Document of The World Bank FOR OFFICIAL USE ONLY Report No: 52886-BD PROJECT APPRAISAL DOCUMENT ON A PROPOSED CREDIT IN THE AMOUNT OF SDR 96.4 MILLION (US$150 MILLION EQUIVALENT) TO THE THE PEOPLE'S REPUBLIC OF BANGLADESH FOR AN EMPLOYMENT GENERATION PROGRAM FOR THE POOREST PROJECT October 27, 2010 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (Exchange Rate Effective: September 30, 2010) US$1 = BDT 68.5 BDT 1 = US$ 0.0146 US$1.55619 = SDR 1 FISCAL YEAR July 1 ­ June 30 ABBREVIATIONS AND ACRONYMS ADP Annual Development Plan NFPCSP National Food Policy Capacity Strengthening Programme BRAC Bangladesh Rural Advancement NSAPR National Strategy for Accelerated Committee Poverty Reduction CAS Country Assistance Strategy NSC National Steering Committee CPD Centre for Policy Dialogue OPD Office of the Project Director DHS Demographic and Health Survey DLI Disbursement Linked Indicators OSC Operational Support Component DMRD Disaster Management and Relief PEFA Public Expenditure and Financial Division Accountability EGPP Employment Generation Program PFM Public Financial Management for the Poorest EMF Environmental Management PIC Project Implementation Committee Framework ERD Economic Relations Department PIO Project Implementation Officer FM Financial Management PIU Project Implementation Unit GOB Government of Bangladesh RPF Resettlement Policy Framework HIES Household Income and Expenditure RP Resettlement Plan Survey IBAS SMF Social Management Framework IP Indigenous People SWAp Sector Wide Approach IPPF Indigenous People's Planning PRS Poverty Reduction Strategy Framework IUFR Interim Audited Financial Report TA Technical Assistance MDTF Multi-Donor Trust Fund TAPP Technical Assistance Project Proposal MOFDM Ministry of Food and Disaster UAO Upazila Accounts Office Management Project MPC Main Program Component UNO Upazila Nirbhahi Officer MTBF Medium Term Budget Framework Vice President: Isabel Guerrero Country Director: Ellen Goldstein Sector Director Michal Rutkowski Sector Manager: Mansoora Rashid Task Team Leader: Maitreyi B Das BANGLADESH Bangladesh - Employment Generation Program for the Poorest Project CONTENTS I. STRATEGIC CONTEXT AND RATIONALE ............................................................. 1 A. Country and sector issues.................................................................................................... 1 B. Rationale for Bank involvement ......................................................................................... 3 C. Higher level objectives to which the project contributes .................................................... 4 II. PROJECT DESCRIPTION ............................................................................................. 4 A. Lending Instrument ............................................................................................................. 4 B. Project development objective and key indicators (See Annex 4) ..................................... 5 C. Project components ............................................................................................................. 5 D. Lessons learned and reflected in the program design ....................................................... 11 E. Alternatives considered and reasons for rejection ............................................................ 12 III. IMPLEMENTATION .................................................................................................... 12 A. Partnership arrangements .................................................................................................. 12 B. Institutional and implementation arrangements ................................................................ 12 C. Monitoring and evaluation of outcomes/results ................................................................ 13 D. Sustainability..................................................................................................................... 13 E. Critical risks and possible controversial aspects ............................................................... 14 F. Loan/credit conditions and covenants ............................................................................... 16 IV. APPRAISAL SUMMARY ............................................................................................. 16 A. Economic and financial analyses ...................................................................................... 16 B. Technical ........................................................................................................................... 17 C. Fiduciary ........................................................................................................................... 17 D. Social................................................................................................................................. 18 E. Environment...................................................................................................................... 20 F. Safeguard policies ............................................................................................................. 20 G. Policy Exceptions and Readiness...................................................................................... 21 Annex 1: Sector and Program Background ............................................................................. 22 Annex 2: Major Related Projects Financed by the Bank and/or other Agencies ................. 31 Annex 3: Results Framework and Monitoring ........................................................................ 31 Annex 4: Detailed Project Description ...................................................................................... 40 Annex 5: Project Costs ............................................................................................................... 63 Annex 6: Implementation Arrangements ................................................................................. 64 Annex 7: Financial Management and Disbursement Arrangements ..................................... 71 Annex 8: Procurement Arrangements ...................................................................................... 79 Annex 9: Economic and Financial Analysis ............................................................................. 85 Annex 10: Safeguard Policy Issues ............................................................................................ 93 Annex 11: Project Preparation and Supervision ..................................................................... 96 Annex 12: Documents in the Project File ................................................................................. 97 Annex 13: Statement of Loans and Credits .............................................................................. 98 Annex 14: Country at a Glance ............................................................................................... 100 Map............................................................................................................................................. 102 BANGLADESH EMPLOYMENT GENERATION PROGRAM FOR THE POOREST PROJECT PROJECT APPRAISAL DOCUMENT SOUTH ASIA SASSP Date: October 27, 2010 Team Leader: Maitreyi Das Country Director: Ellen A. Goldstein Sectors: Public administration ­ other social Sector Director: Michal Rutkowski services (100%) Sector Manager: Mansoora Rashid Themes: Social safety nets (70%); improving Project ID: P118701 labor markets (30%) Lending Instrument: Specific Investment Loan Environmental category: Partial Assessment Project Financing Data [ ] Loan [X] Credit [ ] Grant [ ] Guarantee [ ] Other: For Loans/Credits/Others: Total Bank financing (US$m.): 150.00 Proposed terms: standard IDA terms with 40 years maturity including 10 years grace period Financing Plan (US$m) Source Local Foreign Total BORROWER/RECIPIENT 0.00 0.00 0.00 International Development Association 0.00 150.00 150.00 (IDA) Total: 0.00 150.00 150.00 Borrower: Economic Relations Division (ERD), Ministry of Finance Government of Bangladesh Responsible Agency: Disaster Management & Relief Division (DMRD), Ministry of Food & Disaster Management Government of Bangladesh Estimated disbursements (Bank FY/US$m) FY 2011 2012 2013 2014 Annual 23.21 28.68 49.34 48.77 Cumulative 23.21 51.89 101.23 150.00 Project implementation period: Start: January 15, 2011 / End: December 31, 2013 Expected effectiveness date: January 15, 2011 Expected closing date: June 30, 2014 Does the project depart from the CAS in content or other significant respects? [ ]Yes [X] No Does the project require any exceptions from Bank policies? [ ]Yes [X] No Have these been approved by Bank management? [ ]Yes [ ] No Is approval for any policy exception sought from the Board? [ ]Yes [X] No Does the project include any critical risks rated "substantial" or "high"? [X]Yes [ ] No Does the project meet the Regional criteria for readiness for implementation? [X]Yes [ ] No Project development objective To provide short-term employment on community sub-projects to enable households to better cope with vulnerability, while strengthening program implementation. Project description The project would provide financing of the Employment Generation Program for the Poorest and operational support to the implementing agency, the Disaster Management and Relief Division. The project would expand the Program's reach and increase its potential for innovation through a) strengthening the Program's targeting approach to better reach vulnerable households in extreme poverty, and especially women; b) providing better governance as reflected in revised rules that are transparently applied and publicly disclosed; and c) strengthening program monitoring and evaluation systems. Which safeguard policies are triggered, if any? Environmental Assessment (OP/BP 4.01): The nature of the community sub-projects expected to be implemented under the project may lead to negative environmental impacts. Indigenous Peoples (OP/BP 4.10): The community sub-projects may be implemented in tribal areas or areas where indigenous peoples (IPs) are present. Involuntary Resettlement (OP/BP 4.12): The community sub-projects may trigger involuntary resettlement because activities under the expected community sub-projects may require land acquisition resulting in temporary or permanent loss of land, crops, and other means of income generation. Significant, non-standard conditions, if any, for: Board presentation: None Loan/credit effectiveness: None Covenants applicable to project implementation: The Government of Bangladesh, through the DMRD, shall carry out its parts of the program in accordance with Program Guidelines, including the Environmental Management Framework (EMF) and the Social Management Framework (SMF). The Government of Bangladesh, through the DMRD, shall establish and maintain throughout the period of program implementation, the National Steering Committee, the Technical Project Review Committee and the Office of the Project Director, along with necessary functions including appropriate fiduciary resources and staffing that satisfy the Association. The Government of Bangladesh, through the DMRD, and the program institutions shall maintain financial management systems and carry out the audits in a timely manner, in accordance with terms of reference that satisfy the Association. The Government of Bangladesh, through the DMRD, and the program institutions, shall ensure that implementing agency carries out all procurement under the program in accordance with the World Bank's Procurement Guidelines and Consultant Selection Guidelines. I. STRATEGIC CONTEXT AND RATIONALE A. Country and sector issues 1. Bangladesh has seen an impressive decline in poverty over the last twenty years, but a large number of households still live below the poverty line. About 35 million people live in extreme poverty. The poor have benefited from Bangladesh's impressive growth during the 1990s, with the national poverty rate declining roughly one percentage point per annum over the last 15 years. However, according to the 2005 Household Income and Expenditure Survey (HIES) 40 percent of the population still lives in poverty and 25 percent of the population in extreme poverty (HIES 2005). These poverty rates are very high by international standards and severely impact a number of human development outcomes. The poor are less educated and have lower school attendance rates. Only 40 percent of household heads among the poor has received more than 5 years of education and only 6 percent more than 10 years (HIES 2005). Literacy rates for poor male workers aged 15 and older stand at 52 percent and 33 percent for poor female workers. Besides being less educated, the poor are also less healthy. The Demographic and Health Survey (DHS) (2004) finds that, on average, 16.7 percent of poor children suffer severe stunting. 2. The obduracy of Bangladesh's poverty is deepened by its chronic natural disasters, which undermine both economic gains and the resilience of poor households. Bangladesh is often cited as the most "disaster prone" country in the world. Climate change is likely to exacerbate this situation. The parts of the country that endured the main brunt of the floods of July-September 2007 and the cyclone Sidr in November 2007 saw their situation further aggravated by the following the hike in food and fuel prices. In 2009 cyclone Aila severely damaged the southern coast, and as recently as May 2010, cyclone Laila was expected to wreak havoc on the south-western coast of Bangladesh. Areas in the north are less prone to the vagaries of cyclones, but experience frequent floods from numerous rivers. The ecologically fragile areas along the riverbeds of the north are also notoriously prone to seasonal monga, when household food stocks dip to starvation levels. In all, natural disasters and an unpredictable climate have become a part of life in Bangladesh's poorest and ecologically fragile areas. While the Government of Bangladesh (GOB) has been very proactive with its relief efforts, the long-term impact of these disasters on the poorest households is known to be severe. Mitigating these long-term impacts requires effective safety nets on the one hand, and opportunities for people to graduate from extreme poverty on the other. 3. The food crisis of 2007-08 revealed the vulnerability of Bangladesh's poor households to shocks. The sheer size of the population living around the poverty line1 means that even a small shock can push large numbers of individuals into poverty, and many poor into extreme poverty. High levels of food inflation in 2007-08 eroded real incomes of the poor much faster than that of others because of the higher share of food in their consumption basket. A recent World Bank survey suggests that during the crisis a majority of households reduced their food and non-food intakes as well as switching to food of inferior quality. Between December 2007 and March 2008 the number of households that either skipped a meal or ate scarcer meals at least once a month rose from 43 percent to 58 percent. Roughly 8 percent of households with school-going children took their children out of school; most of these children took up jobs. Clearly shocks, whether economic or in the form of natural disasters, can frustrate improvements made in nutrition and education. 4. Bangladesh's poverty profile is further complicated by the need to employ a large and growing labor force. While open unemployment rates are low, serious under-employment exists, especially among young men. Female employment levels have remained low, despite increasing levels of education and 1 As reflected by the distribution of consumption in HIES 2005. See Bangladesh 2008 Poverty Assessment for detailed analysis. 1 low levels of fertility. Most women are employed in household enterprises as unpaid workers. Even women from poor households, who would be expected to engage in manual labor, are unlikely to be employed for a wage. They do not seem to find casual employment either. Self-employment is the default employment type all around and only a very small proportion of all workers are employed in wage jobs. Each successive government realizes the importance of the employment agenda, and there is continuous political pressure to come up with policies and programs that will protect vulnerable workers and promote employment. 5. In response to these complex factors, the government in Bangladesh implements a wide range of safety net programs targeted to the poor. This is in keeping with Bangladesh's successive national strategies. Both the previous Poverty Reduction Strategy (PRS) and the current draft National Strategy for Accelerated Poverty Reduction II (NSAPR II) focus on the social protection agenda. They focus in particular on providing employment to the poorest households and regions, so as to prevent the most vulnerable from falling into extreme poverty. In line with this national vision, the Government has a range of cash and in-kind (or food) programs. This is reflected in the categories of safety net programs: (i) labor intensive public works programs where creation of infrastructure is a secondary objective; (ii) livelihood programs for income generation, group training and building awareness on health, nutrition and legal rights; (iii) education and health programs that provide (food or cash) incentives for children's education and for seeking maternal health care; (iv) relief programs designed to mitigate the consequences of disasters, and (v) programs and services for disadvantaged groups like the elderly, the widowed, the disabled, and freedom fighters. The larger of these programs include the Vulnerable Group Feeding (VGF) program, the Old Age Allowance, Vulnerable Group Development (VGD) and Test Relief (TR). 6. Until most recently, Bangladesh had very low coverage of safety nets, despite a plethora of initiatives. In 2005, about 22 percent of households in the lowest consumption quintile received benefits from safety net programs. Even among the bottom 10 percent of the population, the combined coverage of all safety net programs is just 23 percent, and for targeted programs it is only 16 percent. However, the percentage of households who benefit from targeted programs declines progressively as one moves to higher quintiles. Such progressive benefit incidence is a positive feature, but a strong area of concern is the considerable level of inclusion errors across programs 7. In response to the international food and fuel crisis of 2008, the GOB launched its largest public works program: the 100 Day Employment Program. Managed by the Ministry of Food and Disaster Management (MOFDM) in partnership with local governments, the program was intended to cushion the poor against the rise in food prices. The program's design drew on Bangladesh's successful experience with workfare and was a major expansion of the existing cash-for-work model, with only minor adjustments. The program was designed to run during the lean season in rural areas. 8. The Employment Generation Program for the Hardcore Poor (EGPHP) was launched in 2009, close on the heels of the 100 Day Employment Generation Program. The current project ­ the Employment Generation Program for the Poorest (EGPP) is the IDA's support to the EGPHP. The program is intended to run during two lean season phases ­60 days typically from September to November and 40 days from March to April. It takes the lessons of the 100 Day Employment Program into account, especially the issues of targeting, accountability and transparency. In the first phase (which began in September 2009), the program created 16,555,325 person days of employment. The second phase (from March 2009) has just been concluded. Lessons from the design and implementation of the first phase have informed the IDA support to this program. The involvement of the Bank team in the assessment of the first phase has also led to continuous updating of the implementation guidelines. 2 9. Based on the success of the EGPP, the GOB has allocated Tk. 10 billion to the EGPP in the current fiscal year (2011). This includes the proposed IDA support. In FY10, the program had just been introduced and given the urgency for public works in the areas affected by Cyclone Aila and in monga areas, the program was implemented in only 16 districts. Therefore, while the allocation for last fiscal year was 15 percent higher than this year's allocation, expenditures were far lower. In the current fiscal year, the program is expected to extend nationwide, and it is expected that allocations from GOB's side will be increased based on expenditures in the first phase. 10. The design of the EGPP has benefitted from independent assessments of the 100 Day Employment Generation Program conducted by the World Bank (2009), the Centre for Policy Dialogue (CPD) (2009) and the NFPCSP-FAO-BRAC (2008). The assessments were supportive of the program that was implemented during the food crisis, and concur about the lessons learned for future improvements to successor programs. The main lessons, shown in Box 1 in Annex 1, can be grouped under the areas of targeting and beneficiary selection, sub-project selection, accountability, payment mechanisms, and monitoring and evaluation. In addition, there have also been assessments of other food for work and public works programs in Bangladesh, notably by the World Bank (del Ninno, 2003) and by IFPRI (2009). 11. While public works programs in Bangladesh are reasonably well targeted, several assessments of food and cash for work programs have suggested options for improvement. These include increased community input into beneficiary selection and the use of easily observable and verifiable selection criteria. The assessments also call for setting wages (whether in food or in cash) correctly so that the wage rate serves a self-targeting function in which only the poor choose to work. Self-targeted programs could allow for more rapid scale-up in response to crises compared to programs with more extensive selection processes. In addition to wage setting, the assessments also focus on delays in cash payments and suggested exploring a greater use of technology, including mobile phone-based payments, as a potential solution to these delays. The assessments also identified the phenomenon of `ghost workers' whereby a larger number of workers are shown on the rolls than actually work. This allows the project implementation committees at the local level to claim larger resources and is one of the main sources of leakages in public works programs in Bangladesh (and in India). One way of improving the efficiency of the program and reducing these leakages is by increasing the amount of monitoring and supervision at sub-project sites. B. Rationale for Bank involvement 12. The Bank has had a long standing engagement with the GOB on expanding and reforming safety nets. This has been in the form of analytical and advisory work, just-in-time technical assistance and bringing international evidence to policy makers in Bangladesh. Elsewhere in the region as well, the Bank has an expansive dialogue and investments in safety nets and cross-country learning within the region has been effectively fostered. In the case of a workfare program, this intra-region learning is particularly important since several countries have similar delivery systems and a similar history of public works programs. 13. Given previous engagements with the Bank, the Ministry of Finance (MOF) requested financial and technical support for the EGPP. The MOF and the MOFDM are seeking the Bank's involvement not only for financial assistance, but also to make use of the Bank's technical expertise as it reforms the design and implementation of one of the country's largest safety net programs. The MOFDM specifically wants to address poor governance and leakage of funds, and is keen to make use of international experience in doing so. Support for the improvement of safety net programs exists at the highest levels of government. The expectation of the Government and the Bank team is that improving the capacity of the 3 MOFDM to design and implement safety net programs such as the EGPP will benefit a large number of similar programs across the country. 14. The Bank is well placed to assist the Government of Bangladesh in better design and implementation of the EGPP for a number of reasons: It has an ongoing dialogue with the MOFDM and has already contributed to the design of the EGPP; Economic and Sector Work on poverty and safety nets has led to the adoption of poverty maps as the basis for geographical targeting in the EGPP and has contributed to a more refined understanding of safety nets programs in general; The Bank's experience in improving governance and accountability will allow it to bring a wealth of global experience to the design and implementation of the EGPP; and The Bank has substantial expertise in designing and implementing safety net programs and can provide strategic advice on developing a unified, objective and transparent targeting system. C. Higher level objectives to which the project contributes 15. The EGPP, which the IDA proposes to support, is in keeping with the priorities of the GOB as reflected in the draft National Strategy for Accelerated Poverty Reduction II (NSAPR II). The NSAPR II seeks to expand social protection to reduce the vulnerability of the poor and in the process to bring about substantial systemic reforms. In its request to the Bank for assistance, the government has emphasized the importance of receiving and learning from global best practice in improving the governance and implementation of the EGPP. 16. The proposed support is also in congruence with recently approved Country Assistance Strategy (CAS) of the Bank (2011-2014). The CAS intends to provide support to building systems and governance arrangements in the myriad safety nets programs of the GOB and in particular to reach women from the poorest households. The budget speech of the Finance Minister for FY11 highlights the government's imperative to expand safety nets and to place their implementation on a more sound administrative footing. 17. Support to the EGPP is expected to be a long-term programmatic engagement, improving over time the design and implementation of the MOFDM's programs. It is also expected to have a favorable impact on local programs more generally. At least two other Bank projects work at the local level. The Local Governance Support Project (LGSP) supports and creates incentives for decentralization and local accountability. "Notun Jibon" is intended to provide community-based solutions to local problems and issues. The EGPP will work in tandem with both projects and a number of other programs of the GOB. To support this strategic process, the project aims to strengthen the analytical, empirical and institutional base to further develop its social protection policy as well as instruments. II. PROJECT DESCRIPTION A. Lending Instrument 18. Support to the EGPP is envisaged as a Credit of US$150 million from the International Development Association (IDA) through a Specific Investment Loan (SIL). Disbursements will be linked to a set of outcome indicators that are laid out in Annex 3. This mechanism creates incentives for efficiency and allows for innovations. The national discourse after the announcement of the Budget of 2010-11 has focused heavily on the poor implementation of development programs in general. In 4 recognition of this, IDA would finance an operational support component that will assist with the design and implementation of the program. 19. IDA support to EGPP will finance up to 70% of wage costs over the three-year period (2011- 2013) up to a maximum of US$144.75 million. There is also provision for retroactive financing if DLIs are achieved before effectiveness. If any amount remain undisbursed at the end of the IDA support, due to the non-achievement of DLIs, the Bank may authorize an additional disbursement, after the final disbursement for FY2014, in an amount not to exceed the lowest of the following: (i) total undisbursed funds under the MPC; (ii) 70% of total wages over the entire project period less total payments to date; and (iii) US$20 million. In addition, US$5.25 million is earmarked for technical and operational support through a separate window (Component 2 below). This operational support will go towards capacity enhancement that will enable the GOB to achieve the agreed results. In the current financial year (2010- 11), the total allocation to the EGPP is BDT 10 billion, and if this allocation remains stable in future years, a little over one-third of the EGPP budget is expected to come through IDA financing. However, actual disbursements will be based on expenditures and the achievement of the results. B. Project development objective and key indicators (See Annex 4) 20. The development objective is to provide short-term employment on community sub-projects to enable households to better cope with vulnerability, while strengthening program implementation. 21. The PDO will be achieved by i) ensuring that the program benefits are targeted at the poorest areas and households; ii) improving governance through better transparency and accountability; and iii) improving capacity for monitoring and evaluation. 22. Overall, if the same level of allocation to the program is maintained as for the FY10-11 it is expected that the program will create approximately 200 million person days of employment over the three years of the project. Actual performance will be measured by the following outcome indicators, which are indicated in the results framework on Annex 3: short-term employment targeted to vulnerable households in extreme poverty, and especially to women; better governance of the program reflected in revised rules that are transparently applied and publicly disclosed; and program monitoring and evaluation system strengthened to allow timely monitoring of program implementation and third-party verification. C. Project components 23. As mentioned above, the EGPP provides short-term employment to manual workers during the lean season over two phases that add up to 100 days in a year. Wages are meant to be set below the prevailing market wage for unskilled manual workers and this contributes to self-selection of the poorest households into the program. 24. Two components are proposed for this programmatic operation: i. The Main Program Component (MPC) aims to support expansion of and innovations to the existing program through three Focus Areas while financing short-term employment. ii. The Operational Support Component (OSC) aims to provide direct support to the MOFDM for implementation and capacity strengthening. 5 25. The MPC can be viewed as a component that reimburses the GOB for results, while the OSC can be seen as a financing instrument for the inputs needed to achieve the results indicated in the MPC. In terms of the activities, the MPC will finance wages, based on a set of agreed results. The OSC will finance all inputs like monitoring systems, personnel, hardware like motorcycles and computers and consultants. OSC therefore is akin to a technical assistance window. 26. IDA Credit for the MPC would flow into the Consolidated Fund and then onward to the MOFDM. The OSC would disburse separately from the MPC, in that a Designated Account would be set up with the MOFDM and the funds would be disbursed in accordance with the traditional SIL procedures. The Designated Account would be managed directly by the MOFDM. 27. COMPONENT 1: Main Program Component (MPC) (total estimated cost US$144.75 million). This will support expansion of and innovations to the existing program through three Focus Areas. Focus Area 1: Rigorous Targeting 28. Bangladesh has a number of targeted safety net programs that face real challenges in ensuring that the maximum numbers of the poorest households receive benefits. Prior to the implementation of the EGPP and its precursor program there were clear mismatches between the spatial distribution of poverty and program coverage. Data from HIES 2005 showed that the three poorest divisions (Khulna, Rajshahi and Barisal) had lower coverage of safety nets programs than two of the better off divisions (Dhaka and Sylhet). An independent assessment of the precursor to the EGPP found that while 67 percent of the overall program benefits went to the poor, just above 37 percent of the resources reached the extreme poor (NFPCSP-FAO-BRAC 2008). 29. Reaching the poorest is the cornerstone of the EGPP. This includes both the poorest areas and the poorest households. There is a three-fold targeting methodology in the program: Geographical or area based targeting based on the poverty map, where the bulk of the resources go to the poorest upazilas. Household targeting: o Filter 1: The first filter that will ensure the participation of the poorest is the setting of wages. The current wage of Tk. 120 per day is below the prevailing market wage in most areas, ensuring that only the poorest enrol for the program. The program will ensure that wages are set below the prevailing market wage so that the program is not attractive to the non-poor households. o Filter 2: The second filter is a proxy for household poverty ­ individuals from those households where the head is a manual laborer and the household has less than ½ acre of land will be eligible; In the poorest areas, it is likely that there may be a larger turnout of potential beneficiaries than there is budget allocation or capacity to implement. Guidelines to deal with such over-subscription will emphasize the importance of equity, participatory and transparent selection of beneficiaries and publicizing the results of the selection. On the other hand, in richer areas or areas where alternative employment is available, there may be lower turnout. These trends will also inform the allocations in successive phases. Gender targeting where 33 percent of the beneficiaries are intended to be women during the course of the project period. 6 Focus Area 2: Improved governance through clearer rules and enhanced transparency 30. As pointed out earlier, IDA support to EGPP is meant to improve its design and implementation in a way that enhances the governance of Bangladesh's largest safety net program. Focus Area 2 proposes a set of results that enhance clarity of rules and transparency of implementation. 31. Clarity and accessibility of rules and procedures: Program rules are currently quite generic and many field level functionaries are unaware of many changes that were made to the rules and procedures. Rules seldom percolate down the chain to the Union Parishad and hardly ever reach the PIC. The project therefore proposes a set of results (see Annex 3) that require changes to the program document and for the revised rules and procedures to be publicly available all the way down the chain. 32. Information and public disclosure: Currently, communities have limited knowledge of the structure and functioning of the program. Evidence from across the world and from the Sirajganj pilot project in Bangladesh indicates that community level notice boards, displaying regularly updated, reliable and correct information, enhance community confidence and participation. Community notice boards will be mandated under the project. PICs and Union Committees would also be required to make selection criteria and beneficiary lists publicly available. Adherence to this provision would be monitored through independent third party validations and process assessments conducted under Focus Area 3 of the project. 33. Provision of non-wage costs: Typically, public works programs in Bangladesh do not have a provision for non-wage costs and the EGPP is no exception. Providing for costs associated with the purchase of materials, hardware or skilled labor is central to the quality of infrastructure and governance of the program. Lack of such provision could have several negative impacts. First, it restricts the type of projects to mud-based or "earth" works. Second, if communities are expected to contribute non-wage costs, it affects poorer communities disproportionately. Finally, lack of provision for non-wage costs can create incentives for irregularities, especially when combined with strict monitoring (pressure to deliver the program). In such cases, more "enterprising" PICs may divert wage costs to material costs through the use of `ghost workers'. During the project period, guidelines will be issued to earmark up to 20 percent of the cost of each EGPP subproject to purchase materials and skilled labor. This is in keeping with global practice. For most infrastructure projects at least 10 percent of the cost would be mandated for materials and other non-wage costs. Disbursement projections have assumed an average of 20 percent for non-wage costs. While IDA will reimburse only wage costs, the inclusion of non-wage costs in the GOB's program will be an important innovation, which will be linked to disbursement. 34. Better system for payment of wages: Another major innovation to be introduced to the program through IDA support concerns the way wages are paid. Currently EGPP workers are paid by the PIC in cash transactions on a weekly basis. This violates the important principle that resource management and payment processing should be kept separate. The lack of separation accounts partially for the aforementioned irregularities. Furthermore, the recordkeeping systems are dated, processing is done manually, and monitoring is weak. This leaves considerable exposure to misappropriation, payment mismanagement, and abuse in individual payment cases (delays, bribes, collusion, etc.). If administration and payments were separated, if parties other than the PIC made payments directly to the workers, and if all such transactions were adequately automated, it would significantly reduce fiduciary risk and empower the poorest. New robust payment systems therefore need to be introduced and innovations in cash delivery piloted where necessary. In fact, several programs in Bangladesh already pay cash benefits through formal financial channels. Annex 4 details the options that were considered in the reform of payment systems. Given the constraints under which the poorest citizens are served, IDA support to the 7 EGPP envisages that direct payments to workers using formal channels will be introduced and this result will be linked to disbursement. It is proposed that during an initial phase, following all the required preparations, a minimum of 80 percent of those participating unions that have banks located within them will receive payments through banking channels, which will grow to 85 percent in the second year. By the final year of IDA support to the program, it is expected that 95 percent of those UPs which have banks located within them will be paying out wages through banks or other formal financial channels. Importantly, the project will support development of comprehensive manual covering operations of the bank payments program. Delivery of such a manual will be an important indicator for disbursement of IDA funds for payment of wages. 35. Furthermore, where access to the formal financial mechanisms remains significant challenge (e.g., remote rural areas), and with a view to further the progress of expansion of coverage of the new payment systems, IDA will support a pilot to adapt some innovative payment solutions to the operational needs of the program. The pilots will cover a minimum of three upazilas and will be subject to a rigorous evaluation to fine-tune the proposed solutions, bring awareness about good practices, and promote standards in benefit delivery. 36. Enhanced capacity at the field level for better supervision: IDA support to EGPP is fundamentally about strengthening systems of monitoring and implementation. The Government's current capacity is very limited and this creates a potential for malpractice. IDA support to the EGPP therefore provides for field level supervisors in the poorest 334 upazilas. Moreover, upazila level PIOs and the new field level supervisors will be provided with transport in the form of a motorcycle in each of the poorest 334 upazilas. The financing for staff and transportation costs of these supervisors will be done under the Operational Support Component of the project. In addition, while this is not a disbursement indicator, there is a provision for training field supervisors and PIOs. Finally, it is anticipated that the new payment systems will considerably improve the quality of record-keeping, and thus comprehensively strengthen program monitoring at the field level. For example, by allowing for better directed audits. 37. Grievance redress: The project will also support the development of an enhanced grievance redress system. The first point where complaints will be accepted is proposed to be the upazila. This is the level of the project closest to the community that is not directly involved in beneficiary selection or wage payments, two common sources of grievances. However, complaints can be accepted at any of the three levels - Upazila, district or the Ministry. At the upazila level, the Upazila Nirbahi Officer (UNO) will be the Grievance Redress Officer (GRO). At the district level, the Deputy Commissioner (DC) will serve as the GRO. The Deputy Project Director (DPD) in the office of the Project Director will act as the GRO at the central level. A feedback loop that includes informing the complainant about the action taken on their complaints, and feeding this information into the implementation of the program, will be a key aspect of the grievance redress system. Focus Area 3: Better Capacity for Monitoring and Evaluation 38. IDA support to the program is intended to assist the implementing Ministry in refining their monitoring strategy for the EGPP and would provide support to develop the monitoring capacity of the MOFDM at the upazila level and at the Center. The program would support the development of a set of standardized summary indicators and monitoring reports and would provide financing for critical staffing at the central and local levels and investments to introduce a digitized MIS. The details on the proposed investments in this area are provided below. 39. Streamlined data collection and flow: IDA will support the redesign of existing monitoring forms and reports to facilitate monitoring of inputs, outputs, intermediate outcomes and compliance with 8 program guidelines. Data on a set of streamlined indicators will be collected at the sub-project level. In order to ensure that the data collected is available for real-time decision-making, a number of institutional and technical enhancements will be made under focus area three of the program. The program will provide for MIS (in the form of hardware, software, and data entry clerks) at the upazila or district level, as well as connectivity that will allow information to be transferred electronically from the local level directly to the center. The monitoring cell in the Project Director's Office will receive additional support and training. Resources that will allow greater supervision at the sub-project level will also be provided. Providing contract staff and access to transport for PIOs and Tag Officers will further strengthen Monitoring of program implementation at the upazila level (as discussed in Focus Area 2 above). 40. External Validation: Independent validation of the program implementation would be conducted in a random sample of approximately three percent of union parishads each program phase by a firm contracted by the World Bank and financed by trust funds. The independent third party validations would examine whether program implementation guidelines are followed and whether program benefits reach the intended beneficiaries. 41. Impact evaluation: The World Bank would partner with a reputable local survey firm to undertake baseline and follow-up surveys of program participants and non-participants in a representative sample of upazilas. This will be funded by the Trust Fund on Environmentally and Socially Sustainable Development, which has been approved. The survey data will be used to examine program targeting and the determinants of program participation, including whether women or individuals from households in the lowest income quintiles are (i) more likely to receive job cards; and (ii) more likely to work under the program. The individual, household, and community characteristics (e.g., literacy, other wage-earning opportunities, shocks suffered, and distance from the nearest worksite) that influence both the ability to access job cards under the program and the days worked by beneficiaries in the EGPP will also be examined. Program impacts to be examined at the household level include: net household incomes (measured relative to the household's prior year income or relative to otherwise similar non-participant households); responses to lean season vulnerability, including the use of coping strategies such as borrowing, migration, or reductions in the number and quality of meals eaten per day; changes in household assets, especially in relation to possible distress sales, and impacts of the program on human capital through children's school enrolment and household health spending. The survey data will be merged with administrative data detailing the extent and types of workfare activities in the upazilas where households were surveyed. Program targeting performance, new payment system efficiency, and the net income gains to participating households will be identified. 42. Qualitative assessments: To complement the quantitative work, beneficiary and non-beneficiary assessments will be conducted. This work would examine how well the program is working, including perceptions of targeting, of leakages/corruption, the quality of works produced under the EGPP, the value of the community sub-projects (both to participants and to non-participants), and impacts of the program on internal and external migration as well as on social empowerment. 43. The independent assessments (independent third party validations and impact evaluation, including qualitative assessments) under this focus area would not be funded out of the program OSC. Funding for the impact evaluation household surveys has been secured from the TFESSD, and trust funds and external resources from development partners are currently being sought to finance the independent third party validations and qualitative assessments. 44. COMPONENT 2. Operational Support Component (OSC) (total estimated cost US$5.25 million). As pointed out earlier, the MPC can be viewed as a component that reimburses the GOB for results, while the OSC can be seen as a financing instrument for the inputs needed to achieve the results indicated in the MPC. 9 45. Overall objective: Provide operational support to the Disaster Management and Relief Division, MOFDM, in delivering EGPP with enhanced efficiency in terms of better targeting, better benefit delivery mechanism, greater transparency and improved monitoring. 46. Specific objectives: i) Strengthen reporting and monitoring of the EGPP operations ii) Provide training to staff engaged in the EGPP to enhance their capacity for targeting beneficiaries and delivering benefits 47. The activities under this component will contribute to preparing a solid base for the implementation of the Main Program Component (MPC) of the proposed EGPP. But the support through the OSC is expected to go beyond the EGPP. The capacity developed through the OSC at the central and upazila level will be utilized in the implementation of other programs of the MOFDM. The experience of the EGPP, especially the innovations introduced are also likely to influence new initiatives on the ground. The benefits of the OSC are thus likely to percolate to other social protection initiatives implemented by the MOFDM. OSC activities will also strengthen cross-sectoral collaboration among the ministries and with other stakeholders. 48. Cost: The total cost of the OSC is estimated at Bangladeshi Taka 3660.409 lakh or US$5.305 million. Of this amount, Bangladeshi Taka 3623.039 lakh or US$5.25 million will be financed by IDA as a component of the EGPP financing. The detailed cost estimates are provided in Table-1. 49. Expected outputs: Services of financial management and procurement personnel availed. Services of Field Supervisors availed. Management information system (MIS) established. Transport facility for Field Supervisors provided. Systematic targeting methodology established. Training for beneficiary targeting and enrolment process provided. A team of Government officials received exposure to global best practices. The Ministry's overall capacity to deliver quality social protection projects improved. Table 1: Overall Governance Arrangement to Ensure that Benefits Reach the Workers Type of Purpose Actors Check/Assessment Impact Evaluation To assess the impact of the program (i) DMRD, MOFDM / OPD Survey To provide insights on the extent to which Bangladeshi (ii) World Bank households are vulnerable to multiple seasonal, climatic (iii) Firm contracted for Survey and idiosyncratic shocks, how they cope and specifically how the EGPP contributes as a coping strategy Independent third To examine whether program implementation guidelines (i) DMRD, MOFDM / OPD party validations are followed and whether program benefits reach the (ii)World Bank intended beneficiaries Local think tank (iii) (iv)Organization contracted for independent third party validations Financial Audit To ensure implementation of a sound financial (i) Comptroller and Auditor management and monitoring system General of Bangladesh 10 To ensure that the program's (MPC+OSC) financial (ii) DMRD, MOFDM / OPD statements are audited annually (iii) World Bank Grievance To provide a platform for the beneficiaries to voice their (i) NSC headed by the mechanism grievances Secretary, DMRD-MOFDM To deal with the grievance and complaints of workers (ii) Deputy Project Director, under the EGPP OPD (iii) Deputy Commissioners- Grievance Redress Officer (GRO) for the concerned districts (iv) UNO - GRO for the concerned districts (v) Beneficiaries Community level To enhance knowledge about the structure and (i) DMRD, MOFDM / OPD disclosure functioning of the program amongst community member (ii) Union Committee & beneficiaries (iii) Project Implementation To ensure information & public disclosure for enhancing Committee community confidence and participation (iv) Organization contracted for independent third party validations for verification Feedback to To ensure that information about the program generated (i) DMRD, MOFDM / OPD implementation through different sources has an impact on agencies implementation and course corrections are made D. Lessons learned and reflected in the program design 50. Two sets of lessons have informed the design of the EGPP. The first set of lessons draws from recent experiences in Bangladesh, especially the 100 Day Employment Program that preceded the EGPP. The second set of lessons draws from international experience in the design and implementation of public works programs. Box 1 in Annex 1 lays out the lessons learned from the 100 Day Program. a. Bangladesh experience: There have been three independent assessments of the 100 Day Employment Program. They indicate that the targeting methods have worked well overall, but need some refinement to ensure that errors of exclusion and inclusion are minimized. The assessments also showed that communication of the program rules to the local level was lacking, and that the program guidelines were too generic and needed to be made more concrete. In particular, the assessment by the Centre for Policy Dialogue (CPD) (2008) exhorted government to communicate better with the communities and to allow the latter to voice their needs. Almost all recommendations of the CPD assessment have been included in the design of IDA support to the EGPP. In addition, data from the 2005 HIES has been used in the poverty maps that are being used for the geographical targeting of the EGPP. 51. Although program preparation has proceeded at a fast pace, the team undertook some rapid assessments of the first two phases of the EGPP. In particular, in the absence of real time wage data, the team commissioned telephone interviews of Union Parishad Chairpersons to understand whether the wages set under EGPP were below the prevailing market wage. The assessment showed that even at 120 Tk. per day, wages are well below the market. Second, the team commissioned an assessment of options for paying wages through formal financial channels and found that this is desirable but will need to be sequenced, given the number of steps needed before it can be rolled out nationally. b. International experience: The following are the lessons from international experience: 11 Wages should be set slightly below than the prevailing market wage for unskilled labor to promote self-selection If rationing becomes inevitable, use poverty mapping for geographical targeting and then use communities for individual selection Reasonable labor intensity, generally not lower than 40 percent Launch during times when the opportunity cost of labor is low (agricultural slack seasons) Community involvement in selection of projects Adapt design features to encourage female participation Keep transaction costs of participation low Maintain assets created Almost all the principles that emerge from international learning have been incorporated in the design of the EGPP. E. Alternatives considered and reasons for rejection 52. Two alternatives were considered while designing this operation. The first was a development policy loan (DPL) and the second was doing two "standard" SILs. The idea of a DPL was evaluated and rejected because the present operation is intended to disburse against results of a national program, not policy reforms. In fact, safety nets policy in Bangladesh is fairly robust but implementation and program design have mixed results. The present operation therefore, is intended to assist the Government on the one hand through an operational support component and on the other to reward results achieved through such implementation support. 53. The alternative of two independent SILs was rejected to ensure that objectives were complementary and that the program is treated as a single entity without breaking it down in parts to make an artificial distinction between implementation and results. III. IMPLEMENTATION A. Partnership arrangements 54. The proposed program foresees no formal co-financing agreements with other donors or organizations. Nonetheless, the World Bank has a robust dialogue with a number of development partners and will explore opportunities for collaboration in the course of program implementation. B. Institutional and implementation arrangements 55. The Disaster Management and Relief Division (DMRD) of the MOFDM will implement the program along existing institutional structures and implementation arrangements, but with a strong focus on strengthening its systems for a more effective implementation and monitoring of the program. A program implementation team led by a full-time Project Director, not below the rank of Joint Secretary, will support the program. An Office of the Project Director (OPD) will be established in the DMRD for the duration of IDA support to the program, to ensure efficient and timely implementation. One full-time Deputy Project Director, not below the rank of Deputy Secretary, will assist the PD in the implementation of the project. The OPD will comprise of a Project Director, Deputy Project Director, Financial Management Specialist, Procurement Specialist, MIS Specialist, Payment Systems Specialist, Training Specialist, 334 Field Supervisors and the support staff. The Project Director will be responsible for overall management of the program, providing guidance on planning, implementation, operation, 12 monitoring and supervision. He will be assisted by the Deputy Project Director, who will be responsible for field supervision and coordination, administration, finance and monitoring. 56. The OPD will be advised, guided and supervised by a National Steering Committee (NSC), headed by the Secretary DMRD. The NSC will provide policy advice and oversight to ensure the effective implementation of the program. A Technical Project Review (TPR) Committee, headed by the Project Director, will further assist in implementing the program. This Committee will oversee overall program implementation as per the GOB and IDA rules and regulations. The Project Implementation Officer (PIO) of every upazila will be the focal point for the implementation of the program at the upazila level. S/he will be assisted by a Field Supervisor, hired on a contractual basis by the program. At the upazila level, the Upazila Committee (UC) - headed by the Upazila Nirbahi Officer (UNO) - will be responsible for the implementation and supervision of the program. At the Union Parishad (UP) level, the Union Committee - headed by UP Chairman - will have this responsibility. Union Committees may form Project Implementation Committees (PICs) for each community sub-project. These will be responsible for the day-to-day implementation of the community sub-projects, in close collaboration with the Union Committee. In order to deal with the grievances and complaints of workers under the EGPP, a Grievance Redress System will be set up at the national, district and upazila levels. C. Monitoring and evaluation of outcomes/results 57. A range of M&E tools will be employed in assessing the EGPP's implementation performance, outputs, and outcomes. Results monitoring for the program is expected to utilize the strengthened program monitoring system in combination with independent assessments, including both independent third party validations and a program evaluation. Each tool is described in Annex 3. As indicated in Focus Area 3 of Component 1, the following monitoring tools will be employed: (a) Program monitoring. The program will support the development of a set of standardized summary indicators and monitoring reports. (b) External validation. The World Bank will partner with a reputable local think-tank to conduct independent checks for a random sample of approximately three percent of union parishads during each phase of program implementation. (c) Impact evaluation. In order to assess the impact of the program for participants, the World Bank will partner with a reputable local survey firm to undertake baseline and follow-up surveys of program participants and non-participants in a representative sample of upazilas. D. Sustainability 58. IDA is supporting an ongoing GOB program, and the program is expected to remain a key priority of the Government beyond the duration of the proposed program. A 100 Day Employment Program was started by the previous government, and was scaled up after some redesign as the EGPP by the current government. There is strong political support for and ownership of the program across the political spectrum. The EGPP is also in keeping with the priorities of the GOB as reflected in the draft NSAPR II, and furthermore, the current government is keen to make the EGPP a flagship program of the GOB. 59. The institutional sustainability of the Program is supported by recent strengthening of the implementing agency. The GOB recently restructured the MOFDM to create DMRD under which the Ministry's cash-based safety net programs have been placed. Although the Ministry is still limited in its staffing and technical capacity, the DMRD has significant experience in implementing public works programs. The Program would also strengthen the institutional capacity of the DMRD through the technical assistance provided through the OSC. 13 60. The Program design includes mechanisms to assess the Program's progress and impacts to ensure that future decisions about its sustainability incorporate information from assessments. As discussed earlier, the Program would include external assessments of the institutional arrangements during each program phase and a thorough impact evaluation to measure the impact of the EGPP on households' net incomes and ability to cope with shocks. The monitoring mechanisms would provide accurate information of the financial flows and actual expenditures to track program progress. The process assessments would provide timely information about the Program which would provide analytical support for proposed measures that improve its functioning. In other country contexts, such monitoring and evaluation mechanisms have been useful both in strengthening the functioning of programs but also in sustaining support for the program through different political cycles. E. Critical risks and possible controversial aspects 61. The program design is robust and builds on the lessons of previous assessments and international evidence. In addition, an institutional assessment of the MOFDM, a procurement capacity assessment and a financial management assessment were carried out. Discussions have also been held with a variety of stakeholders on potential risks in the Program. 62. The overall risk rating for the EGPP is Substantial. This rating is driven by risks in implementation. The main driver for the risks is that the implementing agency has limited capacity to provide cash payments to such a large number of workers and to monitor program implementation. Thus, two major risks are improper beneficiary targeting and wages misappropriation. The mitigation of these risks is, in a sense, through the project design itself: IDA support to the EGPP is intended to strengthen implementation and monitoring capacity. The Operational Support Component of the EGPP will support improvements in targeting, benefit delivery mechanisms, large-scale investment in transparency and accountability mechanisms, and enhanced monitoring of the Program. Key innovations in program design, such as the introduction of non-wage costs and moving toward payments through formal financial channels, are also expected to mitigate the risks. 63. In addition to the above program-related risks, there are implementing agency risks associated with the limited capacity. The agency has no prior experience with World Bank procurement or financial management, limited exposure to other development partner procurement or financial management processes and guidelines, and needs to boost its staffing. There are several mitigating measures which are planned to address these weaknesses, including the strengthening of capacity by hiring a Financial Management Specialist and a payment specialist. 64. Table 2 below describes the overall risk environment of the project ­ not just in terms of Governance and Corruption risks, but the broader risks that may affect achievement of the project's development objectives: 14 Table 2: Key Risks and Mitigation Measures Risk Rating Description of Risk Mitigation Measures with Mitigation Weak system for payment of The project will support the progressive payment of wages S wages, with large amounts of through banks and other formal financial channels (DLI). wages in cash currently being The OSC will provide technical support for electronic distributed, leading to possible database of beneficiaries which will make beneficiaries irregularities in wage and payments easier to monitor (DLI). distribution through the use of The OSC will also provide for the services of a payments `ghost workers' and short specialist to design and monitor wage payments. payment of wages. Non-wage costs for community sub-projects will be borne by the Government during the Project term to ensure that wage cost allocations are not diverted to provide for these expenses (DLI). Lack of adequate staffing and Bank support through the OSC provides for Field S resources at field level to Supervisors with vehicles to support PIOs, plus computers monitor implementation of the and connectivity to assist in monitoring and reporting numerous community sub- (DLI). projects, including compliance The OSC will provide technical support for electronic of beneficiary targeting and database of beneficiaries which will make compliance with selection with program guidelines easier to monitor (DLI). guidelines. The Bank will support external validation checks at the sub-project level. Standard selection criteria and processes for beneficiaries, with a focus on community involvement to ensure transparency, will be incorporated into program guidelines and compliance with household targeting will be closely monitored (DLI). Program oversight currently The central unit will be staffed throughout the year during N rests with a Project the Project period and formalized as the Office of the Management Cell (PMC) that Project Director (OPD). is understaffed and only The OSC will provide for the services of additional seasonally functional. specialists to support the OPD. Beneficiary grievances are Formal grievance address procedures will be introduced in M currently addressed informally the revised program guidelines. on a case by case basis. No prior experience with Service of an experienced Procurement Specialist to assist S World Bank procurement and the DMRD in preparation of bidding documents, request FM procedures, lack of for proposals, bid evaluation, technical evaluation, contract adequate staffing for negotiation and contract management. procurement and FM, and Service of an experienced Financial Management limited experience with Specialist to assist the DRMD in ensuring smooth flow of managing local procurement project funds, preparation of IUFRs, preparation of annual processes. financial statements, coordination of project audit, among other relevant tasks. A functional webpage with procurement related information accessible to the public. A system for handling complaints and a database for recording, monitoring and following up all project procurement activities. Training on ICB procurement, internal control, documentation, information dissemination, FM, etc. provided for OPD. 15 Risk Rating Description of Risk Mitigation Measures with Mitigation Procurement guidelines will be issued to the local level to ensure clarity on procedure and training will be provided to upazila level staff. Use of report-based disbursement directly into the Government consolidated fund following independent verification of DLI achievement. No prior experience with Training will be provided to central and upazila level staff M World Bank social and on safeguards requirements and how to carry out screening environmental safeguards procedures for the SMF and EMF. requirements. Third party validation checks will be performed on a random sample of sub-projects to ensure compliance with safeguards requirements. Overall Risk Rating: Substantial Risk Ratings N ­ Low or negligible risk; M ­ Modest Risk; S ­ Substantial Risk; H ­ High Risk F. Loan/credit conditions and covenants 65. Covenants applicable to program implementation: i) The Government of Bangladesh, through the DMRD, shall carry out its parts of the program in accordance with Program Guidelines, including the Environmental Management Framework (EMF) and the Social Management Framework (SMF). ii) The Government of Bangladesh, through the DMRD, shall establish and maintain throughout the period of program implementation, the National Steering Committee the Technical Project Review Committee and the Office of the Project Director, along with necessary functions including appropriate fiduciary resources and staffing that satisfy the Association. iii) The Government of Bangladesh, through the DMRD, and the program institutions shall maintain financial management systems and carry out the audits in a timely manner, in accordance with terms of reference that satisfy the Association. iv) The Government of Bangladesh, through the DMRD, and the program institutions, shall ensure that implementing agencies carry out all procurement under the program in accordance with the World Bank's Procurement Guidelines and Consultant Selection Guidelines. IV. APPRAISAL SUMMARY A. Economic and financial analyses 66. The economic analysis presents conservative estimates of the cost effectiveness ratio of the program and concludes that the Program is economically sustainable. The analysis quantifies the expected impact of the EGPP on poverty and estimates how cost effective the workfare program is in raising incomes of the poor compared to a counterfactual program in which every household receives a uniform transfer (following the approach in Ravallion 1998).2 Under this comparison, a workfare scheme should not be supported if the total gain to the poor as a proportion of a given budget allocation is less 2 The longer term benefits of public works programs on consumption smoothing, asset protection and the avoidance of negative coping behaviors are harder to quantify, and it is thus difficult to calculate an overall rate of return to such programs, as most evidence from workfare programs cannot measure the expected reduction in the long-term transmission of poverty and vulnerability. 16 than the percentage of households who are poor, as in that case an untargeted transfer to the population would provide superior poverty alleviation. The cost effectiveness analysis of the program considers (i) the extent to which the program reaches the beneficiary population, (ii) the benefits of the program in terms of net income transfers to poor households, and (iii) the value to poor households of community assets created under the program. Cost effectiveness ratios calculated using conservative but realistic assumptions show that the EGPP can be justified versus the counterfactual program even if the future value of the assets created is low. 67. The fiscal impact of the current EGPP is consistent with the fiscal costs of safety net programs in countries with comparable levels of income and with other countries in the region. The EGPP budget is adequately funded and fits within the Government's overall macroeconomic framework. Therefore, the fiscal impacts of the EGPP seem manageable. B. Technical 68. IDA support to the GOB's EGPP is intended to provide technical and financial inputs that would lead to reform of the program and increase the capacity of the implementing agency. Even during the preparation phase, the involvement of the preparation team has led to changes in program rules. The design of the program will benefit from the following principles: A robust targeting system that will ensure that the benefits got to the most vulnerable households Innovations in design that include wage payments through formal channels rather than through cash transactions, provision of non-wage costs so that the possibility of "ghost workers" is minimized, systems of public disclosure and accountability and an electronic beneficiary database. Enhanced monitoring and supervision capacity through training, personnel, computers and transportation. 69. The program breaks new ground in the way public works programs in Bangladesh are designed and implemented. It is expected that the positive outcomes of this program will influence the design of other similar programs. C. Fiduciary Financial Management 70. A financial management assessment of the EGPP was carried out as part of the preparation of the program. The assessment has been carried out in accordance with OP/BP 10.02, and follows the principles laid down in the Financial Management Practices Manual. 71. The funds' flow and the financial management arrangements of the existing program would continue without any significant change required for IDA financing. To enable this, the major chunk of IDA financing associated with the MPC would directly go to the Government's Consolidated Fund and the disbursement would be linked to achievement of outcome targets ­ disbursement linked indicators. For the OSC, a CONTASA account (designated account) would be established at the disposal of the Project Director for making payment pertaining to the TAPP. Separate withdrawal applications would be required for disbursement under each component. 72. The IBAS system is capable of generating program-related financial reports using the 13 digit chart of accounts comprising four separate elements for the legal, function, operational/program, and 17 economic codes consistently used for budgeting as well as accounting. Relevant budget appropriation for wages would be provided under code 1-4801-0006-5965 (employment scheme for extremely poor people - special grant). This enables generating a financial report indicating expenditures by upazilas. There would be no separate accounting required, as the MPC would finance the wages as the only eligible expenditure under the IDA Credit. 73. An FM staff would be engaged for accounting, disbursement, and financial reporting purposes. The Bank requires that the program's (MPC+OSC) financial statements be audited annually and provided to the Bank no later than 6 months after the end of the reporting period. 74. The initial financial management risk of the program is assessed as Substantial. While some of the mitigating measures have been incorporated into program design, a number of measures identified are yet to be taken up for implementation. The main driver of the risk rating is the large number of wage payments. Procurement 75. A procurement capacity assessment was carried out in the DMRD with the web based Procurement Risk Assessment Management System (P-RAMS). The agency will require adequate procurement staff to manage procurements under the OSC. Furthermore, the DMRD needs more training and experience in international competitive bidding (ICB) following Bank guidelines. Like any other Ministry or government agency, the DMRD is not immune to systemic issues affecting procurement efficiency and performance. In addition to adequate staffing for procurement needs, emphasis also needs to be put on internal control, documentation, information dissemination, administration of contracts - including delivery follow up, payments, handling complaints, etc. The program is rated as "Substantial- Risk" from a procurement operation and contract administration viewpoint. Several measures will need to be introduced to minimize the risk during the implementation of the program. 76. Measures for Improving Governance in Procurement: Several measures have been introduced to and agreed with the Government to improve governance and minimize the risks of procurement under this program. The major action plans are: (i) Identify Procurement Focal Points (PFP) in DMRD (ii) Hire an experienced procurement consultant to help the agency with the preparation of bidding documents / requests for proposal, bid evolution / technical evaluation, contract negotiation and contract management. (iii) Establish a functional webpage for the DMRD where procurement related information is made accessible to the public; (iv) Establish a system for handling complaints and a database to record, monitor and follow up on all the procurement activities under the program; (v) Introduce a procurement risk mitigation plan (PRMP) through reports submitted to IDA on a periodic (semi-annual or quarterly) basis. D. Social 77. Sub-projects are intended to be very small scale and implemented over brief spells to protect the poorest. No private land acquisition is expected to take place under the Program. Since most sub-projects will be carried out in rural settings where squatters are not widely found, no large scale displacement is expected. However, some sub- projects may require limited land acquisition or may displace encroachers/squatters as a result of project works. The program thus triggers the safeguards policy on involuntary resettlement (OP4.12). Given the nation-wide approach of the program it is also probable 18 that the program will touch upon areas with indigenous people (IPs), thereby triggering the safeguards policy for Indigenous Peoples (OP 4.10). 78. A Social Management Framework has been produced that includes a Resettlement Policy Framework (RPF) and an Indigenous Peoples Planning Framework (IPPF). Due to very large number and nature of sub-projects (up to 37,000 or more) to be implemented within the project period, the implementing agency will use the screening procedures outlined in the SMF to identify, assess, evaluate, mitigate and monitor social impacts of each sub-project. The Social Management Framework (SMF) approach complies with World Bank policies as well as national policies, while considering the practical aspects of the program's implementation. The SMF include the following key points: a. The PIOs will be responsible for screening all of their sub-projects using a check-list to identify possible social impacts related to either land acquisition resettlement or the presence of IPs in the project areas. b. No involuntary land acquisition is anticipated under the project, but if small quantities of land are required for some sub-projects, it may be acquired on a voluntary basis on willing buyer-willing seller terms where permissible. All land transactions will be recorded. Where direct purchase is not permissible/feasible and land acquisition is needed, all consultation and compensation for land transactions will be carried out in accordance with the National Ordinance on Land Acquisitions and Requisition 1982, and World Bank OP 4.12 on Involuntary Resettlement. The RPF provides guidelines to comply with both processes. c. Although no land acquisition or significant displacement of squatters is anticipated under this project (as most projects will take place in rural settings where squatters are not widely found), if the sub-projects trigger land acquisition and /or are expected to impact squatters or economically displace less than fifty families, an Abbreviated Resettlement Plan (RP) will be prepared by the implementing agency using the procedures and guidelines outlined in the SMF. It is expected that the vast majority of sub-projects will fall in this category. PIOs will be responsible for ensuring compliance with Bank policies (incorporated in the Resettlement Policy Framework), including producing the RP and keeping records of actions undertaken. The Bank will sponsor spot-checks through a random audit based on a sample of projects to ensure that adequate safeguards were in place for any sub-projects triggering land acquisition and/or displacement. d. For sub-projects that trigger the displacement of more than fifty families, a brief Social Impact Assessment will also be carried out, based on which a more detailed RP will be produced. The implementing agency may hire specialist consultants to assist the PIO in preparing the RP, which will be cleared by the PD's office. For this batch of projects as well as those involving impacts on Indigenous People (IP) a third-Party audit will be undertaken to assess compliance with the requirements. The PIO will be responsible for carrying out implementation of the RPF. The Bank will review the results of the audits and conduct spot-checks on a sample basis. e. During the periods of the employment generation scheme implementation, a "mobile M&E" team will be employed to support capacity building and assist with implementation of safeguards mitigation plans. The M&E team can be staffed with project personnel and/or external consultants as needed. 79. The SMF indicates what actions are needed and what processes required to implement the safeguards approach described above. It includes (i) the key principles to follow during implementation; (ii) a check-list to identify the possible impact of land acquisition, resettlement or the presence of IPs in project areas; (iii) guidance on how to implement the approach in proportion to the nature and the scale of the impact; and (iv) a detailed process for audits. 19 80. The approach suggested above builds on the best practice of the Bangladesh Local Government Support Project (LGSP) and the Bangladesh Social Investment Program Project (SIPP). The SMF will be disclosed on the website of the DMRD after the requisite clearances are obtained. E. Environment 81. The majority of the EGPP's approximately 37,000 sub-projects will be earthworks such as: (i) canal (re-) excavating, (ii) earth dam (re-) construction, (iii) rural road (re-) construction, (iv) drain (re-) construction, (v) land-fill for community institutions like schools, graveyards, prayer grounds, etc., (vi) earthen shelters to protect animals against cyclones, (vii) excavation of public ponds or fish farms, (viii) organic fertilizer production; (ix) further development of markets or helipads; and (x) (re-) construction of water reservoirs. 82. Environmental issues that could arise in these sub-projects are likely to be minor. But to avoid adverse environmental impacts and enhance environmental outcomes of the individual sub-projects, the World Bank Operational Policy on Environmental Assessment (OP 4.01) has been triggered for this project. The project has been identified as Category `B' since the nature of sub-projects expected to be implemented under the project may lead to negative environmental impacts. Sub-project may also have an adverse environmental impact if they are not properly designed or executed, or mitigation measures are not implemented. The implementing agency already has an Environmental Management Framework (EMF), which provides: screening tools to identify potential environmental risks in sub-projects; procedures to mitigate potentially adverse environmental impact (if any), and a monitoring mechanism. The EMF will be disclosed on the website of the DMRD after the requisite clearances are obtained. 83. All sub-projects will be subject to an environmental screening by the Union Parishad Project Implementation Committee (PIC) during the project selection process. The upazila Field Supervisor will review the environmental screening, suggest any required mitigation measures and prevent activities with significant negative impact from being implemented. As part of the monitoring of the sub-project implementation, the Project Implementation Officer (PIO) of each upazila will monitor compliance on environmental issues. The PIO at the upazila level and the Deputy Director at Office of the Project Director (OPD) will maintain a database for sub-project specific environmental screening, and mitigation measures. The project's M&E system will capture that information as well. In addition, the project's independent assessment/third party assessments (conducted by a reputed local think-tank) will include a brief environmental audit to assess and evaluate the quality of environmental compliance of the sub- projects. F. Safeguard policies Safeguard Policies Triggered by the Project Yes No Environmental Assessment (OP/BP 4.01) [X] [ ] Natural Habitats (OP/BP 4.04) [] [X] Pest Management (OP 4.09) [] [X] Cultural Property (OPN 11.03, being revised as OP 4.11) [] [X] Involuntary Resettlement (OP/BP 4.12) [X] [] Indigenous Peoples (OP/BP 4.10) [X] [] Forests (OP/BP 4.36) [] [X] Safety of Dams (OP/BP 4.37) [] [X] Projects in Disputed Areas (OP/BP 7.60)* [] [X] Projects on International Waterways (OP/BP 7.50) [] [X] * By supporting the proposed project, the Bank does not intend to prejudice the final determination of the parties' claims on the disputed areas 20 G. Policy Exceptions and Readiness 84. No policy exceptions are sought. 85. Readiness: The GOB has shown a high level of commitment to the program as evidenced by the continued budget allocation in the current fiscal year. In addition, MOFDM has worked with the IDA team to rapidly prepare the program. The MOFDM's Technical Project Proposal, which conforms to the OSC of this program, has been approved in principle by the GOB. As the program operates on the basis of the institutional framework of an ongoing government program, the program structure is in place. For the additional positions of Project Director, Deputy Project Director and core PIU staff, TORs have been formulated and the recruitment process had started at the time the PAD was being written. This process will be completed before the program becomes effective. At that moment the recruitment of the 334 Field Supervisors will be ongoing. 21 Annex 1: Sector and Program Background Bangladesh - Employment Generation Program for the Poorest Project Poverty and the Outreach of Safety Net Programs 1. Bangladesh has seen impressive decline in poverty over the last twenty years, but a large number of households still live below the poverty line. That the poor have benefited from Bangladesh's impressive growth during the 1990s is clear from the fact that the national poverty rate declined at the rate of roughly one percentage point per annum over the last 15 years. However, 40 percent of the population using the upper poverty line, still lives in poverty (HIES 2005). This is extremely high by international standards and plays out in a number of adverse human development outcomes. The poor are less educated, and have lower school attendance rates than the non-poor. Only 40% of household heads among the poor have received more than 5 years of schooling, and only 6% have had more than 10 years of education (HIES). Literacy rates for workers aged 15 and older are only 52% among poor male workers and 33% for female workers. Aside from education, the poor also have worse health outcomes. The Demographic and Health Survey (DHS) (2004) finds that, on average, 16.7% of poor children are severely stunted. 2. The obduracy of Bangladesh's poverty is deepened by chronic natural disasters which undermine both economic gains and the resilience of poor households. Bangladesh is often cited as the most disaster prone country in the world, a situation that will likely be exacerbated through the impact of climate change. In parts of the country that endured the main brunt of the floods of July-September 2007 and the cyclone Sidr in November 2007, the hike in food and fuel prices meant a further exacerbation of the adverse impact of such natural shocks on their lives and incomes. In 2009, cyclone Aila affected the southern coast and as recently as May 2010, the exact trajectory of cyclone Laila was unclear but it was expected to wreak some havoc on the south-western coast of Bangladesh. Areas in the north are less prone to the vagaries of cyclones, but experience frequent floods from numerous rivers. Ecologically fragile areas along the riverbeds of the north are also notoriously prone to seasonal monga, when household food stocks dip to starvation levels. Therefore, natural disasters and the unpredictability of the climate have become a way of life in Bangladesh's poorest and ecologically fragile areas. While the Government of Bangladesh has been very proactive in relief efforts, the long-term impact of these disasters on the poorest households is known to be large. This needs effective safety nets on the one hand, and the ability to graduate out of extreme poverty on the other. 3. Lessons from the last food crisis of 2007-08 display the intense vulnerability of Bangladesh's poor households to shocks. The sheer size of the population living around the poverty line3 implies that a small shock can push a large number of individuals into poverty, and many who are already poor, into extreme poverty. High levels of food inflation in 2007-08 eroded real incomes of the poor much faster than that of others because of the higher share of food in their consumption basket. A recent World Bank survey suggests that a majority of households reduced their food and non-food intakes as well as switched to inferior quality food during the crisis. Roughly eight percent of households with school going children took their children out of school: most of these children took up jobs. Further, the number of households who skipped a meal or ate less sometimes or even more frequently in a month rose from 43 percent to 58 percent between December 2007 and March 2008. This shows that shocks ­ both economic and in the form of natural disasters can derail the improvements in nutrition and education outcomes. 3 As reflected by the distribution of consumption in HIES 2005. See Bangladesh 2008 Poverty Assessment for detailed analysis. 22 4. Bangladesh's poverty profile is further complicated by the development objective of employing a large and growing labor force. Bangladesh's labor force is growing rapidly and while open unemployment rates are low, there is serious under-employment, especially among young men. Female employment levels in Bangladesh have remained low despite growing education attainment for women and low levels of fertility. Since self-employment is the default employment type, a very small proportion of all workers are employed in waged occupations, women are especially less likely to receive wages. Most women are employed in household enterprises as unpaid workers. Even poor women have very low likelihoods of being employed for wage, and do not find employment even in casual work (World Bank, 2008). As a result they seem to swell the ranks of domestic workers in the homes of richer households. Each successive government has realized the importance of the employment agenda and has been under political pressure to come up with policies and programs that both protect vulnerable workers and promote employment. 5. In response to these complex factors, the government in Bangladesh implements a wide range of safety net programs targeted to the poor. This is in keeping with Bangladesh's successive national strategies. Both the previous PRS and the current draft NSAPR II focus on the importance of the social protection agenda as a whole but particularly on providing employment to the poorest households and regions, so as to prevent the most vulnerable from falling into extreme poverty. In keeping with this national vision, the government has a range of cash and in kind (or food) programs. This is reflected in the following broad categories of safety net programs: (i) labor intensive public works programs Table 1.1: Households participating in at least one where creation of infrastructure is a secondary safety net program (%) objective; (ii) livelihood programs for income generation, training to groups and awareness building Quintiles Non- Targeted Pension Total regarding health, nutrition and legal rights; (iii) Targeted education and health programs that provide Lowest 2.3 15.7 3.6 21.6 nd incentives for children's education and for seeking 2 2.7 10.6 2.2 15.4 maternal health care. These incentives could be in 3rd 3.3 7.9 2.2 13.4 the form of food or cash; (iv) relief programs th 4 2.3 5.3 2.2 9.8 designed to mitigate the consequences of disasters; th and (v) programs and services for disadvantaged 5 1.6 2.2 0.6 4.4 groups like the elderly, the widowed, disabled, and Total 2.4 8.1 2.1 12.6 freedom fighters. The larger programs include the Bottom 2.4 16.0 4.6 23.1 Vulnerable Group Feeding (VGF) program which has 10% the highest coverage, followed by Old Age Pension, Source: HIES 2005 in Ahmed (2007). Vulnerable Group Development (VGD) and Test Relief (TR) programs. 6. Despite the plethora of programs, safety net programs have low coverage and perhaps due to the plethora, there are inefficiencies in targeting. Despite impressive gains in poverty reduction in Bangladesh in recent years (the incidence of poverty fell from 57 percent in 1991-92 to 49 percent in 2000 and to 40 percent in 2005), the number of extreme poor in Bangladesh still remain at a staggering 35 million. The latest extreme poverty rate is calculated at 25 percent nationally, and 14.6 percent in the urban areas (forthcoming Bangladesh Poverty Assessment). In response to its extreme poverty conditions and to mitigate the risk of households falling into (or further into) poverty as a result of shocks, Bangladesh implements a wide range of government social safety net programs targeted to the most vulnerable and extreme poor sections of the population. However, these programs operate mostly in the rural areas. Coverage rate among rural households is 15 percent compared to only 5 percent among urban households (Ahmed, 2007).4 They have poor coverage when it comes to serving extreme poor 4 Ahmed, S. (2007) "Social Safety Nets in Bangladesh". Draft Paper. The World Bank. 23 households. About 13 percent of all households benefit from at least one safety net program Targeted programs benefit only around 8 percent of all households (Table 1.1). The larger programs include Vulnerable Group Feeding (VGF) program which has the highest coverage among all programs, followed by old age pensions, Vulnerable Group Development (VGD) and Test Relief (TR) that are located in rural areas. 7. Before the 100 Day Employment Program and the current EGPP were introduced, Bangladesh had very low coverage of safety nets programs, despite a plethora of such initiatives. In 2005, about 22 percent of households in the lowest consumption quintile received benefits from safety net programs. Even among the bottom 10 percent of the population, the combined coverage of all safety net programs is just 23 percent, and for targeted programs it is only 16 percent. There is also an urban-rural imbalance in terms of safety net coverage: 15 percent of rural households report being a member of at least one safety net program compared to only 5 percent among urban households. However, the percentage of households who benefit from targeted programs decline progressively for higher quintiles. While such progressive incidence of coverage of benefits is a positive feature, a strong area of concern is the considerable level of inclusion errors across programs. These numbers would have changed in fragile areas with the introduction of the public works programs mentioned above, but coverage would not have expanded very much in the aggregate. Table 1.2: Distribution of beneficiaries across consumption groups Lowest 2nd 3rd 4th Top Program Quintile Quintile Quintile Quintile Quintile VGD 31.7 29.1 19.4 14.3 5.5 TR 38.9 22.2 18.9 13.3 6.7 VGF 36.1 25.0 20.7 13.0 5.2 Old Age Pension 31.9 20.0 21.1 20.5 6.5 Total (targeted) 34.2 24.6 20.0 15.3 6.0 Total (non-targeted) 30.9 23.8 21.6 16.2 7.5 Source: HIES 2005 in Ahmed, 2007. 8. In spite of overall coverage being progressive, a sizeable share of the benefits goes to the non- poor. The targeting outcomes, however, differ quite considerably across programs (Table 1.2). For example, 48 percent of beneficiaries of old age pensions are in the top 3 quintiles compared with 39 percent of TR, VGF and VGD. Further, 41 percent of the benefits provided by targeted programs go to the top 3 quintiles. The same group receives 45 percent of the benefits of the non-targeted programs. This distribution of beneficiaries across consumption groups suggests that means tests based targeting mechanisms currently employed by the majority of the safety net programs do not achieve much efficiency gains over untargeted programs. 9. Regional program coverage is not correlated with regional poverty. The coverage of safety net programs varies significantly by division and is not well correlated with divisional level poverty rates (Table 6). For example, Sylhet has a poverty rate much lower than the national rate but nevertheless has the highest coverage of safety nets among all divisions. In contrast Khulna, which has the second-highest poverty rate in the country, has the least coverage of safety nets. Low coverage among the total population of the relatively poorer districts also translates to low coverage among the poorest. Around 41 and 28 percent of households from the poorest decile participate in safety net programs in Sylhet and Chittagong respectively, compared to 15 percent in Barisal and Khulna (Ahmed, 2007). 24 10. Targeting inefficiencies are also associated with means tests. The high leakage rates of certain programs appear to be in part due to weaknesses in targeting mechanisms. First, program allocations do not take into account the geographic variation in poverty across the country. Instead, the general targeting strategy involves an initial guideline prepared by the implementing ministry, which sets the targeting criteria, the total number of beneficiaries, the type of beneficiaries (including caps on male and female beneficiaries) per union, and the amount of transfer per beneficiary. The selection criteria usually include income level, asset and household structure, and demographic features. Based on the criteria, union parishads,5 in consultation with other local agencies and the local community, identify beneficiaries. Table 1.3: Poverty Headcount and the Distribution of Safety Net Beneficiary Households (%) Poverty Headcount Distribution of Beneficiary Households Division 2005 2005 2005 2000 2005 Total Rural Urban Dhaka 46.7 32.0 14.27 19.98 4.94 Barisal 53.1 52.0 13.34 14.79 5.00 Chittagong 45.7 34.0 11.06 12.89 5.72 Khulna 45.1 45.7 9.58 11.03 4.23 Rajshahi 56.7 51.2 12.07 13.02 6.71 Sylhet 42.4 33.8 22.42 24.31 11.25 National 48.9 40.0 13.02 15.54 5.45 Source: HIES 2000 and 2005 in Ahmed, 2007. 11. Similar programs sometimes use different criteria for targeting benefits, and these are not applied universally. Some programs such as VGD, VGF and Old Age Allowance target similar low income groups but use different sets of criteria. Surveys show that selected individuals rarely fulfill all the criteria for a specific program.6 For example, for the Income Generation for Vulnerable Group Development (IGVGD) program, an overwhelming proportion of households do not fulfill the criteria related to monthly household income and occupational category. While 95 percent of IGVGD households fulfill at least 2 of the 6 principal criteria for selection, 39 percent fulfill at least 3 criteria, 23 percent fulfill 4 or more criteria, and only 2.5 percent fulfill all the criteria. Furthermore, some of the criteria used for targeting are not very useful. Using household income, for example, as is widely known, is problematic since income is difficult to measure or verify, particularly for households engaged in the informal sector. The various safety net programs currently run by the Government of Bangladesh appear to face problems that are generally associated with means testing in developing countries.7 Given the inaccuracy of simple means tests and the administrative difficulties associated with them, the idea of using observable and verifiable household characteristics as proxies for income is attractive. 12. Several ministries are involved in the administering safety net programs, many of which have overlapping objectives and target similar beneficiaries. As a consequence, many households may access benefits from multiple programs, while some may not be able to participate in even a single program suggesting the need for better targeting and coordination to increase coverage. The array of social safety net programs and targeting criteria merits rethinking of the institutional delivery mechanisms in order to use the country's scarce administrative capacities in an efficient and streamlined way. 5 The lowest unit of administration in the local government 6 Ahmed, S. (2005) Delivery Mechanism of Cash Transfer Programs to the Poor in Bangladesh. Social Protection Discussion Paper 0520. The World Bank. 7 See Grosh and Baker (1995) for a discussion on the problems associated with means tests. 25 13. The Government also recognizes the need to ensure adequate funding for social protection and improve the administration of safety net programs. The Government's commitment to social protection is evident in the increase of social safety net expenditures. Table 1.4 provides a summary of major safety net programs and expenditures over time. Expenditures on these programs, stagnant till the mid-1990s, have increased steadily since, as economic growth has provided the fiscal space for re-distributing the benefits of growth to the poor. Expenditure rose to 1.21 percent of GDP in 2006-07, and to 1.60 percent of GDP (or 10.58 percent of the total budget) in 2007 (Ahmed, 2007). Bangladesh has not only increased budgetary expenditures on these programs but also attempted to improve their functioning by shifting from food (in-kind) to cash and by introducing conditionality in their in-kind program - roughly one percent of GDP. Over the past couple of decades, the ability to cope with "flood-induced" vulnerability has improved and the country has been able to manage the national food security situation well (del Ninno, Dorosh and Subbarao, 2007). The above responses have resulted in a significant increase in safety nets spending to 2.8 percent of GDP in FY09 compared to 0.4 percent of GDP in FY02 (see Table 1.4). However, projected high global food prices in the medium term imply such increased spending on safety nets may have to be sustained for some years to come. Table 1.4: Social Safety Net Programs with Incremental Financial Allocations (Million Taka in Real Terms) Year 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 Food for Work (FFW) and 8,060 8,682 6,728 4,123 2,047 4,040 2,197 2,638 2,609 10,581 other workfare Gratuitous Relief (GR) & 2,280 2,315 2,319 2,201 1,896 1,623 2,687 2,602 3,019 5,586 Test Relief (TR) Vulnerable Group 2,720 1,952 2,367 1,865 1,751 1,736 2,447 2,580 2,859 5,611 Development (VGD) Vulnerable Group Feeding 2,290 2,914 1,250 992 1,490 811 1,883 3,040 5,643 4,749 (VGF) Allowance for Widow 250 245 239 370 777 964 1,022 1,102 1,522 1,811 Honorarium for Freedom 150 147 275 263 155 175 318 424 503 579 Fighters Old Age Allowance 490 490 477 686 1,555 2,109 2,385 2,712 3,136 4,024 Primary Education Stipend 0 1,099 954 5,485 3,749 4,219 3,356 3,305 3,270 3,139 Project (PESP) Total 16,240 17,844 14,609 15,986 13,422 15,677 16,294 18,402 22,562 36,081 Percent Increase in real 9.9 -18.1 9.4 -16.0 16.8 3.9 12.9 22.6 59.9 terms Source: World Bank, 2008. Public Works Programs 14. Bangladesh has longstanding experience in implementing public works programs. While prototypes of the present-day public works programs were first introduced in the pre-British era, they solidified into larger welfare programs during times of famine. Detailed "Famine Codes" laid out the targeting and implementation criteria and civil servants at the field level were historically trained in the implementation of such programs. However, in recent years, these programs have evolved and Bangladesh has been keen to apply the lessons of international best practice to the design of its new initiatives. 15. The Food for Work (FFW) and Test Relief (TR) are two public works programs which pursue a two-fold objective of (a) mitigating the adverse impact of seasonal decline in labor demand on food security of poor households, and (b) improving community infrastructure. Both FFW and TR offer employment during the slack season in exchange for food wages (wheat or rice). FFW, in operation 26 since 1975, involves the rural poor in infrastructure project, e.g. construction and maintenance of irrigation systems, roads, river embankments etc. Majority of the FFW projects are implemented primarily during the dry season. The selection of eligible projects is made by the Upazila Parishad and approved by the District Coordination committee. The program targets landless day laborers and temporary workers lacking productive assets and earning under TK 300 per month; households headed by widowed, deserted, and destitute women are also targeted. The TR program is administered under the Rural Infrastructure Maintenance Program. In contrast to the FFW program, TR program is promotes employment in smaller scale projects with lighter labor requirement. Majority TR projects are carried out during the rainy season and address immediate infrastructure needs, such as repair of roads and village facilities. Rather than targeting households or individuals, TR targets locations. Payment for both FFW and TR is in the form of foodgrains. 16. While the scope of these two programs is nation-wide, they cover a rather small proportion of the poor. In 2005 FFW covered 1 million beneficiaries, and TR covered approximately 100,000 persons. In addition, inclusion and exclusion errors may further limit their effectiveness in reaching the poor, although the data on these issues are limited; however, a 2003 assessment of the FFW program reported a 26% inclusion error.8 17. In response to the international food and fuel crisis, the GOB launched the largest public works program called the 100 Day Employment Program, in 2008. Managed by the Ministry of Food and Disaster Management in partnership with local governments, the program was intended to cushion the poor against the rise in food prices. The program design drew on Bangladesh's successful experience with workfare and was a major expansion of the existing cash-for-work model with some minor adjustments. The program was designed to run during the lean season in rural areas only. The FY09 budget provided for 100 days of work for up to 2 million people with an initial allocation of approximately US$300 million. There were three independent assessments of this program including one by the Word Bank. In addition, the Government undertook its own assessment. The results of the assessments were encouraging, showing reasonably good targeting and efficient delivery of the program. The assessments also identified areas where the program could be improved upon. 18. Close on the heels of the 100 Day Employment Program, the GOB launched the Employment Generation Program for the Poorest (EGPP). The EGPP is intended to run during two lean season phases ­ for 60 days from September to November and for 40 days from March to April. This program, to which the current IDA support is proposed, has built on the lessons of the 100 Day Employment Program, especially on issues of targeting, accountability and transparency. This first phase of year one (FY 2009-10) also coincided with Cyclone Aila, which hit the south west in May 2009 and led the GOB to take the decision to roll out the program first in 16 districts that were affected by the cyclone or were traditionally monga areas. In the first phase (which began in September 2009), the program created 16,555,325 person days of employment. The second phase (from March 2009) was concluding just as this PAD document was in its final stages. Lessons from the design and implementation of the first phase have informed the IDA support to the program. The involvement of the Bank team in the assessment of the first phase also has led to continuous updates to the implementation guidelines. 19. In FY10-11, the GOB has allocated Tk. 10 billion to the program. This includes the expected IDA contribution and also represents a 15% decrease over last year's allocation. This decrease reflects 8 World Bank (2005) Social safety nets in Bangladesh: An assessment. Report No. 33411-BD; Kumar, Pravesh. 2010. "Institutional Capacity and Training Assessment of the Disaster Management and Relief Division (DMRD)" (Draft). 27 the fact that last year the program was implemented in only 16 districts, and while it is expected from this current year to extend nationwide, yet the MOF is realistic about the implementation constraints. Box 1.1: Main recommendations from independent assessments of the 100 Day Employment Program Targeting and beneficiary selection: Increased emphasis on self-targeting through the wage rate Introduce a minimum quota for women's participation Provide drinking water and crèche facilities to make participation more attractive for women Raise the maximum age of participation from 50 to 65 to better dovetail employment generation programs with the old age allowance program (which starts at age 65) Sub-projects: Increase community consultation in the selection of sub-projects Conduct ex ante specialized technical assessments of the feasibility and sustainability of sub- projects Integrate selection of sub-projects with the local development planning process Ensure sub-projects address environmental concerns Accountability: Improve awareness of the program through IEC activities Make program guidelines more accessible and available to citizens and officials at the local level Adjust administrative costs to align with the size of the program in an upazila or union (rather than a flat rate for each upazila or union) Provide additional human resources at the local level for administration of the program Institute a systematic grievance recording mechanism for addressing complaints Payments: Develop an electronic beneficiary database to facilitate transfers Make payments through banks or post offices Take advantage of technological innovations in making payments (mobile phones?) Monitoring and evaluation: Improve the monitoring information collected by the Government Enhance the effectiveness of Government monitoring processes Strengthen the evidence base for the program through a rigorous impact evaluation Sources: BRAC and NFPCSP. 2008. Study on the first phase of the 100-Day Employment Generation Programme (EGP). Khatun, F., K.M. Rahman, and A. Nabi. 100-Day Employment Generation Program in Bangladesh: Challenges of Effective Implementation. Paper presented at the Center for Policy Dialogue (CDP) Conference on "Development with Equity and Justice: Immediate Tasks for the Newly Elected Government," March 28-29, 2009, Dhaka. World Bank, 2009. An Assessment of the First Phase of the 100-Day Employment Generation Program in Bangladesh. Policy Note, SASEP and SASHD. 28 Wages for manual labor in rural Bangladesh 20. During the design of this program, the Bank team triangulated data from the Monthly Statistical Bulletin of the Bangladesh Bureau of Statistics (BBS) with a telephone survey of randomly selected UP Chairmen in March 2010. Forty-two chairmen in all 7 divisions of Bangladesh, with 6 chairmen in each division were interviewed. 21. UP Chairmen's Survey: UP chairmen reported ­ when possible ­ information regarding three categories of wages: agricultural, non-agricultural unskilled (earth and field work), and skilled non- agricultural (construction work, which also included carpentry, and handicrafts) for men and women. These wages are reported for 1 shift per day, with a shift lasting approximately 5 hours (e.g., from 8 AM to 1 PM). In most cases they reported a range within which the daily market wage rate falls. The team considered the lower boundaries of the UP-assessed range, since the rule of thumb in setting the benefit level in employment generation schemes is to ensure that the EGP wage is below the market wage rate. 22. BBS Statistical Monthly Bulletin: The BBS dataset contains monthly average wages for 22 regions in 5 divisions from October of 2008 through May of 2009. The survey only covers agricultural workers and reports their wage in two ways: wage without food and wage with food. The latter type of remuneration for work consists in providing meals to the workers. Food wage tends to be lower. This report looks only at wages without food thus fully reflecting the monetary value of agricultural work. Key Results For all the surveyed unions, men earned a mean salary Tk. 150 per day in agricultural employment and women earned Tk. 116; in earth work, men earned Tk. 175, while women's rate averaged Tk. 130 per day.9 In construction, men earned Tk. 225 and women earned Tk. 160. The higher wage rate in construction work reflects a greater skill level of the laborers. Women tend to participate in wage work less frequently than men, especially in construction. One-third of UP chairmen reported that either none of hardly any women worked in agriculture or non-agricultural employment; especially in construction. The wage disparity between men and women was 23 percent in agriculture, meaning that women's mean wage was just three quarters of men's mean wage. In earth work women earned 25 percent less than men, and in construction they earned 29 percent less. These figures are broadly consistent with the national average gender gap in agricultural employment reported in the MBS. Across unions, the magnitude of the gender differential in agricultural and earthwork wages varied considerably from 60% (i.e., women's wage was only 40 percent of men's wage) to 0 (no difference between men's and women's rates). In construction, the gender gap ranged from 64 percent of men's wage to no difference at all. At the division level, mean men's wages in agriculture were highest in Chittagong (Tk. 192) and Dhaka (Tk. 183) and lowest in Khulna (Tk. 112) and Rajshahi (Tk. 127). The agricultural wage rates reported by the UP chairmen are overall consistent with the wages reported in the BBS survey. Specifically, according to the Bangladesh Bureau of Statistics, the mean agricultural wage rate for men was Tk. 150. Divisional mean wages for men in agricultural employment reported differ from the UP Chairmen's Survey, which is due to the limited representativeness of the latter on the one hand, and a 10 month gap between the time the UP Chairmen Survey was carried out (March of 2010) and the period covered by the most recent available month in the BBS data (May of 2009). In 9 The rates are per 1 shift which is understood by most chairmen to equal 5 hours (e.g. from 8 am to 1 pm) 29 light of these limitations, the wage numbers reported by the UP chairmen and the BBS should be considered roughly comparable. There is considerable UP-level variation: men's agricultural wage ranges from 100 to 250,10 the wages in earth work vary from 100 to 300,11 and in construction range was from 120 to 300.12 In communities where women work in construction, they tend to work as help, in some communities receiving less pay then they would doing earth work. Even in construction work, which is associated with greater skill levels and greater pay, in many communities women employed in construction were hired as helpers, receiving less pay then they would doing earth work. Further, in two communities the UP chairmen indicated that the prospect of employments with the EGP would be more attractive to tribal women. There is great variation in women's wages across UPs. The lowest paying UPs are Sakhipur and Gumaniganj, where female agricultural workers earn Tk. 60, and the highest paying UPs are Gauripur (Tk. 180) and Haluaghat (250). In addition to geographical variations, there are also seasonal variations. Thirty six out of 38 well-informed UP chairmen reported seasonal swings in wages. In most cases the wages went up during the harvest and plantation time and declined during the monsoon season. Men's13 wage increased from the low-season to the high season on average 81 percent,14 although it varied considerably from a small 15% increase in Dakshin Paschim Baniachong in Sylhet division to a dramatic 200% increase in Tangbhanga, Rangpur. Figure 1.1: Correspondence between UP chairmen-reported and BBS-reported men's wages in agricultural employment, at division level. 250 192 183 200 183 165 167 153 144 150 133 127 112 100 50 0 Barisal Chittagong Dhaka Khulna Rajshahi BBS UP Chairmen Survey Notes: (1) The other two divisions ­ Rangpur and Sylhet were not reported in the BBS data. Source: UP Chairmen Survey and Monthly Statistical Bulletin, Bangladesh. 10 TK .100 in Asharkandi, Dhamoirhat, Goga, Koijuri, Noapara, Pirojpur, Sakhipur, Ulania and TK 250 in Haluaghat, Jhauail, Rowangchhari. 11 TK. 100 in Haripur, Noapara, and Sakhipur and TK 300 in Sourjomoni and Tangbhanga. 12 TK. 100 in Koijuri and TK 300 in Chhaygaon, Gauripur, Mahilara, Rowangchhari, Singhhchapair, and Sourjomoni. 13 Women's wage was available for a handful of unions only. 14 Among 16 communities, for which both the low season and the high season rates were reported. 30 Annex 2: Major Related Projects Financed by the Bank and/or other Agencies Bangladesh - Employment Generation Program for the Poorest Project Sector Issue Project Name and cost Latest Supervision Rating Bank-financed Projects (PSR/ISR) Implementation Development Progress (IP) Objectives (DO) Promoting decentralization and strengthening MS MS Local Governance the lowest tiers of government while focusing Support Project (LGSP) on transparency and accountability $111.50 mil. Promoting "pro poor" economic growth by Northern Areas Under preparation Under providing trained labor to the readymade Reduction of Poverty preparation garment sector focusing on vulnerable Initiative (NARI) women from the poorest region of Bangladesh. To develop effective and efficient financing Social Investment MS S and institutional arrangements for improving Program Project (SIPP) access to local infrastructure and basic $101.2 mil services through the implementation of community-driven small scale infrastructure works and social assistance programs. Other Development Agencies Project Name and cost Description Remarks USAID The REAL Project Support to the cyclone (Reconstruction Sidr affected population Economic Development towards alternative and Livelihoods) sustainable income $3.5 million generation opportunities such as cash for work and homestead gardening, value added agricultural processing techniques, etc. Nutritional Support to Emergency response Mothers and Children in activity for nutritional the Sidr Affected rehabilitation in the regions, cost $100,000. cyclone Sidr-impacted coastal areas of Barisal Division. DFID Char Livelihood Support to people in the Program islands (chars) through 50 mil. Pounds provision of livestock, seeds and other items World Food Program Vulnerable Group Support to most Development vulnerable women by Programme providing them food assistance and training for alternative livelihood. 31 Other Development Agencies Project Name and cost Description Remarks Food for Education Food provision conditional on school attendance Cash for Work (CFW) Targeting the extreme poor in 2 districts affected by cyclone Sidr European Commission Rural Employment The objectives are to Opportunities for Public increase income of Assets (REOPA) destitute women by Euro 20, 700,000 providing employment, improving public assets for the benefit of rural communities and strengthening local government institutions for better response to pro poor growth. Asian Development Bank Emergency Disaster Rehabilitation of Damage Rehabilitation economic and social (Sector) Project activity in the 51 districts US$120.00 million for 25 million affected people affected by the 2007 cyclone. Second Chittagong Hill The objective will be to Tracts Rural (i) improve rural Development Project infrastructure and (CHTRDP II) sustainable natural Bangladesh Japan resources management Special Fund and monitoring (ii) US$900,000/ Water increase rural incomes Financing Facility- and include rural women Multidonor TF from remote areas and US$265,000 disadvantaged groups through increased economic opportunities and (iii) support institutional strengthening of the Ministry of Chittagong Hill Tracts and relevant institutions to plan, implement and monitor sustainable rural development in Chittagong Hill Tract. 32 Annex 3: Results Framework and Monitoring Bangladesh - Employment Generation Program for the Poorest Project Institutional arrangements and financing for results monitoring 1. Results monitoring for the project would utilize the strengthened program monitoring system in combination with independent assessments, including both independent third party validations and a program evaluation. 2. Under focus area three, the project would assist the implementing Ministry in refining their monitoring strategy for the EGPP and would provide support to develop the monitoring capacity of the MOFDM at the upazila level and at the Center. The project would support the development of a set of standardized summary indicators and monitoring reports and would provide financing for critical staffing at the central and local levels and investments in MIS capacity. This technical assistance will be provided through the Operational Support Component (OSC) of the program and through Bank supervision missions 3. The reforms to the program MIS would be supplemented with external independent third party validations. The World Bank would partner with a reputable local think-tank to conduct independent third party validations of program implementation and record-keeping at a sample of union parishads during each phase of program implementation. Trust fund resources are currently being sought to finance the independent third party validations. 4. The evaluation strategy for the EGPP aims to provide information on the effectiveness of targeting and the impact of the program on household net incomes and coping strategies in response to shocks. In order to assess the impact of the program for participants, the World Bank would partner with a reputable local data collection firm to undertake baseline and follow-up surveys of program participants and non-participants in a representative sample of upazilas. In addition, qualitative work would be commissioned to examine impacts on outcomes such as social empowerment and leakages/corruption. The evaluations will not be funded out of the program OSC; trust fund and external resources from development partners are currently being sought to finance the evaluations. Monitoring and evaluation tools and sources 5. A range of M&E tools are expected to be employed to assess EGPP implementation performance, outputs, and outcomes. 6. Program monitoring. As part of the Ministry's regular program monitoring, the monitoring cell will produce a user-friendly biweekly summary report for senior management. Under the OSC, the existing monitoring forms and reports would be redesigned to facilitate monitoring on inputs, outputs, intermediate outcomes and compliance with program guidelines. The set of streamlined indicators to be collected at the sub-project level will include information on the flow of funds, beneficiary profile, wages paid, and project type. Financial indicators to be collected will include data on how much has been disbursed to and paid out in each project and spending on administrative and non-wage project costs. Indicators on beneficiaries will include: how many job cards have been issued in each upazila to men and to women, the characteristics of program beneficiaries (e.g., the proportion of beneficiaries who are manual laborers and who have less than 0.5 acres of land), how many days of work have been created on each project, and the percent of person-days of work by female beneficiaries. Wage indicators will provide the average wages paid per worker per day and, the total wage payments made, and the amount of 31 time elapsed between work and payment. Infrastructure indicators will list the type of projects (e.g., road construction, drainage system maintenance) under implementation. 7. Independent third party validations. External independent validations would be conducted in a random sample of approximately three percent of union parishads each program phase. The sample of union parishads where independent third party validations will take place will be selected randomly by the firm before each program phase. The independent third party validations will examine whether program implementation guidelines are followed and whether program benefits reach the intended beneficiaries. The employees of the firm contracted for independent third party validations will visit the sampled upazilas each program phase and will examining the hard copy (and electronic, if available) documentation on program implementation, beneficiary selection, payments, and other financial arrangements at the community sub-project level. During the independent third party validations, the employees will also conduct focus group discussions or interviews with beneficiaries, other community members, program staff and other union parishad and upazila officials. The information obtained from independent third party validation visits will be verified against program monitoring and financial reports received at the center, and the implementing agency will have the opportunity to cross-check findings at the verification stage. The independent third party validation firm will produce a report of findings each phase (i.e., twice per year) to allow for immediate feedback into the program. 8. Impact evaluation. The evaluation strategy for the EGPP aims to provide information on the effectiveness of targeting and the impact of the program on household net incomes and coping strategies in response to shocks. In order to assess the impact of the program for participants, the World Bank would partner with a reputable local survey firm to undertake baseline and follow-up surveys of program participants and non-participants in a representative sample of upazilas. The surveys will collect information on household demographics, assets, income generating activities, risk coping strategies, and participation in safety net programs. Information on individual and household characteristics will be used to construct, using analytical techniques such as propensity score matching, a comparison group from among the non-participant households who are surveyed who are otherwise similar to participant households. Approximately 3,000 households will be surveyed at the baseline, which is planned for November 2010. A smaller sample of approximately 1,000 households surveyed at baseline will be re- surveyed in November 2011 if resources are available. In November 2012, the larger follow-up survey of 3,000 households, including the 1,000 panel households, will be carried out. This impact evaluation work will be supplemented with analysis of data from nationally representative surveys from Bangladesh. Questions about EGPP participation by household members have been included in the 2010 Household Income and Expenditure Survey (HIES) that is currently in the field. Even if the HIES sample size does not permit analysis of EGPP outcomes, the HIES will be very useful in identifying the key sources of risk, vulnerability and coping mechanisms for poor Bangladeshi households. 9. Qualitative work. Qualitative assessments would also be commissioned to complement the quantitative work. The qualitative work would examine beneficiary and non-beneficiary assessments of how well the program is working, including perceptions of targeting, of leakages/corruption, the quality of works produced under the EGPP, the value of the infrastructure projects (both to participants and to non-participants), and impacts of the program on internal and external migration as well as on social empowerment. This work would take the form of focus group discussions and key informant interviews with EGPP participants and non-participants as well as with local elite and administrators of the EGPP. 32 Results Framework PDO Project Results Indicators Use of Project Results Information The project development objective is to provide - Number of individuals participating in short- - To track the number of beneficiaries of the short-term employment on community sub- term employment created by the program program. projects to enable households to better cope with (male/female) vulnerability, while strengthening Program - Number of person-days of short-term - To monitor the volume of employment implementation. employment generated generated Intermediate Outcomes Intermediate Results Indicators Use of Intermediate Results Information Focus Area 1: Rigorous Targeting Short-term employment is targeted to vulnerable -Percentage of wage costs allocated to the "high - To ensure that funds are flowing to the poorest households in extreme poverty, and especially to poverty density" upazilas with 35% or higher regions women. poverty rates. - Percentage of beneficiaries that are both - To ensure that the extreme poor benefit from engaged in manual labor and own less than 0.5 the program acres of land. - Percentage of EGPP participants that are - To monitor women's participation in the women. program Focus Area 2: Improved Governance through Better Transparency and Accountability Better governance of the program is reflected in - Percentage of sub-projects that have signboards - To ensure transparency in targeting and revised rules that are transparently applied and at worksites and publicly available beneficiary program management publicly disclosed. lists. - Percentage of UPs that have banks where - To make payment systems more efficient and payments are made through banks or formal verifiable. financial channels. - Percentage of sub-projects that are using the - To ensure sub-projects are allocating adequate allocation for non-wage costs. resources for non-wage costs. Focus Area 3: Better Capacity for Monitoring and Evaluation The program monitoring and evaluation system is strengthened to allow timely monitoring of program - Percentage of sub-projects whose progress - To ensure information on the implementation implementation and third-party verification reports were prepared using standardized of sub-projects is provided consistently. (including evaluation of program impacts and monitoring formats in each program phase. independent process documentation). - Annual household surveys conducted for a - To help inform follow-on activities and assess representative sample of households (Y/N) program impacts. - To verify the quality/ validity of M&E - Number of union parishads where independent information and provide information on process documentation activities took place. program implementation. - Number of participating upazilas that have entered beneficiary lists into an electronic - To ensure beneficiaries can be tracked. database 33 Disbursement Linked Indicators (DLIs) for the Main Program Component (MPC) TARGETS FOR DISBURSEMENT VALUE COMMENTS FOCUS Million NO OBJECTIVE BASELINE AREA 2011* 2012 2013 US$ (% of MPC) Program guidelines issued to Actual allocation from Actual allocation from reflect allocation formula: Ministry to upazilas to Ministry to upazilas to 100 Day EGP covered % Upazilas with remain as per program remain as per program Wage costs all 64 districts and 480 program extreme guidelines AND a guidelines AND a allocated and 20.68 Geographi upazilas; resources resources: poverty rates: minimum of 40% of minimum of 45% of 1. utilized according c targeting were not explicitly actual expenditures on actual expenditures on to upazila level 50% 35% and higher (14.29%) targeted to higher wage costs in each phase wage costs in each phase poverty rates. poverty upazilas. 40% 21-34% going to upazilas with going to upazilas with extreme poverty rates extreme poverty rates 35% 10% 20% and lower 35% and higher. and higher. PD's office provides copies of all issued guidelines, office PD's office provides report certifying budget allocation Verification Mechanism: orders, circulars, and budget from Ministry to upazila level for each upazila for each allocation from Ministry to program phase. upazila level. 94% of beneficiaries in 100 Day EGP have less (1) Program guidelines issued If 2011 targets Household than 0.5 acres land; requiring that beneficiaries have been met targeting Use of household 73% of heads of eligible for EGPP participation At least 75% of At least 80% of before project based on targeting criteria beneficiary households be from households owning less beneficiaries in each beneficiaries in each phase 20.68 effectiveness, 2. well- for beneficiary are manual laborers than 0.5 acres of land and phase are both engaged in are both engaged in the 2011 DLI established selection. (Baseline values from engaged in manual labor; (2) manual labor and own manual labor and own less (14.29%) can qualify for poverty WB survey in six Operations Manual reflects less than 0.5 acres of land. than 0.5 acres of land. up-front correlates districts in the Rajshahi program rules laying out financing. division, a monga beneficiary selection criteria. prone area). PD's office provides community sub-project-level PD's office provides copies of program monitoring reports from every program phase all issued guidelines, office Verification Mechanism: that certify percentage of participating beneficiaries orders, and circulars as well as with less than 0.5 acres land and engaged in manual Program operations manual. labor on each sub-project. 34 TARGETS FOR DISBURSEMENT VALUE COMMENTS FOCUS Million NO OBJECTIVE BASELINE AREA 2011* 2012 2013 US$ (% of MPC) If 2011 targets have been met 10% of beneficiaries Program guidelines issued before project Increase At least 30% of EGPP in 100 Day EGP requiring that a minimum At least 33% of EGPP 20.68 effectiveness, Gender women's participants are women 3. were women. 33% of participants are participants are women the 2011 DLI targeting participation in in each phase. (14.29%) can qualify for women. in each phase. program. up-front financing. PD's office provides copies of PD's office provides community sub-project-level all issued guidelines, office Verification Mechanism: program monitoring reports from every program phase orders, and circulars and that certify the number of women beneficiaries. reports of each phase. (a) Program guidelines issued All payments in at least All payments in at least Develop, test, and All payments are cash- and (b) all payments in at least Efficient 85% of UPs which have 95% of UPs which have 20.68 define strategy for based, with significant 80% of UPs which have banks to 4. payment banks to be made through banks to be made through efficient payment fiduciary risk. be made through banks or systems banks or formal financial banks or formal financial (14.29%) mechanisms. formal financial channels in each channels in each phase. channels in each phase. phase. PD's office provides guidelines PD's office provides PD's office provides developed and report certifying report certifying cash report certifying cash Verification Mechanism: cash payments through banks in payments through banks payments through banks in each program phase. in each program phase. each program phase. Electronic beneficiary Electronic beneficiary Efficient Specifications for the beneficiary Implement robust database and back-end database and back-end record- database and back-end system 20.68 and transparent Record-keeping system developed and system fully developed 5. keeping designed and contracting record-keeping systems are manual. tested in those 334 and installed in all 334 mechanis initiated for database (14.29%) procedures. upazilas that receive OSC upazilas that receive OSC ms development. assistance. assistance. PD's office provides report on PD's office provides and manual developed for the PD's office provides report certifying final back Verification Mechanism: database and back-end system database prototype and end database with testing specification. testing report. report. 35 TARGETS FOR DISBURSEMENT VALUE COMMENTS FOCUS Million NO OBJECTIVE BASELINE AREA 2011* 2012 2013 US$ (% of MPC) Program guidelines issued Between 10% and requiring community sub- 20% of community projects to spend between 10% Non-wage costs are 10% Non-wage costs are 10% Provision 20.68 sub-project costs No provision for non- and 20% of the total allocation to 20% of actual to 20% of actual 6. of non- used for material wage costs. on materials and skilled labor, expenditures for each expenditures for each wage costs (14.29%) and skilled labor and to specify flow of funds for phase. phase. inputs. non-wage costs and procurement processes. PD's office provides copies of PD's office certifies non-wage expenditures and total Verification Mechanism: all issued guidelines, office expenditures for each phase. orders and circulars. Adequate personnel Additional Supervision of sub- 90% of Field Supervisor 90% of Field Supervisor 20.68 Selection process completed for 7. for personnel in place projects weak due to positions are filled during positions are filled during all Field Supervisor posts. supervisio at the upazila level. lack of personnel. each phase. each phase. (14.29%) n PD's office provides PD's office certifies presence of Field Supervisors Verification Mechanism: confirmation of appointment during each phase. letters issued for all Field Supervisor posts. *IDA funding for the program is expected to run from January 2011 through December 2013 with the last disbursements occurring in early 2014. 36 Arrangements for results monitoring Target Values Data Collection and Reporting Outcome Indicators Baseline YR1 YR2 YR3 Frequency and Data Collection Responsibility for Reports Instruments Data Collection Number of individuals 1,997,075 600,000 600,000 600,000 Reports on Report from MOFDM PD's participating in short-term beneficiaries beneficiaries each beneficiaries beneficiaries each number of MOFDM office employment created by the registered for the phase each phase phase (assuming registered project (male/female) each implemented phase (assuming same same GoB budget beneficiaries phase of the 100 Day GoB budget allocation as in compiled twice EGP (during food allocation as in year 1) per year at the crisis) year 1) completion of each phase Number of person-days of 90 million person 60 million person 60 million 60 million person Reports on Report from MOFDM PD's short-term employment days in 100 Day days per year person days per days per year person-days MOFDM office created EGP (during food year (assuming (assuming same compiled twice crisis) same GoB GoB budget per year at the budget allocation as in completion of allocation as in year 1) each phase year 1) Results Indicators for Each Focus Area Focus Area One: 100 Day EGP Program guidelines Actual Actual allocation Each phase Expenditure report MOFDM records Percentage of wage costs covered all 64 issued to reflect allocation from from Ministry to (twice per year) from MOFDM and audit results. allocated to the "high poverty districts and 480 allocation formula: Ministry to upazilas to density" upazilas with 35% or upazilas; resources upazila upazilas to remain as per higher poverty rates. were not explicitly extreme remain as per program targeted to higher % poverty program guidelines AND a poverty upazilas funds rate guidelines AND minimum of 45% 50% 35% & a minimum of of actual higher 40% of actual expenditures on expenditures on wage costs in 40% 21-34% wage costs in each phase going each phase to upazilas with 10% 20% & going to extreme poverty below upazilas with rates 35% and extreme poverty higher. rates 35% and and actual allocation higher. from Ministry to upazilas reflects program guidelines. 37 Target Values Data Collection and Reporting Results Indicators for Each Baseline YR1 YR2 YR3 Frequency and Data Collection Responsibility for Focus Area Reports Instruments Data Collection Percentage of households of 94% of At least 60% of At least 75% of At least 80% of Each phase Monitoring reports MOFDM EGPP beneficiaries that are beneficiaries in beneficiaries in each beneficiaries in beneficiaries in (twice per year) from MOFDM, monitoring reports, from households owning less 100 Day EGP have phase are both each phase are each phase are for independent independent third private sector firms than 0.5 acres of land and less than 0.5 acres engaged in manual both engaged in both engaged in third party party validations contracted for engaged in manual labor land labor and own less manual labor and manual labor validations and and household independent third 73% of heads of than 0.5 acres of own less than 0.5 and own less monitoring surveys party validations beneficiary land. acres of land than 0.5 acres of reports from and for surveys households are land MOFDM, manual laborers annually for (Baseline values household from WB survey in surveys six districts in monga prone areas of Rajshahi division) Percentage of EGPP 10% of At least 25% of At least 30% of At least 33% of Each phase MOFDM MOFDM participants that are women beneficiaries in EGPP participants EGPP EGPP (twice per year) monitoring reports, monitoring reports, 100 Day EGP were are women participants are participants are for MOFDM independent third private sector firms women women women monitoring party validations contracted for reports and and household independent third independent surveys party validations third party and for surveys validations, annually for household surveys Focus Area Two: No requirements 50% of sub-projects 60% of worksites 75% of sub- Each phase Independent third Private sector firm Percentage of sub-projects that for signboards or have signboards and sub-projects have projects have (twice per year) party validations contracted for have signboards at worksites publicly available publicly available signboards and signboards and independent third and publicly available beneficiary lists in beneficiary lists publicly available publicly party validations beneficiary lists. guidelines for 100 beneficiary lists available Day EGP beneficiary lists Percentage of EGPP 0% (Payments are Program guidelines Payments in 85% Payments in Each phase Reports from MOFDM PD's participants receiving entirely cash-based issued and payments of UPs which 95% of UPs (twice per year) MOFDM and office payments through formal and are vulnerable in 80% of UPs have banks to be which have banks/formal financial channels. to fiduciary risks.) which have banks to made through banks to be financial be made through banks or formal made through institutions banks or formal financial channels banks or formal financial channels in in each phase financial each phase channels in each phase 38 Target Values Data Collection and Reporting Results Indicators for Each Baseline YR1 YR2 YR3 Frequency and Data Collection Responsibility for Focus Area Reports Instruments Data Collection Number of contracted field No field personnel 90% of Field 90% of Field Each phase Report from MOFDM PD's Selection process supervisors in upazilas with to supervise sub- Supervisor Supervisor (twice per year) MOFDM office completed for all poverty rates of 21% or higher projects positions are positions are Field Supervisor filled during filled during each posts. each phase. phase. Percentage of sub-projects No provision for Program guidelines 50% of sub- 75% of sub- Each phase Report from MOFDM PD's using 10-20% of project non-wage costs issued to specify projects utilize projects utilize (twice per year) MOFDM office resources for non-wage costs that community sub- non-wage costs non-wage costs of projects should of 10% to 20% 10% to 20% of spend between 10% of actual actual and 20% of the total expenditures. expenditures. allocation on materials and skilled labor and to specify flow of funds for non-wage costs and procurement processes Percentage of upazilas where Program guidelines Grievance Grievance redress Each phase Independent third private sector firm grievance redress system is issued to specify redress system system functional (twice per year) party validations contracted for functioning establishment and functional in in 75% of independent third function of 50% of upazilas upazilas party validations grievance redress system Focus Area Three: Upazila-level Community sub- Community sub- Community sub- Each phase MOFDM MOFDM and Percentage of community sub- progress reports project-level project-level project-level (twice per year) monitoring reports private sector firm projects whose progress prepared progress reports progress reports progress reports and independent contracted for reports were prepared using prepared each phase prepared each prepared each third party independent third standardized monitoring for 60% of sub- phase for 75% phase for 90% of validations party validations formats each program phase. projects of sub-projects sub-projects Number of participating No beneficiary MIS specialist All unions that All unions that Each phase MOFDM MOFDM PD's upazilas that have input information recruited and will make will make (twice per year) beneficiary office beneficiary lists into an digitized standard monitoring payments payments through database electronic beneficiary-level formats developed through banks banks will have database for community sub- will have beneficiary lists project-level reports beneficiary lists into an electronic that include land into an beneficiary and manual laborer electronic database at the household targeting beneficiary upazila level criteria. database at the upazila level 39 Annex 4: Detailed Project Description Bangladesh - Employment Generation Program for the Poorest Project 1. The Employment Generation program for the Poorest (EGPP), implemented by the Ministry of Food and Disaster Management (MOFDM), is an ongoing program of the GOB that provides short- term employment to manual workers during the lean season over two phases that adds up to 100 days in a year. The first phase typically runs from September to November providing 60 days of employment and the second phase runs from March to April providing 40 days of employment. Wages are meant to be set below the prevailing market wage for unskilled manual workers and this contributes to self-selection of the poorest households into the program. 2. The project development objective is to provide short-term employment on community sub- projects to enable households to better cope with vulnerability, while strengthening program implementation. In the current financial year (2010-11) of the GOB has allocated BDT 10 billion in its budget. Of this a little over one-third is expected to come through IDA. Overall, if GOB maintains the same level of allocation, we should expect the program to have the capacity of reaching 2 million individuals and creating 200 million person days of employment over the three years of the project. The reason why we say "capacity" is because if there is very rapid decline in poverty, we can expect fewer individuals to avail of this safety net, or if there are other global or local shocks we could expect an increase. Performance will be measured by the following outcome indicators: Short-term employment targeted to vulnerable households in extreme poverty, and especially to women Better governance of the program reflected in revised rules that are transparently applied and publicly disclosed Program monitoring and evaluation system strengthened to allow timely monitoring of program implementation and third-party verification. 3. Support to the EGPP is envisaged as credit from the International Development Association (IDA) through a programmatic Specific Investment Loan (SIL). Disbursements will be linked to a set of outcome indicators that are laid out in Annex 3. This mechanism allows for innovations to take place, in addition to rewarding efficiency. In the recent country discourse after the announcement of the Budget of 2010-11, there has been a great deal of discussion on the implementation capacity for all development programs. IDA will also support an operational support component that will provide technical support on design and implementation and the realization of results. 4. IDA support to EGPP will be in the form of financing wage costs to a maximum of US $144.75 million, but not exceeding 70 percent of wages paid, over the three-year period (2011-2013). Disbursements are linked to results which aim at improving efficiency, accountability and transparency. There is also provision for retroactive financing if up to four specified DLIs are achieved before effectiveness. However, if any amounts remain undisbursed due to the non- achievement of DLIs the Bank may authorize an additional disbursement, after the final disbursement for based on 2013 results. This would be equivalent to the sum of the undisbursed amounts or US$20.00 million, whichever is lower, on meeting the 2013 targets for the DLI for efficient payment systems and the total disbursement does not cross 70 percent of wage costs. In addition, US$5.25 million will be earmarked for technical and operational support through a separate window (Component 2 below). This technical support will go towards capacity enhancement to enable the GOB to achieve the results have been agreed upon. Project components 5. IDA support to the EGPP will flow through two components: 40 The Main Program Component (MPC) would support expansion of and reforms to the existing program through three Focus Areas. The Operational Support Component (OSC) would provide direct support to the MOFDM for implementation support and capacity strengthening based on the needs. 6. Therefore, the MPC can be viewed as a component that reimburses the GOB for results, while the OSC can be seen as a financing instrument for the inputs needed to achieve the results indicated in the MPC. 7. IDA Credit for the MPC would flow into the Consolidated Fund and then onward to the MOFDM. The OSC would disburse separately from the MPC, in that a special account would be set up with the MOFDM and the monies would be in direct control of the Ministry. In effect, OSC would function as a standard sector investment loan (SIL) and not as a results-based financing instrument. 8. The design of the project hinges on a number of innovations in design and implementation. Improved governance is an underlying theme of both components and the major result being pursued by the project. Main Program Component (MPC) Focus Area 1: Rigorous Targeting: Background 9. As pointed out earlier, Bangladesh has a number of targeted safety net programs with real challenges in ensuring that the maximum numbers of the poorest households receive benefits. Among the bottom 10 percent of the population the coverage of all safety net programs is 23 percent, which includes 16 percent covered by targeted programs. While the overall coverage of safety net programs is pro-poor, a sizeable number of non-poor households also receive benefits. For example, 41 percent of the beneficiaries of targeted safety net programs are non-poor. Apart from low coverage, benefit amounts tend to be small relative to household needs. For example, benefits offered by standard cash transfer programs were only at about 30 percent of the lower poverty line (World Bank 2008). More general issues in the targeting of safety nets programs are discussed in Annex 1. 10. In any targeting system, errors of inclusion in particular point to elite capture and potential malpractice. Therefore, better targeting of a program is likely to improve governance across the board. This includes clear rules and norms, eligibility criteria, transparent system of beneficiary selection and systems of public disclosure, discussed in Focus Area 2. 11. When both food and cash benefit programs are combined, the MOFDM is responsible for delivering the largest share of safety net programs in Bangladesh. The EGPP, implemented by the MOFDM, is currently one of the largest ongoing safety net programs in the country. Reforms to the targeting system in EGPP are likely to have significant spill over benefits to other programs implemented by the MOFDM and beyond those implemented by the MOFDM. 12. Prior to the implementation of the EGPP and its precursor program, there were clear mismatches between the spatial distribution of poverty and program coverage. Data from HIES 2005 showed that the three poorest divisions (Khulna, Rajshahi and Barisal) had lower coverage of safety nets programs than two of the better off divisions (Dhaka and Sylhet). For example, about 12 percent of all households in Rajshahi division, which has a poverty headcount rate of 51 percent, are beneficiaries of at least one safety nets program. In contrast, about 22.4 percent of all households in Sylhet division are beneficiaries of at least one safety nets program, though poverty incidence is 34 percent there (World Bank 2008). 13. An independent assessment of the precursor to the EGPP, the 100 Day Employment Generation Scheme, found that 67% of the benefits went to the poor, a proportion that is comparable 41 to other safety net programs in Bangladesh. However, the same study found that just above 37% of the resources reached the extreme poor (NFPCSP-FAO-BRAC 2008). A World Bank assessment carried out in the north-west of Bangladesh showed that the household heads of 73 percent of EGP participating households worked as unskilled day labor, as opposed to 33 percent of the non- participating households. According to HIES 2005, this employment category is a very strong correlate of poverty in rural Bangladesh. The incidence of poverty among households with daily wage workers is extremely high, with 67 percent being poor and 47 percent being extremely poor. Another strong correlate of poverty in rural areas is ownership of cultivable land. Data from the HIES (2005) suggests that about 55 percent of the rural population with cultivable land below 0.5 aces live in poverty, while 38 percent live in extreme poverty. For practical purposes the cutoff of 0.5 acre ownership of cultivable land is commonly used as a proxy for determining household poverty status in rural areas of Bangladesh. Microfinance institutions such as Grameen Bank and BRAC use this criterion to determine households eligible for loans. The census data indicates that EGP participant households are more likely to own less than 0.5 acres of cultivable land than non-EGP participant households. Description of current EGPP targeting system 14. Geographic targeting to the upazila (sub-district) level is based on the poverty map. Currently, the following allocation system is in place: 40% of the budget to the 81 upazilas with extreme poverty rates of 40% and higher, 40% of the budget to the 254 upazilas with extreme poverty rates of 21-39%, and 20% of the budget to the 146 upazilas with extreme poverty rates of 20% and below. 15. Within the upazila, the Upazila Committee further divides the budget among union parishads. According to the program guidelines, the Upazila Committee is supposed to allocate a larger share of program resources to high poverty areas within the upazila. Once the Union Committee has received their allocation, they decide on the wards where the sub-projects will be implemented, and the PIC committee selects beneficiaries to work on each sub-project. 16. According to the program guidelines, the PIC should select beneficiaries based on the following criteria: one beneficiary per household can receive benefits, beneficiary households must have less than ½ an acre of landholdings, the household head must be employed as a wage laborer in agriculture, beneficiaries must not getting concurrent assistance from other government-run safety nets programs (other than educational stipend programs), and beneficiaries must be between the ages of 18 and 60. 17. Currently 30 percent of beneficiaries are also supposed to be women. Proposed revisions to the targeting system 18. Reaching the poorest is the cornerstone of the EGPP. This includes both the poorest areas and the poorest households. a. Geographical or area based targeting based on the poverty map, where the bulk of resources go to the poorest upazilas. 19. The current geographic allocation formula produces a very low coverage rate for the large group of upazilas with poverty rates of 21-39% and hence some reform is required. Under the current geographical allocation system described above, and using the program budget for FY11, the EGPP 42 can cover 2.6% of extreme poor individuals in high poverty density upazilas, 1.4% in moderate poverty density upazilas, and 2.6% in low poverty density upazilas. Starting from the September 2010 project phase, and in preparation of Bank support to the EGPP, the Ministry has proposed to change the definition of the `high' and `moderate' poverty density slabs, as currently the moderate slab of upazilas is much larger than both the high and low slabs. 20. Implications of change in definition of high and moderate poverty upazilas: The group of 133 upazilas with extreme poverty rates of 35% and above would then be considered the `high poverty density' slab, The group of 202 upazilas with extreme poverty rates of 21-34% would be the `moderate poverty density' slab, and The group of 146 upazilas with extreme poverty rates of 0-20% would remain the `low poverty density' slab. 21. The allocation formula would also be changed during the project period.15 50% of the budget will now go to 133 upazilas with poverty rates of 35% and higher, 40% of the budget will now go to the 202 upazilas with poverty rates of 21-34%, 10% of budget will now go to the 146 upazilas with poverty rates of 20% or below. 22. The change in the `slabs' of upazilas and in the allocation formula would increase program coverage in upazilas with moderate poverty density and decrease coverage in upazilas with low poverty density. Under the proposed changes, and using the program budget for FY 2011, the EGPP would cover 2.3% of extreme poor individuals in high poverty density upazilas, 1.9% in moderate poverty density upazilas, and 1.3% in low poverty density upazilas. b. Household targeting o Filter 1: The first filter that will ensure the participation of the poorest is the setting of wages. Public works programs are most successful when they are self-targeted, with only those willing to work at a given wage rate self-selecting into the program. For the EGPP, the wage rate is set nationally at 120 takas per day. Surveys of local wage rates in a random sample of Union Parishads find that this wage rate is slightly lower than wages for manual laborers in agriculture and non-agricultural work in all but the most remote areas (see Annex 1). However, because the budget allocation for the program only covers a small proportion of poor households, there is in practice oversubscription in the poorest areas that necessitates other mechanisms to select beneficiaries. The program will ensure that wages are set below the prevailing market wage so that the program is not attractive to the non- poor households. o Filter 2: The second filter is a proxy for household poverty ­ individuals from those households where the head is a manual laborer and the household has less than ½ acre of land will be eligible. The combination of these two criteria have been found to be very good predictors of poverty in rural Bangladesh, as 72% of agricultural manual workers (19% of households in rural areas) and 59% of non-agricultural manual workers (12% of the rural population) are poor. In addition, 62% of the rural population has less than 0.5 acres of land and around 54% of them are poor (HIES, 2005). Hence a combination of both manual laborer and less than 0.5 acres captures a very significant share of the poor. 15 The definition of poverty density slabs and the allocations to those slabs may be subject to change based on revisions to the poverty map. 43 c. Gender Targeting Independent assessments of the precursor to the EGPP, the 100 Day Employment Generation Scheme, found that about 10 percent of the beneficiaries were women. The project will progressively increase the number of women participants in the program to reach 33 percent by the final year. In principle cards will be distributed according to the land/labor criteria described above. Then the project will ensure that one-third of cardholders who actually show up to work are women. Already, in the course of the dialogue during project preparation, the participation of women has reached 25 percent. The project will ensure that the proportion of beneficiaries who are female is recorded through the regular monitoring reports and verified through the impact surveys and independent third party validations. In addition, implementing agencies will receive training in enrolling more women into the program and in devising sub-projects that would encourage more women to enroll. Focus Area 2: Improved governance through clearer rules and enhanced transparency 23. As pointed out earlier, IDA support to EGPP is meant to improve its design and implementation in a way that enhances the governance of Bangladesh's largest safety net program. This is indeed the main focus of the GOB's request to the Bank. Focus Area 1 in the previous section showed how improved targeting can maximize the gains to the poorest households and regions, while reducing the scope for discretion. Similarly, Focus Area 3 provides support to better monitoring mechanisms including independent assessments. Focus Area 2 therefore proposes a set of results that enhance clarity of rules and transparency of implementation. Clarity and accessibility of rules and procedures 24. Currently the program rules are fairly generic and are supplemented from time to time by circulars. However, many upazila level functionaries are unaware of the changes to rules and procedures. These rules seldom percolate further down the chain to the Union Parishad level and hardly ever reach the PIC level. Even if many of the rules are not directly relevant to the PICs or the communities, knowledge of the norms and procedures strengthens the hands of communities to demand accountability from the PICs and the Union Parishads. The project therefore proposes a set of results (see Annex 3) that require changes to the program document and for the revised rules and procedures to be publicly available at the lowest levels. Information and public disclosure 25. In addition to clarity and accessibility of rules and procedures, information about implementation should also be available to the community. Currently, communities seem to know about wage rates and type of work they are supposed to do, but have little idea about allocations to their wards, criteria for beneficiary selection, or governance arrangements at the sub-project level. They are also not aware of their entitlements or the responsibilities of the tiers above. For instance, beneficiaries need to know that they can inspect the muster rolls (attendance sheets) at designated times or at the end of a work day. To improve the availability of information at the local level, a system of public disclosure is included in the results monitoring arrangements for the project. 26. Evidence from across the world and from the Sirajganj pilot project in Bangladesh indicates that community level sign-boards or notice boards that display regularly updated, reliable and correct information enhances community confidence and participation. Below is a sample of the type of information that needs to be publicly disclosed. Much of this is in keeping with the norms of the Local Governance Support Project and the UP level has previous experience with such disclosure. However, unless disclosure is mandated it seldom takes place in the poorest and most disempowered areas ­ which are exactly the areas where EGPP operates. 44 Table 4.1: Illustrative content for community sign-boards To be displayed on sub-project To be displayed outside UP office site or public place Responsibility Project implementation committee UP Chairman Content Name of sub-project Allocations to the UP from Budget allocated EGPP Start and end date End of phase expenditures Physical specifications at UP level Name of chairman of implementation committee Number of laborers Contact details & telephone number of OPD 27. Currently, information on how beneficiaries are selected and the final beneficiary lists are also not easily available to communities. Under the project, PICs and Union Committees would be required to make selection criteria and beneficiary lists publicly available at the union parishad headquarters. Adherence to this provision would be monitored through independent third party validations and process assessments conducted under Focus Area 3 of the project. Non-wage costs 28. Typically, public works programs in Bangladesh do not have a provision for non-wage costs and the EGPP is no exception. Providing for costs associated with the purchase of materials, hardware or skilled labor is central to the quality of infrastructure and governance of the program. Lack of such provision thus, can have the following negative impacts. First, it restricts the type of projects that can be undertaken to those which are primarily mud based or "earth works". Second, if communities are expected to contribute non-wage costs, it affects poorer communities disproportionately. Finally, lack of provision for non-wage costs can create incentives for irregularities, especially when combined with strict upwards monitoring (pressure to deliver the program). In such cases, more "enterprising" PICs may divert wage costs into material costs through the use of `ghost workers'. 29. During the project period, guidelines will be issued to earmark up to 20 percent of the cost of subprojects under EGPP to purchase materials and skilled labor, in keeping with global practice. For most infrastructure projects at least 10 percent of the cost would be mandated for materials and other non-wage costs. This minimum level of non-wage expenditures will provide encouragement for PICs to select sub-projects that move beyond pure earthworks and into more durable asset creation. While IDA will reimburse only wage costs, the inclusion of non-wage costs in the GOB's program will be an important innovation which will be linked to disbursement. Globally, non-wage costs are typically included in public works projects and the proportion of non-wage costs for labor intensive public works projects ranges from 20 percent to over 60 percent. Better system for payment of wages 30. One of the biggest innovations of the project will be in wage payments. Currently workers on EGPP are paid through cash transactions on a weekly basis by the PIC. In principle, this violates an important principle of separation of tasks of resource management and payment processing (which in part is responsible for the irregularities discussed in the above paragraph). Furthermore, the recordkeeping systems are dated, processes are manual, and monitoring is weak. This leaves considerable risk exposures in funds misappropriation, payment mismanagement, and abuses in individual payment cases (delays, bribes, collusion, etc.). If administration and payments were 45 separated, if payments were made directly to workers by parties other than PIC, and if all such transactions were adequately automated, it would significantly reduce the fiduciary risk on the one hand and empower the poorest on the other. In fact, several programs in Bangladesh already pay cash benefits through formal financial channels. Table 4.2 in this Annex details the options that were considered in the reform of payment systems. Given the constraints under which the poorest citizens are served, IDA support to the EGPP envisages that direct payments to workers using formal channels will be introduced and this result will be linked to disbursement. It is proposed that during an initial phase, following all the required preparations, a minimum of 80 percent of those participating unions that have banks located within them will receive payments through banking channels, which will grow to 85 percent in the second year. By the final year of IDA support to the program, it is expected that 95 percent of those UPs which have banks located within them will be paying out wages through banks or other formal financial channels. Importantly, the project will support development of comprehensive manual covering operations of the bank payments program. Delivery of such a manual will be an important indicator for disbursement of IDA funds for payment of wages. 31. Furthermore, where access to the formal financial mechanisms remains significant challenge (e.g., remote rural areas), and with a view to further the progress of expansion of coverage of the new payment systems, IDA will support a pilot to adapt some innovative payment solutions to the operational needs of the program. The pilots will cover a minimum of 3 upazilas and will be subject to a rigorous evaluation to fine-tune the proposed solutions, bring awareness about good practices, and promote standards in benefit delivery. 32. Two sets of assessments undertaken during the preparation phase indicate that several actions are necessary before a comprehensive roll out of direct payment systems can be achieved. While Bangladesh has made great strides in payment and communication technology, linking the implementing Ministry, the service provider (bank, mobile phone or other technology partner), the PIC and the worker would require several intermediate steps, potentially including changes or clarifications to existing banking regulations. Recent evidence from a similar program in India shows that when payments through formal channels are mandated before requisite systems are in place, it leads to significant hardships for workers (Khera, 2010). On the other hand, several programs in Bangladesh, notably the Secondary School Stipend Program do pay out cash through formal financial channels, although transactions in those programs only happen twice per year rather than the more frequent payments needed for the EGPP. Table 4.2: Available options of the EGPP Payment Platform Conventional Banking Mobile Phone Banking Biometric Smart Cards Low banking penetration across Banks are not yet ready for a Requires comprehensive planning rural areas where most of the complete technology solution for and implementation strategy to target beneficiaries belong. In mobile phone banking. Some perform biometric finger capture order to accommodate EGPP banks are in the process of of beneficiaries, card printing, beneficiaries banks would have to upgrading their IT systems to distribution, identification of expand their client base in certain incorporate necessary mobile point of service, training etc areas, which may not be phone banking functionalities. immediately feasible. It may be impractical to launch Most of the banks do not have that solution if it serves the Data quality in the existing paper G2P license (regulatory purpose of EGPP for less than based practice of some banks requirements) for launching this 100 days a year, considering roll limits effective monitoring of the solution. out and maintenance costs. payment process. In fact, the However, synergies with other banks such as Shonali and Agrani Some beneficiaries may not have programs may strengthen the case banks having wider coverage mobile phones, although there are for smart cards. across rural areas are still alternative mechanisms. operating manually. Requires comprehensive planning May be impractical to open and implementation strategy to conventional accounts for appoint the implementing partner, 46 Conventional Banking Mobile Phone Banking Biometric Smart Cards beneficiaries that collect their point of service. EGPP payments only during a very short period throughout the Without a comprehensive year, especially if it comes at beneficiary list, the transmittal of high maintenance costs. information from the PIC to the PIO and the banks could imply From the beneficiaries' delays in payments perspective, distance from the banks could have a high opportunity cost. Need for a comprehensive beneficiary list, since the transmittal of information from the PIC to the PIO and the banks could imply delays in payments Enhanced capacity at the field level for better supervision 33. IDA support to EGPP is fundamentally about strengthening systems of monitoring and implementation. Currently, as noted elsewhere, field level capacity to monitor the program is very limited. The upazila level Project Implementation Officer (PIO) is responsible for a number of programs and during times of frequent disaster, becomes totally consumed in relief efforts. S/he therefore relies on the PIC and the UP to monitor its own programs, which is not the basis for a sound system. S/he also receives help from "tag officers" who are officers belonging to different departments that have field level presence, but who are "tagged" to monitor programs in a Union Parishad that may not be mapped to their own department. Clearly, they have little incentive to monitor programs that are not part of their core mandate. Poor monitoring on the ground creates the conditions for potential malpractice. 34. IDA support to the EGPP therefore provides for field level supervisors in the poorest 334 upazilas. Moreover, PIOs and the new field level supervisors will also be provided with transport in the form of a motorcycle for each of the poorest 334 upazilas. The financing for staff and transport costs will be through the Operational Support Component. Recruitment of the field supervisors and ensuring that they are in place is a key disbursement indicator. In addition, while not a disbursement indicator, there is also provision for training of the field supervisors and PIOs. Finally, it is anticipated that the new payment systems would considerably improve quality of the record-keeping, and so would comprehensively strengthen the function of program monitoring at the field level by allowing, e.g., for better directed audits, etc. Grievance redress 35. The project will also support the development of an enhanced grievance redress system. The first point of acceptance of complaints is proposed to be the upazila since this is the level of the project closest to the community that is not directly involved in beneficiary selection or wage payments, two common sources of grievances. However, complaints can be accepted at any of the three levels ­ upazila, district or the Ministry. At the upazila level, the Upazila Nirbahi Officer (UNO) will be the Grievance Redress Officer (GRO). At the district level, the Deputy Commissioner (DC) will serve as the GRO. The Deputy Project Director (DPD) in the office of the Project Director will act as the GRO at the central level. 36. Complaints will be entered in a Grievance Redress Register at each level and complainants will be given a receipt. All grievances should ideally be disposed within 15 days, failing which the complainants can appeal before the appellate authority. The DC of the respective district will be the appellate authority for the upazila level complaints, while the DPD of the project will be the appellate 47 authority for the district level complaints. The National Steering Committee (NSC) headed by the Secretary, DMRD will act as the final appellate authority for grievance redress. The Office of the Project Director will be responsible for developing appropriate formats for complaints and redress as well as disseminating information about the Grievance redress system. The importance of a feedback loop that includes informing the complainant about the action taken on their complaints and feeding this information into the implementation of the program will be a key aspect of the implementation of the grievance redress system. 37. Through its system of checks and balances, the program will collect a considerable amount of information on its functioning. The Office of the PD will assign program staff to specific districts and they would be expected to tour their areas of responsibility and monitor the progress from the ground up, including receiving feedback on actions taken on complaints. Focus Area 3: Better Capacity for Monitoring and Evaluation Background 38. The MOFDM, has an annual budget of almost US$350 million, and implements the bulk of food and cash for work programs in Bangladesh. The EGPP is currently one of the largest ongoing safety net programs in the country. However, the Ministry has limited monitoring capacity to oversee the EGPP and its other programs. The current monitoring system and the MIS for the EGPP is based on monitoring systems for disaster relief assistance, where the emphasis has traditionally been on ensuring timely distribution of resources in crisis situations rather than on outcome monitoring. Improvements in the Ministry's monitoring capacity generated through improvements in EGPP program monitoring would have large spillovers for strengthening the capacity to monitor other large safety nets programs administered by the Ministry. 39. The Ministry's monitoring capacity would be developed under the Focus Area 3 of the proposed project - improving monitoring and evaluation systems for the Bangladesh EGPP. Monitoring activities would involve collection of streamlined indicators on finances, sub-projects, and beneficiaries, while process assessments, institutional performance, targeting, leakage rates and the impact of the program would be examined through independent assessments. Please see Supplementary Annex 4 for more details on the planned Program Monitoring and Evaluation. Current monitoring system 40. The current program MIS focuses on monitoring disbursements of wages and days of work created. The limited information is aggregated to the upazila level by the time it reaches the center, so it is difficult to assess progress and outcomes at the sub-upazila level. Additionally, because no beneficiary information is digitized, it is also difficult for the PD's office to ascertain whether the program is reaching the poorest of the poor. Under the current monitoring system, there are also few personnel to monitor the sub-projects or to undertake technical feasibility assessments. Monitoring system under implementation 41. Under focus area three, the proposed project would assist the implementing Ministry in refining their monitoring strategy for the EGPP and would provide support to develop the monitoring capacity at the upazila level and at the Center. The project would support the development of a set of standardized summary indicators and monitoring reports and would provide financing for critical staffing at the central and local levels and investments in MIS capacity. 42. The existing monitoring forms and reports would be redesigned to facilitate monitoring on inputs, outputs, intermediate outcomes and compliance with program guidelines. The set of streamlined indicators to be collected at the sub-project level would include information on the flow of funds, beneficiary profile, wages paid, and sub-project type. Financial indicators to be collected 48 will include data on how much has been disbursed to and paid out in each sub-project and spending on administrative and non-wage program costs. Indicators on beneficiaries will include: how many job cards have been issued for each sub-project to men and to women, the characteristics of program beneficiaries (e.g., the proportion of beneficiaries who are manual laborers and who have less than 0.5 acres of land), how many days of work have been created on each sub-project, and the percent of person-days of work by female beneficiaries. Wage indicators will provide the average wages paid per worker per day, and the total wage payments made, and the amount of time elapsed between work and payment. Infrastructure indicators will list the type of sub-projects (e.g., road construction, drainage system maintenance) under implementation. These indicators will be used to produce a user- friendly biweekly or monthly summary report for senior management. 43. In order to ensure that the data collected on the monitoring indicators is available for use in real-time decision-making, a number of institutional and technical enhancements would be made under focus area three of the project. The project would provide MIS capacity (in the form of computers and data entry clerks) at the upazila level, as well as modems so that information can be transferred electronically from the local level directly to the center. Additional support and training would also be provided to the monitoring cell in the central PIU. Resources would also be provided to allow for greater supervision at the sub-project level. Monitoring of program implementation at the local level would be strengthened by providing for contractual hires of staff and access to transport to support the PIOs and Tag Officers at the upazila level. Independent assessments 44. The independent assessments for the EGPP would aim to provide information on program implementation, the effectiveness of targeting and the impact of the program on household net incomes and coping strategies in response to shocks. Assessments would include both external independent third party validations and an impact assessment based on surveys of beneficiary and non-beneficiary households. 45. The World Bank would partner with a local think tank to conduct independent third party validations in a sample of union parishads. External independent third party validations would be conducted in a random sample of approximately three percent of union parishads each program phase. The sample of union parishads where independent third party validations will take place would be selected randomly by the partner organization before each program phase. The independent third party validations would examine whether program implementation guidelines are followed and whether program benefits reach the intended beneficiaries. 46. The employees of the organization contracted for independent third party validations would visit the sampled upazilas each program phase and would examining the hard copy (and electronic, if available) documentation on program implementation, beneficiary selection, payments, and other financial arrangements at the community sub-project level. During the independent third party validations, the employees would also conduct focus group discussions or interviews with beneficiaries, other community members, program staff and other union parishad and upazila officials. The information obtained from independent third party validation visits would be verified against program monitoring and financial reports received at the center, and the implementing agency would have the opportunity to cross-check findings at the verification stage. The independent third party validation firm would produce a report of findings each phase (i.e., twice per year) to allow for immediate feedback into the program. 47. In order to assess program targeting and the impact of the program for participants, the World Bank would also partner with a reputable local survey firm to undertake baseline and follow-up surveys of program participants and non-participants in a representative sample of upazilas. The surveys will collect information on household demographics, assets, income generating activities, risk coping strategies, and participation in safety net programs. Information on individual and household characteristics will be used to construct, using analytical techniques such as propensity score 49 matching, a comparison group from among the non-participant households who are surveyed who are otherwise similar to participant households. Approximately 3,000 households would be surveyed at the baseline, which is planned for November 2010. A smaller sample of approximately 1,000 households surveyed at baseline would be re-surveyed in November 2011 if resources are available. In November 2012, the larger follow-up survey of 3,000 households, including the 1,000 panel households, would be carried out. 48. The survey data will be used to examine program targeting and the determinants of program participation, including whether women or individuals from households in the lowest income quintiles are (a) more likely to receive job cards and (b) more likely to work under the program. The individual, household, and community characteristics (e.g., literacy, other wage-earning opportunities, shocks suffered, and distance from the nearest worksite) that influence both the ability to access job cards under the program and the days worked by beneficiaries in the EGPP will also be examined. Program impacts to be examined at the household level include net household incomes (measured relative to the household's prior year income or relative to otherwise similar non-participant households), responses to lean season vulnerability, including the use of coping strategies such as borrowing, migration, or reductions in the number of meals eaten per day, changes in household assets, especially in relation to possible distress sales, and impacts of the program on human capital through children's school enrolment and household health spending. 49. The impact evaluation surveys would be supplemented with analysis of data from nationally representative surveys from Bangladesh, including the 2010 Household Income and Expenditure Survey (HIES).16 In all of the analysis, the survey data will be merged with administrative data detailing the extent and types of workfare activities in the upazilas where households were surveyed. Program outcomes, including targeting performance, leakage rates, and the net income gains to participating households, will be identified using several analytical techniques, including propensity score matching. 50. Qualitative assessments would also be commissioned to complement the quantitative work. The qualitative work would examine beneficiary and non-beneficiary assessments of how well the program is working, including perceptions of targeting, of leakages/corruption, the quality of works produced under the EGPP, the value of the community sub-projects (both to participants and to non- participants), and impacts of the program on internal and external migration as well as on social empowerment. This work would take the form of focus group discussions and key informant interviews with EGPP participants and non-participants as well as with local elite and administrators of the EGPP. 51. The independent assessments under this focus area would not be funded out of the project OSC. Funding for the impact evaluation household surveys has been secured from the TFESSD and trust fund and external resources from development partners are currently being sought to finance the independent third party validations and qualitative assessments. 16 Even if the HIES sample size does not permit analysis of EGPP outcomes (which will be examined in the impact evaluation work described above), the HIES will be very useful in identifying the key sources of risk, vulnerability and coping mechanisms for poor Bangladeshi households. 50 Table 4.3: Overall Governance Arrangements to Ensure Benefits Reach the Workers Type of Purpose Actors Check/Assessment Impact Evaluation To assess the impact of the program (iv) DMRD, MOFDM / OPD Survey To provide insights on the extent to which (v) World Bank Bangladeshi households are vulnerable to multiple (vi) Firm contracted for seasonal, climatic and idiosyncratic shocks, how they Survey cope and specifically how the EGPP contributes as a coping strategy Independent third To examine whether program implementation (v) DMRD, MOFDM / OPD party validations guidelines are followed and whether program benefits (vi) World Bank reach the intended beneficiaries (vii) Local think tank (viii) Organization contracted for independent third party validations Financial Audit To ensure implementation of the sound financial (iv) Comptroller and Auditor management and monitoring system General of Bangladesh To ensure that the program (MPC+OSC) financial (v) DMRD, MOFDM / OPD statements be audited annually (vi) World Bank Grievance To provide a platform to the beneficiaries for voicing (vi) NSC headed by the mechanism their grievances Secretary, DMRD- To deal with the grievance and complaints of workers MOFDM under the EGPP (vii) Deputy Project Director, OPD (viii) Deputy Commissioners- Grievance Redress Officer (GRO) for the concerned districts (ix) UNO - GRO for the concerned districts (x) Beneficiaries Community level To enhance knowledge about the structure and (v) DMRD, MOFDM / OPD disclosure functioning of the program amongst community (vi) Union Committee member & beneficiaries (vii) Project Implementation To ensure information & public disclosure for Committee enhancing community confidence and participation (viii) Organization contracted for independent third party validations for verification Feedback to To ensure that information about the program (i) DMRD, MOFDM / OPD implementation generated through different sources has an impact on agencies implementation and course corrections are made II. Operational Support Component: Detailed Description 52. As pointed out earlier, the MPC can be viewed as a component that reimburses the GOB for results, while the OSC can be seen as a financing instrument for the inputs needed to achieve the results indicated in the MPC. Background 53. The Ministry of Food and Disaster Management (MOFDM) implements some of the largest food for work and cash for work programs in Bangladesh. The Ministry is very concerned that such a large budget has such limited resources for monitoring. This OSC therefore proposes to strengthen monitoring capacity by providing the upazila with a field supervisor who will be contracted and a motorcycle to enable movement and supervision. The OSC also proposes to strengthen monitoring 51 capacity at the PIO level by providing access to computers with internet access so that regular reports can be transmitted to the Relief Control Center. The monitoring capacity strengthened through this OSC will improve capacity for monitoring scheme implementation and compliance with EGPP guidelines on an ongoing basis and will also have spillovers for strengthening the capacity to monitor other large safety net programs implemented by the Ministry. 54. Overall objective: Provide operational support to the Disaster Management and Relief Division, Ministry of Food and Disaster Management in delivering EGPP with enhanced efficiency in terms of better targeting, better benefit delivery mechanism, greater transparency and improved monitoring. 55. Specific objectives: i) Strengthen reporting and monitoring of the EGPP operations ii) Provide training to staff engaged in EGPP to enhance their capacity in targeting beneficiaries and delivering benefits 56. The activities under this operations support will contribute to preparing a solid base for the implementation of the Main Program Component (MPC) of the proposed EGPP. Further, the support through the OSC will go beyond EGPP and is likely to percolate to other social protection initiatives implemented by the MOFDM including Food for Work (FFW) and Test Relief (TR). These activities will also strengthen cross-sectoral collaboration across the ministries and with other stakeholders. 57. Cost: The total cost of the OSC is estimated at Bangladeshi Taka 3660.409 lakh or US$5.305 million. Of this amount, Bangladeshi Taka 3623.039 lakh or US$5.251 million will be financed by IDA as a component of the EGPP financing. The detailed cost estimates are provided in Table 4.5. 58. Expected outputs: Services of financial management and procurement personnel availed. Services of Field Supervisors are availed. Digitalized reporting, monitoring and supervision system is established. Transport facility for the Field Supervisors provided. Systematic targeting methodology is established. Training is provided for beneficiary targeting and enrolment process. A team of Government officials will have received exposure to global best practices. The Ministry's overall capacity to deliver quality social protection projects is improved. Table 4.4: Job Description for Key Staff of PIU Position Key tasks Education/ Experience and Skills Qualifications Project Supervise preparation of an Preferably a Extensive field experience in Director annual work plan and budget post-graduate heading / supervising rural and submit these for reflection degree in Social development programs. A senior in the ADP and National Science or Previous experience as Project official of the Budget. Business Director or team member of Government of Supervise preparation of an Studies World Bank supported Bangladesh, effective annual implementation including project(s) would be an not below the manual. Economics / additional qualification. rank of Joint Monitor and ensure timely Commerce / Secretary, will implementation according to the Development be appointed work plan, manual and budget Economics / as Project and supervise the set up of an Public Director. efficient MIS with linkage to all Administration the tiers. or other 52 Position Key tasks Education/ Experience and Skills Qualifications Supervise operating the special relevant Accounts and prepare subjects. withdrawal applications for forwarding to IDA. Ensure that consolidated project accounts are maintained and arrange for annual audits. Arrange for evaluation of project progress. Undertake regular field visits to Project sites and swiftly resolve identified problems. Hold regular meetings with field personnel of the Project and promote their effectiveness. Deputy Project Working in the Office of the Preferably a Extensive field experience in Director Project Director (OPD) and post-graduate heading / supervising rural mainly responsible for handling degree in Social development programs. A senior the field implementation, Science or Previous experience as Project official of the administration and finance Business Director or team member of Government of issues decided by the National Studies World Bank supported Bangladesh at Steering Committee (NSC). including project(s) would be an the rank of Assist the PD in his duties as Economics / additional qualification. Deputy Member Secretary of the NSC Commerce / Secretary will and the Technical Review Development be appointed Committee (TRC). Economics / as Deputy Act as Focal Point on Public Project environmental and social Administration Director. safeguards issues. or other Prepare an annual work plan relevant with budget and ensure subjects. submission and clearance by Disaster Management and Relief Division and subsequent reflection in the ADP and National Budget. Supervise operating the special Accounts and prepare withdrawal applications for forwarding to IDA. Ensure that consolidated project accounts are maintained and arrange for annual audits. Participate in the preparation work of the annual implementation manual of the Project. Maintain close contact with Deputy Commissioners, Upazila Committees and Union Parishads of the program areas on a regular basis to solicit necessary support for smooth transfer and allotment of funds at field level. Monitor and ensure timely implementation according to the 53 Position Key tasks Education/ Experience and Skills Qualifications work plan, manual and budget. Ensure and pursue timely submission of progress reports from the field. Undertake extensive field visits, spending at least 10 days a month in the field and report findings on a regular basis. Hold regular meetings with field personnel of the Project and promote their effectiveness. Facilitate training of all field personnel, viz. foundation training from new personnel and refresher training for all other officials and staff. Supervise training workshops at national and field level. Arrange for evaluation of project progress. Provide regular feedback to the PD and seek guidance from him when needed. MIS Specialist Review existing formats and Bachelors At least 5 years experience in provide recommendations for Degree in ICT / relevant fields. the modification of the existing CSE / CST / Sound understanding of MIS reporting system to meet the EEE / ECE; or methods and approaches. agreed requirements of the MBA; or Post- Experience in data management Project. Graduate and analysis using MS Access, Develop specific MIS Programs Degree / Four Oracle, Visual Basic, SPSS, for all units. years Honors Stata and other related Ensure the finalization and Degree in software. distribution of relevant reporting Applied Understanding of and templates and guidance / Physics & experience in public sector explanatory notes to all agencies Electronics / and/or non- governmental affected by the reporting system Physics / sector and/or financial sector (including UPs, Upazilas, Mathematics / reporting systems. district and central government). Statistics / IT / Track record on information Act as a resource person to all Computer analysis, report writing and stakeholders, and provide inputs Engineering / publications. to the development of training Computer Proficient in English and curricula. Science / Bangla writing and speaking. Library and Ability to work in a team, Information deliver under pressure and Science / MIS; within deadlines. or B.Sc. Engineering with proven experience in Software / MIS development. Training Assess the capacity- Masters in At least 7 years of experience Specialist strengthening needs of the OPD Social Science, in training and capacity and field level functionaries Social Work or building on social protection including PIOs and Field Sociology or programs Supervisors. other relevant Extensive knowledge on socio- Develop the annual training plan areas. economic structure of 54 Position Key tasks Education/ Experience and Skills Qualifications for the project based on the Bangladesh, human capacity-strengthening needs development, poverty reduction assessment. programs, etc. Develop training curriculum and Proven ability in providing modules on relevant issues. training to large groups Organize training and capacity participants. building programs. Proven skills and experience in Develop tools for evaluation and training needs assessment, impact assessment of training curricula/module development, programs. planning and conducting Liaise between OPD and all participatory training and other stakeholders for capacity building programs; strengthening training and Excellent skills and experience capacity building of personnel. in conducting evaluation and Facilitate meetings and impact assessment of training workshops on relevant subjects. programs and writing related Prepare monthly performance reports. reports. Payment Provide comprehensive Bachelors At least 7 years working Systems technical assistance in designing Degree in ICT / experience in payment systems. Specialist and implementing a payment CSE / CST / Proven skills on developing system through formal financial EEE / ECE; or strategy of operational channels. MBA; or Post- integration for the payment Assess needs in implementing Graduate system for the public sector / pilots on payment system. Degree / Four social security / social safety Train field level functionaries years Honors nets schemes. and other stakeholders on their Degree in Understanding and experience roles and responsibilities in Applied in public sector and/or non- strengthening the payment Physics & governmental sector and/or system. Electronics / financial sector reporting Monitor implementation of Physics / systems. payment system pilot. Mathematics / Track record on information Work with MIS Specialist on Statistics / IT / analysis, report writing and development of MIS System Computer publications. and facilitate integration of Engineering / Proficient in English and indicators relating to payment Computer Bangla writing and speaking. system. Science / Ability to work in a team, Library and deliver under pressure and Information within deadlines. Science / MIS; or B.Sc. Engineering with proven experience in Software / payment system Financial Financial management of the Professional 8-10 years of working Management Main Program Window (MPW) Accounting experience in Specialist and the Operational Support Qualifications national/international/multinati Window (OSW) funds. (FCA or FMA). onal agencies in financial Deliver requisites and ensure management, accounting, smooth flow of funds to both the banking and international MPW and OSW. transactions. Preparation of interim un-audited At least 3 years in a managerial financial statements. position in government / semi Preparation of annual project government organization, or financial statements. development projects. 55 Position Key tasks Education/ Experience and Skills Qualifications Coordination of project audit. Knowledgeable in GOB budgeting, fund release and reporting formalities, World Bank's financial management and disbursement procedures. Procurement Assist the project operation "Procurement At least 5 years work Specialist through implementing all Specialist" experience as Procurement policies and procedures related Graduate/Post Specialist or related to procurement activities and -graduate responsibilities with a ensuring that the World Bank qualification substantial procurement procurement Guidelines are At least a component. followed in all transactions Bachelors in Undertake procurement Engineering activities, such as assisting / or a Masters reviewing technical in specifications for goods and procurement / terms of reference for commerce / consulting services, preparing finance / bidding documents, writing business evaluation reports and similar management / activities law Lead the development of a project procurement plan and provide guidance and training to all staff involved in procurement processing 56 Table 4.5: Economic code and Sub code wise description of component and year-wise breakdown of cost 57 Supplementary Annex 4: Monitoring and Evaluation Bangladesh - Employment Generation Program for the Poorest Project Background 1. The MOFDM, which has an annual budget that exceeds 2,400 crore taka, implements the bulk of food and cash for work programs in Bangladesh, and its EGPP is currently one of the largest ongoing safety net programs in the country. However, the Ministry has limited monitoring capacity to oversee the EGPP and other programs. The current monitoring system and MIS for the EGPP is based on the MOFDM's monitoring systems for disaster relief assistance, where the emphasis has traditionally been on ensuring timely distribution of resources in crisis situations rather than ongoing outcome monitoring. Improvements in the Ministry's monitoring capacity generated through improvements in EGPP program monitoring would have large spillovers for strengthening the capacity to monitor other large safety nets programs administered by the Ministry. 2. The Ministry's monitoring capacity will be developed under the third focus area of the project, improving monitoring and evaluation systems for the Bangladesh EGPP. Monitoring activities will center on the collection of streamlined indicators on finances, sub-projects, and beneficiaries, while assessing targeting performance, leakage rates and the impact of the program will be part of the evaluation activities. The current monitoring system is described before the proposed changes to the monitoring system and the evaluation strategy that will be implemented with World Bank support are discussed. Current monitoring system 3. At the local level there are insufficient personnel to monitor project implementation and outcomes. The Program Implementation Officer at the upazila level has an allocation for two staff, neither with technical qualifications. These staff positions may not be filled in some upazilas. Each UP has also been assigned to a "Tag Officer" based in the upazila and who has the responsibility of overseeing the UP or UPs assigned to him/her. However, although the PIO and Tag Officer must travel to sometimes far-flung locations to monitor sub-projects, neither the PIO nor the Tag Officers have a budget for visiting sub-project sites. 4. The current program MIS focuses on monitoring disbursements of wages and days of work created. Information on ongoing program progress, in terms of the percentage of the program allocation for wages that has been spent, the number of work days that have been created, and the number of job cards issued, is recorded using a standard template. The PIO aggregates sub-project-level information to the upazila level then faxes or phones in the upazila level reports to the district DRO, as even upazilas with computer facilities may not have internet connections. The current report tracks the number of male and female job cards issued and the number of days of employment created for men and women in each upazila as well as the amount of wages paid out. Transcription or data entry then typically occurs at the district level. The DRO then typically faxes or emails the district report to the Relief Control Center at the MOFDM. At the Relief Control Center, district reports received in non-electronic form are re- entered. The information is then used to generate a variety of reports, charts and GIS maps to assess program progress at the upazila level. 5. Information on program beneficiaries is recorded in hard copy on standardized Worker Registration Cards. Often, the Worker Card information is incomplete, with only the name, id number, signature, and daily wage fields filled in. Each beneficiary (or a designated labor leader for the sub- project) is supposed to keep his or her Registration Card. The name and identification number from the Worker Registration Card is transcribed to the master role and attendance log for each sub-project, which 58 are both kept in hard copy. The upazila PIO holds a copy of the muster roll. Attendance logs are completed at the sub-project level. Only the information from the muster roll necessary for processing the weekly payment (total number of days of work during the previous week) is passed up to the upazila level. 6. Because the information is aggregated to the upazila level by the time it reaches the center, it is difficult to assess progress and outcomes at the sub-upazila level. Additionally, because limited information is passed up from the local level, and none of the worker-level information from job cards, muster roll and attendance sheets is digitized, it is also difficult for the center to ascertain whether the program is reaching the poorest of the poor or identify potential problem areas at the sub-upazila level. Monitoring system under implementation 7. Under focus area three, the project will assist the implementing Ministry in refining their monitoring strategy for the EGPP and will provide support to develop the monitoring capacity of the MOFDM at the upazila level and at the Center. The project will support the development and refinement of a set of standardized summary indicators and monitoring reports and will provide financing for critical staffing at the central and local levels and investments in MIS capacity. Indicators and reporting 8. In order to strengthen the monitoring system to enable the Government to assess, on an ongoing basis, the extent to which the program is meeting its objectives, the existing monitoring forms and reports will be redesigned to facilitate monitoring on inputs, outputs, intermediate outcomes and compliance with program guidelines. The primary goal will be the production of a user-friendly biweekly or monthly summary report for senior management. 9. Following discussions with our counterparts we expect that current versions of program monitoring forms will be streamlined to eliminate superfluous information and enhanced to include information necessary to assess the project's development objectives. For instance, the job card will collect information on whether the beneficiary is a manual laborer and the total amount of land owned by the participant's household (including homestead land), which are both relatively easy to verify indicators of household economic status and vulnerability to shocks. 10. The set of streamlined indicators to be collected at the sub-project level will include information on the flow of funds, beneficiary profile, wages, and sub-project types. Financial indicators to be collected will include data on how much has been disbursed to and paid out in each sub-project and spending on administrative and non-wage project costs. Indicators on beneficiaries will include: how many job cards have been issued for each sub-project to men and to women, the characteristics of program beneficiaries (e.g., the proportion of beneficiaries who are manual laborers and who have less than 0.5 acres of land), how many days of work have been created on each sub-project, and the percent of person-days of work by female beneficiaries. Wage indicators will provide the average wages paid per worker per day and the total wage payments made. Infrastructure indicators will provide the type of sub- projects (e.g., road construction, irrigation system maintenance). 11. Upazilas will also begin the process of creating electronic beneficiary registries of all job card holders registered for work each program phase. It is expected that all upazilas will have electronic registries of beneficiaries by the last year of the project. 12. Several other monitoring checks will also be built into the monitoring system. For instance, indicators of worker characteristics that can be used to assess targeting performance (whether a 59 beneficiary is a manual laborer and the amount of land owned) can then be compiled at the sub-project level and average worker characteristics for each sub-project can be reported to the center. Further, independent third party validations will be used to examine whether the worker characteristics information has been completed on both job cards and monitoring reports and verify whether the information is accurate. Institutional arrangements for monitoring and reporting 13. The institutional arrangements for monitoring and transferring monitoring data will also be strengthened to improve the flow of information from the sub-project level and the utilization of the information at the center. The project will provide resources to allow for increased supervision at the sub- project level, investments in MIS and connectivity at the local level to facilitate the flow of information, and investments that strengthen the analytical capacities of the PIU. 14. In order to ensure that the data collected on the monitoring indicators is available for use in real- time decision-making, a number of institutional and technical enhancements will be made. Revisions to the institutional arrangements must strike a balance between the risk that too much monitoring data is collected that it is then not used and the need to monitor program implementation on an ongoing basis. Currently, because limited information is passed up from the local level, and no worker-level information is digitized, it is also difficult for the center to ascertain whether the program is reaching the poorest of the poor or identify potential problem areas at the sub-upazila level. In the current program MIS, information on sub-projects is aggregated at the upazila level and is passed up through the district to the Relief Control Center at the MOFDM. As even upazilas with computer facilities and data entry personnel do not typically have internet connections, reports are often phoned or faxed to the center, where the upazila level information is entered and is then used to generate a variety of reports, charts and GIS maps to assess program progress at the upazila level. 15. This component of the project would provide allocations for strengthening MIS capacity at the upazila level.17 By providing MIS capacity (in the form of computers and data entry clerks) at the upazila level, as well as modems so that information can be transferred electronically from the local level directly to the center, the flow of information can be streamlined, duplication in data entry work can be eliminated, and more detailed monitoring information at the sub-project level can be provided to the center. If information duplication is eliminated, it will also free up time at the central level to focus on analysis and reporting rather than data entry. Additional support and training will also be provided to the monitoring cell in the central PIU. 16. Resources will also be provided to allow for greater supervision at the sub-project level. Under the current monitoring system, there are few personnel to monitor the sub-projects or to undertake technical feasibility assessments. Monitoring of project implementation at the local level will be strengthened by providing for contractual hires of staff and access to transport to support the PIOs and Tag Officers at the upazila level. Independent assessments 17. The independent assessments for the EGPP would aim to provide information on program implementation, the effectiveness of targeting and the impact of the program on household net incomes and coping strategies in response to shocks. Assessments would include both external independent third party validations and an impact assessment based on surveys of beneficiary and non-beneficiary 17 Successful experiences of remote monitoring using video and computer technology has been successfully piloted recently in Bank projects in Afghanistan. 60 households as well as qualitative work. The independent assessments will not be funded out of the project OSC. Funding for the impact evaluation household surveys has been secured from the TFESSD and trust fund and external resources from development partners are currently being sought to finance the independent third party validations and qualitative assessments. Independent third party validations 18. The World Bank would partner with a local think tank to conduct independent third party validations in a sample of union parishads. External independent third party validations would be conducted in a random sample of approximately three percent of union parishads each program phase. The sample of union parishads where independent third party validations will take place would be selected randomly by the partner organization before each program phase. The independent third party validations would examine whether program implementation guidelines are followed and whether program benefits reach the intended beneficiaries. 19. The employees of the organization contracted for independent third party validations would visit the sampled upazilas each program phase and would examining the hard copy (and electronic, if available) documentation on sub-project implementation, beneficiary selection, payments, and other financial arrangements at the community sub-project level. During the independent third party validations, the employees would also conduct focus group discussions or interviews with beneficiaries, other community members, program staff and other union parishad and upazila officials. The information obtained from independent third party validation visits would be verified against program monitoring and financial reports received at the center, and the implementing agency would have the opportunity to cross- check findings at the verification stage. The independent third party validation firm would produce a report of findings each phase (i.e., twice per year) to allow for immediate feedback into the program. Impact evaluation 20. In order to assess the impact of the program for participants, the World Bank plans to partner with a reputable local survey firm to undertake baseline and follow-up surveys of program participants and non-participants in a representative sample of upazilas. The impact evaluation will provide insights on the extent to which Bangladeshi households are vulnerable to multiple seasonal, climatic and idiosyncratic shocks, how they cope and specifically how the EGPP contributes as a coping strategy. 21. The data from each round of surveys will be used to examine program targeting and the determinants of program participation. We will examine whether women or individuals from households in the lowest income quintiles are (a) more likely to receive job cards, and (b) more likely to work under the program. We will also examine what are the individual, household, and community characteristics that influence both the ability to access job cards under the program and the days worked by beneficiaries in the EGPP. These characteristics could include literacy, other wage-earning opportunities, shocks suffered, and distance from the nearest worksite, among others. We will also examine whether there is evidence of potential sources of leakage of program benefits to the non-poor, including, for example, short payment of wages. 22. Program impacts for beneficiaries will be examined primarily at the household level. We will examine the impact of the program on net household incomes (measured relative to the household's prior year income or relative to otherwise similar non-participant households) and on responses to lean season vulnerability, including the use of coping strategies such as borrowing, migration, or reductions in the number and quality of meals eaten per day. We will also examine changes in household assets, especially in relation to possible distress sales, and look for any impacts of the program on human capital through the channels of children's school enrolment and household health spending. 61 23. Program impacts will be identified using several analytical techniques, including propensity score matching. With the latter approach, survey information on individual and household characteristics is used to construct a comparison group of eligible non-participants and differences in outcomes are compared for otherwise similar households that did and did not participate in the program. 24. In total, the impact evaluation will include baseline and follow-up surveys of a sample of approximately 3,000 representative households in upazilas (sub-districts) in which the program is currently operating and potentially in nearby upazilas where the program is set to expand in future years. Approximately 1,000 of the households surveyed at baseline will be re-surveyed in all follow-ups, forming a panel sample. 25. The baseline survey would take place in November 2010. The smaller panel sample of around 1,000 households will be surveyed a year later. The second follow-up survey of 3,000 representative households, including the 1,000 panel households, will be carried out approximately two years later, which will allow the longer term impacts of program participation and the impact of intervening shocks to households to be examined. 26. The survey will collect information on household demographics, assets, income generating activities, consumption, risk coping strategies, and participation in safety net programs. A listing exercise will also be conducted to determine the sampling frame for the household surveys. Beneficiary and non- beneficiary households will be randomly selected from the list of all households in sampled areas. In the analysis, the survey data will be merged with administrative data detailing the extent and types of workfare activities in the upazilas where households were surveyed. Other data sources 27. The impact evaluation work will be supplemented with analysis of data from nationally representative surveys from Bangladesh. Questions about EGPP participation by household members have been included in the 2010 Household Income and Expenditure Survey (HIES) that is currently in the field. This nationally representative sample survey, which also includes a wide range of variables measuring poverty, can also be used to construct a comparison group of eligible non-participants for an evaluation of EGPP outcomes. Depending on the sample number of EGPP participants, the HIES sample will be used to examine targeting performance, leakage rates, and the net income gains to participating households. However, even if the HIES sample size does not permit analysis of EGPP outcomes (which will be examined in the impact evaluation work described above), the HIES will be very useful in identifying the key sources of risk, vulnerability and coping mechanisms for poor Bangladeshi households. This type of data will feed into the broader context of the intervention strategy as well as provide information on the geographical, household and individual characteristics of poverty. Qualitative work 28. Qualitative assessments will also be commissioned to complement the quantitative work. The qualitative work will examine beneficiary and non-beneficiary assessments of how well the program is working, including perceptions of targeting, of leakages/corruption, the quality of works produced under the EGPP, the value of the community sub-projects (both to participants and to non-participants), and impacts of the program on internal and external migration as well as on social empowerment. This work will take the form of focus group discussions and key informant interviews with EGPP participants and non-participants as well as with local elite and administrators of the EGPP. 62 Annex 5: Project Costs Bangladesh - Employment Generation Program for the Poorest Project Project Cost by Activity Amount (US$) 1. MAIN PROGRAM COMPONENT Wage Payments (to a maximum of) 144,750,000 2. OPERATIONAL SUPPORT COMPONENT Individual Consultant Services Financial Management Specialist 78,000 Procurement Specialist 26,000 Management Information System Specialist 39,000 Payment Systems Specialist 78,000 Training Specialist 39,000 Salaries Office of the Project Director 53,000 Field Offices 1,526,000 Operational & Maintenance Costs Office of the Project Director 41,000 Field Offices 944,000 Training (Local) 237,000 Hardware Vehicles (for head office and field offices) 790,000 Motorcycles (for field offices) 518,000 Computer Equipment 352,000 Other Goods & Equipment 151,000 Management Information System 334,000 3. CONTINGENCIES 44,000 TOTAL 150,000,000 63 Annex 6: Implementation Arrangements Bangladesh - Employment Generation Program for the Poorest Project Project Management Structure 1. EGPP seeks to fund a set of innovations to the ongoing Employment Generation Program through two components - the Main Program Component (MPC) that would support expansion of and reforms to the existing program through three focus areas and the Operational Support Component (OSC) that would provide direct support to MOFDM for implementation support and capacity strengthening based on the needs. In order to implement the OSC, a Technical Assistance Project Proposal (TPP) prepared by the MOFDM has been approved by the Government. The TPP aims at providing operational support to the MOFDM in delivering EGPP with enhanced efficiency in terms of better targeting, better benefit delivery mechanism, greater transparency and improved monitoring. This will help MOFDM in strengthening its mechanisms for effective implementation of the MPC of the project. 2. The Disaster Management and Relief Division (DMRD) of the MOFDM will implement the project along the lines of the existing institutional structures and implementation arrangements, but with a strong focus on strengthening its systems for a more effective implementation and monitoring of the project. A project implementation team led by a full-time Project Director, not below the rank of Joint Secretary, will support the project. An Office of the Project Director (OPD) will be established in the DMRD for the duration of the project, to ensure efficient and timely implementation. One full-time Deputy Project Director, not below the rank of Deputy Secretary, will assist the PD in the implementation of the project. The OPD will comprise of a Project Director, Deputy Project Director, Financial Management Specialist, Procurement Specialist, MIS Specialist, Payment Systems Specialist, Training Specialist, 334 Field Supervisors and the support staff (six Assistants, four Data Entry Operators and three Drivers). The Project Director will be responsible for overall management of the project, providing guidance on planning, implementation, operation, monitoring and supervision. He will be assisted by the Deputy Project Director, who will be responsible for field supervision and coordination, administration, finance and monitoring. 3. The Office of the Project Director (OPD) will be advised, guided and supervised by the committees as follows. A National Steering Committee (NSC) would be set up in the Disaster Management and Relief Division (DMRD) of the Ministry of Food & Disaster Management. Chaired by the Secretary DMRD, the NSC would be responsible for: Providing policy advice and operational guidelines; Approving the Operations Manual and all rules and regulations governing its technical, financial and administrative way of operation; Reviewing the physical and financial progress of the project; Providing guidance to solve any implementation problem and grievances; Overseeing proper integration of results and findings of activities into program implementation; Providing any other necessary guidance and directions for the effective implementation of the project. 4. The NSC will include representatives of the Ministry of Finance, Economic Relations Divisions, Planning Commission, Implementing Monitoring and Evaluations Division of the Ministry of Planning, Local Government Division, Ministry of Social Welfare, Ministry of Women and Child Affairs, and any other Government agencies which the NSC finds useful in coordinating and determining policy issues of the project. The Project Director will be the Member Secretary of the NSC. The Additional Secretary and the Joint Secretary (Relief) of the Disaster Management and Relief Division, Deputy Secretary 64 (Relief), Deputy Chief Planning (Disaster Management and Relief Division), the Deputy Project Director, and the concerned Desk Officer will also be members of the NSC. The NSC would have powers to co- opt member(s), if necessary. The Committee will meet quarterly or at other times of necessity to discharge its functions. 5. Implementation of the technical assistance project will be further assisted by a Technical Project Review (TPR) Committee headed by the Project Director and the Deputy Project Director (DPD) will act as the Member Secretary. The other members of this committee will include representatives of Finance Division, Directorate of Relief & Rehabilitation, Deputy Secretary (Relief), Deputy and Assistant Chief Planning of the DMRD and the World Bank. The TPR Committee will meet quarterly or at other times of necessity to discharge its following functions: Oversee overall project implementation as per the GOB and IDA rules and regulations; Supervise functions of the Office of the Project Director (OPD) to ensure proper implementation of the project; Oversee the management of project activities, including performance of project personnel and quality of deliverables; Advise for timely implementation of the project activities; Review, evaluate and monitor the progress of the project from time to time and suggest course correction; Resolving issues and conflicts that may emerge during project implementation; Facilitating coordination and convergence with other line Ministries/Departments; and Keep NSC apprised about project performance and other key issues relating to the effective implementation of this technical assistance. 6. In order to deal with the grievance and complaints of workers under the EGPP, a Grievance Redress System will be set up at national, district and upazila levels. At the upazila level, the Upazila Nirbahi Officer (UNO) will be the Grievance Redress Officer (GRO), while at the district level the Deputy Commissioner (DC) will be the GRO. The Deputy Project Director (DPD) in the office of the Project Director will act as the GRO at the central level. At each level, there will be a Grievance Redress Register, where complaints are entered petitioners given a receipt. All grievances should ideally be disposed within 15 days at the upazila level, failing which the petitioner can appeal before the appellate authority. The DC of respective district will be the appellate authority for the upazila level complaints, while the DPD of the project will be the appellate authority for the district level complaints. The National Steering Committee (NSC) headed by the Secretary, DMRD will act as the final appellate authority for grievance redress. The Office of the Project Director will be responsible for developing appropriate formats for complaints and redress as well as disseminating information about the Grievance redress system. Complainants can however send letters of complaint to any level and these will be acted upon by the level where they are received. The OPD will also provide a dedicated phone number for strengthening the grievance redress mechanism and this number along with contact details of the OPD will be prominently displayed at all the community sub-project work-sites. Implementation Arrangements 7. The project will operate in 334 upazilas of the country with poverty rate over and above 20 percent. The Office of the Project Director (OPD) will be responsible to manage, administer and coordinate the project, and more specifically: (a) to handle procurement of all goods, works, and services, (b) to ensure and monitor the proper allocation and distribution of the these goods, and (c) to carry out financial control of all operations. The OPD will manage the project in keeping with the Government's obligation to use the project funds with due regard to economy and efficiency and only for the purpose for 65 which project financing was provided. Chart - 1 below presents institutional arrangements for implementation and coordination of the program. 8. At the upazila level, the Upazila Committee (UC) will be responsible for the finalization and approval of the Union Plans containing projects and list of beneficiaries submitted by the Union Committee. The UC will also be responsible for supervision & monitoring of projects, liaising with the Office of the Project Director (OPD) and Union Committee, supporting PIO and Union Committee with necessary technical inputs, timely transfer of funds to Union Committee as well as preparing and forwarding upazila level progress reports to OPD. 9. The UC will compile and submit (via the UNO) its Upazila Plan to the Deputy Commissioner (DC) and the OPD for formal approval. The Upazila Plan will contain union-wise details of planned projects and beneficiaries for each phase of the project. In the event that no comments are received from the DC within one month, the UC will assume that its annual plan/budget has been formally approved. 10. The UC will be chaired by the respective Upazila Nirbahi Officer (UNO) and the Project Implementation Officer (PIO) will act as its Member Secretary. The respective Member of Parliament (MP) will act as the Chief Advisor, while the Upazila Chairman will act as the Advisor. Other members will include Upazila Vice Chairmen, all Union Parishad (UP) Chairmen of the respective Upazila, Upazila Health and Family Planning Officer, Upazila Engineer LGED, Upazila Women Affairs Officer, Upazila Education Officer, Upazila Agricultural Officer, Upazila Cooperative Officer, Upazila Livestock Officer, Upazila Fisheries Officer, Social Welfare Officer, Ansar and VDP officer, Upazila BRDB Officer, all Union Parishad Chairmen of the respective upazila, eminent persons of the upazila (Selected by the Deputy Commissioner), and one teacher and a representative from the voluntary organization (Selected by the Deputy Commissioner). The UC will have powers to co-opt more members from the upazila, if necessary. 11. The Project Implementation Officer (PIO) of every upazila will be the focal point for the implementation of the project at the upazila level and he will be assisted by a Field Supervisor hired on a contractual basis by the project. The Field Supervisor will be responsible for: (i) Facilitating the overall implementation process of the program; (ii) Supporting upazila and union-level public administration in efficiently executing the program; (iii) Coordinating regularly with the head of local administration and relevant government officials; (iv) Participating in local-level meetings where program maters are discussed; (v) Conducting and facilitating the targeting and enrolment processes in the assigned location; (vi) Monitoring and supervising public workfare schemes; (vii) Preparing and submitting all assigned reports from the local-level on a regular basis; and (vii) Assisting the Program Implementation Officer (PIO) with other tasks as requested. 12. At the Union Parishad (UP) level, the Union Committee will be responsible for the selection of all projects and beneficiaries in consultation with the community and the ward members, preparation and regular updating of the beneficiary muster roll, conducting environmental and social assessment of projects. The Union Committee will be responsible for the periodic supervision of projects, timely payment to the beneficiaries, liaison with the UC and OPD, maintenance of all necessary records including site registers, registration/ job cards, receipts & payment, etc., with the help of the Project Implementation Committees (PICs). The Union Committee will also coordinate with the PICs for ensuring timely reporting including data collection and preparation of activity / progress reports. 13. The Union Committee will compile its Union Plan on the basis of list of projects and beneficiaries received from each ward and submit the same to the Upazila Committee (UC) for approval and necessary action. The Union Plan will contain ward-wise details of planned projects and beneficiaries for each phase of the project. 66 14. The Union Committee will be chaired by the UP Chairman and the UP Secretary will act as its Member Secretary. Other members of the Union Committee will include all members of the UP, Sub- assistant Agricultural Officer (Block Supervisor), One representative from LGED, Field Assistant, BRDB, one eminent person from each ward of the union (Selected by the Upazila Nirbahi Officer), one teacher, one women representative from the union (Selected by the Upazila Nirbahi Officer) and a representative of the voluntary organization from the union (Selected by the Upazila Nirbahi Officer). Where necessary, a representative from the Water Development Board and any one as appropriate for the committee may be co-opted in the committee. 15. A Union Committee will be formed in each UP in a community meeting in the presence of UP Chairman, UP members and with community participation from all wards organized by UP. In choosing Union Committee members, the following principles will be respected: (a) The members of the Union Committee should be residents of the ward (or village) that benefits most directly from the proposed investment. Every effort should be made to ensure that the representatives (at least two in every Committee) of beneficiary communities are included in the Union Committees ­ as this maximizes local ownership and increases accountability; (b) At least two members of the Union Committee shall be women. 16. Selection of beneficiaries will be done through involvement of the local people and beneficiaries by arranging a planning meeting. The Union Committee will issue notice for the registration of beneficiaries through the concerned ward members, local Banks, Post Offices, Schools, Colleges, community members and eminent persons of the society. This will be done by issuing a letter to the concerned persons as well as making public announcement about the registration process and locations. The notice for registration will also be prominently displayed at the office of the Union Parishad, Post Offices, Banks, Schools, Colleges and any other public place for disseminating information to the general masses. In the notice and public announcements, the detailed eligibility criteria for the registration of beneficiaries will be clearly mentioned. 17. Any eligible man/woman can make an application to the concerned ward member(s) for being registered as a beneficiary under the program. The ward members with active involvement of one teacher, one religious leader and one eminent person from the community will review the received applications on the basis of eligibility criteria and prepare a list of probable beneficiaries. On the date of ward level Planning Meeting, the list of probable beneficiaries will be discussed and finalized in an open meeting in presence of local people and beneficiaries. Reasons for non-selection of applicants failing to meet the eligibility criteria will be explained during the meeting and any query or clarifications sought by such applicants will be clarified during the meeting. 18. The ward-wise list of beneficiaries will be submitted to the Union Committee that will prepare a final list of beneficiaries for the union after reviewing the list of beneficiaries received from each ward. The union-wise list of beneficiaries will be submitted to the Upazila Committee that will compile an Upazila Plan consisting of union-wise list of beneficiaries and projects and send the same for the approval of the concerned Deputy Commissioner (DC). Upon approval of the list of beneficiaries from the Upazila Committee and the DC, the ward Members will convene a ward-level meeting to disseminate all information regarding approved and unapproved beneficiaries to the members of the community and the applicants. An unemployed man / woman selected as beneficiary under the program shall register their name and obtain Job Card by visiting the concerned Union Parishad Office. 19. Selection of sub-projects will also be done through involvement of the local people and beneficiaries during the ward level planning meeting based on the local demand/ need. The ward members will prepare a list of proposed projects to be undertaken for the ward in consultation with local 67 people including women following the selection criteria laid down in the guidelines issued by the DMRD ­ MOFDM. A ward-wise priority list of projects will be submitted to the Union Committee along with records of the proceedings, discussions and decision of the ward level Planning Meeting. On the recommendations of the Union Committee, the upazila Committee will approve the projects and send the same to the Deputy Commissioner (DC) and the Office of the Project Director (OPD) for the formal approval. On receipt of the approval from the DC and OPD, the Upazila Committee will issue an order to the respective Union Committee for the implementation of projects. 20. Every Union Committee may form Project Implementation Committees (PICs) consisting of 5-7 members for each project. Each PIC will be responsible for day to day implementation of the projects in close collaboration with the Union Committee. They will also be responsible for timely payment to the beneficiaries, liaison with the Union Committee, maintenance of all necessary records including site registers, registration/ job cards, receipts and payment, etc. The day to day reporting including data collection and preparation of activity / progress reports of the concerned project will also be carried out by the PIC. 21. The Project Implementation Committee (PIC) will be formed in a community meeting in the presence of UP Chairman, UP members and with community participation from all wards organized by UP. The members of the PIC should be residents of the ward (or village) that benefits most directly from the proposed project. A concerned ward member or women ward member will be selected as the Chairperson of PIC during this meeting. Other members of the PIC will be nominated by the community at ward level open meeting organized by the ward members. One school teacher and an eminent person from the locality shall be the committee members in addition to the local ward members. One member nominated by the committee will act as the Member-Secretary. The following principles will be respected in the process of selection and functioning: (a) A person will not hold the position of the Chair in more than one PIC; (b) At least 30% of the members of PIC will be women. 22. Grievances of the community will be addressed through a redress mechanism, with the UNO being the first point of receiving and investigating complaints and is described in Focus Area 2 of Annex 4. 68 Chart ­ 1: Project Management Arrangements NATIONAL STEERING COMMITTEE (NSC) Policy advice and operational guidelines Beneficiary feedback & Grievance redress Review the physical and financial progress of the project Project implementation as per GOB and IDA rules and regulations TECHNICAL PROJECT REVIEW (TPR) Guidance to solve any implementation problems COMMITTEE Oversee integration of results and findings of activities into program implementation Disbursement of funds OFFICE OF THE PROJECT DIRECTOR Policy advocacy, sensitization of stakeholders, training and capacity building of service providers (DMRD) Prepare project progress reports Upazila Accounts Officer Budget release (Consolidated Fund) Approval of projects and list of beneficiaries submitted by the OFFICE OF THE DEPUTY Union Committee COMMISSIONER & DRRO Supervision & monitoring of projects Liaison with the Office of the Project Director (OPD) and Union Committee UPAZILA COMMITTEE (UC) Support PIO and Union Committee with necessary technical Joint Bank A/c UC Chairperson & inputs Member Secretary Beneficiary feedback & Grievance redress Timely transfer of funds to Union Committee Monitoring & Reporting Prepare and forward progress reports on all subprojects to OPD Fund Flow Selection of projects and beneficiaries OFFICE OF THE PROJECT Preparation and regular updating of the beneficiary muster roll IMPLEMENTION OFFICER (PIO) Environmental and social assessment Day to day implementation & supervision of projects Joint Bank A/c Union Committee Timely payment to the beneficiaries Chairperson & UNION COMMITTEE Liaison with the Upazila Committee and OPD Member Secretary Maintenance of all necessary records including site registers, registration/ job cards, receipts & payment, etc. PROJECT IMPLEMENTATION COMMITTEE Monitoring & reporting (PIC) PIC Bank A/c Collecting data, preparing activity reports BENEFICIARIES 69 Ward Members & Chart ­ 2: Local Level Community members Implementation including women from Selection of Projects & beneficiaries community Initiates Request for Payment List of selected Projects & Project (RFP) to Union Committee Implementation beneficiaries sent to Union Committee (PIC) Committee Receives funds from Union Committee & makes payment to beneficiaries Review of list received from each Issues Job Cards to Union Committee Ward beneficiaries Prepare Union Plan with proposed Reviews RFP received from list of Projects & beneficiaries PIC & sends it to UC Send Union Plan to the Upazila Receives funds from UC & Committee for approval transfers the payment to PIC Bank A/c Upazila Committee (UC) Review of Union Plans from all Reviews RFP & transfers UPs funds to Union Committee Bank A/c Prepare Upazila Plan with proposed list of Project & beneficiaries Send Upazila Plan to DC & OPD for approval Deputy Review of Upazila Plan with the Commissioner (DC) help of DRRO Approve the Upazila Plan & conveys approval to the UNO & OPD Office of the Project Review of Upazila Plan Director (OPD) Conveys approval & Releases DMRD, Ministry of Food & Budget Disaster Management 70 Annex 7: Financial Management and Disbursement Arrangements Bangladesh - Employment Generation Program for the Poorest Project Overview 1. A financial management (FM) assessment of the employment generation program for the poorest was carried out as part of the preparation for IDA support to the EGPP. The assessment has been carried out in accordance with OP/BP 10.02, and following the principles laid down in the Financial Management Practices Manual. 2. There would be two components for the project: (i) the Main Program Component (MPC) would support expansion of and innovations to the existing program through three Focus Areas while financing short-term employment; and (ii) the Operational Support Component (OSC) would provide direct support to the Disaster Management and Relief Division (DMRD) for implementation support and capacity strengthening. 3. The funds flow and FM arrangements of the existing program would continue without any significant change required for IDA financing. To enable this, the IDA financing associated with the MPC would directly go to the Government Consolidated Fund and the disbursement would be linked to the achievement of a set of indicators called "disbursement linked indicators" (DLIs) in this document. For the OSC, a CONTASA account (designated account) operated by the Project Director would be used to make payments pertaining to the Government's Technical Assistance Project Proposal (TPP). Separate withdrawal applications would be required for disbursement under each component. 4. The IBAS system is capable of generating program related financial reports using the thirteen digit chart of accounts comprising four separate elements for the legal, function, operational/project, and economic codes, consistently used for budgeting as well as accounting. Relevant budget appropriation for wages would be provided under the head 1-4801-0006-5965 (employment scheme for extremely poor people - special grant), which enables generation of a financial report indicating expenditures by Upazilas. There would be no separate accounting required, as the MPC would finance the wages as the only eligible expenditure under the IDA credit. 5. A FM Specialist would be engaged by the DMRD in the Office of the Project Director (OPD) for accounting, disbursement, and financial reporting purposes. The Bank requires that the program (MPC and OSC) financial statements be audited annually and provided to it no later than six months after the end of the reporting period. Country PFM Issues 6. While PEFA Assessment (2005) identifies many weaknesses in the public financial management (PFM) systems of the country, since then there had been a lot of activity in agreeing upon a comprehensive reform strategy and preparing for the implementation of these reforms through a Multi- Donor Trust Fund (MDTF) for PFM improvement. The MDTF would provide support for (a) deepening medium-term budget framework (MTBF) and strengthening financial accountability, (b) strengthening the office of the Controller and Auditor General, and (c) strengthening legislative and public oversight. Nevertheless, strengthening of the employment generation program through enhanced social accountability and administrative monitoring by various means, the fiduciary risks will be minimized to a great extent. Hence, the Bank funding would leverage further improvements in the design and implementation of the program through different DLIs to ensure effective utilization of resources. 71 Risk Assessment and Mitigation Measures 7. The financial management risk of the program is assessed as Substantial. While some of the mitigating measures have been incorporated into the project design, a number of other measures identified are yet to be undertaken. The major driver of the risk rating is the large number of wage payments. 8. The constituent elements of the overall risk, together with their respective mitigating measures, are highlighted in the table below: Risk Assessment Matrix Condition of Risk Mitigating Measures Incorporated into Residual Risk Negotiations, Risk Program Design/to be implemented during Risk Rating Board or project processing/implementation Rating Effectiveness Inherent Risk H S Country/Local Level H -Reforms in country PFM systems are ongoing - S (weaknesses in the Integrated use of country FM systems PFM systems) Entity Level S -FM specialist to be hired to assist the DMRD M (no prior experience with Bank FM; lack of adequate staffing) - Project Level H - Use of outcome-based disbursement S (weaknesses in - Payment specialist to be hired payment of wages) - Clearly defined and inclusive structure of Project Implementation Committees at local level Control Risk S M - Budgeting M - Identification of the relevant budget and L account code for wages -Approval of the required amount in the relevant budget line in the national budget on the basis of allocations per Upazila - Accounting S - Timely adjustment of advances M -Maintain adequate records for assets created Unspent amounts are returned to treasury and not accounted - Internal Control S - Payment through direct bank transfers to the S workers - Use of third party validation of business processes -Ensure use of MPC fund on projects meeting local priority FM Framework to be in place for use of non wage fund -Assign ownership and responsibility for maintenance of the assets created under the program to the UPs - Funds Flow S - Use of report-based disbursement directly into M the Government consolidated fund -Disbursements will follow independent verification of achievement of DLIs - Timely release of fund to Upazila level -Agreeing on modality including service 72 Condition of Risk Mitigating Measures Incorporated into Residual Risk Negotiations, Risk Program Design/to be implemented during Risk Rating Board or project processing/implementation Rating Effectiveness standard for transfer of fund from treasury to PIC's bank account - Financial Reporting M - Agreement on the format of interim financial L reports of the project with the Bank. -Engagement of a FM Specialist - Auditing and S - Agreement on general scope of audit of the M follow-up on issues project annual financial statements Agreement with CAG on Statement of Audit Needs Addressing CAG's capacity constraint for timely and adequate coverage of program activities under audit Overall Risk H S Roles, Responsibilities, and Staffing 9. While the allocation of funds up to Upazila level is centralized, the operation of program activities is completely decentralized and participatory through engagement of a wide-range of people from elected representatives, government officials, and community. The Upazila Committee allocates the number of job cards/beneficiaries among various wards. The Union Committee recommends the projects as well as the list of workers to be approved by the Upazila Committee. At the lowest level, the Project Implementation Committee (PIC) actively manages and supervises community works. 10. For each community workfare scheme, the Chairman and the Member Secretary of the PIC jointly operate the bank account. To identify the bank account as a community account, the title of account must include the name of ward and respective ward number. 11. As the project is designed to support government's own program already under implementation, existing funds flow and FM arrangements would be maintained and IDA funds would be channeled through the Government's consolidated fund. The financial reports generated from the IBAS system would form the basis for disbursement of IDA funds. The assessment indicates that IBAS system is capable of generating such consolidated reports using the 13 digit chart of accounts comprising four separate elements for the legal, function, operational/project, and economic codes consistently used for budgeting as well as accounting. 12. Under the current system, 25 percent of annual allocation to an upazila is drawn from the concerned Upazila Accounts Office (UAO) as an advance and the check deposited in the joint bank account of Upazila Nirbhai Officer (UNO) and Project Implementation Officer (PIO) of DMRD. Further advance to UNO/PIO account is made on submission of an expenditure statement to UAO. From the UNO/PIO account, onwards advances to PICs are made on the basis of expected work for five days, and replenished after the weekly certification of work by the Tag officers. Any unspent amount remaining with PIC is transferred back to UNO/PIO account on completion of the scheme. Unspent amount with UNO/PIO at the end of each phase is deposited back to the treasury. Since the program implementation is seasonal (typically September to November and March to April), it is expected that all advances will be settled before the close of each six-months period during a fiscal year and financial reports generated from IBAS would indicate the actual expenditures. 73 Disbursements Flow GOVT OF WORLD BANGLADESH BANK IFRs & AFS Government Office of Project Upazilawise Consolidated Budget Director Fund Distribution DMRD Chief Accounts Officer Bill submitted for advance/statement of expenditures UAO Upazila Treasury Cheque issued Nirbhai by UAO to UNO Officer UNO/PIO Project Account Implementation Officer (DMRD) Request for funds PIC Account Project Funds transferred from UNO/PIO Account Implementation to PIC Account for the implementation of Committee schemes on account of wage, nonwage, and supervision costs. Direct payment of wages to workers' bank accounts is being considered. 13. The fundamentals of the disbursement mechanism are as follows: For the MPC (i.e., wage payments) Bank funds will flow into the Government Consolidated Fund and then released under the revenue budget to the DMRD through the Ministry of Finance. For the OSC (i.e., individual consultant services, training, salaries, operational and maintenance costs, hardware and management information system) Bank funds will flow into a Designated Account directly to DMRD, and this component will function like a standard Bank specific investment loan. Such a programmatic operation disburses against eligible expenditures. Focusing on expenditures in this way ensures that the program receives and spends budgeted amounts and by doing so effectively, it also meets the DLIs. 74 The programmatic operation must also fulfill a number of DLIs in order to qualify to disburse against program expenditures. DLIs are priced so that each DLI is then treated as independent and missing one does not hold back disbursement of others that are met. There are a maximum of eight disbursements under the credit. Other than the first disbursement, i.e., retroactive financing, which can occur immediately after effectiveness, disbursement would be within one month of the end of each phase (i.e. about every January and July) against expenditures incurred which are confirmed at the end of each phase. Disbursements will be for a maximum of 70% of wage costs incurred during each phase, and further subject to the ceiling for the amount to be disbursed in each period. This is done to optimize performance incentives over the course of the loan. (See indicative disbursement schedule below). The first disbursement scheduled for January 2011 (i.e., immediately after expected effectiveness) would finance the early achievement of three DLIs and part a) of a fourth DLI. If any amounts remain undisbursed due to the non-achievement of DLIs the Bank may authorize an additional disbursement, after the final disbursement for achievement of 2013 DLIs. This would be equivalent to the sum of the undisbursed amounts or US$20.00 million, whichever is lower, on meeting the 2013 targets for DLI 4 (efficient payment systems) and ensuring that the total disbursement does not cross 70 percent of wage costs incurred over the entire project period. 14. The Table below presents the allocated IDA financing of the two disbursement categories under the project, and the allocated amounts represent the 100 percent capped expenditure limits from IDA: Amount Expenditure Category (US$ million) 1. Wage Payments (MPC) 144.75 2. Individual Consultant Services, Training, Salaries, Operational and Maintenance 5.25 Costs, Hardware, Management Information System (OSC) Total IDA Financing 150.00 Taxes will be financed by IDA. 15. Report-based disbursement method, semi-annually, will be used for withdrawal of Credit Funds using the program financial reports generated from the IBAS system. The template of Interim Unaudited Financial Reports (IUFRs) has been agreed for both MPC and OSC. IUFRs under MPC and OSC will be due one month after the end of each program phase. 16. The starting point for determination of eligible expenditures under MPC for the purpose of ascertaining the `financing' amount would be the actual expenditures for FY 2010-11 to date. A report on wage costs by upazilas can be generated from the system and a trial run has been done. 17. The periodicity of future advances and disbursements will then follow the principle of bi-annual IUFRs. For MPC the IUFRs prepared by the Chief Accounts Officer of DMRD through IBAS will be validated by the Finance Division of the Ministry of Finance before being sent to the Bank together with a withdrawal application. The IUFR for expenditures under the first phase of the Program in FY 2010-11 would serve as the basis for the first disbursement to the Government subject to the Government achieving the agreed DLI milestones. For OSC the IUFRs would be prepared by the DMRD and directly send to the Bank along with the withdrawal application. 75 18. The following table provides the project's indicative disbursement schedule: US$ (m) MPC Cost 144.75 Value per DLI 20.68 ( MPC Cost / number of DLIs = 144.75 m / 7 ) Value per DLI per Calendar Year 6.89 ( Value per DLI / number of Calendar Years = 20.68 m / 3 ) Value per DLI for SepNov Phase (60 days) 4.14 ( Value per DLI per Calendar Year x 60% = 6.89 m x 60% ) Value per DLI for MarApr Phase (40 days) 2.76 ( Value per DLI per Calendar Year x 40% = 6.89 m x 40% ) 2010 2011 2012 2013 ( Reporting / DisbursementLinked Indicator SepNov 2010 MarApr 2011 SepNov 2011 MarApr 2012 SepNov 2012 MarApr 2013 SepNov 2013 Target Periods ) 1 1 Geographic Targeting 6.89 2.76 4.14 2.76 4.14 1 2 Household Targeting 6.89 2.76 4.14 2.76 4.14 1 3 Gender Targeting 6.89 2.76 4.14 2.76 4.14 1 ; 2 4 Efficient Payment Systems 1.00 1.76 4.14 2.76 4.14 2.76 4.14 5 Efficient RecordKeeping Mechanisms 6.89 6.89 6.89 6 Provision of NonWage Costs 6.89 6.89 6.89 7 Adequate Personnel for Supervision 6.89 2.76 4.14 2.76 4.14 FY 2011 FY 2012 FY 2013 FY 2014 ( Disbursement MPC Disbursement (US$ m) Jan2011 Jul2011 Jan2012 Jul2012 Jan2013 Jul2013 Jan2014 Periods ) 21.68 1.76 24.81 13.79 34.46 13.79 34.46 144.75 1 First scheduled disbursement against achievement of 2011 targets by Effectiveness 2 Disbursements (from the second payment onwards) will be proportional to percentage of target achieved in paying wages through bank accounts (value shown is for 100% achievement) F Y 2 0 1 1 F Y 2 0 1 2 F Y 2 0 1 3 F Y 2 0 1 4 OSC Disbursement (US$ m) * 1.53 2.10 1.09 0.52 5.25 * based on TPP estimates F Y 2 0 1 1 F Y 2 0 1 2 F Y 2 0 1 3 F Y 2 0 1 4 Jan2011 Jul2011 Jan2012 Jul2012 Jan2013 Jul2013 Jan2014 Total Disbursement (US$ m) 23.21 28.68 49.34 48.77 ...as % of total Credit 15.47% 19.12% 32.90% 32.51% Cumulative Disbursement (US$ m) 23.21 51.89 101.23 150.00 ...as % of total Credit 15.47% 34.59% 67.49% 100.00% 19. Except for DLI 4 (efficient payment systems), there are no payments for partial achievement of DLI targets. For example, if DLI 3 target on gender targeting requires that 30% of participants are women in a given phase, if only 25% of participants are women, then no portion of the payment for achievement of that DLI target in that phase will be made. For DLI 4 on efficient payment systems, partial reimbursement is allowed, with payments prorated based on the proportion of the target achieved. For example, if the DLI target specifies that 80% of payments in UPs with banks are through formal financial channels but only 50% of payments are achieved in a phase, then 5/8 of the payment for that phase will be made. 20. The indicative payment schedule assumes that all DLI targets are met as scheduled. However, an eight disbursement could occur at the end of the project term if there are undisbursed funds from unmet DLI targets. An additional `bonus' payment for achievement of the 2013 DLI 4 target on efficient payment systems can be disbursed in an amount not to exceed the lowest of the following: (i) total pot of undisbursed funds under the MPC; (ii) 70% of total wages over the entire project period less total payments to date; and (iii) US$20 million. 21. The disbursement schedule includes both DLI targets that are `floating' and `fixed'. `Floating' DLI targets can be withdrawn at a later date if the target is not met by the indicative disbursement date. `Fixed' DLI targets must be achieved within a given program implementation phase for disbursement to 76 occur. If fixed targets are not met in a given phase, at that point the undisbursed funds go into the pot of undisbursed funds from which the bonus payment could be made. Floating targets must be met by the end of the project, otherwise the undisbursed funds from those DLIs will also go into the pot for the bonus payment. The following details which DLI targets are fixed and floating: DLI 1 (geographic targeting): The 2011 target can be achieved any time through 2013. The 2012 and 2013 targets must be achieved in each phase. DLI 2 (household targeting): The 2011 target can be achieved any time through 2013. The 2012 and FY13 targets must be achieved in each phase. DLI 3 (gender targeting): The 2011 target can be achieved any time through 2013. The 2012 and 2013 targets must be achieved in each phase. DLI 4 (efficient payment systems): Part (a) of the 2011 target on guideline changes can be achieved any time through 2013. Part (b) of the 2011 target and the 2012 and 2013 targets must be achieved in each phase. DLI 5 (efficient record keeping): The 2011, 2012 and 2013 targets can be achieved any time through 2013. DLI 6 (non-wage costs): The 2011 target can be achieved any time through 2013. The 2012 and 2013 targets must be achieved in each phase. DLI 7 (adequate personnel for supervision): The 2011 target must be achieved by the end of 2011. The 2012 and 2013 targets must be achieved in each phase. Budgeting and Accounting 22. As the project proposes to support the Government's own initiative, the program would continue to make maximum use of country systems for FM and disbursement. While the project would finance a proportion of wages under the program (revenue budget) in addition to the specific OSC (which is implemented via the Government's TPP), the activities related to the program management and monitoring, accounting and financial reporting would remain mainstreamed under the Government system. The existing system of budgeting, fund release, and accounting would continue. This would mean identifying wage costs from existing heads within the Government chart of accounts. Nevertheless, the DMRD would be needed to prepare semi-annual and annual financial reports based on the data received from the office of Controller General of Accounts on the payment of wages. 23. A procurement and FM framework would be established for use of non wage funds. The PICs would maintain records of total amount of wage and non-wage cost incurred in creating community assets. The total wage and non-wage costs incurred in creating these community assets will be assessed relative to the value of works undertaken. The ministry would take a decision that these assets and the related documents would be handed over to the UPs for maintenance and appropriate use. UPs financial statements would include such assets in the UP's list of assets with locations so that the community assets are accounted for when selecting future community schemes by the UPs funded by any means. 24. Expenditures in respect of the OSC will be separately tracked through a Designated (CONTASA) Account with the DMRD. Actual need and amount for this component will be identified in TAPP and annual allocations will be provided under the ADP for each fiscal year. Project Annual Financial Statements and Annual Audit 25. The Bank requires that the program (Wages + Operational Support) financial statements be audited annually and provided to it no later than six months after the end of the fiscal year. The financial statements would be audited by the Comptroller and Auditor General of Bangladesh. While using of 77 country systems prevents duplication of audit effort, there seems to be little scope for conventional audits to be effective given the outreach and spread of activity involved. So there is a need to open a dialogue on how C&AG could provide the requisite assurance to stakeholders (including the Bank) on the use of funds for purposes intended. There might be a need for greater use of other evaluation work such as third-part verification envisaged under the program. There are no outstanding audit reports or ineligible expenditures for Bank projects being implemented by MOFDM. Audit Report Due Date Program Annual Financial Statements (inclusive of MPC & OSC December 31 of each year of the Bank credit) for the fiscal year ending June 30 26. The normal country systems for resolution and settlement of audit observations will be equally applicable to this project. The ministry would support resolution of Serious Financial Irregularities pointed out by the auditors on an urgent basis and keep the Bank informed on the process. Following SWAp principles, the Bank funds would not be separately tracked and the Bank will accommodate withdrawal applications from the Credit as long as the overall expenditures eligible (wages) more than or equal to the amount withdrawn from the Credit. FM Staffing 27. The project implementation unit under the DMRD would require adequate FM capacity for the key FM tasks including preparation of interim unaudited financial statements, annual project financial statements, coordination for project audit, and financial management of Operational Support funds. For this engagement of a suitably experienced FM Specialist would be a key action during project preparation. Supervision Plan 28. The project will require intensive supervision support in the initial year of implementation given the challenges and the capacity of the ministry's FM staffing. During project implementation, the Bank will review: (a) the project IUFRs and audited financial statements, including the budget execution report, together with the management letters; and (b) the project's FM and disbursement arrangements to ensure compliance with the agreed requirements. With the implementation of the sound FM and monitoring system by the professional staff proposed for the ministry, the Bank's normal implementation review procedures will suffice. 78 Annex 8: Procurement Arrangements Bangladesh - Employment Generation Program for the Poorest Project A. General 1. The total value of the project is US$150 million which will be financed by an IDA Credit. The total procurement of US$2.64 million under the project will involve goods of US$1.81 million, followed by consultancy services of US$0.83 million. 2. Procurement for the proposed project would be carried out in accordance with the World Bank's "Guidelines: Procurement Under IBRD Loans and IDA Credits" dated May 2004 (Revised October 2006 and May 2010); and "Guidelines: Selection and Employment of Consultants by World Bank Borrowers" dated May 2004 (Revised October 2006 and May 2010) (cumulatively know as "Bank Guidelines"), and the provisions stipulated in the Financing Agreement. Within the overall context of the Bank Guidelines, local procurement of goods, works, and consultant services (for which shortlist entirely comprised of national consultants) will follow the Recipient's Procurement Laws (PPA). For the purpose of the project, Procurement Laws shall be deemed to collectively be references to the Public Procurement Act 2006, the Public Procurement Rules 2008- PPR (as amended in August 2009), and the Public Procurement Act (1st Amendment) 2009, with the exceptions as outlined in Paragraph 3 below. In case of any conflicts between the Bank Guidelines and Procurement Laws or ambiguous/ confusing interpretation in between, the Bank Guidelines shall prevail. 3. For the purpose of National Competitive Bidding (goods and works), the following shall apply: (a) post bidding negotiations shall not be allowed with the lowest evaluated or any other bidder; (b) bids should be submitted and opened in public in one location immediately after the deadline for submission; (c) rebidding shall not be carried out, except with the Association's prior agreement; (d) lottery in award of contracts shall not be allowed; (e) bidders' qualification / experience requirement shall be mandatory; (f) bids shall not be invited on the basis of percentage above or below the estimated cost and contract award shall be based on the lowest evaluated bid price of compliant bid from eligible and qualified bidder; and (g) single-stage two-envelope procurement system shall not be allowed. 4. All expected major procurement of goods and consultants' services will be announced in the General Procurement Notice (GPN), published in the dgMarket and United Nations Development Business. 5. Procurement of Works: Procurement of works is not foreseen for the project. However, if at a later stage procurement of works become apparent for project execution, DMRD will carry out such procurement using the Bank's standard bidding documents (SBD) for all ICB and followed by NCB and shopping using PPA. 6. Procurement of Goods: There will be few Goods procurement using ICB and NCB method. DMRD will procure as deemed required goods, e.g., vehicles, motorcycles, computers, furniture, office equipments. etc. The procurement of goods will be conducted using the Bank's SBD for all ICB, followed by NCB and national shopping for small value contracts using PPA. 7. Procurement of non-consulting services: Procurement of non-consulting services is not foreseen for the project. However, if at a later stage non-consulting services become apparent for project 79 execution, DMRD will carry out such procurement using bidding procedure acceptable to the Bank and as agreed in the Procurement Plan. 8. Selection of Consultants: The project would contract a Management Information System (MIS) firm, individual consultants to support project implementation, and possibly services for local training and MIS data entry. For services, short lists of consultants may be composed entirely of national consultants, in which case provisions of paragraph 2.7 of the Consultant Guidelines will apply. For the purpose of the project, Quality and Cost Based Selection (QCBS) Quality Based Selection (QBS), Fixed Budget Selection (FBS), Consultants' Qualification (CQ), Single Source Selection (SSS), Least Cost Selection (LCS) and Individual Consultant Selection (ICS) will be used. The Procurement Plan shall specify the circumstances and threshold under which specific methods and Guidelines/ Rules will be applicable for selection of consultants. 9. Operating Costs: These costs will include incremental operating costs for office utilities, office supplies and stationeries, souvenirs, events, bank charges, advertising costs, and salaries and contractual allowances of contracted staff but excluding salaries of Government officials. B. Assessment of the Agency's Capacity to Implement Procurement 10. Procurement Environment: Following recommendations of the Country Procurement Assessment Report (CPAR) 2002, the Government implemented a Public Procurement Reform Project (PPRP) between 2002-2007 to improve the procurement environment of the country by addressing the identified deficiencies. The Government established a nodal procurement policy agency with its revenue budget, mandated the Public Procurement Act (PPA) 2006 with associated Public Procurement Rules (PPR) 2008 for all public procuring entities containing most features of international good procurement practices; developed standard documents for procurement of goods, works, and services; created a critical mass of procurement professionals (25 national trainers), provided training to about 1800 staff of about 300 organizations. To sustain the reform, the Government has been implementing a follow-on project (PPRP II) since late 2007, focusing largely on the implementation and monitoring of PPA at key sectoral agencies including capacity development, e-tendering, and social accountability. Till date, about 900 staffs have been trained. 11. Notwithstanding the above progress over the past years in reshaping the procurement environment through procurement reform that include harmonized procedures, recently the new Government made few amendments to the PPA that were found not consistent with the Bank's Guidelines and good procurement practices. Bank as well as other key development partners raised concern over those. The Bank for its projects did not accept those provisions as provided in paragraph 3. 12. Procurement Capacity and Risk Assessment: The procurement capacity assessment was carried out in DMRD with web-based Procurement Risk Assessment Management System (P-RAM S). The agency will require adequate procurement staff to manage procurements under the OSC. Furthermore, DMRD needs more training and experience in international competitive bidding (ICB) following Bank guidelines. Like any other Ministry or government agency, the DMRD also is not immune to systemic issues affecting procurement efficiency and performance. In addition to adequate staffing for procurement needs, emphasis also needs to be laid on areas of internal control, documentation, information dissemination, administration of contract including delivery follow up, payments, handling complaints, etc. The project is rated as "Substantial-Risk" from procurement operation and contract administration viewpoint. Several measures need to be introduced to minimize the risk during the implementation of the project. 80 13. Measures for Improving Governance in Procurement. Several measures have been introduced and agreed with the Government toward improving governance in procurement and minimizing procurement risks under the project. The major action plans are: (i) Identifying procurement focal points (PFP) of DMRD; (ii) Servicing of an experienced procurement consultant who will help the agency in preparation of bidding document / request for proposal, bid evolution / technical evaluation, contract negotiation and contract management; (iii) Establishing a functional webpage of DMRD with procurement related information accessible to public; (iv) Establishing a system for handling complaints and a database for recording, monitoring and follow up all the procurement activities under the project; and (v) Introducing a procurement risk mitigation plan (PRMP) through reports submitted to IDA on a periodic (semi-annual or quarterly) basis. 14. Establish a functional webpage. All information pertaining to bidding and procurement above the specified thresholds, as per PPR will be published in CPTU's website in addition to DMRD's website. The information, among others, will include: invitation to bid, bid documents and RFPs ( wherever applicable); latest information on procurement plan/contracts; status of evaluation once completed; contract award information including adequate details; and information covering the poor performance of contractors, suppliers and consultants, including list of debarred firms. The website would be accessible to all bidders and interested persons equally and free of charge. 15. Establish a system for handling complaints. A credible system for handling complaints will be put in place in DMRD. The database of DMRD should also record, monitor and follow up the procurement activities under this project. The salient features of the system will be an oversight of a complaint database, a standard protocol with appropriate triggers for carrying out investigations, and actions taken against involved parties. The system will be developed and managed by the implementing agencies. It will be supported by a database containing information about all procurement steps such as official estimates (global unit prices), all bidders (individuals, companies, joint ventures, owners information, etc.), and all bids. 16. The following actions DMRD has to take under this project and status report of these will also be captured in the PRMP periodic report. Such measures are also described below: i) Alert bidders in pre-bid meeting. DMRD through a notification, will alert bidders during pre-bid meeting on consequences of corrupt practices (fraud and corruption, collusion, coercion, etc.). The alert message, among others, will include that if bidders are found to have adopted such practices, they would be debarred from subsequent bidding processes in conformity with PPA and IDA Guidelines as appropriate. In addition, in the pre-bid meeting, the bidders will be clarified for preparation of bids correctly. ii) Alert internal officers/staff. DMRD, in reference to the provisions in the PPA and the legal agreement to be signed with IDA, will issue alert letter(s) notifying about the fraud and corruption indicators and the possible consequences of corrupt and similar behavior in procurement practices and actions to be taken against the official staff if they get involved in such practices. Moreover, DMRD will highlight that in case of noncompliance or material deviation from IDA'S Procurement Guidelines, IDA may take remedial actions (i.e., withdrawal of fund, declaration of misprocurement) for concerned contracts. 81 iii) Multiple dropping. Multiple dropping of bids (bids submitted in more than one location and opened in one location) will not be permissible for any procurement under this project. iv) Bid Opening Committee (BOC) and Bid Evaluation Committee (BEC). A formal structured BOC will be constituted for each contract package, as per the provisions of the PPA. Pursuant to provisions of The Public Procurement Act 2006, the BEC, will have at least five members with two members from outside the procuring entity with proven track record of experience in procurement; depending on the type of procurement such members shall be either from public offices and/or from professional bodies/ beneficiaries. The BOC will include at least one member of the BEC. v) Bid Opening Minutes. During the same day of bid opening, photocopies of the Bid Opening Minutes (BOM) with readout bid prices of participating bidders will be submitted by DMRD for circulation to all concerned. For prior review packages, such BOM will be shared with IDA. vi) Low competition among bidders and high price of bids. The case(s) of low competition (not solely based on number of bidders) in ICB and NCB cases coupled with high priced bids will be inquired by DMRD. vii) Measures to reduce coercive practices. Upon receiving allegations of coercive practices resulting in low competition, DMRD will look into the matter and take appropriate measures. For prior review contracts, observations of DMRD will be shared with IDA, along with the evaluation reports. DMRD may seek assistance from law enforcing agencies to provide adequate security for bidders during bid submission. For ICB contracts, provision for bid submission through International/national courier services will be allowed and confirmation of the receipt of the bid will be informed to the bidders through e-mail. viii) Rebidding. In case of re-bidding, DMRD will enquire into the matter, record and highlight the grounds of re-bidding (i.e., corruption or similar, high bid prices, etc.) along with recommended actions to be taken. For prior review cases, all such detailed reports will be sent to IDA. ix) Filing and record-keeping. DMRD will preserve records and all documents regarding their public procurement in accordance with provisions of the PPA, and these records will be made readily available on request for audit/investigation/review by DMRD. x) Publication of award of contract in websites. Within two weeks of contract award DMRD will publish in its and CPTU's websites the following information: identity of contract package, date of advertisement, number of bids sold, number of submitted bids, number of responsive bids, brief reasons for rejection of bids, name of the winning bidder and the price it offered, date of notification of award, date of contract signing, proposed completion date of contract as well as brief description of the contract awarded. For contracts following International Competitive Bidding (ICB) procedures, DMRD shall publish notice of award of such contracts in UNDB Online and Development Gateway Market (dgMarket) in accordance with the provisions of IDA Guidelines. 17. The overall project risk for procurement is "Substantial". With the above arrangements, the procurement under the project is likely to be effective and transparent resulting in smooth implementation of the project leading to achievement of the project development objectives, and under current assessment residual risk related to procurement is assessed as "Moderate". 82 C. Procurement Plan 18. A draft procurement plan covering the first eighteen months of the project implementation has been prepared which provides the basis for the procurement methods. It will also be available in the project's database and in IDA'S external website for this project. The Procurement Plan will be updated in agreement with DMRD semi-annually or as required to reflect the actual project implementation needs and improvements in institutional capacity. D. Review by IDA of Procurement Decisions 19. Procurement Review: The review by IDA of procurement decisions and selection of consultants will be governed by Appendix 1 of the Bank's Guidelines. For each contract to be financed by the Credit, threshold for prior review requirements and post review contracts will be identified in the Procurement Plan. During the first 18 months of the project, IDA will carry out Prior Review of the following contracts: (i) For Goods: All the ICB contracts and Direct Contracts irrespective of estimated cost. The NCB Contracts estimated cost equivalent or more than US$300,000. (ii) For Works: All the ICB contracts and Direct Contract irrespective of estimated cost. The NCB Contracts estimated cost equivalent or more than US$500,000 and the first NCB contract irrespective of value. (iii) For Consultant's Services: Prior review will be required for consultants' services contracts estimated to cost US$200,000 equivalent or more for firms and US$50,000 equivalent or more for individuals. All single-source contracts will be subject to prior review by and in agreement with IDA. All Terms of references of the consultants are subject to the IDA's prior review. 20. The above threshold for prior review will be updated after 18 months of credit effectiveness based on performance of DMRD as reflected in the early post review carried out by the Bank or Bank appointed firm. E. Frequency of Procurement Supervision 21. In addition to the prior review supervision to be carried out by IDA, the capacity assessment of the Implementing Agency has recommended semi-annual supervision missions to visit the field to carry out post review of procurement actions. A customized review process will be developed for reviewing procurement at the community level. 83 Attachment 1 Details of the Procurement Arrangement involving international competition 1. Overall Procurement Arrangement Table 8.1: General Procurement Arrangements (US$ Millions) Category ICB NCB Other2/ N.B.F. Total 1. Works 0 0 0 0 0 2. Goods net of tax 1.660 0.135 0.016 0 1.811 3. Consulting Services/ Training 0.831 0.831 4. Operating Costs (including salaries) 2.564 2.564 5. Wage cost 144.75 144.75 2. Detailed Procurement Arrangements Goods Review By Estimated Domestic Ref Contract Selection Bank Expected Bid Cost (Million P-Q Preference Comments No. Description Method (Prior/ Opening Date US$) (yes/no) Post) Computer Equipment G1 0.352 ICB No Yes Yes Mar-2011 (printers, UPS, etc.) G2 Motorcycles 0.518 ICB No Yes Yes Mar-2011 Vehicles (1 SUV, 2 G3 0.790 ICB No Yes Yes Mar-2011 microbuses, 20 pick-up trucks) Various office G4 ­ equipment and 0.151 NCB / NS NO No (various) G9 accessories Services Expected Review By Ref Estimated Cost Proposals Contract Description Selection Method Bank (Prior/ Comments No. (Million US$) Submission Post) Date Financial Management 1 0.078 IC Yes Dec-2010 Consultant S2 Procurement Specialist 0.026 IC Yes Dec-2010 S3 MIS Specialist 0.039 IC Yes Feb-2011 Payment Systems S4 0.078 IC Yes Feb-2011 Specialist S5 Training Specialist 0.039 IC Yes Mar-2011 Services for MIS S6 0.304 QCBS Yes Jun- 2011 Development Services for Local S7 0.237 FBS No Jun- 2011 Training Services for MIS Data S8 0.030 FBS No Jun-2011 Entry 84 Annex 9: Economic and Financial Analysis Bangladesh - Employment Generation Program for the Poorest Project Introduction 1. In Bangladesh, where 40 percent of the percent of the population is poor and 25 percent is in extreme poverty, the population at risk of falling into or deeper into poverty is large. A substantial proportion of the population hovers around the poverty line, so a small shock can have a large impact on the poverty rate. Bangladeshi households commonly face both individual and aggregate shocks. More than half of all Bangladeshi households had been affected by shocks in the ten years prior to the survey, with shocks related to illness the most common.18 In addition to household level shocks, Bangladesh is also one of the most vulnerable countries in the world to natural disasters including floods, droughts, and cyclones, and in large areas of north-west Bangladesh, seasonal hunger (monga) occurs regularly and induces high chronic poverty. Workfare programs such as the EGPP can provide protection from shocks and vulnerability. The Employment Generation Program was introduced by the Government of Bangladesh in 2008 to cushion the negative effects of the food crisis on the poor and was then reconstituted as the EGPP to provide short-term employment during agricultural lean seasons to poor and vulnerable households. 2. This annex analyzes the potential benefits of the program using existing evidence from similar programs in other countries and from the precursor to the EGPP, the 100-Day Employment Generation Program. This annex estimates how cost effective the workfare program is in raising incomes of the poor compared to a counterfactual program in which every household receives a uniform transfer (following the approach in Ravallion 1998).19 The cost effectiveness analysis of the program considers (i) the extent to which the program reaches the beneficiary population (ii) the benefits of the program in terms of net income transfers to poor households and (iii) the value to poor households of community assets created under the program. Improvements in household wellbeing as a result of insurance against income risks, such as asset protection and the avoidance of negative coping behaviors, as well as the social benefits of the program are also discussed but are not included in the cost-effectiveness analysis. This analysis thus presents a very conservative estimate of the cost effectiveness ratio (the share of program outlays that benefit the poor) and concludes that the project is economically and financially sustainable. Cost effectiveness analysis 3. This section follows the approach to appraising workfare programs in Ravallion (1998) in which the share of benefits from a workfare program that accrue to poor households is calculated and compared to a counterfactual program in which an unconditional transfer is made to all households. Under this comparison, a workfare scheme should not be supported if the total gain to the poor as a proportion of a given budget allocation is less than the percentage of households who are poor, as in that case an untargeted transfer to the population would provide superior poverty alleviation. 4. In the analysis, the proportion of total public expenditure on the program that accrues as net income gains to the poor is decomposed into the following components, following Ravallion (1998). How much of the gains of the program are realized by the poor depend on several variables: 18 Quisumbing, A. 2007. "Poverty transitions, shocks, and consumption in rural Bangladesh: Preliminary results from a longitudinal household survey." Background paper for Bangladesh Poverty Assessment. Mimeo. The World Bank. 19 Ravallion, Martin, 1998. "Appraising workfare programs." Policy Research Working Paper Series 1955, The World Bank. 85 a) The labor intensity of the program, or wage costs as a share of total operating cost, will determine the share of transfers to beneficiaries. In the EGPP, non-wage costs will be between 10-20% of total costs. These figures lead to a relatively high labor intensity, between 0.80 and 0.90. For the cost effectiveness calculations, a labor intensity of 0.80 is assumed. A recent review of public works programs found that in almost two-thirds of the public works projects they surveyed the labor intensity was higher than 60%, so the estimated labor intensity for the EGPP is very reasonable in the class of public works programs.20 b) The targeting performance of the program, which is defined here as the share of wages that went to beneficiaries in the poorest two deciles of the population by estimated consumption. Evaluations of the 100 Days Program found that the program was well-targeted, especially relative to other safety nets programs in Bangladesh and given the short timeline for preparation and implementation of the emergency program. Approximately 67% of benefits went to the poorest 40% of the population and 37% went to the poorest 20% of the population. Given that the current iteration of the program implements several reforms to targeting procedures, including improved targeting to the upazila level based on the poverty map, it is expected that targeting in the EGPP will improve over time. The EGPP is targeted geographically to vulnerable areas and to particularly poor and vulnerable households, as shocks tend to have a more severe impact on the landless and those with lower endowments of education or low asset ownership.21 In addition, areas at high risk of natural disasters are targeted, as these areas tend to be poorer and have lower access to markets and infrastructure, which can exacerbate the impact of shocks and contribute to poverty traps. The cost-effectiveness calculations assume that 80% of EGPP participants are poor, which is the targeting performance necessary to meet the household targeting DLI. c) The net wage gain, which is household income with program participation minus the income that would have been expected in the absence of program participation, normalized by actual household income. The maximum gross annual transfer under the EGPP is 10,000 taka, or US$143 per household. The transfer amount is over 27 percent of average per capita income in Bangladesh. Evaluations of the Trabajar workfare project in Argentina found that net wage gains were about 50% of the transferred amount,22 and the net income gains from India's NREG program are predicted to be around 51 percent of pre-NREG lean season income for the bottom expenditure quintile.23 The cost effectiveness calculations assume net wage gains of between 0.51 and 0.60, which provides a conservative estimate of the expected net wage gain under EGPP. The EGPP operates only when the private labor market for agricultural labor is slack and targets landless agricultural wage laborers. Thus, the expected income of participants is likely to be significantly higher than in the absence of program participation. 20 del Ninno, Carlo and Kalanidhi Subbarao and Annamaria Milazzo (2009) "How to Make Public Works Work: A Review of the Experiences." World Bank Social Protection Discussion Paper 48567. The World Bank. 21 Ahmed Shaikh. 2007. "Social Safety Nets in Bangladesh." Background paper for Bangladesh Poverty Assessment. Mimeo. The World Bank. 22 Jalan, Jyotsna and Ravallion, Martin, 1999. "Income gains to the poor from workfare: Estimates for Argentina's TRABAJAR Program." Policy Research Working Paper Series 2149, The World Bank. 23 Murgai, Rinku and Ravallion, Martin, 2005. "Is a guaranteed living wage a good anti-poverty policy?" Policy Research Working Paper Series 3640, The World Bank. 86 d) The budget leverage of the program. General revenue funds could leverage other financing to the extent that the government is able to obtain co-financing of projects from the non-poor. The cost-effectiveness calculations assume, conservatively, that all financing is from general revenue. e) The indirect benefits for the poor from the assets created through the program. The benefit of assets comes from both the share of benefits that accrue to the poor from the assets created and the benefit-to-cost ratio of the assets, which is the social benefit of the asset created divided by the cost of asset created. Ravallion (1998) assumes that the share of benefits that accrue to the poor is only one quarter of the total benefits (so three quarters of the benefits from assets created through the program accrue to the non-poor). As the EGPP is geographically targeted to higher poverty areas, the benefits from asset creation are likely to benefit the poor to a greater extent. In the cost effectiveness analysis, the share of benefits that accrue to the poor is conservatively assumed to be half of the total benefits. Ravallion (1998) assumes that the social benefit-to-cost ratio is one half, so the assets are not efficiently created. The introduction of non-wage costs into the EGPP should improve the benefit-to-cost ratio of asset creation under the program. The cost-effectiveness analysis uses a benefit-to-cost ratio of 0.60 to 0.75, which assumes that there is improvement in the ratio due to the introduction of non-wage costs. 5. Evidence from the 100 Days Program and from other impact evaluations of public works programs in low income setting suggest that asset creation of the type expected with the EGPP are likely to offer positive benefits to communities. Table 9.1 provides the distribution of the types of community infrastructure projects completed in the 100 Days Program. The bulk of investments were in earthworks, typically rural road construction and rehabilitation and construction of embankments (for flood control). Table 9.1: Distribution of works undertaken Type of work Percentage of total works Road/ barrage construction /re-construction 69.4% Ground raising 12.3% Compost heap preparation 9.8% Digging /redigging ponds 3.3% Garbage and water removal 1.8% Drain digging 1.6% Others 1.0% Grass and vegetable cultivation 0.3% Household base raising 0.3% Reconstruction of houses affected by natural disasters 0.2% Total 100% Sources: NFPCSP and BRAC, 2008. 87 Though a rigorous cost-benefit analysis was not conducted for the projects completed under the 100 Days Program, a qualitative study of a subset of the infrastructure projects suggests that the chosen projects were generally not ad-hoc and tended to reflect real local needs.24 6. An impact assessment carried out for a sample of projects from the Ethiopia PSNP suggests that the public works created through the PSNP, which are similar to the EGPP projects, are delivering significant economic benefits to communities. Table 9.2 shows the average benefit/cost ratio for some categories of activities. As the roads constructed by the PSNP generally consisted of small additional sections to existing roads (as is also likely to be typical in the Bangladesh EGPP), it did not prove possible to determine benefit-to-cost ratios for these projects. However, data on time-savings due to the PSNP road segments were as follows: reaching health post - 17.8 minutes; kebele office - 7.1 minutes; market - 18 minutes, school - 16.3 minutes. Table 9.2: Sample of Public Works Outputs Completed in 2007 and Benefit/Cost Ratio for Ethiopia PSNP Activity Benefit/Cost ratio weighted average Soil and Water Conservation 1.8 Water Supply Projects 3.7 Health Post Construction 1.8-2.2 Source: PSNP PW Impact Assessment 2009. 7. The results of the cost effectiveness analysis are summarized in Table 9.3. Column 0 of Table 9.3 replicates the base-case assumptions in Ravallion (1998) for a low-income country. Columns 1 through 5 use data and assumptions specific to the EGPP that are justified in the discussion above to estimate the cost-effectiveness of the program. Column 1 uses assumptions for the current Government EGPP implementation with targeting performance drawn from the 100 Days Program, values for labor intensity and indirect benefits that reflect no provision for non-wage costs, and net wage gains conservatively estimated to be 0.51 as in Ravallion and Murgai (2005) for India's NREG program. In Column 1, the cost effectiveness ratio of the current and future value of the assets is slightly less than the poverty rate of the target population (the rural poor), so the EGPP is a slightly less cost effective mechanism for reaching the poor than the counterfactual of an unconditional cash transfer to all households (under which only 44 percent would reach the poor). The cost of transferring a one dollar gain to the poor is estimated at $2.50 under the current Government EGPP. 8. Columns 2 through 5 show cost effectiveness ratios calculated using a range of conservative assumptions for the EGPP after the Bank supported project is implemented. The labor intensity of the EGPP decreases from 0.90 in Column 1 to 0.80 in Columns 2 through 5 due to the introduction of non- wage costs. The provision for non-wage costs in turn alters the social benefit-to-cost ratio by improving the longer-term value of the assets created through the program. In addition, household targeting is assumed to improve with Bank support so that 80 percent of participants in the program are poor (the targeting outcome that would meet the household targeting DLI). Expected improvements in the geographic targeting of the program are reflected in the assumption that the share of benefits from assets created under the EGPP that accrue to the poor rises from 0.25 to 0.50. 9. In the most conservative scenario (Column 2), the cost effectiveness ratio of the current and future value of the assets is 0.63, which is significantly greater than the poverty rate of the target population (0.44). Thus, the EGPP is a more cost effective mechanism for reaching the poor than the counterfactual of a transfer to all households (under which only 44 percent would reach the poor). The 24 Serajuddin, Umar, Tara Vishwanath, Syed Hashemi, Hassan Zaman and Qaiser Khan (2008). "Findings from the initial implementation of the Bangladesh Employment Guarantee Scheme." 88 cost of transferring a one dollar gain to the poor in this case is $1.60. Under slightly less conservative but very reasonable assumptions in Columns 3 through 5, the Bank-supported EGPP has a cost effectiveness ratio of between 0.68 and 0.76 and the cost of transferring one dollar to the poor is between $1.30 and $1.50. In all of these cases, the EGPP significantly outperforms the counterfactual program in reaching the poor. 89 Table 9.3: Cost-effectiveness of the Bangladesh EGPP Column Column Column Column Column Column 0 1 2 3 4 5 Data source Ravallion current 1998 LIC EGPP with Bank support EGPP base case Rural poverty rate: upper rate 0.50 0.44 0.44 0.44 0.44 0.44 Upper poverty line for rural areas 2005 HIES Budget leverage 1.0 1.0 1.0 1.0 1.0 1.0 Assume no co-financing of the program by the non-poor; the project is entirely financed from general revenue Labor intensity 0.50 0.90 0.80 0.80 0.80 0.80 Assuming 10% non-wage and administrative costs in C1; assume 20% non-wage and administrative costs in C2-C5 Targeting 0.75 0.67 0.80 0.80 0.80 0.80 C1 from 100 days program evaluation; C2-C6 assume improvement due to better geographic and household targeting Net wage gain 0.75 0.51 0.51 0.51 0.60 0.60 C1-C3 estimate from Murgai and Ravallion (2005) for India NREG; C4-C5 assume increase due to improvements in program administration Poor people's share of total 0.25 0.25 0.50 0.50 0.50 0.50 C1 assumption from Ravallion (1998) for low-income indirect benefits countries, C2-C6 assume improvement due to geographic targeting Benefit/cost ratio (ratio of 0.50 0.50 0.60 0.75 0.60 0.75 C1 assumption from Ravallion (1998) for low-income social benefits to cost) countries, C2-C6 assume improvement due to introduction of non-wage costs Current + future gains to the 0.41 0.43 0.63 0.70 0.68 0.76 poor per $ of spending Cost of $1 gain to the poor $ 2.50 $ 2.30 $ 1.60 $ 1.40 $ 1.50 $ 1.30 Current earnings gain per $ of 0.28 0.31 0.33 0.33 0.38 0.38 program spending Cost of $1 extra current $ 3.60 $ 3.30 $ 3.10 $ 3.10 $ 2.60 $ 2.60 earnings 90 Other program impacts 10. The EGPP can affect household wellbeing in both the short and long term in several additional ways that are not explicitly accounted for in the cost-effectiveness analysis, including through consumption smoothing, asset protection and the avoidance of negative coping behaviors. Evaluations of the 100 Days Employment Program reported that 77% of surveyed beneficiaries reported an improvement in food consumption, and, on a broader scale, newspapers reported that typically monga-prone areas did not report monga while the program was ongoing (NFPCSP and BRAC, 2008). Evaluations of the 100 Days Program also reported that the program led to investment in productive assets for some beneficiaries. These results concur with evidence from public works programs in other countries. A recent impact evaluation of the Ethiopia PSNP found that public works participants improved food security by 0.40 months relative to comparable households, and the program resulted in a reduction in negative coping strategies and in growth in household livestock holdings. 11. The longer term benefits of public works programs are harder to quantify, and it is difficult to calculate an overall rate of return to such programs, as most evidence from workfare programs cannot measure the expected reduction in the long-term transmission of poverty and vulnerability. Furthermore, increases in human capital induced by the program are difficult to quantify in this context. However, severe shocks often compel the vulnerable to cope with immediate needs by selling productive assets (at low prices), accumulating high interest loans, or removing children from school. All of these adversely impact households' long-term potential. To the extent that workfare programs enable households to address their basic needs in the aftermath of a shock, households would be less likely to take decisions-- such as taking children out of school or selling off productive assets like livestock--that increase the likelihood of chronic poverty in the long run. 12. Workfare programs may potentially also have negative impacts. For instance, public works programs might reduce agricultural production to the extent that program participation diverts labor from productive own farm or other uses. The design of the program mitigates these factors, as the selection criteria explicitly target landless laborers and the program only operates when the private labor market for agricultural labor is slack. Additionally, an evaluation of the Ethiopia PSNP found that participation in the program had no effect on agricultural output, acreage, productivity or fertilizer use, which suggests that public works participation had no disincentive effects on household investments in own production. 13. The EGPP may also provide benefits to some segments of society by helping individuals to overcome social norms that inhibit productive activities or behaviors. Evaluations of the 100 Days Program found that the program was particularly beneficial to women, as women are less likely to be able to migrate for work opportunities (or even to work outside the home). 14. The additional impacts of the program are likely to be positive in aggregate. Thus, the program is likely more cost effective than suggested by the calculations presented in Table 9.3, which focus on the comparison of the EGPP to the counterfactual of an unconditional cash transfer. Fiscal analysis 15. The Government of Bangladesh has increased its commitment to social protection in the wake of recent global food and financial crisis. The pre-cursor program to the EGPP, the 100-Day Employment Generation Program, was launched in September 2008 in response to the food and fuel crises. The initial budget allocation for the 100 Days Program was 20 billion taka, though 9.4 billion taka was actually spent. The initial 100-Day Employment Generation Program served nearly 2 million beneficiaries. If the ongoing program is fully implemented at the current budgeted levels, it will provide 100 days of employment to over 600,000 beneficiaries each year. 91 16. The fiscal impact of the current EGPP is consistent with the fiscal costs of safety net programs in other countries with comparable levels of income and with other countries in the region. If the program is fully implemented then the Government is expected to spend 10 billion taka (US$145 million) for the current fiscal year and in succeeding years. Total public spending on safety net programs is approximately 2.5 percent of GDP. The EGPP budget is adequately funded and fits within the Government's overall macroeconomic framework. Therefore, the fiscal impacts of the EGPP appear manageable. 92 Annex 10: Safeguard Policy Issues Bangladesh - Employment Generation Program for the Poorest Project Introduction 1. The following Annex summarizes the main issues pertaining to environmental and social concerns under the program. It gives a description of the likely risks of the project; the safeguard policies that apply; the main actions already implemented by the executing agencies to mitigate/eliminate the risk, and how the project would provide environmental and social guarantees. Environmental Safeguards 2. The project will support employment generation during the lean period of the year for the poorest sections of the populations in the project sub-districts, through a participatory and community-driven process. Approximately 37,000 sub-projects are envisioned over the duration of the project and most activities would be completed in a span of two to three months. In exceptional cases the subproject implementation period may be extended. Based on the nature and scale of these projects, no significant, large-scale, and irreversibly adverse impacts on the environment are expected. Still, to avoid potentially adverse environmental impacts and enhance environmental outcomes of the activities implemented under individual subprojects, the World Bank Operational Policy on Environmental Assessment (OP 4.01) is triggered. The project has been identified as Category `B', since the specific sub-projects' environmental impacts cannot be precisely identified before sites are selected. The sub-projects may also have environmental impacts if not properly designed, executed or if mitigation measures are not implemented. 3. The sub-projects will be selected by the local committees based on the demand following clear selection criteria. The project will support mainly the earthworks which will help the agriculture productivity, better rural communication, protection during natural disaster etc. The subprojects mainly include: i) canal excavating/re-excavating, ii) earth dam construction/re-construction, iii) rural road construction/re-construction, iii) drain construction/re-construction to address water logging, iv) land filling of community institutions like schools, graveyard, prayer ground etc., v) earthen shelter for animals to project against cyclone, vi) excavation of public ponds/fish firms, vii) organic fertilizer production for agriculture application, viii) further development (height increase) of market/helipad, and ix) water reservoirs construction for rainwater conservation/drinking water. The project is expected to bring specific environmental benefits through improved localized drainage systems, introducing organic soil fertilizer, water conservation, increased vegetation cover, and orientation of government and project staffs in environmental issues and better practices. 4. The activities to be supported by the EGPP cannot be precisely identified ahead of time and according environmental assessment was not possible at this stage. An Environmental Management Framework (EMF) has been developed for the project. The implementation experiences of Social Investment Program Project (SIPP) and Local Governance Support Project (LGSP) have been reviewed and the findings are used in preparing the EMF of the EGPP. A simplified approach has been adopted in the EMF to ensure integration of environmental issues with these small scale sub-projects identified and implemented at grass-root level. The EMF has the following features: (a) procedures for screening; (b) tools for assessment at a subproject level; (c) training requirements, development of communication materials; (d) monitoring requirements; (e) and reporting procedures to be followed. All the subprojects will necessarily be screened based on the requirements of the EMF prior to sanctioning. The EMF prescribes the environmental management parameters to be included in the project monitoring and evaluation plan, and environmental audits to periodically evaluate the adequacy of the EMF and to modify it as required. It is recommended to conduct a supplemental study to explore options for project- 93 supported activities that would enhance the capacity of the local communities for adaptation and living intelligently with disasters. 5. The implementing agency and other stakeholders of the project have limited or no experience on environmental assessment and management. However, considering the project's objective to create emergency employment generation of the rural poorest segment of people, the capacity building initiative will be limited to the EMF implementation. The project staffs will receive briefing on EMF during their orientation for project implementation. 6. The screening of the subproject rests on the Union Parishad Project Implementation Committee (UP-PIC) during the selection of projects in consultation with community and ward members. The Upazila Field Facilitator (hired on a contractual basis by the Project) will review the environmental screening, suggest mitigation measures (if required) and ensure no activities with significant negative impact will be implemented. A simple Environmental Management Plan (EMP) may be required in some of the cases. The Upazila Project Implementation Committee (UPIC) will be responsible for providing necessary technical input to the UP-PIC and overall supervision and monitoring of the subprojects. As part of the monitoring of the sub-project implementation, the Project Implementation Officer (PIO) of each Upazila will monitor the environmental compliance issue. The PIO at Upazila level and the Deputy Director at Office of the Project Director (OPD) will maintain a database for sub-project specific environmental screening, and mitigation measures. The project M&E system will capture that information as well. In addition, the project's third party evaluation (reputed local think-tank) will include a brief environmental audit to assess and evaluate the quality of environmental compliance of the subprojects. Social Safeguards 7. Sub-projects implemented through the Program are intended to be very small scale and implemented over brief spells to protect the poorest. No private land acquisition is expected to take place under the Program. Since most sub-projects will be carried out in rural settings where squatters are not widely found, no large scale displacement is expected. However, some sub- projects may require limited land acquisition or may displace encroachers/squatters as a result of project works. The program thus triggers the safeguards policy on involuntary resettlement (OP4.12). 8. There are approximately 2-3 million indigenous people residing in Bangladesh, belonging to more than 45 communities and tribes. The nation-wide approach of the project makes it likely that the project will touch upon areas with indigenous people (IPs), thereby triggering safeguards policy Indigenous Peoples (OP 4.10). 9. A Social Management Framework (SMF) has been put together that is similar in structure to the EMF. It includes a Resettlement Policy Framework and an Indigenous Peoples Planning Framework. The SMF, like the EMF, follows the best practices from the Local Government Support Project and the SIPP in Bangladesh. The implementing agency is expected to use the screening procedures outlined in the SMF to identify, assess, evaluate, mitigate and monitor social impacts of each sub-project. The Social Management Framework (SMF) approach complies with World Bank policies as well as national policies, while considering the practical aspects of the program's implementation. The Bank team will provide training to core social staff in the PIU on the implementation of social safeguards. 10. The SMF include the following key points: 94 a. The PIOs will be responsible for screening all of their sub-projects using a check-list to identify possible social impacts related to either land acquisition resettlement or the presence of IPs in the project areas. b. No involuntary land acquisition is anticipated under the project, but if small quantities of land are required for some sub-projects, it may be acquired on a voluntary basis on willing buyer- willing seller terms where permissible. All land transactions will be recorded. Where direct purchase is not permissible/feasible and land acquisition is needed, all consultation and compensation for land transactions will be carried out in accordance with the National Ordinance on Land Acquisitions and Requisition 1982, and World Bank OP 4.12 on Involuntary Resettlement. The RPF provides guidelines to comply with both processes. c. Although no land acquisition or significant displacement of squatters is anticipated under this project (as most projects will take place in rural settings where squatters are not widely found), if the sub-projects trigger land acquisition and /or are expected to impact squatters or economically displace less than fifty families, an Abbreviated Resettlement Plan (RP) will be prepared by the implementing agency using the procedures and guidelines outlined in the SMF. It is expected that the vast majority of sub-projects will fall in this category. PIOs will be responsible for ensuring compliance with Bank policies (incorporated in the Resettlement Policy Framework), including producing the RP and keeping records of actions undertaken. The Bank will sponsor spot-checks through a random audit based on a sample of projects to ensure that adequate safeguards were in place for any sub-projects triggering land acquisition and/or displacement. d. For sub-projects that trigger the displacement of more than fifty families, a brief Social Impact Assessment will also be carried out, based on which a more detailed RP will be produced. The implementing agency may hire specialist consultants to assist the PIO in preparing the RP, which will be cleared by the PD's office. For this batch of projects as well as those involving impacts on Indigenous People (IP) a third-Party audit will be undertaken to assess compliance with the requirements. The PIO will be responsible for carrying out implementation of the RPF. The Bank will review the results of the audits and conduct spot- checks on a sample basis. e. During the periods of the employment generation scheme implementation, a "mobile M&E" team will be employed to support capacity building and assist with implementation of safeguards mitigation plans. The M&E team can be staffed with project personnel and/or external consultants as needed. 11. The purpose of the Resettlement Policy Framework (RPF) is to provide a framework for project implementation agencies to follow in order to identify impacts and eligible affected people, and ensure mitigation measures for those impacts, once the subproject sites have been identified. The RPF describes the project, legal framework, applicable laws and policies, the process of consultation, types of impacts and eligibility criteria. It includes an Entitlement Matrix and Implementation Arrangements. 12. The Indigenous Peoples Planning Framework (IPPF) details the process to comply with OP 4.10. It has a particular focus on consultative processes that are culturally acceptable and sensitive to the needs, customs and norms of IPs, while fitting the extreme time constraints the project faces (40/60 days for each phase). 95 Annex 11: Project Preparation and Supervision Bangladesh - Employment Generation Program for the Poorest Project Planned Actual PCN review October 15, 2009 October 15, 2009 Initial PID to PIC November 20, 2009 January 20, 2010 Initial ISDS to PIC November 20, 2009 Appraisal July 12, 2010 July 13, 2010 Negotiations August 10, 2010 October 19-20, 2010 Board/RVP approval November 30, 2010 Planned date of effectiveness January 15, 2011 Planned date of mid-term review September 12, 2012 Planned closing date June 30, 2014 Key institutions responsible for preparation of the project: Ministry of Food and Disaster Management Bank staff and consultants who worked on the project included: Name Title Unit Maitreyi B. Das Sr. SP Specialist/ TTL SASSP Shaikh Shamsuddin Ahmed Sr. Economist SASSP Oleksiy Sluchynskyy Sr. Economist SASSP Jay Pascual Consultant/ Counsel LEGES Thao Le Nguyen Senior Finance Officer CTRFC Chau-Ching Shen Senior Finance Officer CTRFC Hassan Zaman Lead Economist PRMPR Burhanuddin Ahmed Sr. Financial Mgmt. Specialist SARFM Furqan Ahmad Saleem Sr. Financial Specialist SARFM Tanvir Hossain Procurement Specialist SARPS Shakil A. Ferdausi Environmental Specialist SASDI Jessica Leino Young Professional/Economist SASHD Gertrude Cooper Program Assistant SASSP Md. Mahtab Alam Program Assistant SASSP Sandra X. Alborta Program Assistant SASSP Ashiq Aziz ETC/Operations Analyst SASSP Nadia Shamin ETC/ Environmental Specialist SASDI Sabah Moyeen ETC/Social Dev. Analyst SASDS Denis Nikitin ETC/Research Analyst SASSP Nadia Sharmin ETC/Environment Analyst SASDI Pravesh Kumar Consultant/Economist SASHD Ashish Joshi Consultant/MIS/Payment System SASHD Shirin Jahangeer Consultant SASDS Joost Gorter Consultant/M&E SASHD Bank funds expended to date on project preparation: 1. Bank resources: US$308,817 2. Trust funds: US$14,377 3. Total: US$323.194 Estimated Approval and Supervision costs: 1 Remaining costs to approval: US$400,00 1. Estimated annual supervision cost: US$100,000 96 Annex 12: Documents in the Project File Bangladesh - Employment Generation Program for the Poorest Project 1. Bank Staff Assessments Aide Memoire. Identification Mission. July 2009. --------. Preparation Mission. January 2010. --------. Pre-Appraisal Mission. April 2010. Back-to-Office Report. September 2009 Report from a Field Visit to Khulna, April 2010 Bangladesh Country Assistance Strategy. 2006 Project Concept Note. October 2009 Project Information Document. October 2009 Joshi, Ashish (2010). Assessment and Feasibility Study Report: Options to strengthen record keeping and payment systems of the EGPP program Kumar, Pravesh (2010). Institutional Capacity and Training Assessment of the Disaster Management and Relief Division (DMRD), Ministry of Food and Disaster Management, Government of People's Republic of Bangladesh. (Draft). Telephone Survey of UP Chairmen (Draft), 2010 Serajuddin, Umar, Tara Vishwanath, Syed Hashemi, Hassan Zaman and Qaiser Khan (2008). Findings from the initial implementation of the Bangladesh Employment Guarantee Scheme. [PPT] World Bank, 2009. An Assessment of the First Phase of the 100-Day Employment Generation Program in Bangladesh. Policy Note, SASEP and SASHD. 2. Other BRAC and NFPCSP. 2008. Study on the first phase of the 100-Day Employment Generation Programme (EGP). Khatun, F., K.M. Rahman, and A. Nabi. 100-Day Employment Generation Program in Bangladesh: Challenges of Effective Implementation. Paper presented at the Center for Policy Dialogue (CDP). Conference on "Development with Equity and Justice: Immediate Tasks for the Newly Elected Government," March 28-29, 2009, Dhaka. 97 Annex 13: Statement of Loans and Credits Bangladesh - Employment Generation Program for the Poorest Project Difference between expected and actual Original Amount in US$ Millions disbursements Project ID FY Purpose IBRD IDA SF GEF Cancel. Undisb. Orig. Frm. Rev'd P103999 2010 Chittagong Water Supply Improvement 0.00 170.00 0.00 0.00 0.00 170.49 0.00 0.00 P073886 2010 Social Investment Program Project II 0.00 115.00 0.00 0.00 0.00 115.00 0.00 0.00 P090807 2010 Skills and Training Enhancement Project 0.00 79.00 0.00 0.00 0.00 78.78 0.00 0.00 P093988 2009 Dhaka Water Sup &San. Project 0.00 149.00 0.00 0.00 0.00 140.29 22.85 0.00 P095965 2009 Siddhirganj Peaking Power Project 0.00 350.00 0.00 0.00 0.00 338.27 121.80 0.00 P098151 2009 Clean Air and Sustainable Environment 0.00 62.20 0.00 0.00 0.00 58.83 -2.00 0.00 P106161 2009 Secondary Educ Qlty & Access 0.00 130.70 0.00 0.00 0.00 76.56 -19.19 0.00 Enhancement P106216 2009 Higher Education Quality Enhancement 0.00 81.00 0.00 0.00 0.00 77.98 4.34 0.00 P106332 2009 Disability and Children at Risk 0.00 35.00 0.00 0.00 0.00 31.32 5.73 0.00 P111272 2009 Emergency 2007 Cyclone Recovery &Rest 0.00 184.00 0.00 0.00 0.00 171.92 14.59 0.00 Pr P040712 2008 Water Management Improvement Project 0.00 102.26 0.00 0.00 8.08 84.25 11.61 8.88 P098146 2008 Public Procurement Reform Project II 0.00 23.60 0.00 0.00 0.00 14.68 4.03 0.00 P084078 2008 National Agricultural Technology Project 0.00 62.60 0.00 0.00 0.00 43.84 1.93 0.00 P102305 2007 Avian Flu Preparedness 0.00 16.00 0.00 0.00 3.83 10.16 -0.59 0.00 P098273 2006 Local Governance Support Project 0.00 111.50 0.00 0.00 19.96 30.89 6.60 0.00 P089382 2006 Investment Promotion Financing Facility 0.00 307.00 0.00 0.00 0.00 258.93 -15.95 0.00 P074841 2005 HNP Sector Program 0.00 300.00 0.00 0.00 0.00 36.24 21.84 0.00 P086791 2004 Reaching Out of School Children Project 0.00 86.00 0.00 0.00 0.00 40.11 2.00 3.00 P086661 2004 BD - Water Supply Program Project 0.00 40.00 0.00 0.00 25.03 3.55 25.42 2.53 P081969 2004 Enterprise Growth & Bank Modernization 0.00 250.00 0.00 0.00 0.00 62.03 50.85 0.00 P078707 2004 Power Sector Development TA 0.00 15.50 0.00 0.00 0.00 6.27 5.28 0.00 P074966 2004 Primary Education Development Program 0.00 150.00 0.00 0.00 0.13 27.56 18.67 0.00 II P071435 2003 Rural Transport Improvement Project 0.00 210.00 0.00 0.00 0.00 38.89 -0.54 3.36 P053578 2003 Social Investment Program Project 0.00 101.24 0.00 0.00 0.00 54.74 -26.23 24.39 P062916 2003 Central Bank Strengthening Project 0.00 37.00 0.00 0.00 0.00 28.11 23.37 0.00 P071794 2002 Rural Elect. Renewable Energy Dev. 0.00 320.98 0.00 0.00 0.04 44.52 -124.08 -80.31 P041887 1999 Municipal Services 0.00 205.60 0.00 0.00 1.39 45.92 -23.31 -23.97 Total: 0.00 3,695.18 0.00 0.00 58.46 2,090.13 129.02 - 62.12 BANGLADESH STATEMENT OF IFC's Held and Disbursed Portfolio In Millions of US Dollars Committed Disbursed IFC IFC FY Approval Company Loan Equity Quasi Partic. Loan Equity Quasi Partic. 98 2001 BRAC Bank 0.00 1.63 0.00 0.00 0.00 1.60 0.00 0.00 1997 DBH 1.91 0.65 0.00 0.00 1.91 0.65 0.00 0.00 1991 Dynamic Textile 0.00 0.00 0.00 1.48 0.00 0.00 0.00 1.48 GTFP Dhaka Bank 5.00 0.00 0.00 0.00 5.00 0.00 0.00 0.00 GTFP Eastern Bnk 2.59 0.00 0.00 0.00 2.59 0.00 0.00 0.00 2004 GrameenPhone Ltd 24.00 0.00 0.00 0.00 24.00 0.00 0.00 0.00 2006 GrameenPhone Ltd 59.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1998 IPDC 3.13 0.00 0.00 0.00 3.13 0.00 0.00 0.00 1998 Khulna 10.40 0.00 0.00 11.99 10.40 0.00 0.00 11.99 1998 Lafarge/Surma 35.00 10.00 0.00 0.00 35.00 10.00 0.00 0.00 2000 Lafarge/Surma 0.00 0.00 0.00 15.00 0.00 0.00 0.00 15.00 2003 RAK Ceramics 7.20 0.00 0.00 0.00 7.20 0.00 0.00 0.00 2000 United Leasing 2.57 0.00 0.00 0.00 2.57 0.00 0.00 0.00 Total portfolio: 150.80 12.28 0.00 28.47 91.80 12.25 0.00 28.47 Approvals Pending Commitment FY Approval Company Loan Equity Quasi Partic. 2000 USPCL 0.00 0.00 0.00 0.00 1998 Khulna 0.00 0.00 0.00 0.00 Total pending commitment: 0.00 0.00 0.00 0.00 99 Annex 14: Country at a Glance Bangladesh - Employment Generation Program for the Poorest Project Bangladesh at a glance 12/9/09 P O V E R T Y a nd S O C IA L S o ut h Lo w- Development diamond* B a ngla de s h A s ia inc o m e 2008 P o pulatio n, mid-year (millio ns) 160.0 1,543 973 Life expectancy GNI per capita (A tlas metho d, US$ ) 520 986 524 GNI (A tlas metho d, US$ billio ns) 83.4 1,522 510 A v e ra ge a nnua l gro wt h, 2 0 0 2 - 0 8 P o pulatio n (%) 1.5 1.5 2.1 Labo r fo rce (%) 2.4 2.2 2.7 GNI Gross per primary M o s t re c e nt e s t im a t e ( la t e s t ye a r a v a ila ble , 2 0 0 2 - 0 8 ) capita enrollment P o verty (% o f po pulatio n belo w natio nal po verty line) 40 .. .. Urban po pulatio n (% o f to tal po pulatio n) 26 30 29 Life expectancy at birth (years) 66 65 59 Infant mo rtality (per 1,000 live births) 43 59 78 Child malnutritio n (% o f children under 5) 41 41 28 Access to improved water source A ccess to an impro ved water so urce (% o f po pulatio n) 80 87 67 Literacy (% o f po pulatio n age 1 5+) 53 63 64 Gro ss primary enro llment (% o f scho o l-age po pulatio n) 92 108 98 Bangladesh M ale 89 1 11 102 Low-income group Female 94 104 95 KE Y E C O N O M IC R A T IO S a nd LO N G - T E R M T R E N D S 19 8 8 19 9 8 2007 2008 Economic ratios* GDP (US$ billio ns) 25.6 44.1 68.4 79.6 Gro ss capital fo rmatio n/GDP 16.3 21.6 24.5 24.2 Expo rts o f go o ds and services/GDP 5.6 13.3 19.8 20.3 Trade Gro ss do mestic savings/GDP 9.2 16.7 17.5 15.8 Gro ss natio nal savings/GDP 14.7 24.2 34.6 35.7 Current acco unt balance/GDP -4.3 -1.1 1.4 0.9 Interest payments/GDP 0.7 0.4 0.4 0.3 Domestic Capital savings formation To tal debt/GDP 40.8 35.6 31.9 29.7 To tal debt service/expo rts 22.0 8.7 5.1 4.4 P resent value o f debt/GDP .. .. 22.1 18.4 P resent value o f debt/expo rts .. .. 76.6 61.0 Indebtedness 19 8 8 - 9 8 19 9 8 - 0 8 2007 2008 2 0 0 8 - 12 (average annual gro wth) GDP 4.6 5.7 6.4 6.2 5.7 Bangladesh GDP per capita 2.5 4.0 4.9 4.7 4.3 Low-income group Expo rts o f go o ds and services 13.3 1 1 .2 13.0 7.0 10.9 S T R UC T UR E o f t he E C O N O M Y 19 8 8 19 9 8 2007 2008 Growth of capital and GDP (%) (% o f GDP ) 12 A griculture 31.1 25.4 19.2 19.0 9 Industry 21.2 25.8 28.4 28.5 6 M anufacturing 13.2 16.2 17.8 17.8 3 Services 47.8 48.7 52.4 52.5 0 Ho useho ld final co nsumptio n expenditure 86.4 78.6 76.9 78.9 03 04 05 06 07 08 General go v't final co nsumptio n expenditure 4.3 4.7 5.5 5.3 GCF GDP Impo rts o f go o ds and services 12.7 18.3 26.7 28.8 19 8 8 - 9 8 19 9 8 - 0 8 2007 2008 Growth of exports and imports (%) (average annual gro wth) A griculture 2.7 3.5 4.6 3.2 40 Industry 7.2 7.5 8.4 6.8 20 M anufacturing 7.5 7.2 9.7 7.2 Services 4.1 5.9 6.9 6.5 0 Ho useho ld final co nsumptio n expenditure 3.0 4.2 6.7 5.2 03 04 05 06 07 08 -20 General go v't final co nsumptio n expenditure 4.7 8.5 6.4 3.6 Gro ss capital fo rmatio n 7.9 8.1 8.5 1.8 Exports Imports Impo rts o f go o ds and services 8.8 8.4 16.0 -2.1 No te: 2008 data are preliminary estimates. This table was pro duced fro m the Develo pment Eco no mics LDB database. * The diamo nds sho w fo ur key indicato rs in the co untry (in bo ld) co mpared with its inco me-gro up average. If data are missing, the diamo nd will be inco mplete. 100 Bangladesh P R IC E S a nd G O V E R N M E N T F IN A N C E 19 8 8 19 9 8 2007 2008 Inflation (%) D o m e s t ic pric e s 12 (% change) Co nsumer prices 6.3 7.0 7.2 9.6 9 Implicit GDP deflato r 7.6 5.3 6.8 8.8 6 G o v e rnm e nt f ina nc e 3 (% o f GDP , includes current grants) 0 Current revenue 8.9 9.3 10.5 1 .5 1 03 04 05 06 07 08 Current budget balance 2.9 2.1 1.5 2.0 GDP deflator CPI Overall surplus/deficit -4.1 -4.1 -3.1 -3.6 TRADE 19 8 8 19 9 8 2007 2008 Export and import levels (US$ mill.) (US$ millio ns) To tal expo rts (fo b) 1,232 5,103 12,053 4,1 1 51 25,000 Raw jute 81 108 147 165 20,000 Leather and leather pro ducts 147 238 266 284 M anufactures 825 4,416 1 ,1 8 11 3,1 1 53 15,000 To tal impo rts (cif) 2,986 6,772 5,51 1 1 19,481 10,000 Fo o d 704 369 ,91 1 8 3,492 5,000 Fuel and energy 272 543 2,233 2,753 Capital go o ds 1,090 2,072 1,929 1,664 0 Expo rt price index (2000=100) .. .. .. .. 02 03 04 05 06 07 08 Impo rt price index (2000=100) .. .. .. .. Exports Imports Terms o f trade (2000=1 00) .. .. .. .. B A LA N C E o f P A Y M E N T S 19 8 8 19 9 8 2007 2008 Current account balance to GDP (%) (US$ millio ns) Expo rts o f go o ds and services 1,486 5,810 13,537 16,042 2 Impo rts o f go o ds and services 3,252 8,049 18,250 22,897 Reso urce balance -1,765 -2,239 -4,713 -6,855 1 Net inco me -133 -100 -905 -994 Net current transfers 788 1,876 6,554 8,529 0 Current acco unt balance ,1 0 -1 1 -463 936 680 02 03 04 05 06 07 08 Financing items (net) 1,255 545 557 -349 -1 Changes in net reserves -145 -82 -1,493 -331 M emo : Reserves including go ld (US$ millio ns) .. 1,725 3,614 4,171 Co nversio n rate (DEC, lo cal/US$ ) 31.2 45.4 69.1 68.6 E X T E R N A L D E B T a nd R E S O UR C E F LO WS 19 8 8 19 9 8 2007 2008 Composition of 2008 debt (US$ mill.) (US$ millio ns) To tal debt o utstanding and disbursed 10,448 15,692 21,859 23,644 IB RD 66 36 0 0 G: 1,986 IDA 3,188 6,168 10,077 0,61 1 3 F: 535 To tal debt service 502 644 1,008 1,046 E: 3,103 IB RD 5 7 0 0 IDA 36 104 265 290 B: 10,613 Co mpo sitio n o f net reso urce flo ws Official grants 651 657 1,027 1,746 Official credito rs 783 373 406 1,083 P rivate credito rs -22 -29 -23 1 12 Fo reign direct investment (net inflo ws) 2 190 653 973 D: 6,721 P o rtfo lio equity (net inflo ws) 0 -4 153 1 0 C: 686 Wo rld B ank pro gram Co mmitments 225 655 580 1,807 Disbursements 296 350 602 820 A - IBRD E - Bilateral P rincipal repayments 10 66 193 213 B - IDA D - Other multilateral F - Private C - IMF G - Short-term Net flo ws 286 284 408 607 Interest payments 32 45 72 77 Net transfers 254 239 336 530 No te: This table was pro duced fro m the Develo pment Eco no mics LDB database. 12/9/09 101 IBRD 33368R To Gangtok 88ºE 89ºE BHUTAN 90ºE 91ºE B A N GL A D ESH DISTRICT CAPITALS DIVISION CAPITALS To Dispur Panchagar NATIONAL CAPITAL To PANCHAGAR RIVERS Patna BANGLADESH MAIN ROADS LA L L 26ºN Thakurgaon LM L L NILPHAMARI RAILROADS O N N N THAKURGAON Nilphamari Lalmonirhat IR R RH DISTRICT BOUNDARIES A AT A Kurigram Rangpur DIVISION BOUNDARIES To KURIGRAM Katihar Dinajpur RANGPUR INTERNATIONAL BOUNDARIES DINAJPUR To Goalpara To I N DI A Katihar Gaibandha To 92ºE To Dispur Dispur GAIBANDHA JOYPURHAT SERPUR Joypurhat Sunamganj SYLHET 25ºN Serpur 25ºN NAOGAON NETROKONA SUNAMGANJ Sylhet BOGRA Jamalpur Naogaon Bogra JAMALPUR Netrokona RAJSHAHI Mymensingh SYLHET NOWABGANJ To Silchar MYMENSINGH Jamuna Ja na J Nowabganj RAJSHAHI DHAKA Kishorganj MOULVI BAZAR Serajganj HABIGANJ Moulvi Bazar G an Rajshahi Natore SERAJGANJ KISHORGANJ ge NATORE TANGAIL s Habiganj Tangail GAZIPUR PABNA Pabna Gazipur NARSINGDI I N D IA 24ºN a 24ºN gh n Narsingdi Brahmanbaria Me I NDI A KUSHTIA Kushtia Manikanj DHAKA BRAHMAN BARIA Meherpur Rajbari MANIKGANJ NARAYNGANJ MEHERPUR Chuadanga RAJBARI DHAKA Naraynganj Faridpur COMILLA CHUADANGA Jhenaidah Magura MUNSHIGANJ Munshiganj FARIDPUR G a n ge JHENAIDAH s MAGURA Comilla HARI To SARIATPUR CHANDPUR RACH Calcutta Sariatpur Jessore Narail Madaripur Chandpur Khagrachhari KHAG MADARIPUR NARAIL JESSORE GOPALGANJ Gopalganj Feni 23ºN 23ºN KHULNA LUXMIPUR Luxmipur NOAKHALI FENI Khulna BARISAL Noakhali RANGAMATI Satkhira KHULNA Bagerhat PEROJPUR Jhalukathi Barisal Bhola C H I T TA G O N G SATKHIRA JHALUKATHI Rangamati Kaptai To Calcutta BAGERHAT Perojpur Lake CHITTAGONG BARISAL Patuakhali BHOLA a D e l t PATUAKHALI Chittagong s G a n g e Barguna BARGUNA Bandarban 22ºN s Mt. Mowdok 22ºN an s ge (957 m) arb G an d Sun the BANDARBAN of COX'S ths BAZAR Mou Cox's Bazar 0 10 20 30 40 50 Kilometers Bay of Bengal 0 10 20 30 40 50 Miles MYANMAR This map was produced by the Map Design Unit of The World Bank. 21ºN The boundaries, colors, denominations and any other information 21ºN shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. To 89ºE 90ºE 91ºE 92ºE Sittwe MARCH 2008