 4.7 4.9 4.8 12 Percentage of GCA under oilseeds and pulses (%) a. Sodic land Target 0.0 0.0 7.5 7.5 121.3% External Pre & post- Beneficiary exceeded Actual 0.0 0.0 0.0 0.0 0.0 6.3 8.7 9.1 9.1 M&E agency reclamation interviews b. Non sodic land Target 0.0 0.0 4.0 8.0 11.0 11.0 239.1% surveys exceeded Actual 10.4 10.4 14.7 16.8 19.0 21.4 22.4 25.1 26.3 13 Percentage of GCA under vegetables (%) a. Sodic land Target 0.0 0.0 2.0 2.5 116.0% External Pre & post- Beneficiary exceeded Actual 0.0 0.0 0.0 0.0 0.0 1.9 2.7 3.0 2.9 M&E agency reclamation interviews b. Non sodic land Target 0.0 0.0 2.0 3.0 4.0 4.0 155.0% surveys exceeded Actual 3.1 3.1 4.2 5.0 5.9 6.7 6.9 6.9 6.2 Number of producer groups formed (nos) 14 Target - - 30 40 30 100 120.0% exceeded Actual - - 19 59 18 1 1 22 120 Number of SHGs formed (nos) 15 Target - - 700 2,500 2,300 5,500 119.2% exceeded C. Actual - - 919 914 898 1,084 667 957 591 525 6,555 Page 30 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Animesh Srivastava Task Team Leader Paul Singh Sidhu Agricultural Specialist Shyam Sundar Ranjitkar Irrigation Engineer Debabrata Chakraborti Procurement Specialist Anupam Joshi Environmental Specialist Tanuj Mathur Financial Management Specialist Elliot Wamboka Mghenyi Economist Reena Gupta Natural Resource Management Specialist Biswajit Sen Rural Development Specialist Mohammed Hasan Social Development Specialist Melissa Williams Operations Officer Manivannan Pathy Agricultural Specialist Jacqueline Julian Program Assistant Supervision/ICR Abel Lufafa Task Team Leader Sangeeta Patel Procurement Specialist Arvind Prasad Mantha Financial Management Specialist Samuel Thangaraj Social Specialist Anupam Joshi Environmental Specialist B. STAFF TIME AND COST Page 31 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY09 93.090 350,760.31 FY10 0 0.00 Total 93.09 350,760.31 Supervision/ICR FY09 0 0.00 FY10 42.159 149,718.23 FY11 44.889 209,317.25 FY12 28.008 144,409.71 FY13 24.581 156,593.96 FY14 14.377 85,977.54 FY15 8.348 51,818.56 FY16 18.724 106,345.50 FY17 18.006 117,398.76 FY18 17.333 120,373.55 FY19 18.393 105,624.71 Total 234.82 1,247,577.77 Page 32 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) ANNEX 3. PROJECT COST BY COMPONENT Table 3.1. Projected and actual project costs* Amount at Approval Actual at Project Percentage of Approval Components (US$M) Closing (US$M) (US$M) 1. On-Farm Development and 170.70 183.60 107.56 Land Treatment 2. Improvement of Drainage 39.2 23.05 58.80 System 3. Agriculture Support Service 9.40 10.22 108.72 4. Institutional Strengthening and Capacity Building for 9.70 4.28 44.12 Market Access 5. Project Management 22.9 25.08 109.52 Contingencies 20.1 0.00 0.00 Total 272.00 246.23 91.75 * Figures in this table will change after the final IUFR is submitted. Page 33 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) ANNEX 4. EFFICIENCY ANALYSIS Appraisal Economic and Financial Analysis 1. The appraisal analysis projected following benefits. Land reclamation benefits were assumed for 130,000 hectares (ha) of previously sodic land inclusive of 84,500 ha of “Class C”, 32,500 ha of “Class B”, and 13,000 ha of “Class B+” lands. Of the total, 98,000 ha were to receive assured groundwater irrigation. In Without Project (WoP) scenario, “Class C” lands would be Table 4.1 Appraisal projections for sodic land barren, “Class B” lands would produce one crop (paddy in reclamation works kharif) with very low productivity, and “Class B+” lands WoP WiP would produce two crops (wheat and paddy) but at low productivity. In With Project (WiP) scenario, all classes of Cropping area (ha) Cereals 58,000 247,000 reclaimed lands would produce two crops (wheat and paddy), e.g., increasing the baseline cropping intensity at 45 Oilseeds and pulses - 9,750 percent to 200 percent. Depending on the land class, Vegetables - 4,550 respective wheat and paddy yields would increase from 1.1 Sodic land area (ha) 130,000 130,000 to 3.5 t/ha and 1.0 to 3.0 t/ha. Drain benefits would derive Gross cropped area (ha) 58,000 261,300 from expansion of wheat production to 11,300 ha of land Cropping intensity (ha) 45% 201% with reduced waterlogging (at 2.1 t/ha) and 5 percent increase in wheat productivity (over the Class B yield of 2 t/ha) on 215,000 ha of land where flooding duration would be reduced making early planting of wheat possible. Crop diversification benefits would derive from diversification towards non-cereal crops and vegetables on 13,000 ha of “Class B+” sodic lands and 20,000 ha of non-sodic lands. Livestock benefits would derive from breed improvements and animal health care. It was assumed that 8,560 cross-breed lactating cows would be produced annually through the support by 100 animal health camps. Milk productivity for the cross-breed cows was assumed at 8 liter per day (lt/day) over 250 lactation days against 2.5 lt/day over 200 lactation days for indigenous cows. Rural haat (RH) benefits would derive from 4.5 percent increase in price realization and 2.5 percent increase in baseline trade volumes at 260 metric ton (MT) per year in each of the 125 RHs. 2. Incremental net benefits were estimated for a 20-year period, at 2009 prices and 12 percent opportunity cost of capital (OCC). Financial prices were converted into economic values at a standard conversion rate of 0.90. The appraisal estimated Table 4.2 Summary of appraisal EFA results economic rate of return (ERR) at 25.1 percent and Project interventions ENPV (Rs ERR (%) economic net present value (ENPV) of Rs. 9.5 billion. billion) Individually, the land reclamation was expected to Land reclamation 9.7 31.8 generate the highest ERR at 31.8 percent with ENPV of Main drain rehabilitation (0.0) 11.9 Overall Project Rs. 9.7 billion. The drainage investments were to generate Agriculture intensification 6.1 20.9 ERR of 11.8 percent with negative ENPV. The appraisal Plus Diversification 7.7 22.8 analysis was based on unrealistic implementation Plus Drain improvements 9.5 25.0 timelines and as such the incremental returns were Plus Market access 9.5 25.1 overestimated. Page 34 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) ICRR Economic and Financial Analysis 3. Methodology. The ICRR analysis re-estimated the project returns following the appraisal methodology and using actual and updated future costs and benefits. The project M&E data had considerable limitations. To overcome these limitations, the ICRR analysis used secondary data of the line departments, and primary field data collected by the supervision and ICRR missions. The project returns were estimated for 29 years inclusive of nine years of project implementation. 4. Sodic land reclamation benefits. The project reclaimed around 142,677 ha of previously sodic lands. These included 85,287 ha of “Class C”, 43,122 ha of “Class B”, and 14,268 ha of “Class B+” lands.24 Around 68,649 ha of reclaimed land was provided with groundwater irrigation. At closing, 86 percent of “Class C”, 87 percent of “Class B”, and 100 percent of “Class B+” Table 4.3. Sodic land reclamation results lands were utilized. Across all classes of reclaimed WoP WiP lands, cropping intensity increased from baseline of 45 Cropping area (ha) percent to 185 percent. The baseline yields increased Paddy, Class C - 73,286 from 1.36 to 3.36 t/ha for paddy and from 0.8 to 3.38 Paddy, Class B 37,565 37,565 25 t/ha for wheat. Around 2.8 percent of gross cropped Paddy, Class B+ 14,268 14,268 Wheat, Class C - 73,286 areas (GCA) was diversified towards vegetables and Wheat, Class B - 37,565 8.8 percent towards oilseeds and pulses. Combination Wheat, Class B+ 14,268 14,268 of improvements in crop productivity and intensity Oilseeds and pulses, Class B+ - 11,261 and diversification increased crop incomes per ha Vegetables, Class B+ - 2,717 from Rs. 7,113 to Rs. 12,380. At full maturity, the sodic Sodic land area (ha) 142,677 142,677 land reclamation investments are estimated to Gross cropped area (ha) 66,100 264,215 generate incremental annual returns at Rs. 3.84 Cropping intensity (ha) 46% 185% billion. 5. Ravine reclamation. The project reclaimed 23,943 ha of ravine lands. Utilization of reclaimed ravines was not monitored. Therefore, the current utilization rate observed for the reclaimed sodic lands at 88 percent was assumed also for the reclaimed ravine lands. As a result of reclamation and improved agricultural services, yields in ravines increased from 2.3 to 2.8 t/ha for paddy and 2.1 t/ha to 2.7 t/ha for wheat. Cropping intensity increased from 140 percent to 171 percent. These improvements increased crop incomes per from Rs. 9,325 to Rs. 12,800. At full maturity, the ravine reclamation interventions are estimated to generate incremental annual returns at Rs. 0.12 billion. 6. Improving drainage systems. Rehabilitation and improved maintenance of drains benefited around 51,339 ha of sodic and 688,872 ha of non-sodic lands. Agricultural improvements resulting from improved drainage were not monitored. Productivity improvements in reclaimed sodic lands were captured under the sodic land reclamation benefits. Whereas 5 percent increase in wheat yields was assumed for the 20 percent of non-sodic lands. 7. Crop diversification. Diversification changes in the sodic lands are accounted under the sodic land benefits. In non-sodic lands, areas under oilseeds and pulses increased from 10.4 percent to 26.3 percent, while areas under vegetables increased from 3.1 percent to 6.2 percent. General trend in diversification as well as the project attribution to diversification in the non-sodic lands were not monitored. Therefore, only 20 percent incremental returns to diversification in non-sodic lands was assumed to be attributable 24 The ICR estimates of C, B and B+ class lands differ from the UPBSN estimates. The UPBSN estimate is based on overestimated proportions of C class lands and does not account for unutilized proportion of reclaimed lands. 25 Net changes that exclude general productivity trends at 8 percent increase for paddy and 11 percent increase for wheat. Page 35 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) to the project. When these assumptions are applied, the diversification improvements are estimated to generate annual incremental returns at Rs. 0.23 billion. 8. Livestock benefits. Over the project period, 211,410 calves of improved breeds, including 102,603 female calves, were born through improved insemination practices. Combination of improved breed and improved animal husbandry practices led to milk productivity increases from 6.2 lt/day to 7.4 lt/day per lactating buffalo and from 3.7 lt/day to 4.8 lt/day per lactating cow (the appraisal baseline values were underestimated). Whereas, the price premium of Rs. 13,500 was obtained per male calve of improved breed. Cumulative incremental returns from the milk productivity and price premiums are estimated at Rs. 0.34 billion. 9. RH benefits. The project monitored only the values of market transaction. As such no data is available on the incremental benefits in terms of increased price realization and market transactions. Therefore, 20 percent of the current values of market transactions were assumed to represent incremental returns. 10. Costs. Actual project cost was estimated at Rs. 12.7 billion (adjusted to 2009 prices). The government’s budget allocation for maintenance of main drains over the project period was estimated at Rs. 0.91 billion and post-project budget allocation would be at Rs. 0.1 billion per year. 11. Sustainability. The project applied a sequence of farmer-led land reclamation and on-farm development processes to restore the land productivity coupled with provision of assured groundwater irrigation, and adoption of improved farm practices. Maintenance of main drains was assured through the fund allocation from the state budget. WUGs were trained to maintain field drain channels through in- kind as well as membership cash contributions. Sustainability risk to drainage systems is low provided that ID will ensure regular monitoring and technical guidance to the WUG and SIC. There is a risk of resodification of currently unutilized reclaimed sodic lands, if these lands remain unused for up to three years. Sustainability of RH infrastructure and operationality of FPOs cannot be assessed. Therefore, the incremental returns were reduced by 20 percent. 12. Results. Summary of appraisal and ICRR economic results are presented in Table 4.5. The ICRR estimated the overall project ERR Table 4.5 Summary of ICRR EFA results at 19.7 percent and ENPV at Rs. Appraisal ICRR 4.0 billion against the appraisal Project interventions ENPV (Rs ERR (%) ENPV (Rs ERR (%) estimate of 25.1 percent and billion) billion) Rs. 9.5 billion. The divergence Land reclamation 9.70 31.8 3.12 18.7 between the appraisal and ICRR Main drain rehabilitation (0.0) 11.9 (0.23) 11.1 returns was due to: (i) Overall Project overestimation of benefits at Agriculture intensification 6.1 20.9 3.6 19.3 appraisal based on unrealistic Plus Diversification 7.7 22.8 4.1 20.3 project timeframe; (ii) delayed Plus Drain improvements 9.5 25.0 4.1 19.9 benefit realization by three years; Plus Market access 9.5 25.1 4.0 19.7 (iii) lower than projected areas of “Class C land”; and (iv) lower than expected utilization of reclaimed lands. 13. Administrative efficiency: Mainstreaming elements of project implementation into GoUP structures (e.g. through Irrigation Department and at the district level) to reduce administrative overheads (in addition to strengthening ownership and ensuring sustainability) was a positive efficiency attribute. Staff Page 36 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) turnover, especially at the highest levels (Managing Director), gaps in staffing for prolonged periods, and weaknesses in implementation capacity at UPBSN affected the timeliness of implementation. 14. Efficiency rating: Despite delays, the cost per ha of reclaimed land remained within the projected ranges. Estimated economic returns at ICRR are also within the recognized and acceptable range for similar investments elsewhere and are also comparable to those reported for the Phase 2 of UPSLRP. To a certain extent, delays in implementation, tempered down overall project efficiency, which in this case, is rated Substantial. Page 37 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS Subject: Draft ICR for Your Comments From: Managing Director, UPBSN To: Abel Lufafa, TTL, World Bank (alufafa@worldbank.org) Dear Abel, This has reference to your e-mail dated 31-05-2019 vide which draft ICR of UPSLRIIIP has been forwarded with the request to review the same and provide feedback to the WB to finalize the report. The draft ICR report has been reviewed and we generally agree with the contents of the report. Our feedback is as follows:- • Para no. 28: Increased adoption of improved agricultural technology: In this point it is stated that Average milk productivity increased to 8.55 liters per animal per year for buffalos (against an end of project target of 7 liters per animal per year) and 6.33 liters per animal per year (against an end of project target of 5 liters per animal per year). - It appears that inadvertently, “6.33 liters per animal per year” is written in place of “6.33 liters per year per cow” which may be looked into and corrected. - Another important point to be mentioned here is that in the PAD results framework for % increase in milk productivity (ltr/yr./lactating animal) was collectively defined for cow and buffalo as follows: Base line YR 1 YR 2 YR 3 YR 4 YR 5 YR 6 2.5 - - - 6.0 - 8.0 - During World Bank Mission 9-22 September 2015 results framework was updated and for % increase in milk productivity (ltr/yr./lactating animal) separately for Cow and buffalo was revised as follows: Project Outcome (A) Base line (B) PAD Target (C) Actual (C)/(B) (D) PAD end of Indicators for YR 5 Value (Attainment of Target) Project Target Cow =>3.41 4.5 4.5 100% 5.0 Buffalo =>5.70 6.5 7.2 111% 7.0 - Again during World Bank Mission 7-15 December 2016 the results framework was further updated and for % increase in milk productivity (ltr/yr./lactating animal) separately for Cow and buffalo was revised as follows: Project Outcome (A) Base line (B) PAD Target (C) Actual (C)/(B) (D) PAD end of Indicators for YR 6 Value (Attainment of Target) Project Target Cow =>3.41 4.8 4.8 100% 5.0 Buffalo =>5.70 6.8 7.4 143% 7.0 - However, in the Table regarding STATUS OF ACHIEVING OBJECTIVES/OUTCOMES at Sl. No. 11 a., Increase in milk productivity (ltr/day/lactating animal) is mentioned as follows:- Indicator Baseline 2009-10 2010-11 2011-12 2012- 2013- 2014- 2015- 2016-17 Project Achievement 13 14 15 16 end % status Buffalo Target 2-5 2.5 - - 6.0 - 8.0 - - - 92.5 largely Actual 6-2 6.2 6.7 6.9 7.2 7.3 7.3 7.4 7.4 7.4 achieved Cow Target 2.5 2.5 - - 6.0 - 8.0 - - - 60.0 not Actual 3.7 3.7 4.1 4.3 4.4 4.5 4.6 4.7 4.9 4.8 achieved Accordingly, it has been observed that in the revised results framework, project end target for increase in milk productivity for cows is mentioned as 5.0. However, in the concerned table at point 11a the target for increase in milk productivity for cows has been mentioned as 8.0. You will very kindly agree that against the revised target of 5.0 lit, project end achievement will thus be 4.8 which will be much more than 60%. Page 38 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) It is requested that the above facts may be taken into consideration and point no. 11a may be revised accordingly. • Para no. 47: Implementation capacity: As the target for reclamation of sodic lands were changed and fixed on year to year basis, the required number of Implementation Assistants (IAs) was maintained on board. All the more, since IAs worked under the project area on area contract basis, they were hired as per the actual requirements of the project. As far as technical positions are concerned, initially the large gap between Phase–II and Phase-III resulted in the drain of trained manpower to other sectors/ projects. However, through intensive efforts, all technical positions critical to the project implementation were filled-up and kept on board almost throughout the project period. • Para no. 55: M&E Quality Rating We forwarded the Draft ICR report to M/s WAPCOS Ltd., the third-party M&E agency for their comments. In their reply dated 07-06-2019 (copy attached), they have stated that M&E system established for UPSLRP- III was primarily based on the successful experiences of phase-1 and phase-2 of the project. Since beginning, rigorous review of the methodology and M&E implementation plan was conducted by the Mission. Mission seemed satisfied with the methodology and implementation plan and no amendment was advised. During subsequent Missions also, M&E findings along with updated Results Framework, performance indicators, methodology, etc were regularly presented and thoroughly discussed. Missions were mostly in agreement with the overall framework, methodology and findings of M&E system under the project which gave an impression that the extant M&E system and its execution was on right track. Aide Memoires of different Missions are testimony to this fact. Based on the above observations, they have concluded that “M&E was executed with solemnity in accordance with the framework and methodology agreed with the World Bank. It appears justified to believe that rigorous and robust monitoring & evaluation system along with compliance review and enforcement mechanism under the project have duly contributed in achievement of project development objectives which has been rated Satisfactory. Thus, upgrading of ‘M&E quality rating’ may be considered.” We concur with the observations of the External M&E Agency. We further wish to add that after the sad demise of the World Bank M&E expert and Economist Mr. S. Silvarajan in the very early years of the project, the Bank team included no full time Economist to support the M&E activities. The reports as well as findings of the M&E agency were throughout presented & accepted by the Missions. Since no amendment was advised, methodology and M&E implementation plan was executed as agreed with the Bank at the beginning of the project. Accordingly, we request that upgrading of ‘M&E quality rating’ may be considered by the World Bank. • Para no. 63: Technical support to FPOs We wish to inform that the State Government is in the process of nominating UPBSN as the State nodal agency to support FPCs. Thus UPBSN will continue to work in this area. • In the Table under the head STATUS OF ACHIEVING OBJECTIVES/OUTCOMES at Sl. No. 6 - Area reclaimed (ha), the project end target has been mentioned as 135,000 and actual achievement has been mentioned as 142,677. It appears that the target includes sodic as well as ravine area (130,000 sodic and 5000 ravine pilot). However, the actual achievement reflects only the achievement made in sodic area. If it is so then, actual achievement should be 166,631 (142,677 ha sodic and 23,954 ha ravine). This needs reconsideration. • In Annexure 3. Project Cost By Component, Total in column no. 3 - Actual cost at Project closing Page 39 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) (US$M) is mentioned as 251.90. The contingencies cost 20.1 is mentioned in this column but not added to the total. This needs your attention. • As far as Financial disbursement position reflected in the table is concerned, the actual figures will change keeping in view the extended period till 29th June 2019 for clearing the liabilities accrued under the project till closing date. Regards, Ajay Yadav Managing Director Response to Borrower’s Comments a) Para no. 28: Increased adoption of improved agricultural technology: ▪ In this point it is stated that Average milk productivity increased to 8.55 liters per animal per year for buffalos (against an end of project target of 7 liters per animal per year) and 6.33 liters per animal per year (against an end of project target of 5 liters per animal per year). It appears that inadvertently, “6.33 liters per animal per year” is written in place of “6.33 liters per year per cow” which may be looked into and corrected. Response: This error has been addressed. ▪ Another important point to be mentioned here is that in the PAD results framework for % increase in milk productivity (ltr/yr./lactating animal) was collectively defined for cow and buffalo as follows: Base line YR 1 YR 2 YR 3 YR 4 YR 5 YR 6 2.5 - - - 6.0 - 8.0 ▪ During World Bank Mission 9-22 September, 2015 results framework was updated and for % increase in milk productivity (ltr/yr./lactating animal) separately for Cow and buffalo was revised as follows: Project (A) Base line (B) PAD Target (C) Actual Value (C)/(B) (D) PAD end Outcome for YR 5 (Attainment of of Project Indicators Target) Target Cow =>3.41 4.5 4.5 100% 5.0 Buffalo =>5.70 6.5 7.2 111% 7.0 ▪ Again during World Bank Mission 7-15 December, 2016 the results framework was further updated and for % increase in milk productivity (ltr/yr./lactating animal) separately for Cow and buffalo was revised as follows: Project (A) Base line (B) PAD Target (C) Actual Value (C)/(B) (D) PAD end Outcome for YR 6 (Attainment of of Project Indicators Target) Target Cow =>3.41 4.8 4.8 100% 5.0 Buffalo =>5.70 6.8 7.4 143% 7.0 ▪ However, in the Table regarding STATUS OF ACHIEVING OBJECTIVES/OUTCOMES at Sl. No. 11 a., Increase in milk productivity (ltr/day/lactating animal) is mentioned as follows:- Page 40 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Indicator Baseline 2009-10 2010-11 2011-12 2012- 2013- 2014- 2015- 2016-17 Project Achievement 13 14 15 16 end % status Buffalo Target 2-5 2.5 - - 6.0 - 8.0 - - - 92.5 largely Actual 6-2 6.2 6.7 6.9 7.2 7.3 7.3 7.4 7.4 7.4 achieved Cow Target 2.5 2.5 - - 6.0 - 8.0 - - - 60.0 not Actual 3.7 3.7 4.1 4.3 4.4 4.5 4.6 4.7 4.9 4.8 achieved Accordingly, it has been observed that in the revised results framework, project end target for increase in milk productivity for cows is mentioned as 5.0. However, in the concerned table at point 11a the target for increase in milk productivity for cows has been mentioned as 8.0. You will very kindly agree that against the revised target of 5.0 lit, project end achievement will thus be 4.8 which will be much more than 60%. It is requested that the above facts may be taken into consideration and point no. 11a may be revised accordingly. Response: The ICRR evaluates the project against the targets set in the formally approved Results Framework. The referred changes had not been formally processed. Therefore, the target values set in the PAD are used. b) Para no. 47: Implementation capacity: As the target for reclamation of sodic lands were changed and fixed on year to year basis, the required number of Implementation Assistants (IAs) was maintained on board. All the more, since IAs worked under the project area on area contract basis, they were hired as per the actual requirements of the project. As far as technical positions are concerned, initially the large gap between Phase–II and Phase-III resulted in the drain of trained manpower to other sectors/ projects. However, through intensive efforts, all technical positions critical to the project implementation were filled-up and kept on board almost throughout the project period. Response: Staff shortages, including IAs, consistently voiced in supervision reports until June 2013 was one of the contributing factors to inadequate timeliness and quality of implementation. The supervision report from June 2013, for instance, estimated the shortage in IAs between one third to one-fifth. In addition to the shortage, issues related to wages and allowances, additionally contributed to misclassifications of sodic lands including by IAs. Moreover, as stated in the restructuring paper from July 31, 2014, quality issues of the OFD works, which was the basis for the first project extension, were due to the time burdens on field-based staff. c) Para no. 55: M&E quality We forwarded the Draft ICR report to M/s WAPCOS Ltd., the third-party M&E agency for their comments. In their reply dated 07-06-2019 (copy attached), they have stated that M&E system established for UPSLRP- III was primarily based on the successful experiences of phase-1 and phase-2 of the project. Since beginning, rigorous review of the methodology and M&E implementation plan was conducted by the Mission. Mission seemed satisfied with the methodology and implementation plan and no amendment was advised. During subsequent Missions also, M&E findings along with updated Results Framework, performance indicators, methodology, etc were regularly presented and thoroughly discussed. Missions were mostly in agreement with the overall framework, methodology and findings of M&E system under the project which gave an impression that the extant M&E system and its execution was on right track. Aide Memoires of different Missions are testimony to this fact. Page 41 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) Based on the above observations, they have concluded that “M&E was executed with solemnity in accordance with the framework and methodology agreed with the World Bank. It appears justified to believe that rigorous and robust monitoring & evaluation system along with compliance review and enforcement mechanism under the project have duly contributed in achievement of project development objectives which has been rated Satisfactory. Thus, upgrading of ‘M&E quality rating’ may be considered.” We concur with the observations of the External M&E Agency. We further wish to add that after the sad demise of the World Bank M&E expert and Economist Mr. S. Silvarajan in the very early years of the project, the Bank team included no full time Economist to support the M&E activities. The reports as well as findings of the M&E agency were throughout presented & accepted by the Missions. Since no amendment was advised, methodology and M&E implementation plan was executed as agreed with the Bank at the beginning of the project. Accordingly, we request that upgrading of ‘M&E quality rating’ may be considered by the World Bank. Response. In evaluating the project M&E, the ICRR considers: (i) quality of the M&E design at appraisal; (ii) quality of modified M&E design, if this has been done; (iii) observations of the supervision reports on the quality of progress and results monitoring data and reporting; and (d) review of survey designs, M&E data and reports by the ICRR team. Main reasons for the Modest rating are provided in the ICRR. These are: (i) inadequate quality of progress data and reporting: (ii) lack of participatory monitoring; (iii) lack of alternative options to measure general trends; (iv) inadequate quality of survey designs that affected accuracy of assessing results and outcomes and capturing the project’s attribution; (v) limited utilization of the M&E data in management’s decision making; and (vii) quality of M&E reporting. d) Para. 63. Technical support to FPOs. We wish to inform that the State Government is in the process of nominating UPBSN as the State nodal agency to support FPCs. Thus UPBSN will continue to work in this area. e) In the Table under the head STATUS OF ACHIEVING OBJECTIVES/OUTCOMES at Sl. No. 6 - Area reclaimed (ha), the project end target has been mentioned as 135,000 and actual achievement has been mentioned as 142,677. It appears that the target includes sodic as well as ravine area (130,000 sodic and 5000 ravine pilot). However, the actual achievement reflects only the achievement made in sodic area. If it is so then, actual achievement should be 166,631 (142,677 ha sodic and 23,954 ha ravine). This needs reconsideration. Response: The number has been corrected. The end target for ‘area of sodic lands reclaimed’ now reads as 130,000 ha. f) In Annexure 3. Project Cost By Component, Total in column no. 3 - Actual cost at Project closing (US$M) is mentioned as 251.90. The contingencies cost 20.1 is mentioned in this column but not added to the total. This needs your attention. g) As far as Financial disbursement position reflected in the table is concerned, the actual figures will change keeping in view the extended period till 29th June 2019 for clearing the liabilities accrued under the project till closing date. Response: The ICRR template automatically extracts the latest costs from the WB financial system and currently presents the actual costs as of June 14, 2019. Once the final IUFR is processed, the actual costs will be updated. Page 42 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) ANNEX 6. SUPPORTING DOCUMENTS Technical Note on Assessment of the PDO Outcome Indicators As presented in Table 6.1, there is substantial difference between the ICRR estimated and project estimated achievements of outcome indicators. This technical note is prepared to provide clarification on methodology for assessing the PDO outcome indicators in the ICRR and in the project reports and Borrower’s Completion Report (BCR). Table 6.1. Comparison of the Project and ICRR estimated achievements of outcome indicators End Project estimate ICRR estimate Unit of PDO indicators target End actual Achieve- End actual Achieve- measure value value ment value ment 1. Increase in productivity of wheat tons/year 6.40 6.53 102.0% 6.02 82.5% and paddy 2. Increase in cropping intensity % 200% 212% 106.0% 185% 92.6% 3. Percentage increase in crop % BL+71% 451% 635.2% 76% 107.0% income per unit land 4. Increase in C class land value Rs/ha 650,000 1,120,000 172.3% 648,279 99.7% PDO outcome indicator 1. Increase in productivity of wheat and paddy. ICRR estimate presents the With Project (WIP) annual wheat and paddy production on reclaimed (and utilized) sodic lands divided by the total reclaimed sodic lands (refer to the formula below). The WIP wheat and paddy yields are estimated as an average of yields reported by the external M&E as well as Department of Agriculture (DOA) minus general productivity trends observed by the Bank supervision in 2014. The DOA estimated yields through cross cutting experiences (CCE) in the project sites (sodic lands). As the project lost control groups around 2014, monitoring of general productivity trends was discontinued in the subsequent years. Due to unavailability of reliable general trend data, the ICRR adopts the 2014 trend at 8 percent increase for paddy and 11 percent for wheat. The project estimate is based on the WIP annual wheat and paddy production on reclaimed and utilized sodic lands reported by the farmers in survey conducted by the external M&E in 2018. As such, the project estimate does not factor in: (a) the DOA’s CCE yields; (b) general productivity trends for wheat and paddy; and (c) reclaimed but unproductive/unutilized sodic lands. ICRR assessment: Productivitywip = (Pwwip+Ppwip)/(Ar) Where, Pwwip is WIP annual production of wheat on reclaimed sodic lands at project closing estimated as = ̅ × + ̅ × + ̅ (;;+) represents average wheat yield for ̅ + × + Here previously sodic lands (e.g., Class C, Class B, and Class B+) calculated as average of yields recorded by the external M&E agency (surveys) and Department of Agriculture (DOA) (through cross cutting experiments in project sites) less the general productivity trend for wheat observed by the Bank supervision at 11 Page 43 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) percent. (; ; +) represents reclaimed sodic lands (e.g., Class C, Class B, and Class B+) that are currently utilized for wheat production. Ppwip is WIP annual production of paddy on reclaimed sodic lands at project closing estimated as = ̅ × + ̅ × + ̅ (;;+) represents average paddy yield for ̅ + × + Here previously sodic lands (e.g., Class C, Class B, and Class B+) calculated as average of yields recorded by the external M&E agency (surveys) and Department of Agriculture (DOA) (through cross cutting experiments in project sites) less the general productivity trend for wheat observed by the Bank supervision at 8 percent. (; ; +) represents reclaimed sodic lands (e.g., Class C, Class B, and Class B+) that are currently utilized for paddy production. Ar is total reclaimed sodic lands including those that are currently unproductive/unutilized. ---------------------------------------------------------------------------------------------------- PDO outcome indicator 2. Increase in cropping intensity (%): The ICRR estimates the outcome indicator by dividing total areas of previously sodic lands used for production of wheat, paddy, and other crops at project closure (only utilized reclaimed sodic lands) divided by the total area of reclaimed sodic lands (refer to formula below). Whereas, the project presents cropping intensity on reclaimed and utilized sodic lands as cropping intensity for entire reclaimed sodic lands. As such, reclaimed but unutilized lands are not factored in. ++ ICRR assessment: CP = Where, Aw is reclaimed sodic lands under wheat production (reclaimed and utilized); Ap is reclaimed sodic lands under for paddy production (reclaimed and utilized); Ao is reclaimed sodic lands under other crops (reclaimed and utilized); Ar is total reclaimed sodic lands including those that are currently unproductive/unutilized. ------------------------------------------------------------------------------------------------------------------------- PDO indicator 3. Percentage increase in crop income per unit of land. ICRR estimates the outcome indicator in 2009 prices and as percentage change between the baseline (or Without Project - WOP) and current (or With Project- WIP) crop incomes generated by reclaimed and utilized lands (refer to formula below). WOP crop income was calculated using the phase 2 ICRR data (yield, input usage, crop budget, prices) for treatment and control groups as well as the project M&E data. The WIP crop income was calculated using the WIP yields and input usage at 2009 prices. The WIP input usage data is collected by the external M&E agency. The ICRR estimate accounts for incomes from all crops that are wheat, paddy, mustard, pulses and vegetables. The project estimates the outcome indicator based on annual incomes from paddy and wheat per ha reported by farmers in the 2018 survey (refer to formula below). The WOP crop income was estimated based on farmers recollection and perception of WOP yields and usage of inputs in 2018 prices. Whereas, the WIP crop income is annual income from wheat and paddy production generated by the survey farmers in 2018. Both WIP and WOP crop incomes are then then adjusted to 2010-11 prices using deflators. The 2010-11, which corresponds to the project year 2, was used as the baseline year by the project in all estimates. Page 44 of 45 The World Bank Uttar Pradesh Sodic Lands Reclamation III Project (P112033) ICRR assessment: Increase in crop income per ha = (Iwip-Iwop)/Iwop Where, Iwip is WIP crop income per ha generated in reclaimed sodic lands and estimated as = ∑,,+ + ∑,,+ +∑,,+ . Here Au is reclaimed and utilized sodic lands and Ar is total reclaimed sodic lands. ∑,,+ + ∑,,+ +∑,,+ Iwop is WoP crop income per ha estimated as = . Project assessment: Increase in crop income per ha = (Iwip-Iwop)/Iwop Where, Iwip is annual WIP income from production of wheat and paddy on per ha of reclaimed sodic land reported by the farmers in 2018 survey farmers and adjusted to 2010-11 prices using deflators. Iwop is annual WOP income from production of wheat and paddy on per ha of sodic lands in 2010-11. It is estimated based on the farmers recollection of the 2010-11 yields and usage of inputs and using 2018 prices. The annual crop income than is adjusted to 2010-11 prices using deflators. --------------------------------------------------- PDO outcome indicator 4. Increase in Class C land value (Rs/ha). The ICRR estimates the outcome indicator as the farmer perceived market value of the currently productive and previously barren C class land per ha (adjusted to 2009 prices) divided by the total areas of class C sodic lands reclaimed. The project estimate presents the farmer perceived market value of the previously C class land in 2018 prices. List of project reports used for the ICRR Annual and End line Surveys by WAPCOS26 Annual Results Monitoring Reports for 2011-2018. Agricultural and Socio-Economic Impact Study, 2018. Thematic studies by WAPCOS Assessment of adoption of Agriculture Support Services, December 2012. Trends of Crop Diversification in Reclaimed Sodic Villages, 2018. Assessment of Performance and Impact of Farmer Producer Companies, 2017. Impact Assessment of Ravine Treatment Through Engineering and Agronomic Measures in Ravine Pilot, undated. Impact of Sodic Haat on Rural Economy of Sodic Areas, undated. Other reports Borrower’s Project Completion Report by the UPBSN from February 2019. 26 M&E Agency Page 45 of 45