Document of The World Bank FOR OFFICIAL USE ONLY Report No: 48633-ET PROJECT APPRAISAL DOCUMENT ON A PROPOSED GRANT IN THE AMOUNT OF SDR 223.5 MILLION (US$350.0 MILLION EQUIVALENT) AND A PROPOSED CREDIT IN THE AMOUNT OF SDR 83.1 MILLION (US$130.0 MILLION EQUIVALENT) TO THE FEDERAL DEMOCRATIC REPUBLIC OF ETHIOPIA FOR A PRODUCTIVE SAFETY NET APL III PROJECT IN SUPPORT OF THE THIRD PHASE OF THE PRODUCTIVE SAFETY NET PROGRAM September 25, 2009 Social Protection Unit Country Department AFCE3 Africa Region 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 August 31, 2009) Currency Unit = Birr (ETB) US$1 = 12.73 ETB US$1 = SDR 0.639 FISCAL YEAR Ethiopian Fiscal Year (EFY) July 8 ­ July 7 EFY02 IDA FY10 ABBREVIATIONS AND ACRONYMS AED Agricultural Extension Directorate APL Adaptable Program Loan BOARD Bureau of Agriculture and Rural Development BOFED Bureau of Finance and Economic Development BPR Business Process Re-engineering CAS Country Assistance Strategy CBHW Community Based Health Workers CBPWDP Community Based Participatory Watershed Development Planning CFSTF Community Food Security Task Force CGAP Consultative Group to Assist the Poor CIDA Canadian International Development Agency CPAR Country Procurement Assessment Report CSA Central Statistical Agency CSRP Civil Service Reform Plan DAs Development Agents DCT Donor Coordination Team DFID UK Department for International Development DRMFS Disaster Risk Management and Food Security DRMFSS Disaster Risk Management and Food Security Sector DS Direct Support DWG Donor Working Group EC European Commission EFY Ethiopian Fiscal Year EFSRA Emergency Food Security Reserve Administration EIA Environmental Impact Assessment EMCP Expenditure Management and Control Program EMP Environmental Management Plan EPA Environmental Protection Agency ERR Economic Rate of Return ESMF Environmental and Social Management Framework EWRD Early Warning and Response Directorate ii FOR OFFICIAL USE ONLY FIC Federal Information Center FM Financial Management FSCD Food Security Coordination Directorate FSP Food Security Program GDP Gross Domestic Product GoE Government of Ethiopia HABP Household Asset Building Program HIV/AIDS Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome ICB International Competitive Bidding ICR Implementation Completion Report IDA International Development Association IFPRI International Food Policy Research Institute IFR Interim Financial Report IPM Integrated Pest Management ISR Implementation Status Report JCC Joint Coordination Committee JRIS Joint Review and Implementation Support JSOC Joint Strategic Oversight Committee KAC Kebele Appeals Committee KFSTF Kebele Food Security Task Force LEAP Livelihood Early Warning Assessment & Protection LIU Livelihood Integration Unit LCS Least Cost Selection MDG Millennium Development Goals MDTF Multi Donor Trust Fund MFI Microfinance Institution MOLSA Ministry of Labor and Social Affairs M&E Monitoring and Evaluation MOARD Ministry of Agriculture and Rural Development MOFA Ministry of Federal Affairs MOFED Ministry of Finance and Economic Development MOU Memorandum of Understanding MTEFF Medium-Term Expenditure and Financing Framework NCB National Competitive Bidding NGO Non-Governmental Organization NRMD Natural Resource Management Directorate OFAG Office of the Federal Auditor General PASDEP Program for Accelerated and Sustained Development to End Poverty PASS Automated Payroll and Attendance Sheet System PBS Protection of Basic Services PCDP Pastoral Community Development Project PDO Project Development Objective PEFA Public Expenditure and Financial Accountability PFM Public Financial Management PPA Public Procurement Agency PIM Program Implementation Manual PLI Pastoral Livelihood Initiative This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization. iii PSCAP Public Sector Capacity Building Project PSNP Productive Safety Net Program PWCU Public Works Coordination Unit PWFU Public Works Focal Units PWIA Public Works Impact Assessment QCBS Quality and Cost Based Selection RCBP Rural Capacity Building Project RED&FS Rural Economic Development and Food Security REPA Regional Environmental Protection Authority RFQ Request For Quotations RFSCO Regional Food Security Coordination Office RIC Regional Information Center RNE Royal Netherlands Embassy RuSACCO Rural Savings and Credit Cooperative Organization RRM Rapid Response Mechanism RUFIP Rural Financial Intermediation Program SEA Strategic Environmental Assessment SIDA Swedish International Development Cooperation Agency SIL Specific Investment Loan SNNP Southern Nations and Nationalities SWC Soil and Water Conservation TA Technical Assistance TLU Tropical Livestock Units VLSA Village Lending and Savings Association UN United Nations USAID United States Agency for International Development USD United States Dollar WAD Women's Affairs Department WFSD Woreda Food Security Desks WFSTF Woreda Food Security Task Force WMS Welfare Monitoring Survey WOARD Woreda Office of Agriculture and Rural Development WOFED Woreda Office of Finance and Economic Development WFP World Food Program Vice President : Obiageli Katryn Ezekwesili Country Director : Kenichi Ohashi Sector Director : Yaw Ansu Sector Manager : Lynne Sherburne-Benz Task Team Leader : Will Wiseman iv ETHIOPIA Productive Safety Net APL III CONTENTS Page A. STRATEGIC CONTEXT AND RATIONALE ................................................................ 1 1. Country and sector issues.................................................................................................... 1 2. PSNP 2005-2009: Implementation and results ................................................................. 6 3. Rationale for Bank involvement ...................................................................................... 11 4. Higher level objectives to which the project contributes ............................................... 12 B. PROJECT DESCRIPTION ............................................................................................. 13 1. Lending instrument ........................................................................................................... 13 2. Program objective and phases.......................................................................................... 13 3. Project development objective and key indicators ......................................................... 15 4. Project components ........................................................................................................... 15 5. Lessons learned and reflected in the project design ....................................................... 19 6. Alternatives considered and reasons for rejection ......................................................... 21 C. IMPLEMENTATION ....................................................................................................... 22 1. Partnership arrangements ................................................................................................ 22 2. Institutional and implementation arrangements ............................................................ 24 3. Monitoring and evaluation of outcomes/results ............................................................. 26 4. Sustainability ..................................................................................................................... 27 5. Critical risks and possible controversial aspects ............................................................ 28 6. Grant/credit conditions and covenants............................................................................ 31 D. APPRAISAL SUMMARY ................................................................................................ 32 1. Economic and financial analysis ...................................................................................... 32 2. Technical ............................................................................................................................ 33 v 3. Fiduciary ............................................................................................................................ 34 4. Social ................................................................................................................................... 36 5. Environment ...................................................................................................................... 37 6. Safeguard policies .............................................................................................................. 39 7. Policy Exceptions and readiness ...................................................................................... 40 Annex 1: Country and Sector or Program Background ......................................................... 42 Annex 2: Major Related Projects Financed by the Bank and/or other Agencies ................. 51 Annex 3: Results Framework and Monitoring ........................................................................ 55 Annex 4: Detailed Project Description...................................................................................... 68 Annex 5: Project Costs ............................................................................................................... 83 Annex 6: Implementation Arrangements ................................................................................. 86 Annex 7: Financial Management and Disbursement Arrangements ..................................... 96 Annex 8: Procurement Arrangements .................................................................................... 110 Annex 9: Governance Dimensions of the PSNP ..................................................................... 121 Annex 10: Economic and Financial Analysis ......................................................................... 125 Annex 11: Safeguard Policy Issues .......................................................................................... 139 Annex 12: Project Preparation and Supervision ................................................................... 143 Annex 13: Documents in the Project File ............................................................................... 147 Annex 14: Statement of Loans and Credits ............................................................................ 149 Annex 15: Country at a Glance ............................................................................................... 151 Annex 16: Map IBRD 37166 .................................................................................................... 153 vi ETHIOPIA PRODUCTIVE SAFETY NET APL III PROJECT PROJECT APPRAISAL DOCUMENT AFRICA AFTSP Date: September 25, 2009 Team Leader: William Wiseman Country Director: Kenichi Ohashi Sectors: Other social services (50%); Public Sector Director: Yaw Ansu administration- Other social services (40%); Sector Manager: Lynne Sherburne-Benz Central government administration (5%); Sub- national government administration (5%) Themes: Social safety nets (60%); Natural disaster management (20%); Vulnerability assessment and monitoring (20%) Project ID: P113220 Environmental category: Partial Assessment Lending Instrument: Adaptable Program Loan Project Financing Data [ ] Loan [X] Credit [X] Grant [ ] Guarantee [ ] Other: For Loans/Credits/Others (US$m): Grant: 350.00; Credit: 130.00 Total Bank financing (US$m.): 480.00 Proposed terms: Standard Terms for IDA Credit and Grant. Financing Plan (US$m) Source Local Foreign Total BORROWER/RECIPIENT 10.00 0.00 10.00 International Development Association (IDA) 130.00 0.00 130.00 IDA Grant 330.00 20.00 350.00 US: Agency for International Development 530.90 0.00 530.90 (USAID) CANADA: Canadian International 71.00 10.80 81.80 Development Agency (CIDA) UK: British Department for International 286.40 37.70 324.10 Development (DFID) EC: European Commission 75.30 3.40 78.70 IRELAND, Govt. of 78.30 2.30 80.60 NETHERLANDS: Netherlands Development 69.10 2.20 71.30 Association SWEDEN: Swedish Intl. Dev. Cooperation 22.00 1.00 23.00 Agency (SIDA) World Food Program (WFP) 50.00 0.00 50.00 Financing Gap 515.86 10.60 526.46 Total: 2,168.86 88.00 2,256.86 Borrower: Ministry of Finance and Economic Development Addis Ababa, Ethiopia Tel: (251-111) 226-698 Responsible Agency: Disaster Risk Management & Food Security Sector, Ministry of Agriculture & Rural Development Addis Ababa, Ethiopia Tel: (251-111) 503-506 Estimated disbursements (Bank FY/US$m) FY 10 11 12 13 14 15 Annual 100.00 100.00 100.00 100.00 70.00 10.00 Cumulative 100.00 200.00 300.00 400.00 470.00 480.00 Project implementation period: Start: October 22, 2009 End: June 30, 2015 Expected effectiveness date: November 30, 2009 Expected closing date: June 30, 2015 Does the project depart from the CAS in content or other significant respects? [ ]Yes [X] No Ref. PAD A.3. Does the project require any exceptions from Bank policies? Ref. PAD D.7. [ ]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 Ref. PAD C.5. Does the project meet the Regional criteria for readiness for implementation? [X]Yes [ ] No Ref. PAD D.7. Project development objective Ref. PAD B.3., Technical Annex 3 The Project Development Objective for APL III is 'improved effectiveness and efficiency of the Productive Safety Net Program (PSNP) and related Household Asset Building Program (HABP) for chronically food insecure households in rural Ethiopia'. Project description Ref. PAD B.4., Technical Annex 4 APL III includes four Components: Component 1: Safety Net Grants will provide cash and in-kind transfers to chronically food insecure households through: (i) labor intensive public works that provide transfers to able-bodied households; and (ii) direct support that provides transfers to labor-poor households. This component will also provide performance incentives to woredas to improve the timeliness and predictability of transfers to households. Component 2: Drought Risk Financing aims to provide timely resources for transitory food insecurity in response to shocks within the existing program areas. This component will be financed using a contingent grant, which will provide resources for scaling up activities under Component 1 in response to localized or intermediate weather or price-related shocks in PSNP woredas. Component 3: Institutional Support for the PSNP will support institutional strengthening activities focusing on (i) program management at all levels; (ii) capacity building to fill any remaining gaps in general and those specific to the Risk Financing facility; (iii) monitoring and evaluation; and (iv) transparency and accountability. Component 4: Support to the HABP will finance a core set of interventions aimed at (i) strengthening the delivery of demand-driven and market-oriented advice for household investments; (ii) improving the efficiency and effectiveness of financial service delivery to food insecure households; and (iii) supporting program management. Which safeguard policies are triggered, if any? Ref. PAD D.6., Technical Annex 11 APL III triggers four safeguard policies: The Environmental Assessment Policy (OP 4.01) related to the possible impacts arising from the implementation of public works; the Pest Management Policy (OP 4.09), predicated on the possibility of small quantities of pesticides being employed in small, community- level irrigation projects; the Physical Cultural Resources Policy (OP 4.11), because although deemed unlikely in view of the small-scale of the sub-projects, the possibility of 'chance-finds' cannot be ruled out; and the International Waterways Policy (OP 7.50), because of small-scale irrigation projects that may be implemented in watersheds of international waterways. Significant, non-standard conditions, if any: Ref. PAD C.6. Grant/credit effectiveness: · Provide evidence of measures taken to establish and operationalize Regional Information Centers covering Afar, Amhara, Oromiya, SNNP, and Tigray Regions. · Provide evidence of measures taken to ensure use of PASS in all participating woredas in Amhara, Oromiya, SNNP and Tigray Regions. · Adoption of a revised PSNP Project Implementation Manual, satisfactory to the Association, and consolidated PSNP work program and budget for EFY 2002. Covenants applicable to project implementation: Conditions of Disbursement · For Component 4 'Support to Household Asset Building', submission of HABP work plan prior to operational launch. At the Mid-Term Review by June 2012 · Evidence that the National Social Protection Strategy has been finalized. · Evidence of an approved Disaster Risk Management Policy. · Independent process evaluation of experience to date with the Risk Financing within the broader emergency response system. · Independent assessments of PSNP based on a representative household and community-level public works surveys. · Independent assessment of effectiveness of systems development for HABP. · Based on an analysis of an updated MTEFF, an agreed set of actions to address any residual financing gap, including any revisions to the program design as may be necessary. A. STRATEGIC CONTEXT AND RATIONALE 1. Country and sector issues 1.1 The origins of the Productive Safety Net Program: reforming the emergency response system 1. Chronic poverty and food insecurity1 remain pervasive in much of Ethiopia. Despite record rates of economic growth in recent years, the depth of poverty in rural Ethiopia remains high. From 1999/00 - 2004/05 Ethiopia experienced a six-percentage point decline in rural poverty. Yet 38.5 percent of rural households still live below the food poverty line. Most of these households are engaged in subsistence farming on small plots of degraded land where they are vulnerable to weather fluctuations. Traditional agricultural techniques predominate and there is limited production for the market. There has been a secular decline in per capita food production as high population growth has contributed to a decline in farm sizes, while environmental degradation has deepened. Although there have been some improvements, use of fertilizer remains limited and there is even less use of improved seed. 2. Dramatic variations in the climate contribute to food insecurity. Rainfall data for the period 1967 to 2000 indicate that Ethiopia's annual variability in rainfall across different zones is among the highest in the world ranging from a low of 15 percent to a high of 81 percent. The larger the variation in rainfall a household is exposed to, the lower its income and consumption. Yet the use of irrigation remains extremely low. Repeated environmental shocks have severely eroded rural livelihoods, leaving households with little capacity to cope. Beyond rainfall shocks, health risks exacerbate the vulnerability of the poor, driving thousands of people into poverty traps. Many households are not able to fully meet their most basic consumption needs even in years when rainfall is adequate. 3. As a result, every year for over two decades the Government has launched international emergency appeals. Although this humanitarian assistance was substantial (estimated at about US$265 million a year on average between 1997 and 2002) and saved many lives, evaluations have shown that it was unpredictable for both planners and households, often arriving too little, too late. The delays and uncertainties meant that the emergency aid could not be used effectively and did little to protect livelihoods, prevent environmental degradation, generate community assets, or preserve physical or human household assets. As a result, despite the large food aid inflows, household-level food insecurity has remained both widespread and chronic in Ethiopia. In fact there has been an increasing trend in chronic food insecurity in the wake of repeated droughts as vulnerable households fail to cope with shocks and slide deeper into poverty. 4. In 2003, building on its National Food Security Strategy, the Government launched a major consultation process with development partners that aimed to formulate an alternative to crisis response to support the needs of chronically food insecure households, as well as to develop long-term solutions to the problem of food insecurity. This culminated in the New Coalition for Food Security that proposed a Food Security Program (FSP) aimed at shifting 1 Food insecurity is a lack of access, at all times, to enough food for an active, healthy life. Chronic food insecurity is the persistence of this state over time such that households are generally unable to meet their own food needs. 1 households out of the emergency relief system while also enabling them to "graduate" to sustainable food security2. 5. As part of the FSP, in 2005 the Government started a major new initiative - the Productive Safety Net Program (PSNP). The PSNP was designed to complement the existing humanitarian appeal system and became the chief instrument for assisting 4.84 million chronically food insecure people in rural Ethiopia. It was scaled up significantly in 2006 and currently reaches 7.57 million people, roughly 10 percent of the total population. The PSNP aims to provide predictable and timely resources to chronically food insecure households through payments to able-bodied members for participation in labor-intensive public works and through Direct Support to labor-poor, elderly or otherwise incapacitated households. This support assists households to smooth their consumption, avoid asset depletion, and plan with greater certainty. Increasingly the Program has shifted away from giving support in-kind towards providing cash transfers that increase flexibility for households and improve the cost-effectiveness of the Program. The PSNP has also focused on mobilizing multi-annual resources from development partners such that adequate planning, capacity building, and monitoring and evaluation systems can be in place. 6. The PSNP public works program is designed to address a key underlying cause of food insecurity - environmental degradation. In many parts of the country, the land base has been severely damaged through erosion and unsustainable land use practices. Land productivity has subsequently declined and rainfall infiltration has fallen such that many spring and stream sources have disappeared or are no longer perennial. Estimates indicate that the gross erosion from cropland has potentially led to a 2-3 percent drop in annual agricultural Gross Domestic Product (GDP). To address this, PSNP public works have focused on soil and water conservation activities, along with roads and irrigation, developed within a participatory watershed management planning framework3. By promoting, financing and implementing sustainable land management measures, the PSNP continues to represent a unique opportunity to contribute to environmental transformation at scale in Ethiopia. The PSNP Public Works Impact Evaluation concluded that soil and water conservation activities are dramatically reducing surface runoff, increasing infiltration, raising groundwater levels, enhancing spring yields, and increasing stream base flows and vegetation coverage. 7. The PSNP aims to contribute to sustainable graduation from food insecurity for a large number of the chronically food insecure. The objective of the Program is to help households smooth their consumption and build productive community assets through public works. The Government has recognized that while this is clearly a necessary condition for promoting a sustainable solution to food insecurity by providing a much-needed stabilizing environment, it is clearly not sufficient. The Government believes that complementary interventions are also required to directly rebuild household assets to increase household productivity and to promote income diversification. As a result, the PSNP has been conceived as one pillar of the Government's broader Food Security Program. 2 Graduation from the PSNP is defined as a household being able to feed itself for 12 months a year, in the absence of program support, as well as being able to withstand modest shocks. 3 Integrated watershed management is a planning tool to identify and prioritize all public works investments within a given geographic area. 2 1.2 Moving towards a sustainable solution to food insecurity: the broader Food Security Program 8. Beyond the PSNP, the FSP comprises three other components. Firstly, the Household Asset Building Program (HABP) has been financed through a Federal block grant to Regions and the development partner-financed Food Security Project4, amounting together to roughly US$100.0 million per year. Households are provided a one-time highly subsidized credit that ranges from US$200-700 to rebuild their asset base (in the case of the Food Security Project which targets the poorest of the poor) or to purchase "household extension packages". These packages usually consist of various combinations of agricultural inputs based on a business plan developed with support from the extension service. Credit is channeled through multipurpose cooperatives as well as the government administrative system and microfinance institutions (MFIs). Since 2006, PSNP households have been prioritized for support ensuring basic complementarity between the two programs. There is evidence that the combination of the PSNP with HABP can provide a pathway to food security for some households (see Annex 1). 9. Secondly, since 2005 regional governments have made investments of approximately US$200 million in community assets as part of the FSP. Spending has focused on larger-scale, more capital intensive investments such as medium size irrigation projects that are designed to create an enabling environment for food security. This is seen to be particularly important among pastoral communities where the other components of the FSP are perhaps less relevant. 10. Thirdly, the Government has also invested in the Resettlement Program. To date, 188,874 households have been voluntarily resettled to the western parts of Amhara, Oromiya, Southern Nations and Nationalities (SNNP), and Tigray Regions. These represent 43 percent of the Government's target of 440,000 households. The Government views resettlement as an effective instrument for improving the food security status of poor households, although competing demands on dwindling land resources are leading to a re-thinking of this strategy. A review completed in 2009 suggests that the Government will continue the resettlement program but in a significantly scaled-down form that focuses on consolidating investments in existing receiving sites. 11. Taken together the components of the FSP represent an enormous public investment in food security in Ethiopia. Government has demonstrated serious commitment to making food insecurity a public priority with the investment of approximately US$700 million over the last five years. Over the same period development partners have invested approximately US$1.7 billion, largely in the PSNP. 12. At the same time, Government set highly ambitious targets for these programs with the goal of graduating over 5 million individuals from food insecurity by 2009. These targets were not evidenced-based but rather reflected the Government's desire to signal to beneficiaries and implementers that it sought an end to the problems of destitution and dependency. While these motives were understandable, they did not reflect what the FSP could in reality deliver. As of 2009, around 280,000 individuals have graduated from the PSNP. This does not represent a failure of the FSP, but rather suggests that strengthening livelihoods to the extent that households 4 The Food Security Project is financed by the World Bank, CIDA, and Italian Cooperation (see Annex 2). 3 are no longer food insecure and resilient to shocks is a longer and more complex process than was suggested by the initial five-year timeframe of the FSP. 1.3 Promoting sustainable graduation from food insecurity: reformulating the FSP and focusing on the broader enabling environment 13. In 2008, the Government recognized that a longer-term perspective was needed and proposed to development partners a new five-year phase for the FSP, which includes the PSNP, be launched in 2010. For the new phase, Government has again proposed targets for graduation that are unlikely to be fully realized, envisaging a scenario in which up to 80 percent of beneficiaries graduate. In the face of these expectations, it is important to examine why graduation has been limited despite a period of rapid economic growth, and what are the prospects for higher rates of graduation over the coming period. 14. Firstly, interventions supporting graduation are not optimally designed or implemented. While the design of the PSNP and quality of implementation has progressed significantly over the last five years, the same has not been true of other components of the FSP, particularly the HABP. The Program has tended to be top-down and supply-driven with little consideration for household capacity or market opportunities. Credit has often been provided through non-specialized multipurpose cooperatives or government bodies, at a subsidized interest rate, with little capacity to support the client or follow-up on collections. The result has been low repayment rates - with a high of only 72% under the Food Security Project. Such low repayment rates have undermined the sustainability of the Program as well as possibilities for long-term credit provision through alternative service providers. 15. In October 2008 the Government launched a comprehensive process with its development partners to review and reformulate the broader FSP, focusing particularly on the HABP. This was informed by the evidence and lessons learned from the last five years of implementation. Government has demonstrated a desire to reform the Program in ways that should substantially strengthen it. Reforms have focused on the institutional arrangements for financial service delivery and strengthening the extension system to deliver market-led and demand-driven support to households. At the same time, Government has reiterated its commitment to dedicate significant financing for the HABP. 16. Secondly, more focused attention needs to be given to the enabling environment and the broader rural growth strategy. While the combination of PSNP and HABP is expected to significantly strengthen the livelihoods of the chronically food insecure, limitations in the enabling environment ultimately constrain the scale of graduation. Recent high rates of economic growth have not yet translated into widespread improvement in the livelihoods of the poorest. For there to be a step change improvement in the rate of graduation, rural growth will need to accelerate further and its composition particularly will need to evolve. The Government's rural development strategy has made some improvements in land tenure and access to capital, while also reforming service delivery and input and output markets. Further work is needed to increase agricultural productivity and promote diversification, including a move away from a singular focus on on-farm production in a way that can transform rural livelihoods and promote graduation at scale. Issues that require further attention include: 4 (a) Factor markets need further development and strengthening. A "first-phase" land certification process was carried out that has improved land security. However, market- based land transactions are limited and labor mobility is constrained by lingering concerns about land use rights. (b) Access to financial services has improved with the emergence of MFIs, but rural finance is still overly characterized by Government interventions that are distorting both the financial markets (as evidenced by low repayment rates and subsidized interest rates) and associated product markets (such as the tying of input credits to specific suppliers creating an unlevel playing field). (c) There have been efforts to strengthen agricultural extension. Based on a revised curriculum the public extension system was expanded from about 15,000 extension workers in 2002 to more than 60,000 today. Yet the approach to extension remains overly supply driven and further resources are needed to transform public extension workers into agents of change. (d) While there have been some positive developments in agricultural input and output markets, distorting practices, such as the interventions in credit markets mentioned above, need to be removed and the related nonfarm private sector (in areas such as agricultural input or output trade or processing) needs to be further developed. (e) Rural infrastructure remains limited. Improved water management, particularly irrigation, is central to moving away from the current dependence on rain-fed agriculture. This is critical to significantly boosting overall agricultural productivity. Rural feeder roads will also need to be further expanded as the basis for providing much needed connectivity. (f) The reforms above need to be accompanied by an active policy of small town development that fosters the private sector through economies of scale and agglomeration effects. Small town development remains an important mechanism for the promotion of rural nonfarm enterprise because it can serve as both as a complement and ultimately also as an alternative to farming. 17. Many of these issues form part of the ongoing dialogue between the Government and development partners. The Rural Economic Development and Food Security (RED&FS) Government-Development Partner Group is working to improve understanding of the above issues to inform a long-term investment program. Such reforms, along with ongoing investments by the Government and development partners, could significantly strengthen the rural development strategy and economic growth process (see Annex 2). 1.4 Responding to risk and vulnerability in Ethiopia: social protection for the poorest and a scalable safety net 18. Despite these reforms, expectations for graduation may not be realistic. There is significant commitment from Government at all levels to achieve high rates of graduation. However, given the depth of poverty and the nature of risk and vulnerability in Ethiopia, there may be a sizable population in need of support from the Government for some time to come. Moreover, the transformation away from a rural economy based on subsistence agriculture is still some way off. Projections suggest that employment in agriculture will continue to rise and will not fall in absolute terms for another 7 to 10 years. Thus risk-prone livelihoods will continue to dominate in rural areas. Because of this, in many cases graduation must be thought of as a long- term prospect. 5 19. Furthermore, for some households graduation may never be feasible. The implication is a reality nearly all countries face: that Governments need to provide a system of social protection for their most vulnerable citizens. In 2009, Ethiopia endorsed the Social Policy Framework for Africa, which includes commitments to "develop and operationalize costed national plans for social protection based on the concept of a `minimum package'". The "minimum package" would cover: essential health care and benefits for children, informal workers, the unemployed, older persons and persons with disabilities. 20. The Government already has many critical elements of a minimum package in place through the PSNP and the health waiver system but has not brought these together under one umbrella policy or national plan. It is expected that the development of a social protection policy in Ethiopia will situate the PSNP within a long-term social policy framework. While graduation is desirable and may be achievable for many (see Annex 1), the PSNP or a subsequent instrument will be needed in the long term to provide regular support for the most vulnerable citizens. 21. At the same time, livelihoods in Ethiopia remain severely vulnerable to shocks. This was vividly seen in 2008 when Ethiopia faced a crisis that was broader, deeper, and more complex than the food crisis in almost any other country. The crisis was driven by high food price inflation coupled with drought and local food shortages. This caused severe hardship for the country, negatively affecting roughly 12 million chronic and transitory food insecure people. 22. In rural areas the PSNP was a major pillar of the Government response. The PSNP contingency budget provided assistance to 1.49 million transitory food insecure households. As the crisis deepened, the Government then provided additional transfers to 4.43 million existing program beneficiaries negatively affected by the crisis. This was partially resourced through the Drought Risk Financing component of the PSNP APL II. In order to address the diminishing purchasing power of the PSNP cash transfer brought about by inflation, the Government increased the nominal value of the transfer by 25 percent, bringing it to ETB10 (about US$0.90). 23. The scaling-up of the PSNP proved to be an effective and efficient crisis response instrument. As evidenced globally in 2008, countries with well-designed, operational safety nets were able to respond most effectively to the crisis. As a case in point, the PSNP was indispensible in mitigating the shocks' impact on the rural poor in Ethiopia. Such capacity, together with a functioning emergency system, is required in the long-term to guarantee that households have access to support during times of stress. 2. PSNP 2005-2009: Implementation and results 2.1 Program scale and core parameters 24. The PSNP is one of the largest social protection programs in Sub-Saharan Africa, covering 7.57 million rural citizens in 290 of 710 woredas nationwide (Table 1). Woredas are selected based on historic vulnerability, while households participating in the Program are those that are identified as being chronically food insecure. Households receive up to US$137 in transfers per 6 year5. This figure is based on five days of work a month for six months for able-bodied individuals, while those who are unable to work receive a direct transfer. In 2008, over 60 percent of beneficiaries received transfers in cash rather than in kind. 25. The PSNP annual budget is Table 1: 2009 PSNP woredas & beneficiaries roughly US$347 million. Of this, an No. of estimated US$303 million is provided in woredas No. of beneficiaries transfers to beneficiaries; the remaining Afar 32 472,229 budget is spent on capital inputs for Amhara 64 2,519,829 public works, capacity building and Dire Dawa 1 52,614 program management costs. The PSNP Harare 1 16,136 budget is equivalent to 1.2 percent of Oromiya 76 1,438,134 Ethiopia's GDP. At woreda level, the Somali 6 162,671 PSNP is a sizeable source of additional SNNP 79 1,459,160 resources as compared with total woreda Tigray 31 1,453,707 expenditure. On average, PSNP Total 290 7,574,480 resources are equivalent to 60 percent of Source: FSCD PSNP EFY 2001 Annual Plan. total woreda expenditure in PSNP areas (see Annex 10)6. 26. Each year, the PSNP initiates roughly 34,000 public works projects that focus on soil and water conservation, social infrastructure, and roads (Table 2). These projects are planned within an integrated watershed development planning framework. Planning occurs annually at community level using participatory techniques. In 2009, PSNP public works were operational in almost a third of Ethiopian woredas generating an estimated 190 million person-days of labor. Table 2: Sample public works supported through PSNP in 2007 Activity Result Activity Result Soil Embankment Construction (km) 482, 542 Seedlings Produced (no.) 301,778,607 Stone Embankment Construction (km) 443,148 Seedlings Planted (no.) 12,883,657 Pond Construction & maintenance (no.) 88,936 School Classroom Construction (no.) 340 Spring Development (no.) 598 Rural Road Maintenance (km) 20,458 Hand-dug well construction (no.) 491 Rural road construction (km) 8,323 Land rehabilitated Area Closure (Ha) 530 Farmer Training Centre Const. (no.) 119 Small-scale irrigation canals (km) 2,679 Animal Health Post Construction (no.) 71 Tree nursery site establishment (no.) 285 Source: FSCD 2007 Annual Report, January 2008. 27. The PSNP is a model of strong donor collaboration and harmonization. The World Bank and eight other development partners7 have pooled their cash and in-kind financing and developed a unified stream of technical advice to support a single Government-led program coordinated by the Food Security Coordination Directorate (FSCD) in the Ministry of Agriculture and Rural Development (MOARD). 5 Details of this calculation are found in Table 13 in Annex 10. 6 `PSNP resources' refers to the sum of transfers and contingency, capital and administrative budgets. 7 CIDA, DFID, EC, Irish Aid, RNE, SIDA, USAID, WFP. 7 2.2 Program results and outputs Delivering transfers to the poorest citizens in rural areas on a timely basis 28. Evidence shows that the PSNP is well targeted to the poorest households, which have significantly lower incomes, fewer assets, and farm less land than non-beneficiaries. The community-based targeting system is seen to be fair and transparent. A survey of local service delivery in Ethiopia reported that over 85 percent of respondents described the PSNP selection process as being fair. A recent study found that implementers, non-beneficiaries, and beneficiaries widely understood poverty to be the reason for household participation in the PSNP. Table 3 presents economic characteristics of PSNP beneficiaries. Table 3: Household consumption, assets and land access, by PSNP beneficiary status Household Economic Characteristics Direct Support Public Works Non-PSNP Total consumption (birr) per month, average 627 1012 1111 Land (hectares), average 1.0 1.1 1.4 Assets (birr), average 2349 4568 6480 Note: Consumption refers to the value of total consumption (food and non-food) both purchased and the value of production of self-produced goods. Assets are the value of livestock and productive equipment used in agriculture. Source: IFPRI/CSA 2008 Household Survey. 29. In response to major difficulties in delivering timely transfers during the first year of the Program, the Government made significant investments in logistic capacity. Over one thousand contract staff were hired at the woreda level alone. Approximately 76 percent of these contract staff worked as either accountants or cashiers. Large investments were also made to address gaps in physical capacity, particularly transport, computers, generators, communication equipment, and safe boxes. Innovations to computerize the payroll through the automated Payroll and Attendance Sheet System (PASS) were adopted to address a critical bottleneck in the transfer process and improve fiduciary control. Together these initiatives have resulted in strong year-on- year improvements in the timely delivery of transfers to beneficiaries. However, this trend masks significant intra- and inter-regional variability. 30. Overall improvements in program performance are further evidenced by the fact that PSNP beneficiaries are confident that they will receive their transfers. However, transfers still do not always arrive when expected. In 2008, for example, only 27 percent of households reported that they were able to plan ahead on the basis of PSNP transfers. Recent surveys have also highlighted misunderstanding among some households as to their benefit from the PSNP. The design of APL III aims particularly to strengthen the timeliness of payments and beneficiaries' understanding of their entitlements by: (i) making use of the automated payroll system mandatory; (ii) introducing "client cards" and a charter of client rights and responsibilities; and (iii) furthering implementation of the PSNP Communication Strategy to strengthen widespread understanding of program objectives and procedures. 31. The PSNP has also played an increasing roll in response to shocks. The PSNP contingency budgets at woreda and regional levels are used to address transitory needs in PSNP woredas, covering up to 1.5 million transitory food insecure households. These budgets have proven to be 8 an effective mechanism to respond to emerging crises and are highly valued by woreda and regional administrators. Moreover, in 2008, the PSNP provided additional transfers to 4.43 million beneficiaries negatively affected by rapidly rising food prices and drought. This response provided predictable support to these households enabling them to meet their consumption needs until the next harvest. This experience has demonstrated the importance of the scalability of the PSNP, which is institutionalized in the Drought Risk Financing component of APL III. Building productive public works 32. Since 2005, around 7000 Development Agents (DAs) have received regular training in Community-Based Participatory Watershed Development Planning (CBPWDP) through a national training-of-trainers system. Studies confirm high levels of community involvement in the selection of public works. A high and increasing number of households report participation in the project identification and planning exercises. 33. Support for public works at community level tends to be strong, as evidenced by the technical quality of community plans. Responsibility for the oversight and management of public works was shifted from the FSCD to the Natural Resource Sector in 2006 to strengthen the implementation of public works. After some delay, the Public Works Coordination Unit (PWCU) for PSNP has since been established in the Natural Resource Management Directorate (NRMD) and is now fully functioning, significantly strengthening the coordination and oversight of public works. 34. Within the community, public works are widely perceived to be beneficial. In 2008, 92 percent of households indicated that their community benefited from the construction of roads, while 88 percent reported benefiting from soil and water conservation on communal lands. Public works are increasingly perceived to benefit individual households as well. Indications are that public works have increased access to social services and are beginning to transform the natural environment. 35. Public works are generally evaluated to be of a high technical standard. This is true for the soil and water conservation activities that comprise a large percentage of the overall projects. This has not always been the case for roads and water projects. Only 64 percent of roads and 78 percent of water projects met adequate technical standards respectively. Concerns have also been raised regarding the operation and maintenance of these same project types, which are required to achieve long-term sustainability. NRMD is engaged in cross-sectoral dialogue to determine how best to address the issues of technical capacity in key sectors at the local level. In addition, an upgrading of the annual CBPWDP training will incorporate specific modules on maintenance in 2009. Strengthening governance and increasing transparency 36. The PSNP is designed to encourage strong citizen engagement, particularly in the targeting of beneficiaries and planning of public works, in order to ensure transparency and accountability in program delivery at local levels. High rates of community involvement have proven to be an effective means of holding local decision-makers accountable with few reports of manipulation for personal ends or special interests. To guarantee timely and objective treatment for those who might have a grievance, an appeals system was introduced in 2007. Appeals committees were 9 established in 95 percent of the woredas surveyed in 2008 (see Annex 9 for a detailed discussion of Governance issues). APL II also introduced a number of initiatives to strengthen transparency as part of a broader communication strategy. This included the public display of budgets, beneficiary lists, and posters on key program procedures. 2.3 Program outcomes and impacts 37. Recent independent impact evaluations, together with a series of qualitative studies, show that the PSNP is having a positive impact on livelihoods, even during times of crisis, and is significantly enhancing community level infrastructure and contributing to environmental transformation. These same studies highlight, however, that the impact of the PSNP is much more significant and appears across a wider range of indicators when households receive more effective support from the Program. The impact of the PSNP at household level is measured against the outcome indicators for APL II, namely reported food security and asset protection, among other indicators, as follows: (a) Food security: The impact evaluation found that participation in the PSNP significantly improved household food security, as measured by changes in self-reported household food gap. The impact of the PSNP on food security was found to be larger when transfers were predictable and of a higher value8. Households that received these transfers and inputs from HABP experienced an improvement in food security by 0.81 months (or 25 days). Moreover, participation in the PSNP increased caloric acquisition by 19.2 percent among households that received regular, higher value transfers as compared with non-beneficiary households9. (b) Asset creation and protection: When implemented as designed, the PSNP is having a significant impact on asset accumulation10. The growth rate in livestock holdings among PSNP beneficiaries was 28.1 percent faster than among non-beneficiaries. This is supported by evidence from other surveys that shows that PSNP beneficiaries are using cash transfers to invest in farming inputs and livestock. These findings also hold true during times of crises, when the PSNP has proven to be effective at helping households avoid distress sales of assets. (c) Utilization of education and health services: PSNP beneficiaries have increased their use of social services. In 2006, 46.1 percent of PSNP beneficiary households reported that they used health facilities that year to a greater extent as compared with the year before and 76 percent attributed this to the PSNP. In 2008, 26.7 percent of households reported increased use of health facilities over 2007 and 47 percent attributed this increase to the PSNP. This information, together with reports that PSNP beneficiaries use some of their cash transfers to invest in health and education, suggests that the Program is having a positive impact on human capital accumulation. (d) Agricultural productivity: The PSNP is enabling households to take risks that improve household productivity. A major finding of the impact evaluation was a synergy between the HABP and PSNP. While HABP has only a marginal impact on agricultural productivity 8 Higher value is defined as receipt of transfers worth at least 900 ETB between 2006 and 2008. 9 "Non-beneficiary households" refers to a control group established using matching methods. 10 The term "implemented as designed" refers to when PSNP beneficiaries receive a significant proportion of their entitlement, that is, at least 900 ETB between 2006 and 2008, and the transfers arrive on time at regular intervals. This is the basic design of the PSNP. The impact evaluation shows that PSNP is not implemented in this way in all areas. 10 when implemented alone, when combined with the PSNP the results showed a 38 percent increase in maize yields. This suggests that by allowing households to focus on long-term investments and providing more regular cash flow, the PSNP is a critical element of a strategy to effectively improve agricultural productivity. 38. A growing body of evidence shows that the PSNP is having a significant positive impact at community level. The Public Works Impact Assessment carried out in 2008 found that soil and water conservation activities have significantly increased wood and herbaceous vegetation cover. The construction of water conservation structures within the closed areas has reduced surface runoff, increased infiltration and raised groundwater levels, thereby enhancing spring yields and increasing stream base-flows. In several communities, springs last longer into the dry season. Additionally, the number of domestic water supplies has doubled. 39. There is evidence that these community-level benefits are resulting in improved livelihoods. An estimated 34 percent of households surveyed reported significant benefits from the closed areas that had increased the availability of forage for livestock. Up to 87 percent of households reported that family health had improved as a result of access to PSNP water supplies. It is important to note that these benefits accrue to the community at large beyond the immediate target group of the Program. Annex 11 includes a full economic analysis of the PSNP public works. 40. In conclusion, the overarching finding on program outcomes and impacts is that when the PSNP is implemented as designed there are significant positive benefits at household and community levels. The results of these surveys, together with administrative data, show that the Program is being implemented effectively in many areas. However, there is significant variation between Regions and woredas. In many woredas minimum performance standards, particularly those regarding timeliness of transfers to beneficiaries and the technical quality of roads and water projects, are not yet being met. 3. Rationale for Bank involvement 41. The shift to a productive safety net system has been strongly supported by the World Bank, which has, in partnership with a consortium of eight other development partners, worked closely with the Government throughout the design and implementation of the PSNP. The World Bank is supporting the Government through a three-stage APL and has provided significant IDA financing through APL II and I11. The World Bank also: (i) manages substantial co-financing resources; (ii) manages a large technical assistance trust fund on behalf of development partners that ensures a unified stream of technical advice on the Program to the Government; and (iii) houses a donor secretariat that ensures strong coordination among development partners. Building on the original rationale for World Bank engagement, there are four reasons for continued involvement in the PSNP. 42. First, the PSNP remains a central component of both the Government's strategy to combat household vulnerability and of the World Bank's support for that strategy. The PSNP is a key 11 APL I (P087707, 4004-ET, H136-ET), APL II (P098093, H266-ET, H3480-ET). Total IDA financing amounted to US$313.7 million (see Annex 5). 11 pillar in efforts to reduce food insecurity and exposure to shocks as well as address environmental degradation, both of which are objectives outlined in the April 2008 Country Assistance Strategy (CAS) for Ethiopia. 43. Second, the transition to an effective and development-oriented safety net is not yet complete. An essential next step is the development and implementation of an integrated national strategy to assist the poorest households graduate from food insecurity in a sustainable manner. This is underpinned by a review and redesign of the key complementary investments being undertaken through the FSP. The CAS identified the next phase of the PSNP as a logical entry point for dialogue on such reforms. 44. Third, the World Bank is particularly well positioned to engage in these discussions given its other investment in the sector, the Food Security Project12. This Project continues to provide a vehicle for dialogue with the Government on the design of the FSP. The Project will be completed by June 2010, with a view to continuing support to household asset building for the food insecure under this APL III. 45. Fourth, program financing requirements and current indicative commitments show that IDA resources for the medium term are required to help fill a large financing gap. Moreover, IDA's financial contribution continues to be critical in leveraging other development partner contributions. 4. Higher level objectives to which the project contributes 46. Sustainable food security: The higher-level goal to which the PSNP aims to contribute is sustainable graduation from food insecurity for a large number of households in the Program. The PSNP is necessary but not sufficient for this to happen; complementary programs and investments need to be in place as well as the linkages to a broad-based rural economic growth process. 47. Ethiopia's poverty reduction strategy and the Millennium Development Goals (MDGs): The PSNP aims to shift away from a focus solely on short-term food needs met through emergency relief to addressing the underlying causes of household food insecurity. As such it forms a core component of the Government's poverty reduction strategy: the five-year Program for Accelerated and Sustained Development to End Poverty (PASDEP). The PSNP contributes to the reduction in the number of Ethiopians suffering from extreme hunger, malnutrition, and poverty (MDG 1), rehabilitation of the environment, and making agriculture more productive and sustainable (MDG 7). Given the harmonized donor framework adopted for the PSNP, the project also contributes to MDG 8 on partnerships and the Paris Declaration on Aid Effectiveness. 48. World Bank Africa Action Plan: The PSNP is expected to contribute to rural growth by: (i) providing cash rather than food aid to households, which gives them much needed flexibility and stimulates rural markets through demand linkages; (ii) reviving the environmental potential of degraded lands; and (iii) building and maintaining public infrastructure. The Project will also 12 Food Security Project SIL (FY02), P050383. 12 help Ethiopia better manage the impact of shocks by developing a portfolio of financing instruments that allow a more timely response to shocks. Finally, the PSNP contributes significantly to governance and accountability by promoting greater transparency, encouraging citizen feedback, and fostering community participation on decisions related to PSNP resource allocation and use. B. PROJECT DESCRIPTION 1. Lending instrument 49. The World Bank will utilize the third phase of an Adaptable Program Loan (APL III) to provide an IDA grant and credit of US$480.0 million equivalent to the PSNP. It is proposed that the IDA project become effective early in the last calendar quarter of 2009 to run until June 2015. 2. Program objective and phases 50. The development objective of the overall PSNP APL series is to reduce household vulnerability, improve resilience to shocks and promote sustainable community development in food insecure areas of rural Ethiopia. 51. This is to be achieved through continued consolidation of a safety net system that (i) provides timely, predictable, and appropriate transfers to beneficiary households, thereby enabling effective consumption smoothing and avoiding asset depletion; (ii) creates productive and sustainable community assets that contribute to the large-scale rehabilitation of severely degraded areas; (iii) stimulates local markets through demand linkages; (iv) establishes more effective responses to drought shocks to avoid increasing destitution among affected households; and (v) integrates and effectively supports critical interventions that build assets, promote increased productivity, and encourage diversification at the household level. 52. A two-phased APL for the PSNP was initially approved by the World Bank's Executive Board in November 2004. As originally conceived, the first APL phase focusing on Transition was to be followed by a second APL phase focusing on Consolidation. 53. However, it was agreed with the Government during preparation for APL II that strengthening livelihoods to the extent that households become food secure and are resilient to shocks is a process that is longer and more complex than that suggested by the initial five-year timeframe covered by these two phases. The Executive Board therefore approved the addition of a third phase, Integration, to the APL series to span a further five years from the end of APL II, which was envisaged broader support to the household asset building initiatives of the FSP. The revised APL series with three phases is seen broadly as a continuum, as illustrated in Figure 1. 13 Figure 1: The Evolution of the Productive Safety Net Program Effectiveness Predictability Quality & Linkages Productivity EMERGENCY TRANSITION CONSOLIDATION INTEGRATION 2005-2006 2007-2009 2010-2014 Potential for Graduation Potential for Graduation 54. APL Phase I: Transition (2005-2006, US$70.0 million, ICR "satisfactory"). Phase I of the Program assisted the Government to transition from emergency relief to a productive- and development-oriented safety net. Phase I accomplished the following: (i) provided predictable, multi-annual resources to the Government; (ii) replaced food with cash as the primary medium of support; (iii) made resources available for critical capital, technical assistance, and administrative costs to effectively support the public works; (iv) strengthened community involvement by supporting community targeting and local-level participatory planning as core principles of the program; and (v) related public works activities to the underlying causes of food insecurity, especially with respect to soil and water conservation measures. It aimed to put in place all of the essential elements of the new system. 55. APL Phase II: Consolidation (2007-2009, US$175.0 million, Additional Financing US$25.0 million, current IP "satisfactory"). Phase II of the Program aimed to consolidate the progress made under Phase I and continue to strengthen technical capacity for program implementation. Phase II has: (i) improved the efficiency and predictability of transfers by continuing to build capacity in government systems and by strengthening resource planning and mobilization; (ii) strengthened program governance by enhancing existing targeting and grievance systems as well as introducing more transparency in program procedures; (iii) increased the productivity of public works through a systematic focus on community planning using integrated watershed management techniques and enhanced involvement of technical staff from NRMD at all levels; (iv) strengthened monitoring and evaluation systems; and (v) developed more efficient financing instruments for risk management to ensure a more predictable and timely response to shocks. 56. Proposed APL Phase III Integration (timeframe 2010-2014, US$480.0 million). Phase III will continue to consolidate program performance as well as seek to maximize the Program's long-term impacts on food security by ensuring effective integration and coordination with other critical interventions. Phase III will (i) introduce initiatives to further improve the timeliness and predictability of transfers, notably through closer performance monitoring and provision of incentives; (ii) initiate further work to strengthen public work, particularly focusing on regional and federal oversight, coordination, and monitoring; (iii) strengthen program accountability through a number of additional "bottom-up" and "top-down" mechanisms; and (iv) support 14 Government in a re-design of its HABP to significantly increase the efficiency and effectiveness of this intervention to further promote sustainable graduation from food insecurity. 3. Project development objective and key indicators 3.1 Project Development Objective (PDO) 57. The proposed Project Development Objective for APL III is: "Improved effectiveness and efficiency of the Productive Safety Net Program and related Household Asset Building Program for chronically food insecure households in rural Ethiopia." 3.2 Key Performance Indicators 58. Reporting on implementation and program achievements will continue to rely on the Government's FSP M&E system of which the PSNP and HABP. Information from sample household surveys, public works reviews, beneficiary assessments, audits, institutional reports and other reviews will continue to be used to supplement regular monitoring data collected through Government systems. Building on the indicators for APL II, with the addition of indicators specific to HABP, progress towards the achievement of the Project's development objectives will be measured by the following key outcome indicators13: (a) Percent of participants reporting they are able to plan ahead on the basis of PSNP transfers. (b) Percent of households reporting direct benefit from community assets. (c) Percent of PSNP households report that they have developed an on- or off-farm income generating opportunity attributable to HABP. 4. Project components 59. APL III will continue implementation of current program components, including the various modifications already underway or planned in order to further strengthen performance. APL II currently comprises three components: (i) Safety Net Grants for activities including Public Works and Direct Support (ii) Drought Risk Financing to provide additional resources for these activities to allow the Program to scale up in response to shocks; (iii) Institutional Support to PSNP focused on capacity building and management, M&E, and program governance. As indicated at the time of APL II World Bank Executive Board approval, a new component will be added under APL III; and (iv) Support to Household Asset Building, which provides assistance to parts of the Government's HABP to improve the effectiveness and sustainability of this intervention. 60. The PSNP and HABP target 7.57 million chronically food insecure rural citizens (approximately 10 percent of the total population), residing in 290 of 710 woredas nationwide in eight of the ten Regions (see Table 1 above)14. Woredas continue to be selected based on historic 13 See Annex 3 for the complete Results Framework. 14 It is anticipated that the number of beneficiaries and woredas in phase 3 of the PSNP will increase due to the planned roll out of PSNP and HABP programs to communities in the pastoral areas of Somali Region. 15 vulnerability. Households within these woredas are identified based on relative wealth ranking to select the poorest and most food insecure. 61. An overview of each of these components is as follows: 4.1 Component 1: Safety Net Grants (IDA US$398.5 million15; DFID US$218.8 million; EC US$73.2 million; CIDA US$68.0 million; Irish Aid US$68.8 million; USAID US$457.0 million equivalent; WFP US$50 million equivalent; SIDA US$21.5 million; and RNE US$66.3 million). 62. Labor-intensive public works will provide transfers to households whose adults participate in public works. Sub-projects are determined locally through an annual participatory planning process that focuses on integrated watershed management. Public works are timed such that resources are available to households when needed and works are undertaken when weather conditions are appropriate and labor demand from alternative agricultural activities is lowest. 63. Financing for administrative costs and capital inputs will be allocated to woredas alongside the financing for the unskilled labor costs (i.e. the transfers to households) to provide the necessary complementary inputs as well as technical supervision and monitoring for public works activities16. Financing will continue to be transferred from the federal level to regions and woredas before the start of the PSNP implementation season to ensure that tools and materials are procured to initiate activities in a timely way. Technical guidelines and work norms have been developed by MOARD for different types of labor-intensive public works. These are made available to PSNP implementers and supported through an extensive training program. 64. Public works will continue to focus on soil and water conservation, with significant investment in roads, irrigation and social infrastructure. As agreed under APL II, the Government has established a Public Works Focal Unit (PWFU) in the Natural Resources Department of the Bureau of Agriculture and Rural Development (BOARD) in each implementing Region. These are coordinated by NRMD in MOARD. NRMD has established a unit specifically mandated to provide close follow-up and oversight of public works, as well as improve coordination with other key line ministries. Under APL III, NRMD will work with the Natural Resources Departments to ensure (i) strengthened woreda and regional support to public works; (ii) improvements to the annual training on CBPWDP; (iii) improved monitoring of public works; and (iv) better national coordination to address policy or sector-wide capacity gaps. 65. It is expected that this will lead to improvements in the overall sustainability of the public works, notably strengthened implementation of the Environmental and Social Management Framework (ESMF), improved technical quality of projects, particularly in areas that have previously been identified as relatively weak (such as roads and water) and clear arrangements for operations and maintenance of all sub-projects, which will be identified during the planning process, to ensure appropriate management. 15 These are estimated equivalent amounts in US$. However, the term "equivalent" is used throughout this PAD to refer to in-kind contributions only. 16 This includes the financing of small complementary civil works contracts to support the implementation of public works (such as the building of small culverts or bridges in the context of road construction). 16 66. Direct Support (DS) will provide grants to households that are labor poor and cannot undertake public works. Beneficiaries include, but are not limited to, orphans, pregnant and lactating mothers, elderly, labor-poor households with sick individuals, and female-headed households with no other available adult labor. 67. Support to performance management system: This will establish a system of performance incentives and management support to improve program implementation. All woredas that meet minimum performance standards will receive additional financing to be used as part of their administrative and/or capital budgets. The Institutional Support component will provide tailored capacity building support to woredas encountering problems. Underpinning this will be a renewed focus on accurate performance measurement across the Program, based on information generated by the Regional Information Centers (RICs). 68. The PSNP currently includes a pilot program in 18 woredas in pastoral areas of Afar, Oromiya, Somali, and SNNP. The aim of the pilot is to test a range of methods, including targeting, public works and types of transfers, to determine how to effectively deliver a safety net program for pastoral livelihoods. Based on the lessons learned under the pilot program, the PSNP will be scaled up in pastoral areas in 2010 and 2011. 4.2 Component 2: Drought Risk Financing (IDA US$50.0 million; DFID US$31.5 million; and USAID US$73.9 million equivalent). 69. While Component 1 focuses on chronically food insecure households, Component 2 seeks to provide timely resources for transitory food insecurity in response to shocks within existing program areas. This Component will be financed using a contingent grant, which will provide resources for scaling up activities under Component 1 in response to localized or intermediate weather or price-related shocks in PSNP woredas. This component is seen as central to the sustainability of the overall PSNP by providing an early response that can more effectively prevent household asset depletion and increased levels of destitution. 70. Risk Financing can be used to either extend support to current PSNP beneficiaries or support new beneficiaries who have transitory needs17. This activity will focus exclusively on PSNP woredas because of the presence of established institutions and systems for delivering risk financing. Strong linkages between risk financing and the emergency response system are required to ensure a coordinated response to shocks over time. In order to ensure that PSNP systems can be taken to scale on demand, additional support for capacity building will be channeled through the Institutional Support component (see below). 71. The main activities for Risk Financing are: (i) contingent financing through a contingent grant from the World Bank as well as additional in-principle commitments from development partners to be mobilized on the basis of need; (ii) the early warning system will provide ongoing analysis that can trigger the Risk Financing budget in a timely fashion at any point during the year; and (iii) contingency planning will be developed at woreda level to expedite 17 Based on the anticipated size of the risk financing contingent fund, this component could cover up to 3.1 million additional individuals based on PSNP parameters. 17 implementation once the early warning system confirms the likely occurrence of shocks and the release of Risk Financing resources. 4.3 Component 3: Institutional Support to PSNP (IDA US$14.0 million; DFID US$32.0 million; EC US$5.5 million; CIDA US$13.8 million; Irish Aid US$5.5 million; SIDA US$1.5 million; and RNE US$5.0 million). 72. This Component will support institutional strengthening activities. Support will be provided for: (i) program management at regional and federal levels to ensure effective management of the Program by the Food Security, Early Warning, Natural Resources and Finance Directorates; (ii) capacity building to fill any remaining gaps in general, and those specific to the Risk Financing facility, and PWCUs and PWFUs; (iii) monitoring and evaluation to ensure regular monitoring data, with a specific focus on upgrading the monitoring system for public works and establishment of RICs; (iv) implementing the ESMF; and (v) transparency and accountability measures to further strengthen widespread understanding of the Program among key stakeholders and greater accountability of decision-makers, including ensuring program- wide use of PASS and PSNP client cards. 4.4. Component 4: Support to Household Asset Building (IDA US$17.5 million; DFID US$41.8 million; Irish Aid US$6.3 million; and GoE US$10.0 million). 73. The Household Asset Building Program (HABP) is designed to assist food insecure households in PSNP woredas to transform their productive systems by diversifying income sources, improving productivity and increasing productive assets. The Program has been designed to strengthen the extension system and rural service providers to deliver demand-driven and market-oriented assistance to food insecure households. It intends to achieve four outputs: (a) improved identification and development of on- and off-farm investment and income generating activities for food insecure households; (b) enhanced access by such households to sustainable and multiple financial services; (c) enhanced systems for input sourcing, production and delivery; and (d) increased access by food insecure households to product and labor markets. This component will provide support to the Program as follows: 74. Strengthening the delivery of public advisory services in support of household investments: Financing will be provided for capacity building within the Government agricultural extension and micro/small enterprise development programs for the provision of advisory services to food insecure households that are demand-driven and take into consideration market opportunities and conditions, as well as for strengthening input and output markets. Interventions will include, as appropriate, development of education materials/curricula and training to woreda-level experts on market and technical analyses, promotion of stakeholder consultations at the local level on investment opportunities, and technical backstopping for DAs on supporting household business plan development and entrepreneurship. 75. Improving the efficiency and effectiveness of financial service delivery to food insecure households: Development partners will provide assistance to improve the efficiency and effectiveness of credit delivery to food insecure households to bring it in line with internationally accepted norms and best practices. This will include: (i) supporting the development and field testing of multiple financial products (savings, different types of credit) that respond to the needs and capacities of food insecure households; (ii) the dissemination of these products to service 18 providers, complemented by a rolling training program; (iii) the development of financial literacy materials; and (iv) capacity building of both Rural Saving and Credit Cooperatives (RuSACCOs) and MFIs to expand coverage and enhance outreach. Credits to food insecure households, channeled through RuSACCOs and MFIs, will be financed from Government resources only. 76. Supporting program management: This will provide resources to ensure the effective management of the HABP. Management budgets will be provided to each of the key implementing agencies at all levels, particularly the Agricultural Extension Directorate (AED). An appropriate instrument for supporting management of the program within the micro/small- scale enterprise development program and cooperative promotion agency will be developed. Given that many of the above initiatives are new to Ethiopia, study tours and experience sharing events will be organized to facilitate implementation and to create awareness of international best practices. In addition, this component will support monitoring and evaluation and the procurement of physical inputs. 5. Lessons learned and reflected in the project design 77. Since inception, the PSNP has drawn on a growing number of independent studies and assessments to inform implementation. A large regionally representative impact evaluation was completed in 2008. This body of work provides valuable insights into how to improve implementation and maximize the impact of the PSNP overall. Some of the key lessons for the design of APL III are detailed below. Lessons learned from the implementation of the PSNP 78. Providing households with predictable transfers is central to protecting livelihoods. The impact evaluation concluded that the PSNP is an effective instrument to smooth household consumption and protect, and even build, household assets when transfers are predictable. The evaluation found that predictable transfers, even of a low value, effectively protect households from shocks. Thus, when implemented as designed, the PSNP is an effective safety net. 79. Bottom-up accountability mechanisms work best when households are aware of their entitlements. Experience to date demonstrates the need to reinforce the initiatives introduced under APL II to strengthen transparency and accountability in the Program. Studies have shown that while the level of satisfaction with the Program is high (75 percent), households that perceive they have enough information to understand how the program works report even higher rates of satisfaction (90 percent). In order to effectively hold local decision-makers to account, it is critical that all households have this understanding. 80. When implemented together the Safety Net and Household Asset Building programs have a combined impact that is greater than when each program is implemented alone. The results of the 2008 impact evaluation suggest that the PSNP is a critical element of any strategy designed to enhance rural livelihoods, particularly through household asset building. A major finding of the impact evaluation was that although the HABP has a marginal impact on agricultural productivity, as measured by changes in maize yields, when the Program is combined with the PSNP maize yields increased by as much as 38 percent. There is also evidence from the Public 19 Works Impact Assessment that the investments in soil and water conservation on communal lands increase farmer productivity making this a valuable complementary input to household level investments. 81. To improve Program performance, monitoring needs to be part of a responsive management system. During APL II, Government, with support of development partners, made continuous improvements to the PSNP monitoring system, such as streamlining the reporting formats and strengthening the Federal Information Center (FIC), which generates real-time data on program performance. While this resulted in a more complete overview of program implementation, there is little indication that these changes have led to a more responsive management system. Performance remains variable across regions and woredas and it appears that the information generated by the monitoring system is not being used to inform management decisions. To address these issues, APL III will introduce a system of performance incentives, which requires the regular use of performance monitoring. 82. To ensure effective and sustainable service delivery, financial services must be provided through financial institutions. The FSP Review found that credit for the HABP was delivered through diverse channels, including MFIs, RuSACCOs, multipurpose cooperatives and government departments. Rates of repayment tended to be significantly higher in areas serviced by MFIs and RuSACCOs. In addition, government departments did not always enforce sound microfinance practices. APL III will support the increased outreach and coverage of RuSACCOs and MFIs within food insecure woredas to better ensure that Government resources are channeled through these institutions. Lessons from Ethiopian experience 83. Implementing a program with multiple Government ministries requires strong political commitment. In Ethiopia, implementing complex programs with multiple departments or ministries demands a high level of communication and coordination. Coordination across government departments for PSNP public works, for example, needs to be more effective to reduce fragmentation in oversight. Experience shows that a high level of coordination and communication can be achieved when there is strong leadership commitment to program implementation, especially at woreda level. 84. Responsive systems are needed to address limited capacity in Government departments. Despite investments in capacity building across Government ministries, many departments remain weak, particularly at woreda level. To be effective, capacity building strategies need to be designed in a way that is responsive to the very weakness they aim to address. This involves, for example, addressing the high level of staff turnover among woredas by adopting continuous training, coupled with backstopping at region level in order to ensure consistency in implementation. 85. Programs in Ethiopia need to be sensitive to macro risks. Recent experience has again highlighted the vulnerability of development programs to macro risks in Ethiopia. In 2008, for example, the PSNP and emergency system were forced to respond to the fuel and food crisis. For the PSNP, this demanded the flexible use of cash and in-kind resources and the ability to scale up 20 to protect beneficiaries. Going forward, it is important that the PSNP maintains the capacity to respond to such risks. Lessons from international experience 86. Incentives can be effective to improve program management and performance. International evidence from other social protection programs shows that the use of incentives can be a powerful means of improving performance. The Ministry of Social Development in Brazil established incentives for the Bolsa Familia ("family grant") program to improve implementation at local levels. There is evidence to suggest that this has improved performance, while supplementary actions are taken in poorly performing municipalities. This type of approach seems promising for the PSNP, as performance varies significantly across Regions. An estimated 86 percent of households in Amhara reported that they received their transfer on time as compared to 43.5 percent in Tigray. The percentage of beneficiaries who report that they received all the information they need to understand how the Program works ranged from 80.4 percent in SNNPR to 54.7 percent in Oromiya. 87. A spatially-based database and M&E system are important mechanisms for strengthening oversight and management of public works. Evidence from the Loess Plateau project in China points to the importance of a spatially-based database and monitoring system for the effective oversight of projects that rehabilitate the natural environment. Most simply, these systems include maps derived from Geographic Information Systems of each community watershed, which provide a comprehensive and accurate overview of all community projects and track progress. This information is then used to estimate the impact for specific livelihood zones. The PSNP would benefit from more accurate monitoring data such as this to enhance oversight and management. 88. Very poor households need sequenced interventions in order to achieve food security. It is increasingly understood that when sequenced correctly the combination of financial services with safety net programs can be an effective ladder out of food insecurity. The Consultative Group to Assist the Poor (CGAP) has compiled evidence that shows that households can move out of food insecurity when they have access to diverse financial services, including social transfers and micro-grants, along with complementary technical and advisory support, such as financial literacy and business management training for the poorest. This enables households to move more easily along the risk-reward spectrum. Evidence from a number of countries, including Ethiopia, confirms that this is an effective approach to creating food security. Experience from the Ethiopia Food Security Project revealed that for households to transform their productive systems, the most effective approach appears to be staggered investments, offering larger loans with technological innovations to relatively better off households and repeater loans to poorer households, which gradually introduce innovative investments. 6. Alternatives considered and reasons for rejection 89. Restricting financing solely to the safety net component rather than providing additional support to the HABP. It is widely recognized that while the PSNP can stabilize livelihoods, complementary interventions are required to rebuild assets, increase productivity and promote diversification. It was therefore decided to extend support to aspects of the HABP to improve its 21 design. The possibility of continuing to finance the PSNP alone was also considered, but was rejected because it was felt that financing aspects of the HABP offers the opportunity to work with Government and other development partners to enhance the performance of the HABP and to strengthen and harmonize the coordination and management of these two programs. 90. Financing credit within the Government's HABP. The FSP Review found that the impact of the HABP was limited. Based on this, it was decided to provide financing to develop the institutional structures and capacities necessary for effective program delivery. The alternative of also providing financing for credit was considered and rejected. This is because the current level of credit financed by Government is significant. The limited impact of the HABP is due to implementation weaknesses. This suggests that large gains can be made by focusing on institutional development and capacity building. 91. Scaling back World Bank support to the PSNP and alternatively providing additional investments in agricultural growth. While experience shows that the PSNP, together with investments in household assets, can provide a pathway to food security for some households, this strategy has not yet resulted in large-scale graduation. It is therefore proposed to work with Government to reform the HABP by providing support to aspects of this program together with the PSNP. The alternative of scaling back support to the PSNP and instead providing additional investments in agricultural growth programs was considered. This was rejected because although Ethiopia has enjoyed impressive rates of economic growth, livelihoods of the poorest households in rural areas remain extremely vulnerable. In this environment, financing the PSNP remains a necessity and a core element of the Government's food security strategy. This complements and can enhance the ongoing investments in agricultural growth, which are also being scaled up through initiatives such as the Agricultural Growth Program (AGP). 92. Adding an additional phase to the current APL series. It is likely that Ethiopia will require a safety net in the long term as part of the emerging social protection architecture and to provide scalable support to households vulnerable to shocks. As a result, it is anticipated that World Bank financing for the PSNP will continue beyond 2014. The possibility of adding an additional phase to the current APL series was therefore considered. This was rejected because the most appropriate instrument to support social protection and livelihood policies and programs in the long run in Ethiopia is currently unknown. It was therefore agreed that determining the best instrument for this support (e.g. APL, SIL, DPL) would be explored during the Mid-Term Review of APL III. C. IMPLEMENTATION 1. Partnership arrangements 93. The policy shift from emergency response to a productive safety net system has been strongly supported by the World Bank in close partnership with a consortium of eight donors ­ CIDA, DFID, EC, Irish Aid, RNE, SIDA, USAID and WFP. Development partners have pooled their financing ­ both cash and in-kind contributions ­ and formulated a unified stream of technical advice in support of a single program led by the Government. This approach allows for better harmonization and enables enhanced Program supervision and monitoring while avoiding 22 excessive transaction costs for the Government and development partners. This helps ensure that the PSNP achieves impact at scale. 94. A continued commitment to partnership and the principles of the Paris Declaration on Aid Effectiveness is expected for the next phase of the Program. A Memorandum of Understanding (MOU) articulating all rights, obligations and coordination arrangements of the partnership was adopted by all parties in 2005. This MOU is currently being revised to reflect the restructured Government system. 95. To strengthen the effectiveness and overall quality of program interventions, several joint Government-development partner technical committees and taskforces have been established to monitor program implementation and provide technical guidance on component-specific and crosscutting issues. The role of the PSNP Joint Coordinating Committee (JCC), which is chaired by the State Minister for Disaster Risk Management and Food Security (DRMFS) and includes all PSNP development partners, will evolve to include additional issues in the FSP, including linkages with RED&FS. 96. All development partners continue to commit significant resources for implementation support and enhanced supervision and monitoring of the PSNP. The World Bank-managed PSNP Partnership Multi-Donor Trust Fund (MDTF), which was established under PSNP APL I and has resources from CIDA, DFID, EC, Irish Aid, and SIDA, will continue to enable enhanced supervision of PSNP and HABP. Annex 12 provides the details of the supervision strategy for the PSNP and HABP, which is implemented jointly by the Government, the World Bank, and other PSNP development partners. 97. A Donor Working Group (DWG), aided by a Donor Coordination Team (DCT), was established under APL I to harmonize development partner support for the PSNP. Each development partner takes on responsibility for being the Chair of the DWG for a period of six months. The DCT is functioning well and will continue to support coordination among development partners and manage the large volume of studies and technical assistance that are mobilized for the PSNP and HABP. 98. CIDA, EC, and RNE are expected to provide financing for the PSNP and HABP via a World Bank-administered co-financing MDTF. DFID, Irish Aid, and SIDA will provide parallel financing. These development partners will continue to channel their contributions to pooled accounts that are subject to common procurement and financial management processes. USAID and WFP will provide in-kind contributions to the PSNP. 99. Development partner financing amounted to US$409.0 million under APL I. Funding rose to US$1,040.0 million under APL II (see Annex 5). For the third phase of the Program, indicative commitments from development partners are: co-financing of US$219.3 million from CIDA, EC and RNE channeled through a Bank-administered MDTF; parallel financing of US$427.7 million from DFID, Irish Aid and SIDA; and in-kind contributions valued at US$580.9 million from USAID and WFP. Including IDA financing and the Government counterpart contribution, this represents over 77% of total requirements for the third phase. While a significant financing gap remains, Government and development partners have a 23 strategy in place to mobilize additional resources (see paragraph 119). The details of development partner financing for the third phase are in Annex 5 and summarized in Table 4. Table 4: Development Partner Financing for Phase III of PSNP Components Sources of Financing (in USD million) CIDA DFID EC IDA IA RNE SIDA USAID WFP Total 1. Safety Net Grants 68.0 218.8 73.2 398.5 68.8 66.3 21.5 457.0 50.0 1,422.2 2. Risk Financing 0.0 31.5 0.0 50.0 0.0 0.0 0.0 73.9 0.0 155.3 3. Institutional 13.8 32.0 5.5 14.0 5.5 5.0 1.5 0.0 0.0 77.3 Support to PSNP 4. Support to HABP 0.0 41.8 0.0 17.5 6.3 0.0 0.0 0.0 0.0 65.6 Total: 81.8 324.1 78.7 480.0 80.6 71.3 23.0 530.9 50.0 1,720.3 Source: PSNP Donor Working Group. 2. Institutional and implementation arrangements 100. The institutional structures for the PSNP established in APL I and strengthened in APL II will be used for the third phase of the PSNP, while those for the HABP will be developed and strengthened. APL III will be implemented through Government systems, with food security line agencies at every level accountable for oversight and coordination. 101. At the federal level, MOARD is responsible for the management and coordination of the PSNP and HABP with overall coordination vested in the Disaster Risk Management and Food Security Sector (DRMFSS). The Ministry of Finance and Economic Development (MOFED) is responsible for financial management and reporting, and channels resources to Regions. The FSCD ensures the timely transfer of resources to beneficiaries, while the Early Warning and Response Directorate (EWRD) is responsible for the early warning system that informs the PSNP risk financing mechanism and for the transport and monitoring of in-kind resources. The NRMD through its PWCU provides technical coordination and oversight of PSNP public works. The AED coordinates HABP-related services. The Cooperative Promotion Agency oversees capacity building to RuSACCOs. 102. The Joint Government-Development Partner Strategic Oversight Committee (JSOC, previously the JCC) provides ongoing support to and supervision of program implementation. Three Joint Technical Committees report to the JSOC that assesses program implementation. The PSNP Risk Financing Management Committee is responsible for the allocation of transfers to targeted beneficiaries in PSNP woredas through the PSNP risk financing facility. 103. At regional level, the Regional Cabinet approves the safety net and household asset building annual plans and budgets. The Regional Food Security Steering Committee, chaired by the Regional President or his delegate, oversees implementation of the Programs, while the Head of BOARD is responsible for the management of PSNP and HABP and chairs the Regional Food Security Taskforce, to which three Technical Committees report. The Regional Food Security Coordination Office (RFSCO) is responsible for day-to-day coordination of the programs, including the timely delivery of resources to beneficiaries, while the Early Warning and Response Department is responsible for the collection and analysis of early warning data. 24 104. The Bureau of Finance and Economic Development (BOFED) is responsible for overall financial management at Regional level and channels cash transfer to woredas. The Natural Resources Management Department manages the public works component of the PSNP through the PWFU. The Extension Department is responsible for the effective implementation of capacity building activities and systems change to ensure the effective delivery of technical advice to households, market analysis and input sourcing relating to the delivery of the HABP. The Cooperative Promotion Bureau provides technical backstopping to RuSACCOs. 105. At woreda level, the woreda cabinet prepares, and the woreda council approves, the PSNP and HABP annual plans. The council assists in resolving unresolved appeals, while the cabinet ensures that program plans, budgets, listing of appeals and appeals resolutions are posted in public locations. The Woreda Office of Agriculture and Rural Development (WOARD) is responsible to manage the PSNP and HABP. The Woreda Food Security Task Force (WFSTF), with WOARD as chair, reviews kebele annual PSNP and HABP plans and budgets, ensures that contingency plans for PSNP risk financing are in place, participates in monitoring and evaluation activities and provides assistance to kebeles. Three Technical Committees report to the WFSTF. The Woreda Food Security Desk (WFSD) coordinates safety net and household asset building activities. The Woreda Office of Finance and Economic Development (WOFED) ensures that the budgets for the safety net and household asset building programs are received in a timely manner at woreda level and subsequent transfers to beneficiaries are undertaken on a timely basis according to the minimum performance standards. The Woreda Cooperative Promotion Office will assist in capacitating existing RuSACCOs and establishing new ones, and implement directives to improve the regulatory environment. 106. At kebele level, the kebele cabinet approves the beneficiary list for the safety net and related plans for PSNP and HABP. It also assists in establishing and ensuring effective operation of the Kebele Appeals Committee (KAC), which is tasked with hearing and resolving appeals regarding Safety Net and HABP in a timely manner. The council/cabinet posts the lists of beneficiaries, appeals heard and resolved, and program plans and budgets in public locations. The Kebele Food Security Task Force (KFSTF) oversees all planning and implementation of safety net and household asset building activities. 107. At community level, the Community Food Security Task Force (CFSTF) is responsible for identifying beneficiaries of the PSNP. It also participates in mobilizing the communities for participatory planning for public works and HABP activities. The DAs play a facilitating role in PSNP and HABP implementation. DAs are members of the CFSTF and KFSTF. They support communities in the preparation of annual plans, oversee implementation of public works, prepare PSNP payment lists for submission to the WFSD and WOFED, provide training to households on investment opportunities, and assist households to prepare business plans. 108. At beneficiary level, beneficiary and non-beneficiary households participate in public meetings on PSNP and HABP that target beneficiaries and determine multi-year annual plans. Community members work with DAs on an annual basis to determine priority public works and participate in the consultative meetings to identify viable household-level investment opportunities. PSNP beneficiaries participate in public works or direct support, while those 25 engaged in HABP devise business plans, seek support from local financial service providers, and carry out these activities. Beneficiaries and non-beneficiaries alike also play a key role in holding implementers to account through the KAC and public forums. 3. Monitoring and evaluation of outcomes/results 109. The Government has established a single monitoring and evaluation system for the FSP, which includes the PSNP and HABP. This system is designed to assess progress towards higher level objectives while also responding to the realities of collecting regular monitoring data through Government systems. The system is guided by an integrated logical framework (logframe) to ensure progress towards the overall objective of reducing food insecurity in rural Ethiopia. 110. Regular monitoring data is collected through Government implementing bodies and consolidated by the FSCD on a quarterly basis. Monitoring of public works is the responsibility of NRMD, while the AED monitors the implementation of HABP technical services and overall financial reporting is carried out by MOFED. FSCD combines and analyses these data to prepare a comprehensive and continuous assessment of progress in implementation. In addition, FSCD will report on the use of PSNP risk financing resources following the release of these resources and subsequent implementation. 111. Regular monitoring data is augmented with real-time data from a range of sources to inform management decisions, particularly on the PSNP. The FIC collects data on the timeliness of PSNP transfers and market prices from a sample of 80 woredas. RICs will be established as part of APL III. The Rapid Response Mechanism (RRM) is the cornerstone of the Government's Risk Management Strategy for the PSNP. The RRM detects problems that warrant immediate attention and responds rapidly to resolve the problems. 112. A number of systems audits were adopted in APL II to improve information flows on systems and processes, particularly at woreda level, with the aim of strengthening accountability. The Appeals and Procurement Reviews will continue on an annual basis. The Roving Financial Audit will be consolidated with the Annual Audit and will be strengthened in the area of reconciliation of the payroll with household level data. A Commodity Audit of food management systems and practices will be included in APL III. 113. The Program commissions a number of independent studies and reviews to assess progress towards outputs, financed both by Government and the MDTF. This includes annual reviews of public works planning and the technical quality of the design and implementation of public works. A review of PSNP Risk Financing will be carried out each time it is triggered to determine the effectiveness of the response. The wage rate study is conducted annually to ensure that the PSNP provides an appropriate cash wage rate. A social assessment will review the effectiveness of program targeting and assess whether the program continues to target the poorest and most vulnerable. 114. The impact of the FSP, particularly of the PSNP and HABP, is evaluated through a set of independent evaluations. A regionally representative household survey is carried out every two 26 years to assess the impact on direct and indirect beneficiaries. This survey also provides valuable information on implementation progress. A second evaluation assesses the impact of PSNP public works at community level by sampling watersheds. 4. Sustainability 115. The concept of sustainability is central to the PSNP. The genesis of the Program was a recognition that the emergency system was a fundamentally unsustainable mechanism for supporting the chronically food insecure population. An alternative developmental approach was needed in order to provide the basis for a sustainable solution. The addition of "scalability" to the PSNP under APL II further built on this premise. Looking ahead there are four elements of sustainability that need to be considered. These are: 116. Commitment to ensuring a quality program: The recent impact evaluation has confirmed that the quality of program implementation varies significantly between and within regions. It confirms that when the Program is implemented as designed, impacts can be substantial. When implementation is poor, much of the impact is lost. Improving performance is therefore critical to achieving the Program's objectives. As part of APL III, emphasis will be placed on performance monitoring and incentivizing improved performance. The reforms to the HABP have significantly improved the design of that program and promise to further enhance the potential for graduation. But again, it is the actual implementation of this new program - guided by responsive management - that will determine whether the Program will achieve its objectives. 117. Ensuring mechanisms for the long-term operations and maintenance of public works: PSNP public works have already made a sizable contribution to environmental rehabilitation and the provision of social and productive infrastructure. In order for these benefits to be sustained, appropriate operations and maintenance arrangements must be in place. Evidence so far suggests that these mechanisms are largely in place. However, problems persist for a significant minority of projects, particularly in the roads and water sectors. The Project will address these by: (i) upgrading the training that woreda implementers receive; (ii) integrating how operations and maintenance arrangements are determined into the planning process; and (iii) further strengthening the integration of the public works within woreda plans to ensure that necessary complementary budgets are considered. 118. Building broad-based support for the program: At a broader level, continued local and international support for the Program over the long term will require not only good performance but also: (i) a transparent program administration that subjects the Program to public scrutiny and shows that it is fair; and (ii) evidence that the Program is functioning as designed and is generating the results that it aims to achieve. Achieving this requires continued implementation of a robust and transparent M&E system with further emphasis placed on performance monitoring as well as implementation of a comprehensive communication strategy. 119. Ensuring sufficient financing in the medium and long term: The financial sustainability of the Program is affected by a number of diverse factors including, but not limited to, the occurrence of any large shocks, the pace of graduation, and the rate of inflation. The PSNP must also address relatively short-term financial sustainability issues. While all development partners 27 are committed to a multi-annual framework and the next five-year phase of the Program, PSNP is not yet fully financed. Excluding the resources needed for the HABP, the PSNP budget for five years would cost approximately US$2,173.5 million. At this stage US$1,654.8 million has been indicatively committed (see Annex 5). Nevertheless, this represents a significant financing gap. Due to additional development partner financing anticipated before the approval of APL III, this gap is likely to be reduced. The strategy of the Government and development partners to address this will be to ensure that on a rolling basis the next two program years are adequately funded. This strategy was successfully employed during implementation of APL I and II to overcome financing gaps that were highlighted at the time of Executive Board approval. However, there is a risk that the current global economic climate may affect development budgets. Current indications from development partners are that financing for social protection programs is likely to be protected from any budget cuts. 120. Should the PSNP remain underfinanced, measures could be taken to scale back the design of the Program, which is sufficiently flexible to allow such changes. This approach is, however, undesirable given the vulnerability of target households, which if not covered by the PSNP would likely require support through the emergency appeal system. Development partners recognize that the PSNP is a more effective and efficient response than the emergency appeal system providing an additional incentive to ensure that the PSNP is fully financed. 121. With regards to long-term fiscal sustainability, program spending is at present equivalent to around 1.2 percent of GDP. Although currently financed by development partners, this resource envelope is within the limits considered sustainable for a safety net system financed from domestic resources. It was noted that the overall cost of the Program is expected to fall over the medium term at a rate commensurate with the pace of graduation. 5. Critical risks and possible controversial aspects 122. A number of country-level risks have been identified. These include macro-economic risks, particularly mounting balance of payments pressure and the negative effects of the global economic slowdown. Beyond the significant risk present at the country level, there are a number of sector- and project-specific risks that need to be considered. The central challenge for the PSNP is the use of decentralized systems with limited capacity to deliver a targeted cash and in- kind transfer program to millions of beneficiaries living in a large and diverse country. 123. While stakeholder dialogue, implementation experience, and capacity building over the last four years have reduced risk overall, significant risks remain. These are detailed in the matrix below. The overall risk rating of the project is substantial. Risk factors Description of risk Rating Mitigation measures Residual Rating I. Sector Governance, Policies and Institutions Vulnerability Ethiopia remains vulnerable to High The PSNP design aims to effectively Substantial to shocks significant shocks. In 2008, record respond to shocks, including those food price inflation and localized associated with climate change by: (i) drought increased food insecurity for adaptive measures such as soil and water many vulnerable segments of the conservation activities and small scale population. irrigation and the focus on integrated 28 Risk factors Description of risk Rating Mitigation measures Residual Rating watershed management; and (ii) the The possibility of price and/or capacity to scale up in response to weather-related shocks affecting the shocks, guided by woreda-level risk target population during the life of the management plans and financed through Program remains high. Moreover, the the use of contingency budgets at the impacts of climate change are likely woreda and regional level, and risk to exacerbate these vulnerabilities. financing at the federal level. Program Although there is no evidence to date Substantial The Program has developed multiple Moderate governance that this may be an issue, program avenues for citizen participation and & Financial resources may be used for personal voice in the Program with the aim of management ends or special interests, which can strengthening its overall accountability. reduce the availability of resources to Additional to this has been strengthening vulnerable households. This is systems for "upwards" accountability. especially of concern leading up to The following measures have been the 2010 elections, when program implemented to strengthen the Program's resources may be manipulated for governance framework: (i) an political purposes. Such an occurrence independent appeals/ redress procedure; could significantly damage the (ii) a communication campaign focusing credibility of the Program at local and on financial transparency; (iii) fiduciary national levels. It also implies controls verified by the Annual Audit, significant reputational risk for the including the interim Financial Audit, World Bank. the Procurement Review, with actions taken at woredas-level and system-wide Evidence to date suggests that overall based on findings; and (iv) governance of the Program is good computerization of the payroll system and that there are no systematic (PASS) rolled out under APL II will be problems. Concerns around the made mandatory under APL III. accuracy of payments to beneficiaries in some woredas have been investigated and actions agreed to strengthen the system. Fiscal The PSNP is a large program with Substantial The emerging evidence on outcomes and Moderate sustainability evolving financing needs. impacts provides a solid justification for Existing/new development partners existing partners to continue support for and Government may not be in the the Program. An MTEFF for the five- position to maintain the necessary year life of the Program is being agreed, long-term financing, particularly in which will maximize the multi-annual the current climate of economic commitments to the Program and uncertainty. This will undermine the therefore build predictability. ability of the Program to maintain Depending on the outcome of this current levels of support, as well as its exercise, a resource mobilization ability to scale up in response to strategy will be developed to fill the shocks. Currently there is a residual financing gap. A similar strategy substantial financing gap for the was effective at addressing significant Program. financing gaps during APL I and II. Linkages to The FSP (including the PSNP and High As part of APL III, re-design of the Substantial the enabling HABP) is a necessary but not HABP will focus on delivery of environment sufficient condition to enable household credit through MFIs or & growth graduation. The broader enabling RuSACCOs. This approach promises to environment plays a critical role and leverage the diversified package of linkages to it therefore are important financial services that these institutions for meeting the Program's higher- can offer. It will also actively promote level objectives. This includes off-farm income earning opportunities. strengthening multi-sectoral linkages 29 Risk factors Description of risk Rating Mitigation measures Residual Rating to other sources of growth and basic The complementary AGP planned for services. If the broader rural growth FY2010 aims to improve agricultural process remains weak, household systems and prospects for improved level graduation will likely remain agricultural productivity. limited. Graduation Use of ambitious household Substantial The Government has developed an Moderate targets graduation targets by the Government empirical evidence-based definition of may provide perverse incentives at graduation. Regions and woredas have lower-levels and result in premature been instructed not to exclude graduation and therefore increased beneficiaries that have not met the household vulnerability. graduation benchmarks and to include households identified for graduation for one additional year. DRMFSS will also clarify to Regions that beneficiaries have guaranteed access to the Program for at least three years. Impact of the A large number of NGOs participate Moderate The Government has confirmed that Low CSO law on in the PSNP providing support in NGOs working on a Government the PSNP specific woredas to all aspects of program are legally permitted to program implementation. These continue with program activities. The woredas are financed directly and PSNP activities are related to "service solely by USAID. The newly delivery", an activity that is explicitly approved CSO law may affect their permitted under the law. The PSNP ability to be effectively involved in DWG is in the process of revising the the program. MOU with the Government, which will reflect the role of NGOs. III. Operation-specific Risks Technical Design Quality & If the HABP is not an effective Substantial As part of APL III, technical assistance Moderate coverage of complementary intervention to the will be provided to ensure that the Household PSNP, the probability of household modalities for enhancing fund flows to Asset Building graduation will be reduced. financial institutions operating in food Program insecure communities follow sound The Government has re-designed this financial principles. Continuous program and significant support from awareness creation at all levels will APL III is envisaged to help make it ensure widespread understanding of the more effective. Nevertheless, a Program. Furthermore ongoing commitment to implementing the new supervision of the proposed reforms approach is required, particularly in under the HABP will provide terms of promoting market-based opportunities for taking corrective approaches to providing credit and actions as necessary during enhancing the quality of technical implementation. support provided by the extension service. Implementation Capacity And Sustainability Capacity Despite significant improvements and Substantial Several ongoing procurement activities Moderate some important results, capacity related to building local physical constraints have not allowed the capacity are yet to be finalized (for Program to fully meet its objectives. example distribution of generator sets, Capacity gaps still exist with regard to large-scale vehicle procurement etc). coordination, planning, management, These will significantly address capacity results-based monitoring, financial constraints experienced to date. management, and procurement. High 30 Risk factors Description of risk Rating Mitigation measures Residual Rating rates of staff turnover, which are A major new capacity building facility particularly acute at woreda-level, (to be financed by CIDA) will be rolled affect the efficiency and effectiveness out before APL III and will focus on of implementation. building capacity for program management in all areas, particularly at The lack of capacity to effectively regional level. implement the HABP presents a new and substantial challenge that if not In order to address high rates of staff addressed appropriately will turnover, under APL III attention will be undermine the impact of any re- given to strengthening training programs design of this component undertaken for regional, woreda, and kebele staff as part of APL III. and offering these on a rolling basis. Sustainability Although an estimated 80 percent of Moderate The NRMD is now in a position to fulfill Moderate of community public works are rated satisfactory or its mandate to provide oversight of assets created better, the technical quality and public works, and will identify the most by Public maintenance arrangements for a effective way to upgrade capacity in key Works significant minority of projects are sectors. Further cross-sectoral problematic (i.e. roads and water). coordination at the federal level is planned to determine how to best address If public works are not built to identified capacity gaps. minimum technical standards, following good environmental The NRMD is presently engaged in a practice, and with the necessary major review of training and capacity- operations and maintenance building needs for improved arrangements in place, the environmental management and sustainability of public works, the performance of ESMF implementation. state of the environment and the In addition, an upgrading of the annual program impact will be undermined. CBPWDP training will incorporate specific modules on maintenance. These modules will be informed by the new rural road maintenance policy currently under development. 6. Grant/credit conditions and covenants 124. As evidence of completion of the agreed triggers for moving to APL III, prior to Appraisal, the Government of Ethiopia provided evidence of the following, in a form deemed satisfactory to the Association: (i) coverage of the household asset building component of the FSP increased as compared with the baseline in 2006; (ii) the PSNP fiscal cycle has been aligned with the Ethiopia Government Fiscal Year and PSNP plans are integrated with the woredas development plans; (iii) the review of implementation experience under the Access to New Lands program and the implications of any return migration to chronically food insecure areas for the PSNP was completed; (iv) the report "Social Protection in Ethiopia: Status Report and Next Steps", which includes a review of how the Government is addressing vulnerability in urban and rural non- FSP/PSNP woredas was completed; (v) a comprehensive public works impact evaluation was completed and the findings are under discussion with Government; and (vi) the 2008 IFPRI/Central Statistical Agency (CSA) Impact Evaluation of the FSP was completed. 31 6.1 Conditions of Effectiveness · Provide evidence of measures taken to establish and operationalize Regional Information Centers covering Afar, Amhara, Oromiya, SNNP, and Tigray Regions. · Provide evidence of measures taken to ensure use of PASS in all participating woredas in Amhara, Oromiya, SNNP and Tigray Regions. · Adoption of a revised PSNP Project Implementation Manual, satisfactory to the Association, and consolidated PSNP work program and budget for EFY 2002. 6.2 Conditions of Disbursement · For Component 4 "Support to Household Asset Building", submission of HABP work plan prior to operational launch. 6.3 Dated Covenants Mid-Term Review by June 2012 · Evidence that the National Social Protection Strategy has been finalized. · Evidence of an approved Disaster Risk Management Policy. · Independent process evaluation of experience to date with the Risk Financing within the broader emergency response system. · Independent assessments of PSNP based on a representative household and community- level public works surveys. · Independent assessment of effectiveness of systems development for HABP. · Based on an analysis of an updated MTEFF, an agreed set of actions to address any residual financing gap, including any revisions to the program design as may be necessary. D. APPRAISAL SUMMARY 1. Economic and financial analysis 125. The economic benefits of the PSNP are represented by: (i) improvements in household well-being as a result of consumption smoothing, asset protection, and avoidance of negative coping behaviors; (ii) enhanced livelihoods through asset accumulation and increased productivity; and (iii) increased use of social services, market access and agricultural productivity as a result of the infrastructure created through the community public works. PSNP can be seen to provide both protective and productive benefits to households and communities. There are additional economic impacts expected from the HABP. 126. Household-level benefits. The 2008 impact evaluation concluded that the Program is smoothing household consumption and protecting assets, even during times of crisis. The transfers provided to households are equivalent to about 40 percent of annual food needs. While cash transfers are largely used for consumption purposes, roughly 25 percent of funds are invested in productive assets. There is emerging evidence that participation in the PSNP supports 32 households to adopt high risk/high return strategies such as taking credit leading to higher rates of agricultural productivity. This has the potential to transform the livelihoods of households participating in both the PSNP and HABP. 127. Community level benefits. A growing body of evidence shows the positive impact public works are having on rural livelihoods. Soil and water conservation projects have resulted in significant and visible increases in wood and herbaceous vegetation cover and a broader range of plant species diversity. Small-scale irrigation projects were found to have increased incomes between 4-25 percent. In addition to health gains from greater access to clean water, water projects were found to significantly reduce the distance women and children travel to fetch water. An analysis for a sample of projects found positive benefit-to-cost ratios ranging from 3.7 for water supply interventions to 1.8 for soil and water conservation and health infrastructure. A composite economic rate of return is not possible to calculate because the specific public works projects are not identified in advance and, for most of the soil and water conservation and rural road projects, expected rates of return vary widely depending on location. 128. Overall program efficiency. The use of cash transfers creates administrative efficiencies, largely by reducing the costs of transporting food. These gains have been calculated to be over US$21 million annually. PSNP compares favorably with international efficiency benchmarks for safety net and public works programs in terms of labor intensity and targeting. In the PSNP, 7.8 percent of program resources are absorbed in administrative costs, which is below the international benchmark of 10 percent for a well-run safety net program. 129. Economic analysis of the HABP. Preliminary analysis suggests that investing in productive household assets can result in high rates of return. A recent survey found that 85 to 90 percent of households experienced positive rates of financial return from household investments through the FSP. Benefit-to-cost ratios for a sample of household level investments were between 1.97 and 5.63. 130. Fiscal and macroeconomic implications. The PSNP annual budget is currently about 1.2 percent of GDP. In comparison, most developing countries spend between 1-2 percent of GDP on safety net programs. To date, the shift to cash transfers through the PSNP appears to have little, if any, inflationary effect. Analysis found that the recent food price inflation in Ethiopia can be largely explained by overall nominal increases in prices. A 2008 woreda-level analysis similarly concluded that there is no evidence to suggest that increasing the size of PSNP cash transfers can fuel inflation. Of more concern, however, is the impact inflation can have on the purchasing power of the cash transfers. Continuous monitoring is required to ensure that the value of the cash transfer is not eroded by inflation, thereby undermining the move to cash transfers. 2. Technical 131. APL III is built upon the significant technical groundwork laid during the first five years of program implementation as well as an extended formulation process for the new FSP that has strengthened the design of the PSNP and substantially reformed the HABP. 33 132. For the PSNP there is a Program Implementation Manual (PIM) and related Operational Summary and detailed subject-specific guidelines. The PSNP PIM is being revised to reflect design changes agreed as part of APL III. A comprehensive HABP PIM to guide implementation will be developed based on the agreed program document. 133. Under the PSNP, Government continues to conduct training programs on the different guidelines using a cascade training approach. Under APL III further upgrading will be undertaken to ensure a more systematic approach to transferring knowledge and skills to program implementers. This will be informed by efforts to continually evaluate the curricula and quality of the training provided. This initiative is particularly critical in light of the recent Business Process Re-engineering (BPR) and the high levels of staff turnover within the public service. 134. One of the most significant training programs is training on participatory planning processes, which will be delivered in 2009 to over 7,000 DAs. The DA training includes instruction in the work norms and sectoral technical standards (roads, irrigation, soil and water conservation, etc.) that have been developed to ensure the technical quality of the public works projects produced under the Program18. 135. The objective of the Support to Household Asset Building Component is to: (i) improve the ability of poor households to undertake simple budgeting and business planning; (ii) improve the identification of opportunities for households through local market value chain analysis; and (iii) strengthen the technical training and ongoing support provided by DAs to households in the use of new technologies and systems adopted. Overall this package will aim to strengthen the application of well-designed and financially sustainable livelihood opportunities. Ongoing training to local level implementers will ensure that this is systematically applied across the Program. 3. Fiduciary 136. Financial Management: The FM arrangements for the Project rely substantially on the country's regular Public Financial Management (PFM) system. Specifically, for budgeting, internal control and accounting arrangements country systems are used. The regular government Chart of Accounts will be used with necessary modifications to accommodate the specific needs of the Project. 137. The strengths of the country's PFM system, notably the budget process and compliance with financial regulations, and the well defined computerized accounting system, particularly at federal and regional levels are important. However, there are also deficiencies in this system. These include a shortage of qualified accountants and auditors, mainly at woreda level, and the limited attention of the internal audit system, for various reasons combined with the system's limited capacity to effectively function. Program financial reporting requires submission and consolidation of timely and accurate reports from a large number of decentralized institutions. The scale of the Project and complexity arising from the large number of local-level 18 The technical standards take the form of "infotechs" that provide detailed technical specifications for around 70 different project types. 34 implementing institutions pose challenges. Under APL II delays were experienced in the submission and quality of both financial and audit reports. The experience gained under APL II and I have built significant system capacity resulting in both reduced delays in audit and financial reports as well as higher quality reporting overall. 138. Notwithstanding, APL II audit reports noted a number of accounting and internal control weaknesses for which the Government will take actions to rectify. These have been detailed in action plans, the implementation of which will continue to be reviewed by the World Bank and development partners. Actions under APL III to address the weaknesses noted include: (i) ensuring that the FM Manual for the Project is clear and that there is adequate provision of on- going training; (ii) providing FM support, supervision, and monitoring to all local-level implementing entities by MOFED and BOFED; (iii) ensuring timely auditing with the early appointment of auditors; (iv) ensuring a more robust use of the interim audit function (transaction-based system and internal control testing formally known as the roving audit); and (v) linking the interim audit with the final annual financial audit to minimize delays and facilitate more timely completion of the external audit. The FM risk for the project is rated as high, but with the mitigating measures this is reduced to substantial. 139. Procurement: Procurement will largely continue to follow the pattern developed under APL I and II. Procurement will 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 "Guidelines: Selection and Employment of Consultants by World Bank Borrowers" dated May 2004; revised October 2006. The procurement unit within DRMFSS will manage the procurement of all International Competitive Bidding (ICB) packages, including the bulk procurement of food at federal level. The Regional Procurement Coordinators will manage the procurement of materials consolidated into sizeable National Competitive Bidding (NCB) packages while the Purchasers in the WOFEDs will be responsible for the purchase and delivery of tools and materials to woredas, kebeles and communities, for use in the public works program. As part of APL II, procurement was made more efficient at the regional and woreda levels. In particular, the increase in the threshold for ICB procedures for goods to US$500,000 has provided the option of consolidating larger packages at the regional level for procurement using NCB. The stocking of construction materials and cement at regional level is also possible. Procurement Manuals have been prepared for the PSNP and training has been provided at all levels during the implementation of APL II. 140. Procurement efficiency for APL III is expected to improve given the recruitment under APL II of the Regional Procurement Coordinators within each program region and the introduction of the Regional Support Facility for building capacity at regional level. Given this, further emphasis is being placed on decentralizing the activities of capacity building in procurement to the regions. The BPR exercise has also provided additional capacity for procurement directly within the DRMFSS, which should help further streamline procurement at the federal level. As a result of the BPR a new procurement unit specifically for the two Directorates of the DRMFSS has been created. In the past both directorates would receive procurement services from the procurement unit of the MOARD, which was over-stretched. Further emphasis will be placed on early conclusion of pre-procurement activities such as design and technical specifications and the earlier execution of capacity building programs to more 35 effectively support the Program. At the woreda level, the use of Purchasers from the pool has been challenging for prioritizing PSNP purchases. It is agreed that under APL III, there will be one Purchaser designated to focus on PSNP purchases. 4. Social 141. Evidence suggests that the targeting of the PSNP has been good and has improved since 2005; beneficiaries and non-beneficiaries alike report that the targeting process is fair. However, there continue to be reports of exclusion in program woredas, which is attributed to the high rates of poverty in rural areas. As part of APL III, the PSNP communication strategy will be enhanced to ensure that communities are well informed of the targeting criteria for, and objectives of, the Program. At the same time, the KAC will be strengthened and client cards will be introduced to provide evidence of entitlements and proof of payment. These measures are designed to deepen local accountability structures to better enable beneficiaries and non-beneficiaries to hold local- decision makers to account (see Annex 9). 142. The ESMF screening process, which is mandatory for all public works, assesses and minimizes potential impacts such as increased pressures on human settlement, disturbance to religious sites, increased social tensions over asset allocation, and the capacity to enforce measures such as the prohibition of open grazing. Furthermore, for any public works activities that might involve voluntary loss of minor assets, the ESMF incorporates a Voluntary Loss of Assets procedure to ensure that good practice is followed. Public works within or in the vicinity of cultural heritage sites, including community sites of cultural importance, are avoided whenever possible. If the site cannot be avoided, the matter is referred to the regional Environmental Protection Agency (EPA). 143. Women in rural Ethiopia have a heavy workload of both productive and reproductive tasks and their participation in decision-making can be limited. The design of the PSNP has aimed to address these issues. Women who participate in public works are required to work fewer hours than male participants in recognition of their reproductive work. Women who are pregnant or breastfeeding are moved from public works to direct support. A recent study concluded that women are well represented in most PSNP decision-making bodies, particularly at local levels. Despite this, the study noted that some women experience difficulties expressing their views in public forums and accessing the KAC. In response, the Program has ensured that representatives of the Women's Affairs Desks (WAD) are included in the KAC. 144. A study concluded that the PSNP is currently not increasing the vulnerability of beneficiaries to HIV/AIDS. The study found that there are some elements of the PSNP that have the potential to increase the risk of HIV, such as travelling to collect food, which often requires people to spend the night at distribution points, and the increased movement of program staff. However, social norms and the design of the Program suggest that such risks will be low. For example, beneficiaries travel in groups to food distribution sites, while the progressive move toward cash transfers will eliminate the need to travel in the long-term. In spite of this finding, the study recommended that the PSNP adopt a mainstreaming strategy to mitigate any potential risk and to take advantage of the program infrastructure. 36 145. Pastoral and agro-pastoral regions were not included in the rollout of the PSNP in 2005 because of the different livelihood strategies employed by people living in these areas, coupled with some significant capacity constraints. Since 2008, a pilot program has been operational in 18 woredas to inform the full-scale implementation of the PSNP in pastoral areas. The HABP may include risk management activities in addition to household asset building to address the vulnerabilities of pastoral livelihoods. 146. A large body of literature suggests that the provision of micro-credit has the potential to change gender relations within households. Intra-household dynamics will be monitored through the FSP household survey and any adjustments made to ensure that both men and women benefit from HABP. In addition, the FSP review concluded that the access of Muslim households to credit is severely constrained in some areas because of concerns with interest payments. Alternative arrangements are being explored as part of the program design. 5. Environment 147. One of the key objectives of the PSNP is to address the underlying causes of food insecurity, to which environmental degradation is universally agreed to be a major contributor. The adoption of the integrated CBPWDP approach during APL II has already had considerable positive impacts on the environment (see Annex 10). Public works activities under APL III will continue to follow the CBPWDP approach, and are thus expected to constitute a vehicle for continued environmental transformation. Experience shows that these positive environmental impacts will, in turn, enhanced productivity and livelihoods. Thus the emphasis in APL III is environmental transformation coordinated with household-level interventions and opportunities for livelihood enhancement. 148. While these environmentally beneficial impacts from the public works are expected to continue, past mass mobilization efforts in environmental rehabilitation under previous governments in Ethiopia have frequently failed or been abandoned. Such adverse outcomes occur particularly if the location or design of public works does not follow good environmental practice or are incompatible with optimum overall management of the watershed. 149. To ensure that standards are maintained to avoid such scenarios, the approach to the environmental performance and sustainability of public works is three-pronged: (i) public works are derived from a community-based approach to integrated watershed management, supported by a budget to provide technical and material inputs; (ii) the design and implementation of the public works follow the standards set out by MOARD, which are made available along with training to concerned woreda staff and DAs; and (iii) public works projects are screened for possible negative environmental impacts thereby ensuring that these project designs not only incorporate appropriate mitigating measures but also comply with both Ethiopia's Environmental Impact Assessment (EIA) proclamation and the ESMF. 150. While any impact from the community public works would be limited in scale and site- specific, the ESMF is designed to identify and mitigate potential adverse impacts such as: 37 · Community road construction altering drainage patterns and increased flooding and soil erosion from materials excavation sites; · Small-scale irrigation projects causing depletion of surface or groundwater sources; · Afforestation projects causing reduction in tree and plant species diversity arising from the introduction of new plantations and re-vegetation schemes; · School and health facilities causing increased production of human and medical wastes and potential for contamination of water bodies and groundwater. 151. Responsibility for overall coordination of ESMF implementation lies with the NRMD through its PWCU. Responsibility for managing the process and ensuring that there is sufficient capacity at the lower levels lies with the regional PWFUs. Individual public works sub-projects screening is conducted by the DA, and supervised by the Natural Resources Expert in the woreda Natural Resources Desk. 152. The public works reviews under APL II indicate that despite the improvements in ESMF implementation experienced since APL I and the lack of significant concerns about negative environmental impacts associated with public works, there is still a need to strengthen implementation of the ESMF to ensure that standards continue to improve. Consequently, the NRMD has considerably strengthened the PWCU and is further committed under APL III to: (i) strengthening the PWFUs to ensure that the ESMF is implemented to a high standard; (ii) upgrading the ESMF training materials; (iii) providing further training, guidance and support in ESMF implementation to regional and woreda technical staff and DAs; and (iv) strengthening the monitoring of ESMF implementation in the PW M&E system. Since the PSNP does not normally include public works necessitating a separate EIA, it is considered that with support from the newly strengthened PWCU, the Regional EPA will have sufficient capacity to review and determine, whether an EIA is required, and to make any necessary arrangements. 153. Since each household-level activity will be at micro-scale, and as the procedures under the HABP will include the assessment of the agro-ecological suitability of the activity, including screening for potential negative environmental impacts, no significant site-specific negative impacts are expected from these household-level activities. The only environmental concerns might be potential cumulative negative impacts in the longer term of large numbers of households adopting new activities in fragile environments over a number of years. 154. There will be linkages between HABP activities and PSNP public works. Thus public works in a community watershed will have a bearing on which income-generating activities will be viable for the beneficiary households, and in some cases should be a condition for granting the concerned credits. This harmonization will be ensured by collaboration between the Natural Resource Expert in the Woreda Natural Resources Desk and the Natural Resource Expert in the Extension Services Desk. 38 6. 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] [] Physical Cultural Resources (OP/BP 4.11) [X] [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] [] 155. Although a large number of public works activities will be implemented under APL III, they are generally micro-level, and it is expected that any negative impacts will be site-specific; few if any of the impacts would be irreversible, and in most cases mitigating measures can be readily designed. Thus EA Policy OP 4.01 is triggered and the project is designated Category B. Also under OP 4.01, in case the public works should include clinic construction or rehabilitation, a Medical Waste Management Guide for Rural Health Clinics is required to be developed and disclosed. This has been done. The technical advice and training provided under the HABP will include consideration of the agro-ecological suitability of each activity at individual household level, and guidelines for simple environmentally-friendly family business plans to be incorporated in the household-level credit and grant applications procedure. In addition, procedures will be implemented for tracking, mitigating and correcting any potential cumulative impacts that might arise in the long term. The management and monitoring of the implementation of these procedures will be the responsibility of the PWFU, overseen by the PWCU. The details of the management, monitoring, and oversight arrangements for the HABP environmental management system will be finalized by the PWCU when the final operational procedures for the HABP have been completed. The PWFUs will be responsible for ensuring that the PSNP public works and the HABP are fully harmonized in terms of environmental impacts and environmental management. 156. The Pest Management policy OP 4.09 is triggered as there is a possibility of small-scale irrigation projects using pesticides. In such cases, the policy requires the development and disclosure of the Government's Guidelines on the Implementation of Integrated Pest Management (IPM) for Small-Scale Irrigation Schemes in the Productive Safety Net Program, which has been done. These guidelines require an IPM Plan to be developed for each scheme and at minimum be supported by (i) Technical Assistance, training and awareness-creation from the Crop Protection Section of the BOARD; (ii) advice on Pest-Resistant varieties; (iii) supervision by the DA; (iv) technical information on pesticides as established by the Pesticides Registration and Control Decree No. 20/1990; (v) development and Implementation of arrangements for the safe use, handling and storage of pesticides, and the proper use, maintenance and storage of pesticide spraying equipment; and (vi) regular monitoring and reporting, which will be the responsibility of a woreda team of experts. * By supporting the proposed project, the World Bank does not intend to prejudice the final determination of the parties' claims on the disputed areas. 39 157. OP 4.11 concerning Physical Cultural Resources is triggered in the unlikely event that the small sub-projects should encounter any "chance-finds". The policy is addressed by being integrated into the ESMF screening process at three stages: (i) any sub-project located within a known cultural heritage site is earmarked as a sub-project of Environmental Concern, to be referred to the Regional EPA, which will decide if an EIA is required; (ii) assessment for potential disturbance to cultural or religious sites is carried out as part of the site-specific sub- project screening, which also contributes to a decision whether to earmark a sub-project for possible EIA; (iii) inclusion of assessment of potential cultural heritage impacts in the EIA of sub-projects, where EIA is found to be necessary; and (iv) monitoring of sub-project implementation by DAs and woreda staff in liaison with the Regional Bureau of Tourism and Culture. 158. The International Waterways policy OP 7.50 is triggered because some of the public works are likely to be small-scale irrigation projects located in watersheds of a number of international waterways. In accordance with this policy, the World Bank, on behalf of the Government, notified the concerned riparian governments, regarding the five-year period of APL III. The Government of Eritrea expressed concern regarding the initial estimate of potential water abstraction from the Mereb basin. Despite being provided with a more detailed estimate, the Government of Eritrea re-affirmed its objection. Since the accuracy of the abstraction estimate provided to Eritrea was not in dispute, this was not considered a case that required the opinion of independent experts. The World Bank's assessment is that the Project will not cause appreciable harm to any of the riparian countries concerned. 159. The costs of implementing the environmental and social safeguard measures are covered partly by the management budgets at federal and regional levels and by the regular government staffing and overhead costs at all levels. The estimated total implementation cost for the safeguard measures is US$2,222,000 per year. Of this amount, US$1,907,000 will be covered by the provision of regular government staff time and US$315,000 will be covered by the project (see Annex 11). 7. Policy Exceptions and readiness 160. Policy Exceptions: The proposed project does not require any exceptions from World Bank policies. 161. Readiness: As part of the PSNP APL series, the main design parameters and operating systems of the proposed project are the same as those of the first two phases. The design for APL III was strengthened in numerous ways, particularly with respect to: (i) program governance; (ii) technical quality and sustainability of PSNP public works; (iii) program management; (iv) consolidating risk financing and links to the early warning system; (v) strengthening program monitoring for results; and (vi) creating linkages to maximize synergies across programs. Necessary changes to the operational procedures of the PSNP will be captured in a revised PIM, which will be provided to implementers together with associated training prior to program launch. 40 162. Consistent with the original plans envisaged for the final phase of the PSNP APL series and in line with the reformulation of the Government's FSP, the design of APL III has included a new component to support the reform of the Government's HABP with the aim of improving its effectiveness and efficiency. A detailed HABP document has been developed and approved. Based on this document an operational manual will be finalized prior to the start of implementation of this component. 163. The project is deemed ready for implementation, subject to the fulfillment of the conditions for Effectiveness, and Disbursement. 41 Annex 1: Country and Sector or Program Background Achieving Graduation: Pathways, Obstacles and the Wider Framework ETHIOPIA: Productive Safety Net APL III 164. The overall objective of the Government's Food Security Program (FSP) is to achieve food security for chronic and transitory food insecure households in rural Ethiopia19. Despite significant investments in the FSP since 2005, there has been little graduation from food insecurity to date. This Annex explores the nature of food insecurity in rural Ethiopia together with the opportunities and challenges for households seeking to improve their food security status. It concludes that there is significant potential for household graduation if the FSP is implemented as designed, including the recent reforms to make the program more demand- driven and market-oriented. To achieve large-scale graduation a range of program and policy- related obstacles will also need to be overcome. 1.1 Food security and poverty 165. The current state of food insecurity in rural Ethiopia is severe. The 2004/5 Welfare Monitoring Survey (WMS) found that 34.2 percent of rural households suffered food shortages in the previous 12 months and 15.2 percent of rural households had a food gap of longer than four months. 166. Food insecurity is a specific manifestation of the widespread poverty in Ethiopia. In 2004/05, 39.3 percent of the rural population lived below the poverty line20. Other indicators of poverty are equally as stark: 48.5 percent of children in rural areas are stunted, net enrolment rates for primary and secondary schools in rural areas stand at 32.8 percent and 8.3 percent respectively, and the vast majority of rural households reside in single room dwellings, which are usually shared with livestock21. Poverty is multi-dimensional; it affects and is the result of a combination of factors that include income, assets, location, political influence, and risk. 167. Poverty is often measured with household income, an indicator of a household's ability to obtain adequate food or secure other basic needs such as shelter or health care. Low levels of income often reflect limited assets, although the two are not always perfectly related. While the Gini coefficient for income (0.26) suggests limited inequality in rural Ethiopia22, the Gini coefficient for cultivated land is 0.47 without taking into account landless rural households23. Similar disparities exist for livestock holdings: better-off households in Tigray and Amhara had more than four times the livestock holdings of their poor counterparts24. Lack of plough oxen significantly reduces the amount of land a household can cultivate and often forces those without 19 Food security is defined as "access by all people at all times to sufficient food for an active and healthy life". Chronic food insecurity refers to the persistence of this situation over time, even in the absence of idiosyncratic or covariate shocks. 20 The poverty line has been set at ETB1,075 birr per adult per year at 1995/96 national average constant prices. Household Income Consumption and Expenditure Survey (HICES) 2004/05 21 WMS 2004/05. 22 HICES 2004/05. 23 Well-Being and Poverty in Ethiopia: The Role of Agriculture and Agency, World Bank 2005. 24 LIU Data set collected from 2006-2008. Data from the Livelihoods Integration Unit (LIU) of the Disaster Management and Food Security Sector is collected using an approach called the Household Economy Approach (HEA). 42 plough oxen to rent out their land to better-off households through share-cropping agreements. This limits the ability of households to meet their own food needs through agricultural production. As a result, the majority of poor households in food insecure areas are net cereal buyers, purchasing around 30 percent of their food needs from the market25. 168. Assets include not only land and livestock but also skills and labor that can be affected by ill-health and high dependency ratios. Poor households are not only those with low livestock holdings. The poorest households are often those headed by women or an elderly person. Female-headed households make up 47.4 percent of the poorest rural quintile compared to only 11percent of the wealthiest26. 169. Access to land and labor is often related to phases in a household's lifecycle. Young households frequently have less land than those with older household heads27 and there is a growing young landless class. Young households also tend to have high dependency ratios leading to labor shortages. More severe, however, are the labor shortages experienced by elderly households or households whose lifecycle has been disrupted through the death of a spouse or divorce. 170. Poorer people are often found in poor communities whether as a result of natural endowment scarcity, degraded natural resources or remoteness from markets and services. Households living below 1,800 or above 2,400 meters above sea level have lower levels of expenditure than those living within this range28. An estimated 43 percent of households in rural Ethiopia have to travel more than 15 kilometers to access transport services while around 30 percent have to travel a similar distance to access veterinary services or purchase fertilizer29. 171. This geographic remoteness can be compounded by a lack of access to information. For instance, 18.5 percent of rural people have a radio30 and 41.9 percent lack weekly access to any media - radio, television, or newspaper31. This lack of access to information combined with low status and limited participation in social networks severely impedes poor people's involvement in decision-making processes. Furthermore, while indicators related to government effectiveness are increasing (from 12.3 in 1998 to 37.4 in 2007), indicators related to voice and accountability decreased over the same period (from 24.0 percent to 13.5 percent)32. 172. Life in rural Ethiopia is very risk prone and many households are pushed into poverty as a result of shocks. The 2004/5 WMS found that the most commonly reported shocks as perceived by households in rural areas were: food shortage, 28.1 percent; illness, 24.8 percent; drought, 12.8 percent; death of a household member, 7.7 percent. Livestock losses were also frequent with 9.2 percent of households reporting such an event in 2004. The prevalence of such shocks may 25 LIU Data set (ibid). 26 HICE 2004/05. 27 Households with heads 30 years of age have 10% less land than those who are 50. World Bank 2005 ibid. 28 World Bank 2005. 29 WMS 2004/05. 30 WMS 2004/05. 31 Demographic Health Survey (DHS), 2005. 32 www.govindicators.org. Similar findings were also made by a study looking at the Protecting Basic Services Programme: Enhancing Understanding of Local Accountability Mechanisms in Ethiopia, 2008, Alula Pankhurst. 43 explain, in part, the dynamic nature of poverty in Ethiopia. Analysis showed that while 22 percent of the population moved out of poverty between 1994 and 1999, 16 percent fell below the poverty line during the same period33. 1.2 Graduation from food insecurity 173. The FSP aims to enable food insecure households to achieve either food sufficiency34 (graduation from the PSNP) or food security (higher level graduation from the FSP). To this end, the Program has four components: i) The PSNP, which provides transfers to both meet household consumption and protect assets, and also builds community assets through public works; ii) the Household Asset Building Program (HABP) which provides credit and extension services; iii) the Complementary Community Investment Program which undertakes community infrastructure investments in food insecure woredas, and iv) the Resettlement Program which provides transfers, credit and infrastructure for resettled households. The movement of households to food sufficiency or food security these components aim to achieve is commonly described as graduation. 174. Of the four programs mentioned above, access to the PSNP and HABP are targeted to food insecure households in food insecure woredas. These are the focus of the discussion in the section on pathways below. However, the FSP does not operate in isolation and is affected by the wider context in which economic growth, rural-urban migration, population growth, and environmental change play their part. The impact of this wider context on graduation potential is also considered in section 1.5. 1.3 Pathways to household graduation Role of PSNP 175. The PSNP is expected to support graduation by: · Smoothing consumption to prevent households from selling assets; · Allowing some direct investment of PSNP transfers in productive household assets; and · Encouraging households to make higher risk/higher return investments (including taking credit) by reducing risk aversion. 176. In addition, the infrastructure developed through the PSNP public works and the cash injected in the rural economy, as a result of the larger proportion of cash transfers, are likely to have positive community-wide effects. 177. Smoothing Consumption: The mean value of payments received by households from the PSNP in 2007 was ETB 95035. PSNP transfers frequently represent around 40 percent of annual food needs36 and the vast majority of transfers are used for consumption37. Impacts on 33 World Bank 2005 ibid. 34 A household can be deemed food sufficient when, "in the absence of receiving PSNP [or emergency] transfers it can meets its food needs for 12 months and is able to withstand modest shocks". 35 Both food and cash payments, with food converted to a cash equivalent for the month in which it was received. Ethiopia Food Security Program: Report on 2008 Survey, IFPRI, 2009 (a). 36 Calculated from LIU data for Amhara and Tigray. SNNP was excluded as data relates to a mixture of PSNP and the preceding humanitarian assistance. 44 consumption and asset protection are much higher when transfer levels are significant and more regular. The recent impact evaluation found that caloric acquisition was 16.7 percent higher among households receiving regular transfers. Meanwhile distress sales for households receiving a high level of public work transfers were lower as compared to households receiving a low level of transfers38. 178. Investing in Assets: There is evidence that some households invest PSNP transfers in productive assets. A recent study found that 16 percent of public works beneficiaries bought seeds and 12 percent purchased livestock. Households also made larger investments in their children's education (31 percent)39. It is likely that a combination of asset protection, direct investments of transfers and increased available income are contributing to increases in Tropical Livestock Units (TLUs) among public works participants. The impact evaluation concluded that public works participants have a change in their TLU holdings of 0.281 ­ equivalent to nearly three sheep40. 179. Reducing Risk Aversion: It is more difficult to find conclusive evidence that PSNP is enabling households to make higher risk/higher return investments, although indications are positive. There is evidence that households receiving public works transfers are increasing the amount of land they have under cultivation and are experiencing improved yields, but these are not statistically significant. More positive are the significant increases in overlap between participation in the PSNP and uptake of credit through FSP. Overlap in Amhara Region has increased from 29 percent in 2006 to 46 percent in 2008, for example. While these results indicate that people may be becoming less risk averse, improved targeted of FSP interventions to PSNP participants will have also increased overlap. Role of Household Asset Building 180. The most significant, direct impacts on increasing the likelihood that households graduate are expected from the provision of credit and accompanying extension services. Households accessing the credit component of the FSP receive loans of ETB 1,500 to 5,000. These loan sizes represent significant potential investments, as compared with the median asset values of beneficiaries in 1996, which ranged between 712 and 2,471 depending on the Region41. In some Regions, the loan size depends on the technical package adopted (for example, sheep, beekeeping, seeds, or some combination thereof), while others were provided a standard credit amount (e.g., Oromiya provided 2,500 ETB). 181. A number of studies have demonstrated that in most instances the combination of credit and technical support to households has performed well. An end-users' evaluation of the Food 37 Ethiopia's Productive Safety Net Programme (PSNP): 2008 Assessment Report Devereux, S; Sabates-Wheeler, R; Slater, R; Mulgueta Tefera; Brown, T; Amdissa Teshome, 2008 (ODI/IDL Panel Survey). 38 An Impact Evaluation of Ethiopia's Productive Safety Nets Program, IFPRI, 2009 (b). 39 Devereux et al. 2008 ibid. 40 IFPRI, 2009 (b) ibid. 41 Ethiopia Food Security Program: Report on 2006 Baseline Survey, IFPRI, 2007. 45 Security Project in SNNP42 found that 84 percent of households considered their interventions successful and 87 percent reported an increased wealth ranking. An analysis modeling the impacts of the current household asset building program found that the majority of extension packages had significant potential to improve household food security, with 85-90 percent of households achieving positive rates of return43. Table 5 presents the modeled change of income and income-to-cost ratios of assessed packages: Table 5: Cost Components and Cost Effectiveness of Packages ­ 5-Year Totals (ETB) Add Annual Total Credit Total Change in Scenario Capital maintenance income to taken cost total income costs costs cost ratio Ox fattening (ABW-Oromiya) 3,200 0 2,350 5,550 10,933 1.97 Ox-Shoat (GCO-SNNP) 2,440 163 1,519 4,122 8,791 2.13 Bees-Irrigation/vegetables (EPL-Tigray) 1,400 0 7,410 8,810 23,444 2.66 Ox/cow-Crop-Irrigation/vegetables (GBG-Oromiya) 2,700 2,915 2,660 8,275 22,709 2.74 Crop-Irrigation/fruit-Camel-Ox-Shoat- Dairy (RVL-Tigray) 4,600 0 4,119 8,719 28,961 3.32 Poultry-Shoat-Dairy (GHL-Amhara) 2,760 200 175 3,135 16,656 5.31 Irrigation/fruit-Shoat (NMC-Amhara) 2,100 500 762 3,362 18,924 5.63 Source: Coulter and Sutcliffe, 2009. 182. This study found that the packages represented net changes of income between ETB 8,500 and 28,000 and a return on investments of between 2.13 and 5.63. 183. The program impact evaluation suggests more moderate indications of the success of the HABP initiatives. Among households that did not receive support from the PSNP, households receiving FSP irrigation services increased area cultivated by 0.3 hectares while those receiving seed services increased cultivation by 0.69 hectares. Irrigation services were also associated with a 9.7 kg increase in fertilizer use. Combined programming 184. All available evidence highlights the significant increased impacts of combined programming. The Modeling Study indicates that combined intervention packages have significantly greater impacts than single intervention packages44. The authors attribute this finding to the fact that combined packages spread risk and that components can complement each other. This is because if one component of a package fails, income from other components can compensate for this loss and enable a household to complete credit repayment. Examples of such complementarity between components include the fact that water for irrigation also ensures 42 Report on End Users Evaluations of Project Benefits in SNNPR, Million Legesse, Teshome Mekonnen and Associates, 2007. The Food Security Project, which is financed by the World Bank, CIDA and Italian Cooperation, contributes to the Government's Food Security Program. 43 Household Extension Packages: Modeling Impacts and Comparing Alternative Approaches, Coulter, L and Sutcliffe, P, 2009. 44 Coulter and Sutcliffe, 2009 ibid. 46 access to water by bees, while bees can increase the production of fruits and vegetables through cross-pollination. 185. The results of the impact evaluation highlight the benefits of combining PSNP and credit programming. A combination of public works transfers and FSP seed provision through credit resulted in significant increases in the area cultivated and a greater crop yield of 266 kg per hectare. By comparison, results for one intervention alone were much less and not statistically significant45. 186. Overall, the evidence strongly supports high quality programming that consists of: (i) timely delivery of adequate transfers and credit; (ii) a combination of PSNP and FSP credit and services; and (iii) a combination of household investments. It is likely that when these conditions are achieved households will be able to rapidly and significantly improve their food security status. 1.4 Obstacles to household graduation 187. While the section above describes the pathways out of poverty provided by the FSP, pitfalls remain. The results of the modeling discussed above are unlikely to be representative for all targeted beneficiaries or for all time periods. Uptake of larger credits, which are associated with innovative investments, appears to have been among the "better-off poor"46 and credit has been highly subsidized. It is likely that real rates of return are lower and households with greater labor or land constraints will experience even lower returns. Additionally, the past four years has seen above average rainfall and comparatively high levels of production. In years of below average rainfall when production decreases, the costs of maintaining livestock or crop assets will likely increase. The next phase of the FSP aims to address some of these obstacles. It includes market-based credit as well as arrangement for the poorest to better enable credit up take by such households, tailored products and services to ensure that investments are appropriate to the environment in which people live and a combination of household investments to help to spread risk. 188. Furthermore, interventions that can be successful for a few households may be unfeasible or even become problematic when implemented at scale. Degraded areas of rural Ethiopia are unlikely to have sufficient carrying capacity for large increases in livestock as higher livestock numbers may contribute to natural resource degradation. Cut-and-carry feeding may be able to mitigate some of the risks of higher livestock holdings, while investments by PSNP public works will help increase fodder availability. 189. Increased production must coincide with market development if it is to lead to improved food security. Ethiopia has already experienced price collapse following large-scale increases in national production. Without interventions to expand and find new markets, this experience could be repeated in areas supported by the FSP. The new phase of the Program also includes a much greater emphasis on value chains and market development to ensure that increased production can be absorbed in local markets. 45 IFPRI 2009 (b). 46 IFPRI, 2007, ibid. 47 190. It should be noted that significant environmental risks remain and that these continue to challenge household efforts to meet food needs and grow out of poverty. Significant failure of the minor rains in 2008 negatively impacted crop production and livestock. Furthermore, high inflation resulted in cereal prices tripling between July 2006 and July 2008. This was in spite of reported increases in agricultural production during the same period. The ability for the PSNP to scale up will be an important tool to mitigate the impacts of shocks on household consumption and livelihoods. It is expected that the early response facilitated by the PSNP risk financing will better protect livelihoods in the event of a shock. 191. Unsuccessful investments can also pose significant risks to households. Single animal packages, in particular, were found to be high risk as households faced significant setbacks and debt if the animal died. The focus on repeater loans and a combination of investments should mitigate the risks of unsuccessful interventions. This will be achieved by reducing the size of the loan at any one time, diversifying activities, and ensuring that investments are accompanied by better quality extension advice. 1.5 The Wider Framework Disablers of graduation 192. In addition to the obstacles described above there are a number of macro issues that may affect graduation at scale. Population growth remains very high in Ethiopia and has contributed to the fragmentation of land, often to unsustainable levels47. Continued increases in population without significant rural-urban migration will further reduce any remaining viable farm plots and increase significantly the number of rural landless. 193. Environmental degradation remains a concern. Estimates of soil nutrient loss as a result of the removal of dung and crop residue amounted to an annual nutrient loss of 364,050, which exceeds the amount of fertilizer applied48. Forest resources continue to erode as the growing use of land for agriculture and the demand for wood fuel put pressure on the land. All climate change models anticipate rising temperatures that would mean negative effects for production49. 194. Although already mentioned, shocks remain one of the greatest threats to graduation at scale. Climactic shocks have historically had major impacts on economic growth as seen in Figure 2. Moreover, climate change is also anticipated to result in more frequent and severe droughts, which will negatively affect livelihoods. 195. Furthermore, impacts of shocks persist. The 1984/85 drought was found to affect consumption growth well in to the 1990s. Households who suffered the most severe loss of 47 World Bank. 2005, ibid. 48 World Bank. 2005, ibid. 49 Measuring the Economic Impact of Climate Change on Ethiopian Agriculture: Ricardian Approach, Temesgen Tadesse Deressa, 2007. 48 consumption during 1984/85 experienced 16 percent less consumption growth during the 1990s as compared with households who suffered the least50. Figure 2: Growth in Per Capita GNI in Relation to Significant Shocks51 20% 15% Rainfall Shock 10% 5% Severe 0% Catastrophic rainfall drought shock -5% Overthrow of government Rainfall -10% shock 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 Enablers of graduation 196. However, a number of other trends may have strong positive impacts. Growth in GDP has been consistently high in recent years, averaging 6.2 percent between 1997-2007 with a peak of 10.9 percent. A low Gini Coefficient52 means that economic growth is more likely to significantly reduce poverty. 197. Simulations looking at the impact of growth on migration and employment patterns indicate that a decline in the proportion of people in agricultural employment has already begun. If growth rates remain at current levels, not only will reductions in poverty be appreciable, halving the number of people living in poverty in the 15 years between 2005 and 2020, but the absolute number of people in agricultural employment will begin to decline around 2015. This should impact on food insecure populations not only because they are likely to make up a number of the rural-urban migrants but also because increasing rural-urban migration will likely increase the demand for agricultural labor. 198. Although net enrolment rates for primary and secondary education are low, they represent dramatic increases in educational enrolment and attainment over the ten years leading up to 2004/05. According to the WMS, net primary enrolment rates in rural areas have more than doubled from 13.7 percent in 1996 to 32.8 percent in 200453. There is strong evidence correlating improved food security and reduced poverty with even modest levels of educational attainment. 50 Dercon, S. 2004. "Growth and shocks: evidence from rural Ethiopia". Journal of Development Economics 74(2): 309-329. 51 http://ddp-ext.worldbank.org. 52 0.30 for the country as a whole and only 0.26 for rural areas, ibid. 53 Data from the Ministry of Education puts net enrolment rates much higher: 73.9 percent in 2005/06 (General Education Quality Improvement Program PAD, World Bank, 2008). 49 In addition, the benefits of improved education levels are likely to be greater as opportunities for non-agricultural employment grow. 199. In addition to considering the potential negative cumulative effects outlined in section 1.4 it is also important to review positive multiplier effects of poverty reduction. As households become more food secure, they will spend a smaller proportion of their incomes on meeting food needs enabling further investment in productive assets and creating demand for a wider range of goods and services. Fewer poor households will be burdened with providing support for their destitute neighbors and relatives and therefore be able to invest their savings productively. Increased production of goods can lead to flooding of the market, but it can also lead to improved economies of scale. Transaction costs of both inputs and marketing may be reduced, increasing the potential for profit and impacting positively on the size of the market. 1.6 Conclusion 200. Ethiopia suffers from severe poverty caused by a number of interrelated factors that result in low income and assets. As part of its poverty reduction strategy, the Government has put in place a large-scale FSP for poor households. Evidence shows that components of the FSP can have significant positive impacts on household food insecurity. These impacts are greater when the programs are implemented according to design, program components complement each other, and households are encouraged to make a diverse range of investments. The FSP does not operate in isolation, however. Its impacts are strongly affected by a wider context of economic growth, rural-urban migration, expanding populations, and environmental change. While climate change, population growth and Ethiopia's susceptibility to shocks may undermine efforts to improve food security, strong economic growth, low levels of inequality, increasing rural-urban migration, increasing education, and positive multiplier effects should increase the potential for graduation at scale. 50 Annex 2: Major Related Projects Financed by the Bank and/or other Agencies ETHIOPIA: Productive Safety Net APL III 201. This annex provides an overview of major projects financed by the World Bank or other development partners that contribute to achieving food security in rural Ethiopia. In particular, the section focuses on those projects that contribute to rural economic development and strengthen government institutions, systems, and processes. The final section considers projects that have shared objectives or commonalities with the PSNP. 202. Achieving graduation from food insecurity requires not only the PSNP and broader Food Security Program but also relies on the diversification and strengthening of livelihoods, and overall rural economic development. These investments create an enabling environment that will "pull" households into food security. Key projects investing in this are described briefly below: (a) Proposed Agricultural Growth Program (AGP). (Proposed IDA credit of indicative US$200.0 million) Agriculture accounts for 47 percent of national GDP, 83 percent of employment, and the livelihood for 90 percent of poor households. Leading the sector to higher productivity and increased commercialization is fundamental to achieving food security in the long term. The proposed AGP will aim to increase the productivity of selected livestock and crop products in targeted areas and to increase the value of these products marketed. This Program is targeted to "high potential" woredas and thus has a different geographic coverage than the FSP. (b) Improving the Productivity and Market Success of Ethiopian Farmers (IPMS) (Total project cost: Cdn$19.0 million; CIDA grant: Cdn$19.0 million; Timeframe: April 2005- March 2010). IPMS aims to achieve improved and sustainable rural livelihoods by contributing to improved agricultural productivity and production through market-oriented agricultural development. The project is operational in 10 pilot woredas in Amhara, Oromiya, Tigray, and SNNP Regions. The project is implemented by IFPRI in collaboration with MOARD. Experience from this project informed the design of the HABP. (c) PSNP-Plus (Total project cost: US$12.0 million; USAID grant: US$12.0 million; Timeframe: October 2008-September 2011). PSNP assists households graduate from food insecurity by improving the resilience and livelihood assets of 42,414 households. To this end, the project increases the access of households to financial products and services and promotes their engagement in markets. The project is implemented through NGOs. (d) Pastoral Livelihood Initiative, second phase (PLI II) (Total project cost: US$12.0 million; USAID grant: US$12.0 million; Timeframe: May 2009-May 2012) PLI II enhances the livelihoods of 205,774 pastoralists and ex-pastoralists in 15 woredas in the regions of Afar, Oromiya, and Somali. The project is implemented through NGOs to: (i) increase the value and sales of livestock and non-livestock products; (ii) improve rangeland and water management; (iii) improve livestock-based early warning and response systems; and (iv) maximize project and policy impact through quality assessment, documentation and coordination. (e) Sustainable Land Management Project (SLM) (Total project cost USD$37.8 million; IDA Sector Investment Loan: US$20.0 million; GEF grant: US$9.0 million; Government of Ethiopia: US$8.8; GTZ: US$25.0 million; Approval date: April 29, 2008; Effectiveness date: October 10, 2008; Closing date: September 30, 2013). The minimum estimated annual costs of land degradation in Ethiopia range from 2 to 3 percent of agricultural GDP. The 51 SLM Project is a five-year project that aims to reduce land degradation in agricultural landscapes and improve the agricultural productivity of smallholder farmers in 35 watersheds across six regions. The aim is to scale up best practices in sustainable land management for smallholder farmers in "high potential" areas that are increasingly vulnerable to land degradation. This will be complemented by efforts to strengthen the land tenure security for smallholder farmers through the Government's land certification program. The SLM project and PSNP employ a similar approach to environmental rehabilitation but focus on different geographic areas (Latest DO rating: Moderately Satisfactory; IP rating: Moderately Satisfactory). (f) Rural Capacity Building Project (RCBP). (Total project costs: US$71.0 million; IDA Credit: US$54.0 million; CIDA grant: US$17.0 million; Date of approval: June 22, 2006; Effectiveness: December 28, 2006; Closing Date: October 31, 2011). The commercialization of agriculture is a core element of the Government's strategy for poverty reduction. In this context, strengthening the human capital and service provision needed for rural growth and improved productivity is crucial. RCBP is designed to address these issues by building the capacity of the agricultural extension system and the technical and vocational training system, as well as supporting institutional innovation in the National Agriculture and Research System. RCBP also supports the Ethiopian agricultural commodity exchange and development of sanitary and phytosanitary standards. The design of the HABP was informed by the design of and lessons learned from RCBP. As a result there are many similarities between the two Programs. Implementation of HABP will ensure that there is no geographic overlap. (Latest DO rating: Moderately Unsatisfactory; IP rating: Moderately Unsatisfactory). (g) The Food Security Project (FSP) (IDA credit: US$85.0 million; Date of approval: May 30, 2002; Effectiveness: November 26, 2002; Closure: June 30, 2010) is a community-driven development project to help vulnerable, food insecure communities increase their incomes, build assets, lower the real costs of food and improve nutrition levels. The Food Security Project operates in 74 food insecure woredas in Amhara, Oromiya, Tigray, and SNNPR. The design of the HABP built on the lessons learned during the Food Security Project. The closing date of the Food Security Project was extended to ensure overlap with the launch of the new HABP. The Food Security Project is co-financed by the Canadian and Italian governments. (Latest DO rating: Moderately Satisfactory; IP rating: Moderately Satisfactory). 203. The PSNP and HABP are largely implemented through Government systems. As such, other development partner-financed projects that aim to strengthen Government systems and capacity can significantly impact PSNP and HABP performance. At the same time, the HABP will channel resources through the microfinance sector to a range of service providers that participate in and receive support from other interventions. The major projects in this area for the PSNP and HABP are detailed below: (a) Protection of Basic Services, Phase II (PBS II). (Total project costs: US$3,364.1 million; IDA grant: US$309.8 million; IDA Credit: US$230.2 million, GoE ETB1428.6 million with other co-financing from African Development Bank, Austrian Government, Italian Cooperation, CIDA, DFID, EC, Irish Aid, RNE, the Government of Spain, and, the German Development Bank; Date of Approval: May 14, 2008; Effectiveness date: May 22, 2009; Closing date: December 31, 2011) PBS II builds on the experience accrued under PBS I. 52 The multi-donor project contributes to expanding access and improving the quality of basic services in education, health, agriculture, water supply and sanitation, and rural roads delivered by sub-national government, while continuing to deepen transparency and local accountability in service delivery. This is done by financing recurrent costs through the federal block grant system, and a multi-sectoral specific-purpose grant for capital investments on a pilot basis. The project includes a strong focus on transparency and accountability that aims to improve public financial management and procurement capacity along with innovative initiatives to improve bottom-up accountability. (b) The Public Sector Capacity Building Project (PSCAP) (Total Project cost: US$397.8 million; IDA Credit: US$100.0 million; Date of approval: May 11, 2004; Effectiveness: November 22, 2004; Closing Date: July 7, 2009) PSCAP supports Government's efforts to improve state-society relations in three ways: (i) improving the scale, efficiency, and responsiveness of public service delivery at the federal, regional, and local level; (ii) empowering citizens to participate more effectively in shaping their own development; and (iii) promoting good governance and accountability. PSCAP complements the largely equity-driven general purpose "block" grants and food security transfers to regions with a performance-oriented capacity building transfer to help sub-national authorities achieve their institutional transformation goals. (Latest DO rating: Satisfactory; IP rating: Satisfactory). (c) Livelihoods Integration Unit (LIU) project financed by USAID that is part of the Early Warning and Response Directorate within the Disaster Risk Management and Food Security Sector. The main goal of the LIU is to improve the accuracy and objectiveness of seasonal and annual needs assessments in Ethiopia. The LIU has created a countrywide, standardized, quantified, and comparable livelihood baseline and continues to build capacity of federal and regional staff to gather and analyze baseline data and monitor performance. (d) The Rural Financial Intermediation Program (RUFIP) (Total project cost: US$88.7 million; AfDB: US$37.5 million; IFAD Credit: US$25.7; Opening date: 2003; Closing date: 2010). The program established and strengthened rural savings and credit cooperatives with the aim of improving access to financial services for 2 million poor rural households. In part, this is done by establishing about 3,400 saving and credit cooperatives owned and managed by members, and training poor rural people to use financial services. 204. While the combined implementation of the PSNP and HABP is designed to achieve food security, there are other projects that overlap significantly with other aspects of the PSNP and HABP. These projects are expected to create synergies with PSNP and HABP. (a) Pastoral Community Development Project II (PCDP II). (Total Project Cost: US$53.4 million; IDA grant: US$22.3 million; IFAD Credit: US$11.4 million; Effectiveness date: October 9, 2008; closing date: December 31, 2013). The Government recognizes the previous marginalization of pastoral populations and that pastoralists are among the poorest of the poor in Ethiopia. PCDP II increases the resilience of Ethiopian pastoralists to external shocks and works to improve the livelihoods of beneficiary communities. The Project strengthens the decentralized and participatory planning procedures at community and woreda levels through the Community Investment Fund. These steps to strengthen livelihoods are completed by efforts to improve the existing Pastoral Early Warning System, among other initiatives and to strengthen the risk management system in pastoral areas. A MOU was signed between the Ministry of Federal Affairs and MOARD to harmonize 53 implementation of the PSNP, early warning response, and PCDP in pastoral woredas. (Latest DO rating: Moderately Satisfactory; IP rating: Moderately Satisfactory). (b) Nutrition Project (Total project cost US$30.0 million; IDA grant: US$30.0 million; Start date: April 29, 2008; Effectiveness date: September 10, 2008; Closing Date: January 7, 2014) The Nutrition Project supports the National Nutrition Program to improve child and maternal care behavior and increases utilization of key micronutrients among vulnerable groups. To this end, the project supports provision of community-based nutrition and health services, fully utilizing the existing Health Extension Program outreach and model household service provision structure and enhancement of the appropriate utilization of key micronutrients, especially iodine, iron, Vitamin A, and zinc. The project also provides capacity building support to the Ministry of Health in order to more broadly implement the National Nutrition Program. An ongoing pilot is exploring possible linkages between this Project and the PSNP. (Latest DO rating: Satisfactory; IP rating: Satisfactory). 54 Annex 3: Results Framework and Monitoring ETHIOPIA: Productive Safety Net APL III Results Framework PDO Outcome Indicators Use of Outcome Information LONG-TERM PROGRAM DEVELOPMENT OBJECTIVE FOR THE PRODUCTIVE SAFETY NET PROGRAM (2005-2014) Reduced household 1. Average number of months Yr.10: Indicate to what degree vulnerability, improve resilience households report being food the PSNP, as part of the broader to shocks and promote insecure. FSP, is reducing food insecurity sustainable community for program beneficiaries and is development in food insecure 2. % of households with leading to a halt in the trend of areas of rural Ethiopia. consumption below 1800 increasing numbers of food Kcal/person per day. insecure people. 3. % change in household asset The first and second indicators (physical). will capture longer-term impacts of improved food consumption, which is strongly associated with improved food security status. The third indicator will capture changes in household assets. PROJECT DEVELOPMENT OBJECTIVES FOR THIRD PHASE, APL III (2010-2014) Improved effectiveness and 1. % of participants reporting they Y6-Y10: Indicates to what efficiency of the Productive are able to plan ahead on the basis extent the PSNP: (i) provides Safety Net Program and related of PSNP transfers. households with reliable Household Asset Building transfers; (ii) creates productive Program for chronically food 2. % of households reporting direct and sustainable community insecure households in rural benefit from community assets. assets that are beneficial to Ethiopia. community members; and (ii) 3. % of PSNP households report the extent to which households that they have developed an on- or are accessing credit and off-farm income generating advisory services through opportunity attributable to HABP. HABP. Component 1: Safety Net Grants Intermediate Outcomes A. Intermediate Outcome Use of Intermediate Outcome Indicators Monitoring Outcome 1: Appropriate timely 1. Transfers made on time (%)54. Y6-Y10: Indicators one through and predictable transfers three measure the extent to received by households in 2. % of transfers received that have which the PSNP is providing response to chronic an average value of at least 15 kg of transfers to households that are requirements. grain per month. (i) timely and (ii) appropriate. Outcome 2: Quality, new and 3. % of households participating in Indicators four and five assess existing, community assets with the PSNP for 3 consecutive years or the public works in terms of operational management more. technical quality of 54 This indicator is consistent with the World Bank Results Chain for Social Protection in Africa. Timeliness is measured as the number of woredas that deliver 90 percent of transfers to participants within 45 days after the end of the month to which the transfers apply in 4 of the 6 months. 55 mechanisms established. implementation and 4. % of public works reaching sustainability. satisfactory standards and sustainability ratings. 5. % of public works that have an established management mechanism at completion. 6. People in project areas with access to "Improved Water Sources" (number)55. B. Output Indicators56 1. Project beneficiaries (number), of which female (percentage). 2. Person days provided in labor- intensive public work (number). 3. Amount of total transfers (USD). 4. Amount of transfer per beneficiary (average USD). 5. Health facilities constructed, renovated, and/or equipped (number). 6. Classrooms built or rehabilitated (number). 7. Roads rehabilitated, rural (km). 8. Roads constructed, rural (km). 9. Improved community water points constructed or rehabilitated under the project (number). 10. Area with improved land and water management technologies (ha). 11. % of PW screened for ESMF. Component 2: Risk Financing Intermediate Outcomes A. Intermediate Outcome Use of Intermediate Outcome Indicators Monitoring Outcome 1: Transitory cash and 1. % of transfers to participants Y6-Y10: Indicates extent to food needs addressed effectively within 75 days after RF triggered. which the PSNP risk financing is 55 This indicator is consistent with World Bank Core indicators. 56 Indicators 1-10 are consistent with World Bank Core indicators and the Results Chain for Social Protection in Africa. 56 in PSNP woredas, to the limit of providing timely transfers to risk financing resources. B. Output Indicators transitory food insecure 1. No. of woredas with contingency households. plans in place. 2. No. of non-PSNP kebeles with Community and Kebele Food Security Taskforces in place. 3. No. monthly reports generated by the Early Warning Working Group. Component 3: Institutional Support to PSNP Intermediate Outcomes A. Intermediate Outcome Use of Intermediate Outcome Indicators Monitoring Outcome 1: Transparency and 1. % of beneficiaries and non- Y6-Y10: Indicates the accountability of PSNP beneficiaries reporting that the effectiveness of measures to improved. targeting processes are fair. improve the transparency and accountability of the PSNP. Outcome 2: Institutional 2. % of beneficiaries who received capacity to manage the PSNP all information needed to Evaluates whether program strengthened. understand how the program works. procedures, implementation, B. Output Indicators coordination, monitoring and 1. Percent of woredas that have reporting are effective and posted budgets in public places. efficient. 2. Percent of woredas that have PSNP posters on Program objectives, targeting criteria and appeals procedures displayed in public places. 3. Percent of kebeles with functioning appeals committees in place. 4. Percent of woredas using PASS. 5. Percent of federal physical reports submitted on time. 6. Percent of federal financial reports and audits submitted on time. 7. Percent of woredas that have met the cashier/accountant to beneficiary ratio. Component 4: Support to Household Asset Building Program Intermediate Outcomes A. Intermediate Outcome Use of Intermediate Outcome Indicators Monitoring Outcome 1: Advisory services 1. % of HABP beneficiaries report Y6-Y10: Indicates the are demand-driven and market- that they are satisfied that their effectiveness of measures to oriented. business plans reflect their improve the delivery of credit to 57 priorities, needs and capabilities. food insecure households by Outcome 2: Financial services shifting to appropriate financial channeled through appropriate 2. Average repayment rates for service providers and developing institutions. HABP credit. a wider range of financial products. 3. % of credit to food insecure household delivered through MFIs, Assesses the effectiveness of RuSACCOs and VSLAs. systems change to deliver B. Output Indicators advisory services to food 1. Quality of training program to insecure households that are DAs, as measured by client market-oriented and of higher satisfaction. quality. 2. No. of woredas with completed market analysis, including consultative meetings. 3. No. of credit products designed and disseminated. 4. No. of RuSACCOs, MFIs and VSLAs in food insecure woredas. 5. People trained (number) on financial literacy57. 57 This indicator is consistent with World Bank Results Chain for Social Protection in Africa. 58 Arrangements for Results Monitoring Outcome Indicators Baseline Targets Frequency Data Responsibility Y4 Y5 Y6 Y7 Y8 Y9 Y10 collection for data (2008) (2014) Instruments collection Program Average number of months households report PW - 3.64 3.24 Bi-Annual59 Impact CSA/DRMFSS being food insecure58. DS - 3.80 3.42 evaluation % of households with consumption below 27% 13% Bi-Annual Impact CSA/DRMFSS 1800 Kcal/person per day 60. evaluation % change in household physical assets, in real PW - 4,568 Bi-Annual Impact CSA/DRMFSS terms ETB61. DS - 2,349 evaluation APL III % of participants reporting they are able to 27% 35% 50% 70% Bi-Annual Impact CSA/DRMFSS plan ahead on the basis of PSNP transfers62. evaluation % of households reporting direct benefit from PW ­ 86% 90% 95% 95% Bi-Annual Impact CSA/DRMFSS community assets63. DS ­ 67% 80% 90% 95% evaluation % of PSNP households report that they have 36%64 50% 70% 90% Bi-Annual Impact CSA/DRMFSS developed an on- or off-farm income evaluation generating opportunity attributable to HABP. 58 This indicator measures the average number of months that households report being food insecure with the receipt of PSNP transfers. This measure is from the IFPRI/CSA panel survey, which asks households to report the number of months, out of the preceding 12 months, that they had "no problems satisfying the food needs of the household." The target represents a reduction by 4 standard deviations. 59 Every two years. 60 This is measured as kcal/person in 7 days before survey. The detailed methodology for this measure is found in the IFPRI/CSA 2009 Impact Assessment Report. The baseline is the unconditional means for kcal per person in the 7 days before the survey households reported in 2008. 61 Assets are the value of livestock and productive equipment used in agriculture in Birr. The methodology for this calculation follows that used in the IFPRI/CSA analysis of the 2008 panel survey. This indicator reports on agricultural assets only as information on non-agricultural assets was not collected. 62 The baseline from this figure is reported in the 2008 Panel Survey of 8 woredas. This indicator will be captured in the FSP household survey. 63 This indicator is measured with data from the IFPRI/CSA 2008 Descriptive Report on the FSP household survey. The baseline figures were calculated for the APL II results framework. 64 The baseline data is the percent of households participating in the PSNP in 2008 that reported having received support from the Government's OFSP in 2006 or 2008. The data source is IFPRI/CSA 2008 Descriptive Report. 59 Outcome Indicators Baseline Targets Frequency Data Responsibility Y4 Y5 Y6 Y7 Y8 Y9 Y10 collection for data (2008) (2014) Instruments collection Intermediate Outcome Indicators for Each Component Component 1: Safety Net Grants Transfers made on time (percent)65. 6% 30% 40% 50% 60% 70% 80% Monthly, Information DRMFSS Annual Centre % of transfers received that have an average 0% 90% 90% 90% Annual Wage rate DRMFSS value of at least 15 kg of grain per month66. study % of households participating in the PSNP for 47% 60% 75% 85% Bi-annual Impact DRMFSS 3 consecutive years or more67. Evaluation % of public works reaching satisfactory 85% 90% 90% 90% 90% 90% 90% Annual PW Review NRMD standards and sustainability ratings68. and regular monitor % of public works that have an established 94% 95% 95% 95% 95% 95% 95% Annual PW Review NRMD management mechanism at completion69. People in project areas with access to 13,128 Bi-Annual Impact CSA/DRMFSS "improved water sources" (number)70. evaluation 65 This indicator is consistent with the agreed results chain for social protection in Africa. Timeliness is measured as the number of woredas that report delivering 90 percent of transfers to participants within 45 days after the end of the month to which the transfers apply, in 4 out of 6 months. The wording of the target reflects the fact that payments are made following the completion of public works requirements, which are carried-out during the month. The baseline data is based on an analysis of timeliness of payments in the 80 woredas that report to the Federal Information Center. Starting in 2010, Regional Information Centers will collect data from all woredas, which will enable an assessment of the timeliness of transfers across the Program. 66 Rapid food price inflation in 2008 eroded the purchasing power of the cash transfer. In response the Government increased the wage rate in 2009. However, a lag in available data from CSA does not allow this to be calculated at present. The Government is committed ensuring that the cash wage rage can purchase 15 kg of maize per month, which is traditionally the cheapest grain in local markets. Additionally, the wage rate for food and cash is set nationally. The annual wage rate study will be used to calculate this indicator based on an assessment of woreda markets, rather than the national price of maize, as was the case in Y4. 67 The baseline is reported in the IFPRI/CSA 2008 Descriptive Report. 68 The baseline is from the FSCD report to the October 2008 MTR Mission, reporting the findings of the 2nd 2008 Public Works Review. 69 Baseline data is from the 2nd 2008 Public Works Review, as reported in the FSCD report to the October 2008 MTR Mission. 70 This indicator is consistent with World Bank Core indicators. It assumes that the access to "improved water sources" is as a result of the Program. The baseline figures were calculated for the APL III results framework. The baseline was calculated using the percent of households that reported that they had benefited from the construction or maintenance of a well attributed to the PSNP from the FSP 2008 household survey. This was multiplied by the number of households 60 Outcome Indicators Baseline Targets Frequency Data Responsibility Y4 Y5 Y6 Y7 Y8 Y9 Y10 collection for data (2008) (2014) Instruments collection Component 2: Risk Financing (indicators will only be assessed if risk financing is triggered, which is not known) % of transfers to participants within 75 days n/a 60% 65% 70% 75% 80% 85% Following Risk Financing DRMFSS after RF triggered71. RF trigger Review Component 3: Institutional Support to PSNP % of beneficiaries and non-beneficiaries 85% 85% 90% 90% Bi-Annual Impact CSA/DRMFSS reporting that the targeting processes are evaluation fair72. % of beneficiaries who received all 68% 70% 80% 90% Bi-Annual Impact DRMFSS information needed to understand how the evaluation program works73. Component 4: Support to Household Asset Building Program % of HABP beneficiaries report that they are 0 50% 65% 80% Bi-Annual HABP systems DRMFSS & satisfied that their business plans reflect their assessment Ag. Extension priorities, needs and capacities. Directorate Average repayment rates for HABP credit74. 72% 80% 90% 95% Bi-Annual Impact CSA/DRMFSS evaluation % of credit to food insecure households 52% 70% 85% 95% Bi-Annual Impact DRMFSS/CSA delivered through MFIs, RuSACCOs and Assessment VSLAs75. sampled, by an estimated 5 members per household. No target has been set for this indicator because the types of public works constructed through PSNP is determined annually based on a community planning process. 71 75 days is based on the PSNP performance targets that state that transfers will reach woredas from federal level within 4 weeks and then be delivered to beneficiaries within 45 days after the previous month (when public works are carried out). Two additional days have been added to account for communication between the RF Management Committee and implementing bodies, and because RF is a new system. 72 The baseline data is from the 2008 Financial Transparency and Accountability report on the PSNP. 73 This indicator is collected in the IFPRI/CSA panel survey. The baseline data is from the IFPRI/CSA 2008 Descriptive Report and is a simple average. 74 Baseline is from the Review of the Food Security Program, which determined that repayment rates were low across the Program, although the general lack of information on repayment rates under the Government program made it impossible to estimate that actual repayment rates in some cases. Therefore, the Food Security Project, which provided loans to food insecure households on a revolving basis, is being used to measure this baseline. 75 The baseline data for this indicator was calculated based on the proportion of the Government FSP budget allocated to credit through MFIs, RuSACCOs or Village Lending and Savings Associations (VLSA), by Region, as reported in the OFSP Review (2008) and the reported amount allocated by the Food Security Project to household credit, which is allocated through RuSACCOs or MFIs. 61 Monitoring and Evaluation System 205. The Government has established a single monitoring and evaluation system for the Food Security Program that includes detailed assessments of the PSNP and HABP. This system is designed to assess progress towards higher level objectives while also responding to the realities of collecting regular monitoring data through Government systems. The system is guided by an integrated logical framework to ensure concerted progress towards the overall objective of reducing food insecurity in rural Ethiopia. 206. In particular, regarding monitoring of the HABP, the Government confirmed that it will assess progress towards achieving the objectives of the HABP, as outlined in the new program logframe. This will be done using the FSP monitoring system, required assessments as outlined in the Program document and logframe, and the independent FSP household survey, which is undertaken by CSA in collaboration with an international institute every two years. As has been the practice so far, Government will continue to share these assessment reports with development partners. This information will be the basis for management decisions between the Government and development partners to improve the performance of the Program. 207. The FSCD is responsible for the overall monitoring of the Program, while the NRMD, AED, and MOFED respectively oversee the monitoring of the PSNP public works, HABP technical services, and financial management. Program evaluations are carried out by independent bodies. The sections below describe the various components of the M&E system. A summary of the system is found in Table 6. (i) Regular monitoring data 208. Information at output and activity level is collected regularly through Government systems. This includes information on beneficiaries, transfers, public works, and delivery of technical services through the extension system. Public works monitoring records the location, number, type, and design of all implemented public works. In addition, in APL III, watersheds will be mapped at the beginning of the program to track changes over time. Financial reports provide information on budget expended according to agreed line items. In addition, following the release of Risk Financing resources, reports will be consolidated through the Government system. 209. This type of reporting aims to keep stakeholders apprised of expenditures and activities, verify the proper implementation of processes described in the PIM and other program documents, and identify areas where performance does not match expectations so that program managers can take corrective actions. Information collected by the regular monitoring system is expected to flow both "upwards" and "downwards". As each office fulfills its reporting requirements up the chain, it will be expected to also report back to those providing the data. 210. Regular monitoring data is collected through Government implementing bodies. Regular monitoring of public works is the responsibility of the NRMD through its PWCU, while the AED monitors the implementation of HABP technical services and overall financial reporting is carried out by MOFED. FSCD combines and analyzes these data to prepare a comprehensive and continuous assessment of progress in implementation. A summary of responsibilities follows: 62 (a) Community level: Weekly tracking and monthly reporting on food security activities are submitted to kebele. (b) Kebele level: DAs submit monthly reports on implementation progress as compared with the FSP plan to the Food Security Desk. (c) Woreda Level: The Food Security Desk regularly monitors safety net and household asset building activities, compiles the data and reviews implementation against plans. The Finance Office reports on resource utilization. Monthly, quarterly, and annual reports are submitted to the regional food security coordination office; data on PSNP public works is submitted to the Regional PWFU; data on HABP is submitted to the Extension Office; and financial reports are submitted to BOFED. (d) Regional Level: The Regional Food Security Coordination Office compiles data submitted by woredas and reviews progress. Quarterly and annual synthesis reports are submitted to FSCD. The regional Public Works Focal Units analyze data on public works submitted by woredas and undertake additional monitoring activities as required. The Agricultural Extension Office is similarly responsible for the monitoring of HABP technical services. BOFED consolidates and analyzes the woreda financial reports. (e) Federal Level: The FSCD collects reports from the Regions and implementing partners, particularly PWCU, AED, MOFED and NGOs on a quarterly basis and also prepares an annual report. This data is analyzed in a consolidated report to provide a comprehensive overview of progress in implementation. The Public Works Coordination Unit analyses data on public works submitted by Regions. The AED is similarly responsible for data on the delivery of technical services for the HABP. MOFED prepares the interim financial reports on the PSNP and HABP, including in-kind resources. 211. Reporting follows the Government of Ethiopia's fiscal year, which begins on July 8 and ends on July 7. Annual plans for the PSNP and HABP are completed by June 30 and will be used as the basis for preparing quarterly achievement reports of the following fiscal year. (ii) Real-time monitoring data 212. Regular monitoring data is augmented with real-time data from a range of sources, particularly on the PSNP. Real-time monitoring of the PSNP was instituted in APL I to mitigate humanitarian risks and continues to provide a vital source of timely data on Program performance to decision-makers at all levels. 213. The Federal Information Center (FIC) previously collected data every two weeks from about 80 woredas to provide real-time information on the timeliness of PSNP transfers. The FIC compiled and distributed information in response to the needs of decision-makers. As part of APL III, Regional Information Centers (RICs) will be established. They will collect data on transfers and other key indicators for monitoring performance targets and determining performance incentives. They will cover all PSNP woredas in Afar, Amhara, Oromiya, SNNP, and Tigray Regions. A RIC will be established in Somali as the Program rolls out in this Region. Once the RICs are well established, the scope of the FIC will subsequently change to focus on quality control and data auditing, while continuing to provide program-wide analysis on program performance. 63 214. As the cornerstone of Government's Risk Management Strategy for the PSNP, the Rapid Response Mechanism (RRM) addresses critical implementation problems as they occur. The RRM detects problems that warrant immediate attention and responds rapidly to resolve the problems, thus reducing any potentially serious humanitarian or other risk. Rapid Response Teams are regularly deployed to kebele, woreda, and Regional levels to monitor implementation progress. This is done through focus group discussions with implementers, beneficiaries ,and non-beneficiaries and a review of records. Teams also follow up to solve any problems and ensure that consistent follow-up is provided (iii) Systems assessments 215. A number of independent systems audits were adopted in APL II to improve information flows on systems and processes, particularly at woreda level, with the aim of strengthening implementation and overall accountability. 216. The Appeals Review assesses the effectiveness of the appeals system at woreda and kebele levels and disseminates best practices. This review ensures the Kebele Appeals Committee is functioning as expected and that records are being kept. The Independent Procurement Assessment reviews woreda procurement systems and processes to ensure that procedures and guidelines are followed and to make recommendations on how procurement could be improved. 217. In APL III, the Financial Audit will fulfill three functions: audit of accounts, systems audit, and review of transactions to beneficiaries to ensure that funds disbursed are used for purposes intended. Existing roving and annual audits will be amalgamated into one streamlined audit process with quarterly as well as annual reporting. Additionally, a Commodity Audit of in-kind resources will be introduced into the PSNP. This will audit commodity management systems to ensure the proper oversight and management of in-kind resources with the overall aim of strengthening accountability. The audit will include a review of transactions to beneficiaries to ensure that resources are used for the purpose intended. (iv) Assessments of output indicators 218. The Program commissions a number of independent studies and reviews to assess progress towards outputs. These studies aim to assess if the program is on track to achieving its purpose and to identify any adjustments needed. The results of the various studies will also inform adjustments to program activities. 219. A set of annual woreda level reviews provides the main opportunity to assess indicators at this level. The First Annual Public Works Review assesses the adequacy of community public works plans and the integration of these into the woreda plan. The Second Annual Public Works Review, which occurs towards the end of the public works season, assesses the quality of works constructed during the season and reviews project sustainability. 220. Additional assessments of outputs are carried out at the Federal level. A Review of the PSNP Risk Financing will be carried out each time it is triggered to determine the effectiveness of the response. The Wage Rate Study is conducted annually to ensure that the PSNP provides an appropriate cash wage rate by assessing market prices of key cereals in PSNP woredas. 64 221. Building on the review of targeting and studies on HIV/AIDS and Gender completed under APL II, a further Social Assessment focusing on program targeting and other relevant social issues will assess the quality of program targeting and confirm that the most vulnerable continue to be targeted by the program. (v) Impact evaluations 222. A set of impact evaluations aim to measure the changes that are brought about for the direct and indirect beneficiaries and/or their institutional structures as a result of the activities initiated by the program. These are carried out through independent assessments. 223. A regionally representative household survey is carried out every two years to assess the impact on direct and indirect beneficiaries as a result of activities initiated by the Program. The next survey is due to be carried out in 2010. This survey also provides valuable information on implementation progress. Quantitative household-level information is augmented through qualitative assessments, addressing beneficiary perceptions and related social issues. 224. A second evaluation assesses the impact of PSNP public works at community level using a sample of watersheds from across program woredas. This examines the impact of PSNP public works to determine if they have met their objective such as conserving soil or moisture, growing crops through irrigation, or providing market access through road networks. Table 6: PSNP and HABP M&E System Key Reports, Assessments, Audits and Evaluations Types of Information provided Frequency Examples of reports indicators Monitoring Regular collection of information at output and Monthly from - Number of public Reports activity level, including regular financial reports woreda to works completed (IFRs). Regional level; - Volume of transfers Quarterly to delivered Federal level Information Information collection from a sample of woredas Every two weeks - Date and amount of Center largely focused on timeliness of transfers, but also transfers to woredas Reports includes price data. A key set of indicators on the and beneficiaries HABP may also be collected. - Average maize prices Rapid Regular assessments of implementation at kebele, Every two months - Number of Response woreda and regional levels to address critical from Federal level households targeted Mechanism implementation problems as they occur. This includes (regularly from - Beneficiary Report transfers to beneficiaries, public works, capacity issues Regional and satisfaction with and others. below) PSNP Annual -Purchasing power study to set an appropriate wage - Annual - Average prices in Assessments rate for the PSNP PSNP markets over -PW Review (planning) to assess the adequacy of - Annual time PSNP public works plans - Number of public -PW Review (technical) to review the quality and - Annual works meeting sustainability of PSNP PW technical standards -Risk Financing (RF) Review to determine the - As needed - Number of effectiveness of the RF response, if triggered Appeals Committees 65 Types of Information provided Frequency Examples of reports indicators - Appeals Review to assess the functioning of the - Annual established appeals system - Volume of goods -Independent Procurement Assessment to review - Annual procured procurement processes at woreda level Audits -The Financial Audit includes an audit of accounts; - Quarterly - Percent of systems audit; and review of transactions to rolling, annual households receiving beneficiaries to ensure that funds were used for full payment purposes intended. - Quality of food -The Commodity Audit review to ensure in-kind - Annual stock records resources are used for the purpose intended Evaluations -Social Assessment to confirm the effectiveness of - Once -Qualitative review program targeting and assess relevant social issues of targeting -Public Works Impact Assessment to determine if the - Every two years - Benefit-cost objective of the PSNP PW were met assessment of public -Biannual Impact Evaluation, a regionally - Every two years works representative household survey, to assess outcomes - Change in and impacts of all component of FSP household food gap -Risk Financing impact assessment to determine if the - As needed objectives of RF were met Source: FSP Formulation Document 2009. 225. Government and development partners carry out Joint Review and Implementation Support (JRIS) Missions twice a year following an agreed schedule from 2010-2014. The monitoring activities below have been established for each of the JRIS Missions: Semi-Annual Review by June 2010 · Review implementation of PSNP Client Card System and Charter of Client Rights and Responsibilities. · Adopt a revised monitoring system based on the FSP (including PSNP and HABP) logframe and systems review to include, inter alia, PSNP cash and food transfers, public works, HABP technical and financial services, and risk financing. · Review PSNP Rolling Training Program (and each subsequent June JRIS). · Completed semi-annual Public Works Review (planning) and recommendations identified (and each subsequent June JRIS). · Review functioning of federal and regional Information Center system and agreed steps to further strengthen the system, as necessary (and each subsequent June JRIS). · Progress report on program implementation, including woreda performance and public works monitoring, completed and recommendations identified (and each subsequent June JRIS). · Woreda-by-woreda assessment of implementation of PASS in Amhara, Oromiya, SNNP, and Tigray Regions and recommendations agreed (and each subsequent June JRIS). · Review progress of pastoral program roll out and identify recommendations to strengthen implementation (and each subsequent June JRIS). 66 Semi-Annual Review by October 2010 · Agree Appeals Review findings and recommendations (and each subsequent October JRIS). · Agree Procurement Review findings and recommendations (and each subsequent October JRIS). · Semi-Annual Public Works Review (technical) completed and recommendations identified (and each subsequent October JRIS). · Review of the performance management system for all PSNP woredas (and each subsequent October JRIS). · Provide evidence that the revolving funds and credit lines under HABP are being managed according to sound financial principles, including commercial interest rates. Semi-Annual Review by June 2011 · Additional to the above, review findings of 2010 Household and Public Works Impact Assessments and agree modifications to program design, as necessary. 67 Annex 4: Detailed Project Description ETHIOPIA: Productive Safety Net APL III Project Component 1: Safety Net Activities 226. Criteria for Selection of Program Woredas: The PSNP is targeted to woredas that are classified as being chronically food insecure. These are defined as woredas that are (i) in one of 8 regions (Tigray, Amhara, Oromiya, SNNP, Afar, Somali, rural Harari and Dire Dawa)76, and (ii) received food aid in each of the three years preceding the start of the Program. According to the Government, 316 woredas satisfy these two conditions77. The list of woredas is found in the PSNP Program Implementation Manual (PIM). It should be noted that the PSNP is being implemented in rural areas, and that it currently includes a pilot program in 18 woredas in pastoral areas78. Woredas receiving settlers as part of the resettlement component of the Government's FSP are not eligible for the PSNP. 227. Criteria for Selection of Households: There are two types of criteria: 228. General criteria for selection of households: Chronically food insecure households are defined as those households that have faced continuous food shortages (usually three months of food gap or more) in the last three years and who have received food assistance. Additionally, this includes households who suddenly become more food insecure as a result of a severe loss of assets and are unable to support themselves within the last one to two years; and 229. Specific criteria for selection of households: The following criteria can be considered in the selection of eligible household: (a) Households should be members of the community; (b) Status of household assets: land holding, quality of land, food stock, etc; (c) Status of expected household food production compared with household monthly consumption requirements; (d) Income from non-agricultural activities and alternative employment; and (e) Support/remittances from relatives or community. 230. In the first year of the PSNP, communities in identified woredas undertook a targeting exercise that provided a baseline list of chronically food insecure households. Based on lessons learned during the first year's targeting exercise, this was updated significantly as part of the 2006 annual re-targeting. Afar was also included in the program in 2006. The number of chronically food insecure households included in the Program was revised in 2007 with the launch of the pastoral pilot program and again in 2008 (for the 2009 program) with the addition of households in SNNP, which had been assisted with contingency funds for three consecutive years. 76 The ongoing pastoral pilot program will inform the future scale-up of the PSNP in Afar and Somali Regions. 77 The Program currently covers 290 woredas and will expand to 316 under the roll out of the pastoral PSNP. 78 The pastoral pilot covers 21 woreda, of which 18 are currently active. 68 231. Targeting System for PSNP: An administrative and community targeting system is applied in the selection of eligible participants for the PSNP. The roles and responsibilities of each body involved in targeting are outlined in the PIM. The identification of eligible households is the responsibility of each CFSTF. Once a household has been selected to participate in the PSNP, the CFSTF determines whether it participates in the public works or direct support79. 232. Eligibility for public works: Public works are intended for able bodied adult men and women. This includes pregnant women during the first six months of pregnancy, lactating women for a period of 10 months after giving birth, and female-heads of households. It excludes sick or mentally challenged people unable to undertake even light work; pregnant women in their final trimester of pregnancy; lactating women in the first ten months after childbirth; and young people under the age of 16. 233. Eligibility for direct support: Direct support is intended for individuals without labor to participate in public works and who do not have sufficient and reliable support from children or remittances from other relatives living outside the village. Such individuals include some disabled or elderly persons, lactating mothers during the first ten months after birth, pregnant women, and orphaned teenagers. 234. Under APL III, steps will be taken to ensure that beneficiaries and non-beneficiaries are well informed of the eligibility criteria for the PSNP (see section on Transparency and Accountability below). 235. Annual Retargeting Exercise: Each CFSTF updates the list of eligible households annually at the beginning of the program cycle in September/October, which is based on the annual plan submitted in June. The community updates this list within the existing total caseload to (i) correct errors of inclusion and exclusion, and (ii) add households to the list if they have become chronically food insecure because of shocks such as the loss of a breadwinner. 236. Graduation from PSNP: Households graduate from the PSNP when they are food sufficient80. This is determined by comparing household assets to regionally defined asset benchmarks. DAs collect information on household assets and, in some regions, income. This data is then analyzed by woreda officials to identify those households eligible for graduation from the PSNP. This proposal, including the supporting analysis, is verified through community consultations and then finalized. Households are able to appeal against this decision. Once identified for graduation, households will remain in the PSNP for one additional year and continue to access other support services through the FSP until they are food secure. 237. Appeals Process: The PSNP appeals system was strengthened in APL II. This included: widespread dissemination of PSNP targeting rules and appeals procedures in local languages and visually; separation of membership of the targeting structure and the appeals structure; and 79 Non-pilot woredas in Afar receive direct support only until the PSNP is scaled-up in that Region. 80 Food sufficiency is when, in the absence of receiving PSNP transfers, a household can meet its food needs for all 12 months and is able to withstand modest shocks 69 establishment of new guidelines for documentation and follow up for appeals at the kebele, woreda and regional levels. Further, the review function includes a regular assessment of the effectiveness of the appeals process. Under APL III, the Appeals review will continue to provide technical support and guidance to Kebele Appeals Committees, in addition to providing an annual review of the appeals system. This, coupled with initiatives to improve the general awareness among communities on program objectives such as the targeting criteria is expected to deepen local accountability systems. 238. Public Works Activities: Public works are labor-intensive community-based activities that are designed to (i) provide employment for chronically food insecure people who have labor and (ii) create productive sustainable community assets. Resources for this component include grants as wages for labor provided in public works projects as well as capital inputs and administrative/management costs. This includes the financing of small complementary civil works contracts to support the implementation of public works (such as the building of small culverts or bridges in the context of road construction). 239. The selection of sub-projects to be carried out under the public works component is driven by the local planning process in order to identify community needs and prioritize sub-projects. The methodology used for this is the integrated watershed management approach. The planning process is crucial to the success of the public works sub-projects, allowing a pipeline of public works sub-projects to be developed. This will permit the appropriate planning of public works according to seasonal restrictions and the rapid scalability of the public works should needs within a woreda increase due to shocks. As part of APL III, the planning process will be expanded to include identification of appropriate operations and maintenance of public works sub-projects to ensure their long-term sustainability. 240. Planning starts by identifying the key outcomes the community wishes to achieve and then developing an appropriate list of sub-projects that will deliver these outcomes. Examples of desirable outcomes and related projects include: (a) Improved land productivity and soil fertility using methods such as hillside terracing and area closures; (b) Increased land availability through methods such as gully control; (c) Improved market infrastructure through construction of roads, bridges, and market storage facilities; (d) Improved access to drinking and irrigation water through development of springs shallow wells, and water ponds; (e) Increased availability of fodder through vegetative fencing and multi-purpose nurseries; and (f) Improved school and health facilities through repairing and constructing classroom and health facilities. 241. The creation of household assets is not the direct objective of the public works sub-projects and households cannot be paid individually to work on their land. However, given that the integrated watershed management methodology requires a holistic approach to the watershed, work can be undertaken on both household and community land if there are benefits to the 70 community at large. In addition, public works can be undertaken on household land belonging to female-headed households with no labor in order to enhance their productivity directly. In all cases public works sub-projects that take place on household land are permitted if endorsed through the community planning process and by the owner of the land and are within the guideline established for these types of activities. 242. The maintenance requirements for public works sub-projects are identified during the planning process. Once identified, the institutional responsibilities for such maintenance work and associated budgets are detailed. As the maintenance of certain public works sub-projects, such as roads, tends to be labor intensive in nature, this responsibility often falls to the community. 243. Certain sub-projects are not eligible under the public works. These include three main types of sub-projects: (a) Sub-projects that solely benefit private, for-profit organizations; (b) Sub-projects to undertake regular, minimal maintenance activities on assets that the community should already undertake as an in-kind contribution to the protection of those assets; and (c) Sub-projects for military or defense purposes. 244. Staff from relevant line departments will be involved in the technical design of public works for which their department has responsibility. Guidance on appropriate technical design of common public works is available and accurate assessment of non-labor inputs will be made. The Regional Public Works Focal Unit carries out regular monitoring visits. 245. Public works implementation must not undermine normal agricultural activity and therefore the bulk of public works projects should take place during the agricultural slack season to avoid interfering with the peak period of labor demand. It is anticipated that the majority of public works will therefore take place during the dry season. Public works can be undertaken at other times of year. These should be on a reduced scale so as not to compete with the need for labor in agriculture and should be technically feasible in wet conditions. 246. The maximum number of days a household can work per month is based on the size of the household. Households may work up to 5 days per month for each member of the household with a maximum of 20 person-days per month for each individual who is eligible to work. 247. To carry out public works, participants are organized in work teams and each team is assigned tasks. Teams elect their own team leaders. Generally, a team will consist of 15-30 laborers, with the actual number being determined locally based on the nature of the work, the work culture of the people and the type and availability of hand tools. Each team will elect its own team leader with women encouraged to take leadership roles. 248. Where necessary to facilitate the overall management of large numbers of workers within a single large-scale project work, teams may be grouped together in work units, each coordinated by a foreperson. This can also facilitate the distribution and management of resources when a 71 project is operating on a number of scattered sites. Where needed, a labor unit would generally consist of 5 to 10 work teams. 249. Most projects, therefore, will have team leaders, a chief artisan, a DA or site manager and occasionally guards. In a large project, a foreperson may be required and an assistant to DA/site manager may also be assigned. An implementing agency supervisor would normally provide technical and managerial guidance to community projects. Technical supervision is carried out by woreda, zonal, or regional experts depending on the type of activity. 250. All public works should have defined "owners" after completion. These "owners" have user rights and therefore the obligation to maintain and manage assets. In many cases user rights arrangements will involve the creation of group by-laws governing access, management, and obligations. They may also require the establishment of a system to collect user fees. These user rights and maintenance responsibilities need to be defined during the planning process to establish a sense of ownership by the benefitting community or group. Ensuring a common understanding between community asset "owners" and woreda government service providers on what further support "owners" can expect to receive once the asset has been handed over is critical. 251. Direct Support Activities: The direct support activity is a mechanism for providing grants to the poorest members of the community. Direct Support payments are unconditional in recognition of the fact that even though these individuals are unable to perform public works they are entitled to support from the PSNP. Direct Support beneficiaries are identified through the community targeting process as those chronically food insecure households with no available labor and no other means of support. Direct Support beneficiaries tend to include disabled or elderly persons, people who are chronically ill, such as those with HIV and AIDS, pregnant or lactating mothers (in the first ten months after birth), orphaned children up to 16 years of age and female-headed households with no other available adult labor. While the pastoral PSNP is being rolled out in 2010 and 2011, payments to the chronically food insecure population in woredas in Afar not yet covered by the roll out will also be made via Direct Support. See paragraph 263 below for more information on the planned roll out of the pastoral program. 252. Under APL II, a "cap" on the number of days each individual could work under the PSNP was introduced. This "cap" states that for a household with limited labor and multiple dependents the able-bodied members must provide 5 days of labor for each individual in the household up to a maximum of 20 days. Beyond this, households will have access to Direct Support transfers. This is to ensure that households with few able-bodied members can continue to engage in other livelihood activities. 253. At the same time the beneficiary lists for public works and Direct Support were unified to better enable individuals to change their status and to ensure that households received a single stream of support from public works and/or Direct Support. This was particularly relevant for pregnant women, who are transferred from public works to Direct Support in the six months of pregnancy and remain there until ten months after the birth of their child. 72 254. A pilot is ongoing to explore possibilities to link the PSNP with the National Nutrition Program. The pilot is exploring, among other options, the possibility of using the Direct Support component of the PSNP to promote the provision of nutrition outreach and to encourage households to participate in nutrition counseling. The results of the ongoing pilot will inform decisions on how to scale up any linkages between these Programs (see Annex 2 for description of a World Bank support to National Nutrition Program). 255. Program Contingencies: Contingency funding for the program is provided at two levels and is equivalent to 20 percent of the base program cost. BOARD is responsible for managing 15 percent of the contingency fund while WOARD is responsible for managing the other 5 percent. 256. The woreda contingency fund is used to address unexpected needs of chronic food insecure households, such as exclusion from the Program. The regional program contingency funds are used: (a) To address unexpected needs of the chronic food insecure population; and (b) To address transitory food insecurity and should be issued before requests are made for Risk Financing (see Component 2 below) unless it is apparent that the severity and extent of the impending shock is beyond the capacity of the contingency budget. 257. Under these circumstances, the procedures for the use of the contingency funds are outlined in the PIM. At the end of the year, unused contingency funds will be rolled over as part of the contingency funding for the following year. 258. Payment Processes: Households receive their PSNP transfers on a monthly basis either in food or cash, as per the PSNP Annual Plan. The purchasing power of the cash transfer is reviewed annually at the Federal level to ensure parity with the food transfer. In general, food transfers will be made in months when the price of food is high or in areas where markets are poorly functioning. 259. Transfers are sent from the Federal to the Regional and then woreda levels in line with the approved PSNP Annual Plan. Measures such as adopting performance targets have been taken by MOFED and DRMFSS to ensure the timeliness of cash transfers to beneficiaries. Similar steps for food transfers, which are transferred through the DRMFSS system to woredas, are being undertaken. 260. The participation of public works beneficiaries is recorded in an attendance sheet, which is approved by the DA and Kebele Chairman. These are transmitted to the WOFED, which prepares the payroll and makes the cash payments to both public works and direct support beneficiaries. Food payments are made by the DRMFSS structure at woreda level. The attendance sheet and payroll have been computerized to streamline this process. 261. Transfers to beneficiaries are carried out in public locations by the WOFEDs, with support of DAs and other woreda staff. Under APL III, beneficiary cards will be introduced. These will spell out the entitlements of households under the PSNP and also record proof of payment. 73 262. Support to a Performance Management System: This will establish of a system of performance incentives based on performance data generated through the regular monitoring system to improve the timeliness and predictability of transfers to households. It is expected that other performance areas will be added over time. Additional budget will be provided to all woredas meeting minimum performance standards for use as part of their program administrative and/or capital budgets. Payments will be made to any woreda achieving minimum standards (related to basic program systems and fiduciary standards) and the performance target on timeliness of transfers set out in the PSNP performance standards. The performance target will vary across three categories of woredas to recognize that woredas face underlying constraints that affect their implementation performance. Payments to woredas will be equivalent to 30 percent of the existing administrative budget. Payments will be determined annually as part of the budgeting cycle based on the previous year's performance. Underpinning this will be a renewed focus on accurate performance measurement across the Program based on information generated by the RICs and effective dissemination of performance results to all implementing woredas. There will also be closer follow up with woredas that encounter problems in improving their performance through the delivery of tailored capacity building and management support to address bottlenecks. 263. Pastoral Pilot Program and Roll Out: Following a detailed design process involving the Government, development partners, and NGOs, a pilot program was launched in 2008 in 21 woredas in pastoral areas (of which 18 were active in 2009). The aim of the pilot is to test a range of methods, including targeting, public works and types of transfers to determine how to effectively deliver a safety net program for pastoral livelihoods81. The monitoring of the pilot has been subcontracted to ensure regular information gathering and lesson learning from areas that otherwise have limited capacity. Based on the lessons learned under the pilot program, the pastoral PSNP will be scaled up in the pastoral areas of Afar and Somali in 2010 and 2011. Project Component 2: Drought Risk Financing 264. This component will be financed using a contingent grant, which will provide resources for scaling up activities under Component 1 in response to localized or intermediate weather or price-related shocks. This strategy is consistent with the revised Disaster Risk Management Policy, which provides for an early response to emergency needs within a risk management framework. Risk Financing (RF) can be used to either extend support to current PSNP beneficiaries or support new beneficiaries who have transitory needs. It will be used in PSNP woredas only. This is because the institutions and systems for delivering risk financing are already in place in PSNP woredas. Strong linkages between risk financing and the emergency response system are required to ensure a coordinated response to shocks. In order to ensure that PSNP systems can scale up on demand, additional support for capacity building will be channeled through the Institutional Support component (see below). 265. RF consists of routine activities and the phases of the risk financing mechanism itself. These are described in detail below: 81 Parameters tested include food and cash; community targeting and self-targeting; and timing of public works support. 74 266. Routine Activities: These are the activities that must occur on an ongoing basis for the risk financing mechanism to work as designed. To start, contingency plans will be prepared at woreda level to prepare for unforeseen events. Contingency plans are within the needs of RF. Therefore they cover only PSNP woredas, respond to cash and food needs, outline the public works to be mobilized once the RF is triggered and are based on three agreed planning scenarios at woreda level for transitory need. These three scenarios are: (a) Scenario 1: Up to 25 percent of existing PSNP beneficiaries require extension of the response period from 6 to 9 months, plus the number of transitory food insecure people in PSNP woredas is up to 25 percent of the existing PSNP beneficiary number. (b) Scenario 2: Between 25 percent and 50 percent of existing PSNP beneficiaries require extension of the response period from 6 to 9 months, plus the new transitory need is between 25 percent and 50 percent of existing beneficiary number. (c) Scenario 3: Between 50 percent and 75 percent of existing PSNP beneficiaries require extension of the response period from 6 to 9 months, plus the new transitory need is between 50 percent and 75 percent of existing beneficiary number. 267. Action plans will also be formulated at Federal and Regional levels to support the scale up of contingency plans. Contingency plans will be developed for both pastoral and non-pastoral woredas. The Early Warning System will routinely collect and analyze early warning data. At Federal level, data from woredas is triangulated with LEAP and other national indicators. Concurrently, DRMFSS will ensure that the FSCD and EWRD systems have sufficient capacity to run the risk financing mechanism and will maintain this capacity. This includes the risk financing mechanism budget. 268. Phase 1: Early Warning Triggered: This phase describes the actions to be carried out when the Early Warning System triggers a RF response. When the Early Warning System is triggered, a report requesting the release of the RF funds is prepared by the EWRD and sent to the RF Management Committee together with a request for the RF Management Committee to meet. Based on the information provided, the RF Management Committee determines the number of beneficiaries and the duration of support. The RF Management Committee then directly requests the release of funds from the RF account. This stimulates the updating of Contingency Plans and through a process in which bottom-up needs are reconciled with top- down resources, funds are released for onward distribution through regular PSNP channels. 269. Phase 2: Resource Transfer: This phase outlines the process of providing benefits to targeted beneficiaries. Funds are released from the RF Cash Fund by the RF Management Committee either for transfer to the Regions or for food to be purchased by FSCD. Funds are transferred from the Federal level to the Region through the PSNP financial management system. The EW&FSD requests the transfer of funds based on the funds release calculations approved by the RF Management Committee. The Regional PSNP accountant distributes the funds to the appropriate woredas in accordance with the RF budget as per request from the Regional EW&FSO. Woreda PSNP cashiers establish a separate database/payroll using the PASS for RF cash and food distributions. Woreda cashiers make payments to targeted beneficiaries in accordance with PSNP rules and procedures as per the request from Woreda EW and FS desks. 75 Reporting of RF distributions follows the same structure and timetable as for PSNP but separate reports are required. 270. Phase 3: Implementing Contingency Plans: In this phase contingency plans are implemented once resources have been released by the RF Management Committee and distributed through the system to woredas and the contingency plans have been revised. Although the contingency plans are woreda plans, implementation of most activities such as public works will be carried out at kebele level with the involvement of all concerned bodies and with technical support from woreda experts, DAs and sectoral experts. Transfer of cash will be the responsibility of WOFED. Food distribution involves woreda and kebele councils assisted by warehouse staff and KFSTF members. Support and facilitation of cash and food delivery comes from all levels. 271. Phase 4: Exit or Transition: Phase 4 addresses whether the cash/food requirements continue to be provided through the PSNP risk financing mechanism during a shock or through the humanitarian system. A transition of the humanitarian system is considered if: (i) the capacity of the PSNP systems to distribute benefits is exceeded and they cannot be scaled up any further; (ii) the numbers and location of people affected becomes so large that RF is no longer an appropriate mechanism to deal with their needs; (iii) the non-food needs of beneficiaries not addressed by PSNP or RF (e.g. water, health) become sufficiently important; and (iv) the onset of the shock is so fast that RF cannot respond in a timely way. If resources are more readily available through RF than the humanitarian system, a strategy to address the above constraints will be considered. The principles of primacy of transfers will be maintained throughout this phase. 272. Phase 5: Review and Lesson Learning: This outlines the reporting requirements for RF for both physical and financial reports. While RF and PSNP address different needs amongst beneficiaries drawn from the same woredas, separate reporting of both physical and financial activities is required. Funds distributed by RF should be audited as part of the normal PSNP auditing system. Once RF support distribution stops it is important to assess how effective it has been. A post-event report is required on this. Project Component 3: Institutional Support to the PSNP 273. This component will support institutional strengthening activities, including related management costs, during the APL III implementation period. These costs will be covered from different funding sources. Apart from the capacity building and management budgets included in the overall program budget, various development partners will finance other activities directly. For instance, CIDA will pay directly for TA to support the Directorates and as well as the new Regional Support Facility; WFP will support the food management system; and CIDA, DFID, EC, Irish Aid and SIDA will continue to provide funds through an MDTF managed by the World Bank. 274. This component focuses on strengthening all aspects of program implementation, concentrating on three key areas: 76 275. Program Management Costs: This will support the management budgets at Federal and Regional levels for activities related to ensuring effective management of the Program. Budgets are used to finance contract staff and technical assistance, logistics support, training and per diems, among other costs required to support the effective implementation of the Program. In particular these budgets will be used for: (i) planning, implementation, and technical oversight of public works, including the resources necessary for the full functioning of the Public Works Coordination and Public Works Focal Units; and (ii) more effective financial, commodity, and procurement management. At federal and regional level, dedicated management budgets will be given to each of the key implementing agencies (including MOFED, NRMD, EWRD, and FSCD) to ensure adequate overall program management and coordination. These budgets will be determined at the beginning of the fiscal year based on an annual work plan prepared by each of the implementing institutions. The budgets will then be transferred regularly to dedicated accounts. A review of the regional management and woreda administrative budgets will be carried out to determine the adequacy of the current budget formula. 276. Capacity Building: APL III will adopt a more structured and cohesive approach to capacity building based on a rounded analysis of needs, opportunities and gaps experienced to date. There will also be much closer follow-up on the implementation of capacity building efforts to identify areas where progress has been made and where gaps remain. Capacity building will cover the following areas: (a) Human Resource Development: This area is focused on staffing, training, and management skills. Key priorities during this phase will be to (i) review staffing levels following the BPR process to fill any remaining gaps; (ii) ensure effective and appropriate use of contract staff and TAs; (iii) improve the management of rolling trainings (including proper training needs assessment and monitoring and follow up of trainings); and (iv) expand the type of training offered through the Program to include management skills for senior staff and decision-makers. (b) Physical Capacity: A procurement plan has been developed that aims to address current gaps in physical capacity. There will also be an emphasis on ensuring maintenance of existing equipment to expand the lifespan of scarce resources. A key sub-set of this will be ensuring adequate anti-virus software and virus management systems for all computers used for the Program. Additional physical capacity gaps identified during the life of the Program will continue to be addressed, as appropriate, through program procurement. 277. This component will also finance capacity building activities to ensure the effective implementation of the Drought Risk Financing Component. Risk financing will be delivered through existing PSNP systems and structures and, as such, regular capacity building activities that address ongoing institutional and capacity gaps will also benefit the Drought Risk Financing Component. At the same time, a number of systems and processes unique to the Drought Risk Financing Component need be established and strengthened. This includes capacity building for the existing early warning system to ensure that it generates quality data and analysis on time. Support for contingency planning processes will also be provided as well as support for those systems, such as human resources, that are required to ensure the PSNP can scale up at local levels. The capacity building requirements for the Drought Risk Financing have been articulated in the Risk Financing Guidelines, which will be the basis for this component. 77 278. Implementation of the ESMF: This will ensure that there is sufficient capacity at all levels to effectively implement the ESMF. More specifically, this will cover the costs associated with implementing the safeguard measures as outlined in Annex 11. For example, this may include staff time, training for DAs and other relevant staff on the use of the ESMF and the process to revise the ESMF screening formats, as required. Financing will also cover activities required to implement the social and environmental agendas of the Program. 279. Monitoring and Evaluation Framework: This will ensure that the monitoring and evaluation system for the PSNP and HABP is well functioning and delivers the data required to provide a continuous assessment of progress towards achieving the objectives of the Programs, as outlined in the program logframes. In particular, this will: (a) Generate quality regular monitoring data. This will focus on revising the monitoring system to ensure that it generates the data required to monitor the revised program logframe, including revised monitoring formats and staff training. (b) Generate quality real-time data through the strengthening of the Federal Information Center and establishing Regional Information Centers in Afar, Amhara, Oromiya, SNNP, Somali, and Tigray Regions. This includes contracting staff, purchasing equipment and training. (c) Upgrade the monitoring system for public works with a comprehensive system of data management at sub-project level. This includes establishing necessary indicators, formats for reporting and systems to be managed by the Public Works Focal Unit. The establishment of this system will coordinate with that already in operation for the SLM project and WFP's MERET. This will also include financing for the twice annual federal Public Work Review, which includes a planning assessment and a technical review, and a showcase of best practice implementation and regional experience sharing under the public works component. (d) Support systems assessments and audits. These are periodic assessments that complement the data generated through the monitoring system. This includes: (i) an annual Procurement Review, to undertake an assessment of procurement practices at woreda level; (ii) an Annual Appeals Review, to undertake an assessment of the effectiveness of the appeals system. These will complement the other program audits that include; (iii) an Annual Financial Audit; and (iv) an Annual Commodity Audit. Each of the reviews will be accompanied by a Regional workshop. (e) Support the impact evaluation survey, with a household survey and a public works survey, including qualitative assessments. These will follow on from the CSA household survey implemented in 2006 and 2008. 280. Transparency and Accountability: This will finance several initiatives that will ensure widespread understanding of program objectives, procedures, and implementation progress among all key stakeholders. This will include: (a) Strengthening the rollout of the PSNP Communication Strategy. This includes: finalizing the distribution and monitoring of posters related to program objectives, criteria for targeting, and appeal procedures in all implementing kebeles in the Program as well as exploring other 78 innovative means of disseminating issues related to the Program, such as radio and a program website. (b) Establishment of a system of program client cards to increase transparency of entitlements under the PSNP and provide a proof of receipt of transfers. Cards will include the client charter of rights and responsibilities. The charter will also be disseminated through other means as part of the communication strategy. (c) Supporting PASS in all woredas to ensure that it is used correctly and consistently through a Regional IT HelpDesk and training, as required. Project Component 4: Support to the Household Asset Building Program (HABP) 281. Overall Program: The HABP is designed to assist food insecure households in PSNP woredas to transform their productive systems by diversifying income sources, improving productivity, and increasing productive assets. The Program has been designed to strengthen the extension system and rural service providers to deliver demand-driven and market-oriented assistance to food insecure households. It intends to achieve four outputs: (a) improved identification and development of on- and off-farm investment and income generating activities for food insecure households; (b) enhanced access by such households to sustainable and multiple financial services; (c) enhanced systems for input sourcing, production, and delivery; and (d) increased access by food insecure households to product and labor markets. 282. The Program aims to place households that are currently food insecure on a pathway whereby their capacity to take on and repay credit, adopt innovate practices and assume risk improves and they are able to generate incomes sufficient to cover basic food needs and resist shocks. In doing so, it will develop systems to provide support that can address the varying needs of different types of households. The multiple and sequential nature of such support is illustrated in Figure 3 below. 283. Development partners will provide assistance to the HABP for: (i) strengthening the delivery of public advisory services in support of household investments so that these services are more demand-driven and take into consideration market opportunities and conditions; (ii) improving the efficiency and effectiveness of financial service delivery to food insecure households; and (iii) enhancing relevant input and output markets. This will entail building systems and processes as well as enhancing existing capacity with a particular focus on knowledge and skills of frontline staff. The focus of development partners' support to the HABP is detailed below. 284. Strengthening the delivery of public advisory services in support of household investments: While the needs of individual food insecure households differ, they all require support to identify viable investments (including seeking wage employment) and understanding returns to different types of investments as well as to develop viable and appropriate business plans to implement the investments. They also require advice on new technologies and managing credit (if loans are used to finance the investments). Several Government systems provide support in this area, particularly the extension system. The development partner financing will help strengthen these systems according to the following principles: 79 · Consultation to identify investment and IGA opportunities appropriate for food insecure households and supported by expert advice and continual monitoring. · Livelihood zone approach--although consultations will be at woreda and community levels, in order to be relevant the identification of opportunities must be based on livelihood zones. · Simplicity of business plans that should be easily understandable, demand driven, and based on households' technical capacity, market opportunities, and appropriate choice of technologies. · Dynamism: opportunities identified--which will form the basis of technical assistance to target households--as well as technological innovation must remain dynamic and responsive to market changes and generation of new knowledge. Figure 3: Pathway to Graduation from PSNP and overall FSP Food Secure Extension services, including Mainstream credit Asset accumulation (including business planning, input linkages and service Graduates and market advice provision of FSP) Graduation from FSP Food Insecure (including graduates from PSNP) Graduation PSNP from PSNP Transfers Asset stabilization Chronically Food Repeat Insecure access to Community FSP credit assets (CCI & PSNP PW) Ultra Poor Intensive support and tailored products Village savings & credit and financial literacy 285. To help PSNP beneficiaries identify viable investment opportunities, support from development partners will be provided as follows: · Facilitation of stakeholder consultations in all PSNP woredas to identify investment opportunities 80 · Technical input for the identification of investment/IGA opportunities (market and technical analyzes -- specifically market analysis by livelihood zone; and assessments of production potential, enterprise feasibility, production/processing technology, etc.). 286. Capacity building to institutionalize market and technical analyses as described above will be mostly in the form of training managed through a service contract with appropriate institutions at the Regional level. Following the identification of viable investment opportunities, APL III will provide financing for support to food insecure households in the development of business plans appropriate to their particular technical, labor and financial capacity by: · Building technical and facilitation skills (including TA, development of relevant curricula and training modalities, and delivery of training) within relevant institutions at the woreda level and among DAs at the kebele level to ensure that advice to business plans is appropriate in terms of: (i) financial viability; (ii) agro-ecological suitability; (iii) acceptance by the community and individual households; (iv) input supply; and (v) marketability of products/services at woreda and kebele levels. There will also be effective technology dissemination for households interested in innovation; · Undertaking sensitization activities at all levels (Federal, Regional and Woreda) to reorient such support towards a more demand driven approach; · Supporting the capacity of both woreda staff and DAs to follow up and monitor beneficiary households' investments and supervise the business planning process, through the provision of necessary and appropriate equipment, transport, etc.; · Developing training curricula and providing training-of-trainers on new technologies/practices; and · Promoting adaptive research activities and linkages with new sources of technology. 287. Improving the efficiency and effectiveness of financial service delivery to food insecure households: APL III will support the development of a market-based financial system to improve access by food insecure households to a range of financial products. In doing so, it will seek to promote the following principles: · Use of Appropriate Institutions: Financial services should be delivered through sustainable and appropriate financial providers. RuSACCOs are well placed to promote savings and manage credit that is revolved within a community. MFIs are better placed to mobilize external funds and provide credit on a larger scale. They can provide money transfer services for migrant labor. In no case should financial services be provided through Government administrative bodies. · Adoption of Sustainable Interest Rate: The level of interest should not distort the financial market. Service providers will set interest rates that are sustainable. · Prudent Management of Revolving Funds: One form of financial service provision is through revolving funds. Experience shows that to be properly managed, revolving funds should be owned and managed by member-owned entities and should not be an entitlement. Rather, access should be on predetermined criteria linked to performance. 81 · Centrality of Savings: While multiple financial products are required, savings must form the basis of the financial system. Savings mobilization must therefore precede credit provision as this produces financial discipline and experience in managing cash. 288. Development partner financing will be provided for: (i) the development and field testing of multiple financial products (savings, different types of credit) that adhere to sound financial principles including commercial interest rates and are tailored to the differing needs and capacities of food insecure households; (ii) the dissemination of these products to service providers, complemented by a rolling training program; and (iii) the development of financial literacy materials. 289. To establish a strong network of RuSACCOs, APL III will support (i) capacity building of woreda cooperatives promotion desks, and RuSACCOs unions for community mobilization and technical support for the establishment of RuSACCOs in PSNP woredas including the development of by-laws, financial systems, etc, as well as to provide on-going support and guidance to RuSACCOs, (ii) capacity building of RuSACCOs, (iii) support to inter-sectoral coordination and linkages between MFIs and RuSACCOs; and (iv) support to the strengthening of the regulatory and supervisory environment. 290. Government resources only will directly finance credits to food insecure households to be channeled through RuSACCOs and MFIs. 291. Strengthening input and output markets: Sustainable access to input and output markets is critical to promote the innovation required to transform the livelihoods of food insecure households and ensure the success of household investments. Development partner support will therefore also be provided to enhance the delivery of agricultural inputs and strengthen access to output and service markets. The aim is to: (i) promote the role of Government as facilitator rather than supplier of inputs; and (ii) ensure that cooperatives as well as private suppliers provide more effective services. Detailed interventions to support output and service markets will be further developed during project implementation. 292. Support to program management: This will provide resources to ensure the effective management of the HABP at all levels. At the federal level management budgets will be provided to each of the key implementing agencies, particularly AED. A mechanism to provide support to the micro/small enterprise development enterprise and cooperative promotion agency will be agreed upon and outlined in the PIM. The provision of these budgets aims to ensure adequate overall program management and coordination. Budgets will be provided to these line agencies at Regional and woreda level in a similar fashion to support day-to-day management and implementation activities. In addition, this will support monitoring and evaluation, including the establishing of robust reporting systems for the extension system and financial service providers and the procurement of physical inputs such as equipment for Farmer Training Centers and vehicles such as bicycles and motorbikes for DAs and woreda-level staff. 293. Given that many of the above initiatives are new to Ethiopia, study tours and experience sharing events will be organized to facilitate implementation and to create awareness of international best practices. 82 Annex 5: Project Costs ETHIOPIA: Productive Safety Net APL III Program Cost By Component and/or Activity Local Foreign Total US$ million US$ million US$ million Component 1: Safety Net Grants A. Sub-component: Public Works 1,098.30 1,098.30 B. Sub-component: Direct Support 274.60 274.60 Contingencies for Food and Cash Transfers 274.60 274.60 Capital and Administrative Budgets 273.00 1.50 274.50 Performance incentive grants 14.16 14.16 Component 2: Drought Risk Financing 80.00 80.00 160.00 Component 3: Institutional Support to PSNP 71.90 5.50 77.40 Component 4: Support to HABP 79.30 4.00 83.30 Total Baseline Cost 2,165.86 91.00 2,256.86 Physical Contingencies 0 0 0 Price Contingencies 0 0 0 Total Program Costs 2,165.86 91.00 2,256.86 Total Financing Required 2,165.86 91.00 2,256.86 294. Estimated taxes and duties are US$13.65 million, and the total project cost, net of taxes, is US$2,243.15 million. Therefore, the share of project cost net of taxes is 99 percent. 295. Of the total project cost of US$2,256.8 million, IDA will finance US$480.0 million (21 percent); other development partners, US$1,240.3 million (55 percent) of which US$580.9 million from USAID and WFP will be provided in the form of in-kind resources, and Government will provide cash counterpart financing of US$10.0 million. The total financing gap for all the components is estimated at US$526.5 million or 23 percent of total estimated Program costs. 296. This represents a significant financing gap. Due to additional development partner financing anticipated before the approval of APL III, this gap is likely to be reduced. The strategy of the Government and development partners to address this will be to ensure that on a rolling basis the next two program years are adequately funded. This strategy was successfully employed during implementation of APL I and II to overcome financing gaps that were highlighted at the time of Executive Board approval. However, there is a risk that the current global economic climate may affect development budgets. Current indications from development partners are that financing for social protection programs is likely to be protected from any budget cuts. 297. Should the PSNP remain underfinanced, measures could be taken to scale back the design of the Program, which is sufficiently flexible to allow such changes. This approach is, however, 83 undesirable given the vulnerability of target households, which, if not covered by the PSNP would likely require support through the emergency appeal system. Development partners recognize that the PSNP is a more effective and efficient response than the emergency appeal system providing an additional incentive to ensure that the PSNP is fully financed. 298. In addition to the financing for the PSNP and HABP detailed above and in Tables 7 through 9 below, the Government will allocate significant resources to finance the credit component of the HABP, complementary community infrastructure, and the resettlement program. This amounts to 2 billion ETB annually (equivalent to US$160.3 million) for at least the next three years. This budget is allocated to Regions on the basis of a Vulnerability Index and is transferred in the form of a Federal block grant. Regions then allocate the block grant to the initiatives of the Food Security Program depending on their local requirements and priorities. Simultaneously, the Government allocates significant in-kind resources (amounting to an estimated US$53.0 million) to the Program through, for example, the use of dedicated government staff at all levels and logistic support. 299. Historic contributions of development partners to APL I and II are detailed in Table 7: Table 7: Development Partner Financing for Phase 1 and Phase 2 of PSNP Phase Sources of Financing from Partners (in USD million)82 CIDA DFID EC IDA83 Irish RNE SIDA USAID WFP Total84 Aid 1st Phase 34.0 95.9 37.5 113.7 21.3 0 4.3 102.4 0 409.0 2005-2006 2nd Phase 72.5 138.5 187.8 200.0 44.2 34.8 23.0 314.2 25.1 1,040.2 2007-2009 Total: 106.5 234.4 225.3 313.7 65.5 33.7 27.3 416.6 25.1 1,449.2 Source: World Bank Project Document. 300. The details of financing for APL III are detailed below. Table 8: Program Total Cost by Component (including contingencies) (in US$ million) Program Component Total 1. Safety Net Grants 1,936.2 2. Drought Risk Financing 160.0 3. Institutional Support to PSNP 77.4 4. Support to HABP 83.3 Total Program Cost 2,256.8 Source: Government of Ethiopian 2009. 82 Contributions from USAID and WFP are in-kind. The methodology that underpins these calculations is described in the footnotes of Table 9. 83 This includes an IDA grant of US$70 million for APL I plus US$44 from the World Bank's portfolio in Ethiopia (EDR project) and an IDA grant of US$175 million for APL II plus Additional Financing of US$25 million from the GFRP. 84 The reasons for this increase in financing to APL II as compared with APL I are as follows: APL II was one year longer than APL I; PSNP beneficiary numbers increased from 5.4 to 7.57 million; the wage rate was increased twice from 6 to 8 to 10 birr currently; additional resources were provided through the PSNP in 2008; the value of food increased in 2008 and 2009. 84 Table 9: Ethiopia Productive Safety Net Program, APL III (2010-2014) Breakdown of Costs by Sources of Financing (in US$ millions) Sources of Financing (signed and indicative) Project Component Carry- Financing Budget1 GOE IDA DFID EC CIDA Irish Aid USAID2 WFP SIDA RNE Subtotal over Gap COMPONENT 1: SAFETY NET GRANTS 1,936.2 0.0 398.5 218.8 73.2 68.0 68.8 457.0 50.0 21.5 66.3 0.0 1,422.2 -514.0 I. Transfers (i) Public Works Subcomponent 1,098.3 0.0 224.3 122.7 41.2 38.3 38.7 261.1 28.6 12.3 37.3 0.0 804.5 -293.7 (ii) Direct Support Subcomponent 274.6 0.0 56.1 30.7 10.3 9.6 9.7 65.3 7.1 3.1 9.3 0.0 201.1 -73.4 II. Other (i) Contingencies (Cash & Food) 274.6 0.0 56.1 30.7 10.3 9.6 9.7 65.3 7.1 3.1 9.3 0.0 201.1 -73.4 (ii) Capital Budget 205.9 0.0 42.1 23.0 7.7 7.2 7.3 49.0 5.4 2.3 7.0 0.0 150.9 -55.1 (iii) Administrative Budget 68.6 0.0 14.0 7.7 2.6 2.4 2.4 16.3 1.8 0.8 2.3 0.0 50.3 -18.4 (iv) Performance incentive grants3 14.2 0.0 6.0 4.0 1.2 1.0 1.0 0.0 0.0 0.0 1.0 0.0 14.2 0.0 COMPONENT 2: RISK FINANCING4 160.0 0.0 50.0 31.5 0.0 0.0 0.0 73.9 0.0 0.0 0.0 0.0 155.3 -4.7 COMPONENT 3: INSTITUTIONAL SUPPORT5 77.4 0.0 14.0 32.0 5.5 13.8 5.5 0.0 0.0 1.5 5.0 0.0 77.3 0.0 PSNP SUB-TOTAL 2,173.5 0.0 462.5 282.3 78.7 81.8 74.3 530.9 50.0 23.0 71.3 0.0 1,654.8 -518.7 COMPONENT 4: SUPPORT TO HABP 83.3 10.0 17.5 41.8 0.0 0.0 6.3 0.0 0.0 0.0 0.0 0.0 75.6 -7.7 TOTAL PROGRAM: 2,256.8 10.0 480.0 324.1 78.7 81.8 80.6 530.9 50.0 23.0 71.3 0.0 1,730.4 -526.5 Footnotes: (i) The budget for APL III was calculated based on a number of assumptions. These are: a) 40% of program population graduates between 2009 and 2013, with no graduation in pastoral regions; b) the wage rate increases progressively from 10 to 14 birr; and c) a progress move towards cash in highland areas. (ii) Contributions from USAID for the PSNP are stated in-kind. For APL II, in-kind contributions to the PSNP have been valued at the prevailing market rate per year. To calculate the value of the in-kind contribution to APL III, an average price of US$422 per MT was used, which was the average price in 2009. It is also important to note that the USAID contribution to the PSNP is allocated through NGOs and WFP. USAID has also allocated financing to NGOs to support HABP-related activities that are not reflected here. (iii) The performance incentive grant will be allocated annually and will be reported against administrative and capital budget lines. (iv) The overall resource envelope for Risk Financing is US$80 million per year. It is assumed that the Risk Financing will trigger twice during APL III. This is based on historic requirements for emergency response resources among PSNP woredas. (v) The Institutional Support to PSNP component is financed through allocations to Government and to the Multi-Donor Trust Fund (MDTF), a Trust Fund managed by the World Bank. Of the Irish Aid allocation to Institutional Support, an estimated US$300,000 will be channeled through the MDTF. US$12.6 million of the CIDA allocation to this component is channeled through the CIDA Regional Support Facility. 85 Annex 6: Implementation Arrangements ETHIOPIA: Productive Safety Net APL III 301. The Productive Safety Net Program (PSNP) and Household Asset Building Program (HABP) are components of the larger Food Security Program (FSP) of the Government of Ethiopia. Food Security line agencies at every level of Government are accountable for the oversight and coordination of the FSP with implementation being undertaken by line ministries, Government agencies and other partners at all levels. These arrangements are cemented in a Memorandum of Understanding (MOU) between Government and development partners. The roles and responsibilities of implementing partners are summarized below and described in detail in the PSNP and HABP Project Implementation Manuals (PIM). Federal Level 302. The Ministry of Agriculture and Rural Development (MOARD) is responsible for management and coordination of the PSNP and HABP with overall coordination responsibilities vested the Disaster Risk Management and Food Security Sector (see below). MOARD sets the policy direction and provides technical support for planning and implementation of the Programs as necessary and assists in setting the policy direction to which the PSNP and HABP contribute. The MOARD management meetings, attended by the State Ministers of MOARD and chaired by the Minister, advise the Minister on all key decisions including program resource allocation to the various implementers based on the consolidated proposals prepared by the DRMFSS. 303. The Disaster Risk Management and Food Security Sector (DRMFSS) is directly answerable to the Minister of Agricultural and Rural Development for the performance of the PSNP and HABP. The State Minister for DRMFSS chairs the Inter-Ministerial Management Committee for the PSNP and HABP, which is composed of the State Ministers for MOARD, MOFED, the Water Ministry, Roads Authority, and Trade and Industry. This Committee identifies and resolves key implementation issues. The State Minister also chairs the twice monthly FSCD and EWRD planning and monitoring meetings for PSNP risk financing. 304. Within DRMFSS, the Food Security Coordination Directorate (FSCD) is responsible for facilitating the day-to-day management and coordination of the PSNP and HABP. It has direct responsibility shared with the Early Warning and Response Directorate (EWRD) for the transfer components (including risk financing) and co-chairs with the EWRD the Technical Committee on PSNP transfers. It supports the Natural Resource Management Directorate for the public works component. Its key responsibilities include: (i) support to coordination and oversight of the PSNP and HABP; (ii) support to ensuring appropriate linkages of the PSNP and HABP with other FSP and development interventions; (iii) consolidating PSNP and HABP work plans and budget proposals from the Regions, and making resource allocation proposals for decision by the Minister through the State Minister for DRMFSS; (iv) on this basis, allocating PSNP resources to the Regions; (iv) providing technical support to regional food security offices; (v) monitoring overall capacity to implement the PSNP; and (vi) monitoring and evaluating the efficiency, effectiveness and impact of the PSNP and HABP. TAs will be employed to support the implementation of the program, as necessary. 86 305. The Early Warning and Response Directorate (EWRD), which is also under DRMFSS, co-chairs the Technical Committee on PSNP Transfers and plays a critical role in PSNP risk financing by providing both accurate and timely early warning information as well as adequate linkages between PSNP risk financing resourced activities and other activities related to humanitarian response. This includes the collection and analysis of early warning data from Regions and analysis of federal-level triggers; supervision of the Early Warning Working Group (EWWG) in developing a consensual early warning statement on a monthly basis and ensuring that early warning system and structures function at each level. EWRD is also responsible for the procurement, transport, and management of in-kind commodities for the PSNP. 306. The Natural Resources Management Directorate (NRMD) of MOARD, through the Federal Public Works Coordination Unit (PWCU), is responsible for coordination and oversight of the public works (PW) component of the PSNP. Its responsibilities include (i) support for Regional PW Focal Units and awareness creation; (ii) oversight of, and support to, M&E of public works, including the conduct of Public Works Reviews and Impact Assessments; (iii) ensuring satisfactory implementation of the ESMF and review of ESMF design as required; (iv) capacity building for public works, including development of appropriate training materials and conduct of capacity needs analysis; (v) technical support and quality assurance to public works as required, concerning planning, design, operations and maintenance to ensure sustainable public works, as well as the development and potential use of Geographic Information Systems in these functions; (vi) oversight support of the integration of non-Natural Resources sectors into the planning, design, and implementation of public works; and, (vii) liaising with FSCD and other PSNP partner institutions on coordination and management of public works and participation in PSNP design and management forums, including policy issues, the roll out of the pastoral PSNP, and development of exit strategies. 307. The Agricultural Extension Directorate (AED) of MOARD is responsible for managing, coordinating, and overseeing of the technical services component of the HABP. It chairs the Technical Committee on HABP (see below). In this coordination function, the Directorate is responsible for (i) providing guidance to implementing partners; (ii) ensuring appropriate linkages of the HABP with the other components of the Food Security Program and other development interventions; and (iii) ensuring effective coordination across departments and ministries for the successful implementation of the program. One key responsibility of the AED is to implement the reorientation of the Extension Service toward a demand-driven, market- oriented approach. The Directorate will (i) strengthen its system and staff towards a demand- driven extension approach; (ii) chair the special HABP Technical Committee that will provide technical back up to the various implementing agencies concerned; and (iii) issue explicit guidelines that its staff at all levels and particularly woreda extension staff and Development Agents will not be involved in loan disbursement and collection. 308. Ministry of Finance and Economic Development (MOFED) is responsible for disbursing safety net and household asset building resources to Ministries at the Federal level and to Regions in line with requests submitted by FSCD. MOFED is accountable for the overall financial management of the programs, including management of the special and pooled Birr accounts and reporting on the PSNP, risk financing, and HABP. 87 309. The Cooperative Promotion Agency will collaborate with the FSCD, AED and other relevant institutions to strengthen existing RuSACCOs and establish new RuSACCOs where they do not exist. The Agency will work with relevant institutions to make the regulatory environment conducive for RuSACCOs to operate in chronically food insecure areas and to provide guidance to Regional and woreda Cooperative Promotion Bureaus and offices in the implementation of the HABP more generally. 310. A Joint Strategic Oversight Committee (JSOC) is comprised of representatives from the Government of Ethiopia and Development Partners Group. The State Minister for the DRMFSS chairs the Committee, which is delegated to the other MOARD State Ministers as required. The JSOC is responsible for ensuring dialogue and joint oversight of program implementation at a strategic level. Specifically, it is responsible for: (i) making recommendations based on the analysis of the Technical Committees (see below) on the appropriate responses to issues emerging during the implementation of the programs; (ii) making recommendations on strategic decisions concerning program implementation, linkages with emergency interventions, other food security interventions and agricultural growth program, and related policies; and (iii) engaging in policy and strategy dialogue on issues that are of direct relevance to the PSNP and HABP. 311. PSNP Risk Financing Management Committee, which is chaired by the State Minister for the DRMFSS, allocates transfers to targeted beneficiaries in PSNP woredas through the Risk Financing facility. The Committee is formed of PSNP development partners, EWRD, FSCD and MOFED and is chaired by the State Minister for DRMFSS. The RF Management Committee meets at the request of the EWRD when information demonstrates the need for a risk financing response in PSNP woredas. The PSNP Risk Financing Management Committee is responsible for: (i) reviewing the beneficiary numbers; (ii) reviewing and approving the release of financing from the Risk Financing facility based on Early Warning triggers and specific requests from woredas; (iii) monitoring the on-going release of subsequent tranches of funds based on up-to- date reports; (iv) deciding when to transition to an alternative aid modality; and (v) reviewing the post-event report on the effectiveness of the response. 312. Three Joint Technical Committees will be established as part of APL III chaired by the responsible Directorate Head with membership from development partners and other relevant line agencies and stakeholders. These Technical Committees will report regularly to the JSOC and are responsible for: (i) assessing performance and progress toward achievement of agreed benchmarks; (ii) recommending to the JSOC or Minister of MOARD appropriate responses to issues emerging during program implementation; (iii) promoting linkages with other food security programs, agricultural growth initiatives, and emergency interventions, and (iv) managing and overseeing ad hoc measures to support of regional or federal authorities in implementing specific aspects of the programs (see Figure 4 below). 88 Figure 4: FSP Coordination Mechanism (Joint Government-development partners) Minister MOARD Overall RED- FSP Joint Strategic Oversight Committee FS FSCD Chair: MoARD DRMFSS State Minister Committees JOINT TC 3 JOINT TC 1 JOINT TC 2 PW Early Warning & HABP Response/Transfers Co-Chairs: Chair: Director Natural Co-Chairs: EWR & Director Resource (ARD) & FSC Directors Extension MoWR Director Cash Food Extension Financial Srv Working Working Working Working Group Group Group Group Regional Level 313. The Regional Cabinet is responsible for the review and approval of food security, PSNP and HABP annual plans and budgets submitted by woredas and consolidated regional PSNP and HABP work plans and budgets submitted by the Regional Bureau of Agriculture and Rural Development (BOARD). The Cabinet also reviews and approves the annual and biannual progress reports on implementation of the regional PSNP and HABP and budget utilization. The Cabinet reports to the Council on PSNP and HABP implementation, as it does for all activities implemented in the Region, and ensures that the Regional Council's decisions with regards to broad regional development priorities are reflected in the Regional FSP plans, and those for PSNP and HABP. 314. The Regional Food Security Steering Committee (RFSSC), chaired by the Regional President or his delegate, provides advice to ensure the proper implementation of food security strategies and programs at the Regional level based on the recommendations of the Regional FS Task Force chaired by the Head BOARD. It also ensures the effective integration of the Regional PSNP and HABP into the Regional development plan, participates in monitoring and evaluation 89 of Program activities and analyses of the consolidated FSP work plan and budget proposal submission to the Regional Cabinet. 315. The Regional Bureau of Agriculture and Rural Development (BOARD) manages the PSNP and HABP. It oversees the integration of safety net and household asset building activities into the Food Security Program and the Regional Rural Development Strategy. Specific duties include: (i) providing overall guidance to the Regional Food Security Office and line bureaus to ensure coordination on planning and implementation of the Regional PSNP and HABP; (ii) ensuring efficient procurement where applicable; and (iii) reviewing and providing feedback on reports submitted by Regional Food Security Coordination Offices on implementation of safety net interventions. 316. The Regional Food Security Coordination Office (RFSCO)85 reports to BOARD and is also technically accountable to the FSCD. It has the responsibility for supporting the management and coordination of the PSNP and HABP. Its responsibilities include: (i) consolidating annual implementation plans and budgets for the region in line with proposals from woredas and line bureaus, for submission to BOARD and decision on resource allocation by the Regional Cabinet as noted above; (ii) mobilizing technical assistance as needed; (iii) identifying and monitoring capacity to implement PSNP activities at regional, woreda and kebele levels; (iv) supporting the NR Department in overseeing and supporting the implementation of PSNP public works in the Region (including ensuring implementation of the ESMF); (v) holding quarterly review meetings with government and non-governmental agencies involved in implementation of the PSNP and HABP; (vi) approving NGO plans of PSNP and HABP activities; (vii) coordinating monitoring and evaluation activities; (viii) preparing progress reports; and (ix) establishing and implementing the Regional Rapid Response Mechanism. 317. Regional Early Warning and Response Department is responsible for collecting early warning data from the woreda and zone levels, performing a detailed analysis of the data and sending on analyzed data and reporting to the EWRD. It is also responsible for supporting the RFSCO and concerned woredas in managing the scale-up of the PSNP system when the risk financing is activated. It also supports the transport and management of food resources for the PSNP. 318. The Natural Resources Management Department of the BOARD through their Regional Public Works Focal Units acts as secretary to the Regional Technical Coordination Committee. Its responsibilities include: (i) implementation of the public works M&E system, including regular reporting to the federal PWCU on the activities, outcomes, quality, and effectiveness of the public works, as well as participation in Public Works Reviews and Impact Assessments as required; (ii) ensuring implementation of the ESMF through integration of the ESMF in the planning procedures and training for public works; (iii) consolidating public works plans and budgets developed in the woredas; (iv) overseeing woreda supervision of the public works, and providing technical backstopping as required; (v) organizing and delivering annual public works training programs; (vi) assessing the effectiveness of training and undertaking training needs assessments; (vii) reviewing community-level public works planning procedures and formats in conjunction with woreda staff; (viii) overseeing integration of community watershed plans into 85 The post-BPR structure varies in some Regions. The terms used here refer to the most commonly used structure. 90 woreda plans; (ix) liaison between public works planning and non-labor inputs procurement to ensure that materials and expertise are available as required for PW implementation; (x) knowledge management including identifying and disseminating best practices, reviewing standards, work norms, disseminating technical standards, and identifying new technologies to enhance the quality, sustainability, and impact of public works; (xi) liaison with PWCU, FSCD, and other PSNP partner institutions on coordination and management of the public works, and participation in PSNP management forums; and (xii) supporting contingency planning for PSNP risk financing at woreda level. 319. The Regional Extension Department of BOARD is responsible for the effectiveness of the technical services component of the HABP. Its responsibilities include: (i) preparing and reviewing the consultative stakeholder process and providing technical inputs into the market analysis; (ii) consolidating plans and budgets for HABP technical services developed in the woredas; (iii) overseeing integration of market analysis and other HABP-related plans into woreda plans; (iv) overseeing woreda supervision of the HABP technical services and providing technical backstopping; (v) ensuring effective training related to HABP technical services, market analysis, and marketing; (vi) creating linkages for the dissemination of technological innovations; (vii) supporting the M&E system of the RFSCO, especially on the HABP; and (viii) knowledge management including identifying and disseminating best practices. 320. The Bureau of Finance and Economic Development (BOFED) is responsible for disbursing PSNP resources to woredas and line departments in line with requests submitted by the RFSCO. It is responsible for the overall financial management of regional PSNP and HABP budgets including financial reporting and provides technical support to woreda (either directly or through Zonal Departments of Finance and Economic Development). 321. The Bureau of Cooperative Promotion will collaborate with the RFSCO, Regional Extension Department and other relevant institutions to strengthen existing RuSACCOs and establish new RuSACCOs where they do not exist. The Bureau will provide guidance to the woreda Cooperative Promotion Offices in the support that they should give to cooperatives in linking households to the market and assist them for instance in facilitating the establishment of cooperative unions. 322. Three Technical Committees report to the Regional FS Task Force. These Technical Committees will provide oversight and supervision to lower level implementers, ensuring that guidance emanating from the different Regional sectors is coordinated and consistent. The Committees will assist the responsible Departments to ensure synergy between plans and activities and the plans and activities of other FSP components and of other development interventions in the Region. 323. In Amhara, Oromiya, and SNNP Regions, Zones are expected to play a significant role in supporting the implementation of PSNP and HABP. Zones are administratively part of the Regional structure and as such zonal departments have the same responsibilities as the Regional bureaus outlined above. 91 Woreda Level 324. The woreda is the key level of Government that determines needs, undertakes planning, and implements the PSNP and HABP. The Cabinet and Council are the highest woreda-level decision-making bodies. The Cabinet prepares the woreda overall plan and budget, which the Council then approves. Together the two are responsible for the allocation of PSNP and HABP resources to kebeles based on the recommendations of the Woreda Food Security Task Force and an obligation to maximize all resources available to the woreda including PSNP and HABP resources. More broadly, the Cabinet and the Council are responsible for guiding and overseeing the integration of the planning and implementation for the PSNP and HABP, and for the FSP as a whole, in the woreda integrated plan. Moreover in this phase of the FSP the woreda Council will have a stronger role in ensuring accountability for program performance in the woreda. 325. The Woreda Office of Agriculture and Rural Development (WOARD) manages the Safety Net and HABP at woreda level. With the support of the woreda FSP Steering Committee, the woreda FSTF and three specialized Technical Committees, it oversees the integration of the PSNP and HABP activities into the Food Security Program and the woreda rural development strategy. It is also responsible for the implementation and management of the PSNP and HABP, which includes maintaining an accurate record of appeals and appeals resolutions and ensuring that posters are effectively disseminated to ensure broad awareness of rights and responsibilities under the Programs. 326. The Woreda Food Security Task Force (WFSTF) chaired by the head of the WOARD works directly under the guidance of the Woreda FSP Steering Committee chaired by the Woreda Administrator. The Food Security Desk acts as secretary of the Task Force. The WFSTF has a wide-ranging membership, which reflects its broader mandate in relation to the FSP as a whole. In relation to the PSNP and HABP it will establish three specialized Technical Committees. Through these Technical Committees, the Task Force: (i) reviews and recommends kebele annual PSNP and HABP plans for approval; (ii) consolidates annual woreda PSNP and HABP plans and budget and ensures their integration within the overall woreda plan; (iii) ensures that all woreda offices integrate PSNP and HABP activities into their annual work plan; (iv) ensures close collaboration with and regular reporting to the Region; (v) ensures adequate information to the woreda Council and that the Council's decisions are acted upon; (vi) participates in monitoring and evaluation of PSNP and HABP activities; (vii) provides assistance to kebeles in establishing and training KFSTFs; (viii) holds quarterly progress review meeting on PSNP activities; and (ix) reviews monthly progress reports on PSNP and HABP activities. 327. The Woreda Food Security Desk (WFSD) oversees Safety Net and HABP activities and is technically accountable to the RFSCOs. The WFSD functions include: (i) ensuring the preparation of pipeline of projects for PSNP in consultation with the Kebele Food Security Task Force; (ii) mobilizing technical assistance as needed; (iii) ensuring that PSNP risk financing contingency plans are prepared and implemented according to risk financing guidelines (iv) undertaking monitoring and evaluation in coordination with woreda sectoral offices; (v) holding quarterly technical review meetings with implementing agencies; (vi) submitting progress reports to the WOARD; (vii) maintaining accurate records of kebele Safety Net and HABP activities and 92 list of beneficiaries; and (viii) providing information on target areas and selected beneficiaries to sectoral offices and other agencies involved in planning and implementing Safety Net and HABP activities. 328. The Early Warning and Response Desk co-chairs the woreda Technical Committee on early warning and transfers with the FSD. It has a critical role to play with regard to Risk Financing, by providing accurate and timely early warning information, ensuring adequate linkages between Risk Financing resourced activities and other actions related to humanitarian response, and supporting the FSD and the concerned kebeles in managing the scaling up of the PSNP system in case of activation of the Risk Financing facility in the woreda. 329. The Natural Resource Desk co-chairs the woreda Technical Committee on Public Works with the Water Resource desk. It is directly responsible for managing the PSNP public works with the support of the FS Desk. Its responsibilities include: (i) consolidating public works plans and budgets developed in the kebeles; (ii) ensuring integration of community watershed plans into woreda plans and, more broadly, integration of the PSNP public works in the overall woreda plan; (iii) providing assistance to DAs and communities in the planning process; (iv) implementing the ESMF; (v) together with FS Desk, supervising public works and providing technical backstopping; vii) supporting the M&E system especially on the Public Works Review; and (viii) facilitating experience sharing among kebeles. Through the woreda public works Technical Committee it coordinates the interaction and involvement of the relevant line offices/desks and other PSNP actors in the public works program. 330. All concerned Woreda Sector Offices (represented in the WFSTF as noted above) are responsible for: (i) consolidating proposals of the Kebele Food Security Task Force for incorporation in the woreda PSNP and HABP plans; (ii) incorporating PSNP and HABP activities in their yearly program/action-plans, based on the woreda integrated plan including PSNP plans; (iii) preparing activity implementation plans and request budget for implementation; (iv) implementing PSNP and HABP activities at kebele and community levels; (v) providing technical assistance and training to technical personnel and kebele staff; (vi) undertaking project screening in accordance with the ESMF; (viii) conducting monitoring and evaluation of activities; and (ix) preparing quarterly progress and financial reports. 331. The Woreda Office of Finance and Economic Development (WOFED) ensures that: (i) the budget for the PSNP and HABP is received in a timely manner at the woreda level to guarantee smooth implementation of approved plans and activities; (ii) undertakes timely PSNP payments for beneficiary households, supervising personnel, and the purchase of relevant equipment and materials; and (iii) exercises necessary fiduciary controls and reports on fund utilization to Regional BOFEDs. 332. The Woreda Cooperative Promotion Office will assist in capacitating existing RuSACCOs and establishing new ones. It will implement any directive issued from the Federal and Regional Cooperative Promotion Agency to improve the regulatory environment within which RuSACCOs function. The Office will also assist the formation of household groups and cooperatives and assist existing ones in their activities, as a way of strengthening the linkage 93 between households and the market for both input supply and marketing of their products as well as for value addition activities. 333. NGOs also play a role in PSNP and HABP implementation. In addition to being members of the Woreda Food Security Task Force and Regional Food Security Task Forces, they may also, with additional financing, contribute their capacity and expertise to the program. In doing so, they should work with Government structures and abide by the Program Implementation Manuals. Kebele Level 334. The Kebele Cabinet: (i) approves kebele PSNP beneficiaries based on the recommendations of the Community Food Security Task Force; (ii) identifies activities for PSNP purposes; (iii) prepares the kebele PSNP plan; (iv) ensures that the PSNP and HABP are linked, and consistent with, other food security interventions and that PSNP activities and priorities, in particular in relation to public works, are integrated in the broader development plan of the kebele; (v) maintains records on the status of beneficiary households; (vi) reports monthly; (vii) oversees food security activities in the kebele; (viii) participates in the monitoring and evaluation system for the Food Security Program; and (ix) ensures that lists of beneficiaries, heard and resolved appeals along with program plans and budgets are posted in public locations. 335. The Kebele Council, in its overall role of oversight of the Kebele Cabinet, reviews and approves the Cabinet's proposals, notably in relation to the PSNP plan and its integration within the broader kebele plan. The Kebele Council is also directly involved in the functioning of the appeal system (see below) and is responsible for linking up with the woreda Council on this. 336. The Kebele Food Security Task Force (KFSTF) is a decision-making body that oversees all planning and implementation of safety net activities on behalf of the kebele Cabinet. KFSTF members include the Kebele Administration, Development Agents, Community Based Health Workers (CBHW), Teachers and Youth Associations. KFSTF's functions include: (i) community mobilization to identify and prioritize community needs; (ii) supporting DAs in planning work with identified communities following participatory watershed planning guidelines and Line Bureaus specific proposals; (iii) targeting beneficiaries and participants for public works and direct support based on community targeting exercises; (iv) preparing Kebele Safety Net Plan in consultation with woreda sectoral offices; (v) maintaining minutes of KFSTF meetings on Safety Net and HABP issues, Kebele Safety Net activities, list of participants and progress reports; (vi) establishing and training of Community Food Security Task Force; and (vii) participating in monitoring and evaluation of safety net and household asset building activities including the Rapid Response Mechanism. 337. Kebele Appeals Committees (KACs) will be established to hear and resolve appeals regarding Safety Net and HABP matters in a timely manner. KACs will: (i) submit a complete listing of appeals cases, appeals resolutions, and unresolved appeals to the Kebele Council each quarter which will review them and forward them to the Woreda Council and the WOARD every quarter; (ii) convene within one month of the establishment of a new annual listing of beneficiaries to hear appeals submitted in their jurisdiction and will resolve a minimum of 95 94 percent of these cases within the month; and (iii) provide the listing of the appeals and the associated resolutions to the Kebele Council no later than 2 months after the announcement of the beneficiaries listing. Community Level 338. The Community Food Security Task Force's (CFSTF) primary responsibility is the identification of PSNP beneficiaries. It is composed of representatives from the kebele FSTF, a Development Agent, two or three elected female representatives, two or three elected male representatives, an elected youth representative, and an elected representative of the elderly. The functions of the CFSTF include: (i) mobilizing the community for participatory planning exercises; (ii) undertaking a needs assessment identifying those households who can participate in public works and those without sufficient labor or other support who will need direct support; (iii) monitoring the public works; and (iv) participating in the regular review of safety net and household asset building beneficiaries. 339. The Development Agents (DAs) are employees of the Extension Desk of the Office of Agriculture and Rural Development who reside in the kebeles and work to facilitate PSNP and HABP implementation. DAs: (i) are members of the KFSTF and CFSTF; (ii) are responsible for supporting the CFSTF in prioritizing community needs and preparing annual PSNP and HABP plans; (iii) oversee the implementation of public works; (iv) prepare PSNP payments lists for submission to FSD and the Office of Finance; (v) provide training to households on investment opportunities; and (vi) assist households prepare business plans. Beneficiary level 340. Beneficiaries and non-beneficiary households participate in public meetings on PSNP and HABP that target PSNP beneficiaries and determine multi-year annual plans. Community members work with DAs on an annual basis to determine priority public works and participate in the consultative meetings to identify viable household investment opportunities. PSNP beneficiaries participate in public works or direct support while those engaged in HABP devise business plans seek support from local financial service providers and carry out these activities over time. Beneficiary and non-beneficiaries both play a key role in holding implementers to account through the KAC and public forums. 95 Annex 7: Financial Management and Disbursement Arrangements ETHIOPIA: Productive Safety Net APL III Introduction 341. A Financial Management (FM) assessment was conducted from May to June 2009 in accordance with the Financial Management Practices Manual issued by the Financial Management Sector Board on November 3, 2005. The objective of the Assessment was to determine whether the participating institutions have adequate financial management systems and related capacity in place to satisfy the World Bank's Operation Policy/Bank Procedure 10.02 with respect to financial management. Under the policy, borrowers (recipients) and project implementation entities are supposed to have and maintain adequate financial management systems, which include budgeting, accounting, internal controls, funds flow, financial reporting and auditing arrangements, to ensure that they can readily provide accurate and timely information regarding project resources and expenditures. These arrangements are deemed acceptable if they: (a) are capable of correctly and completely recording all financial transactions and balances relating to the project resources; (b) can facilitate the preparation of regular, timely and reliable financial statements; (c) safeguard the project's assets; and (d) are subject to auditing arrangements acceptable to IDA. The Assessment also identified key perceived financial management risks that may affect program implementation and developed mitigation measures against such risks. 342. An effective financial management system is vital for the project because of the need to deliver services quickly to a wide variety of stakeholders. The objectives of the project's financial management system are to: (i) ensure that funds are used only for their intended purposes in an efficient and economical way while implementing agreed activities; (ii) enable the preparation of accurate and timely financial reports; (iii) ensure that funds are properly managed and flow smoothly, rapidly, adequately, regularly and predictably to implementing agencies at all levels; and (iv) enable project management to efficiently monitor project implementation. Executive Summary 343. The FM arrangements for the project will mostly use the country's regular public financial management (PFM) system. The Government Chart of Accounts will be used with necessary modifications to accommodate the project specific reporting requirements. These FM arrangements will cover all program funds such as those financed by the Government and all development partners. 344. The strengths of the PFM system are particularly associated with the budget process, compliance with financial regulations and the well-defined accounting system, including the computerized accounting system at the federal and regional levels. These will all positively assist this project. The staff responsible for the project's FM are experienced in IDA financed projects. There are also deficiencies in the system that may negatively impact on the project, such as a shortage of accountants and auditors mainly at woreda-level, delays in reporting, limited focus, and effectiveness of internal audit. The scale of the project and complexities arising from the large number of implementing institutions can also pose implementation challenges. Financial 96 reporting for the Program requires submission and consolidation of timely and accurate reports from a large number of institutions. This is challenging as there are delays in the submission of quarterly financial reports from woredas, some of which are quite remote, and regions. This may also delay the audit. The APL II audit reports noted a number of accounting and internal control weaknesses for which the Government will take actions to rectify and improve. These have been detailed in action plans, the implementation of which will be reviewed by the World Bank and development partners. Design features of APL III that aim to address these weaknesses include: (a) ensuring that there is a clear and revised FM Manual and provision of necessary and adequate training; (b) providing FM support/supervisions/monitoring and close follow-up by MOFED and BOFEDs to lower levels of Government; (c) appointing auditors early; (d) ensuring a more robust use of the interim audit function (transaction-based system and internal control testing); and (e) linking the interim audit with the final financial audit to minimize delays or facilitate early completion of the external audit. 345. The FM risk for the project is rated as High and is expected to reduce to Substantial when mitigating actions are implemented. APL II had an FM manual. This manual will be reviewed and revised so that the current situation in terms of the FM arrangements are well captured and that new developments, such as the procedures pertaining to the asset building component and other relevant aspects, are included. The update to the FM Manual, which outlines the necessary FM arrangements, will be finalized within two months of Effectiveness of the project. Action plans on the various activities to be completed with regards to FM arrangements have been agreed and documented. 346. The FM-related covenants include: submission of Interim Financial Reports (IFRs) for the project for each fiscal quarter within 60 days of the end of the quarter; quarterly submission of reports on the findings noted during the interim audit due within 60 days of the end each fiscal quarter; submission of annual audited financial statements and audit report within 6 months of the end of each fiscal year; and appointment of project auditors within 2 months after Effectiveness. 347. It is the conclusion of the Bank's FM assessment that the FM arrangements meet World Bank requirements as per OP/BP 10.02. It is adequate to provide, with reasonable assurance, accurate and timely information on the status of the project required by the World Bank. Country Issues 348. The Government has been implementing a comprehensive PFM reform, with support from development partners including the World Bank. The main instrument used for effecting these reforms has been the Expenditure Management and Control sub-program (EMCP) of the Government's Civil Service Reform Program (CSRP). This is being supported by the IDA financed-PSCAP and further support is envisaged under the second PBS Project. The latest PFM study for Ethiopia using the Public Expenditure and Financial Accountability (PEFA) PFM performance measurement framework was completed in 2007. The assessment was conducted at the federal and regional levels (covering 7 regions) and two separate reports were issued. The PFM study notes that Ethiopia has made significant progress in strengthening PFM at both federal and regional levels. Improvements were noted in budgeting and accounting reform but 97 the adequacy and quality of budget reporting needed further improvement. The classification of the budget met international standards. The fiscal relations between the federal Government and the regions were transparent. The budget process is well ordered with the existence of a budget calendar generally adhered to, and a budget circular issued to budgetary institutions. Cash flow management may need improvement. Government posts financial information, including the annual budget, on its website and in its official Gazette. However, there is scope for improvement in public access to fiscal information. The study notes that payroll and procurement controls are satisfactory while controls for non-salary expenditure show some weaknesses. Further improvements are required in financial reporting (reducing delays in year and annual financial reporting), internal audit, external audits and scrutiny of public finances. The first phase of the reform (transaction platform) encompassed budgeting, planning, accounting, and information systems. The second phase of the reform, the policy platform, is continuing at the sub-national level with reforms to the block grant mechanisms and a move towards more performance-based budgeting. Government leadership and ownership regarding ongoing PFM reform efforts are high. 349. The PFM study also notes that regional performance of PFM reform varies. SNNP and Tigray show improvement in the overall public finance function and a consequent reduction in fiduciary risk. Other regions, such as Amhara and Oromiya, have also shown significant progress in PFM. Others that are in an earlier stage of investment in PFM have also demonstrated progress. The Government stressed that now almost all regions are on the same levels of PFM reforms. There have been improvements in budgetary transparency, budget preparation, internal audit scrutiny and follow up, timeliness of in-year and annual financial reports, and mutual supportiveness of the federal and regional Auditor Generals. Nonetheless, the report noted that the quality and nature of internal audit needs to be improved. Untimely clearance of suspense accounts and significant delays in producing timely in-year and end-of-year information in some regions are also major weaknesses. There are capacity issues in reviewing annual budgets by the Legislature in as much as supplementary budgets are endorsed by the Parliament. There continue to be audit-related capacity and staffing issues in all the regions. An insufficient number of qualified professional staff at the sub-national level, particularly at woreda-level, and lack of basic office infrastructure is a major challenge. While there have been improvements in the financial discipline associated with government funds, the use of other parallel FM mechanisms, as in some projects financed by development partners, has the potential to increase fiduciary risk. The use of alternative FM arrangements also creates additional workload in areas where capacity is already stretched. 350. The Government noted that since the last diagnostic work carried out in 2007 (PEFA), the basis upon which much of the above analysis was conducted, there have been significant improvements in the PFM. Further diagnostic works will be conducted by the World Bank and partners in collaboration with the Government. This will help all concerned parties to secure updated information as to PFM status of the country. 98 Table 10: Risk Assessment and Mitigation Conditions Residual Risk Risk Mitigating Measures Incorporated into for Risk Risk Rating Project Design Effectiveness Rating (Y/N)? Inherent Risk Country Level S This is being addressed by the ongoing Civil M N Risk arises from weak Service Reform Program supported by PSCAP capacity, including and PBS II. turnover and shortage of qualified accountants and auditors. Entity Level H The current FM manual stipulates the respective S N There are a number of responsibilities of players in the program. This players implementing the manual will be revised to accommodate new project. Monitoring and aspects of APL III as well as to address enforcement of financial weaknesses noted under APL II. This helps regulations and rules of the remove latent overlaps and confusion and also project needs improvement. helps deal with key issues of monitoring and coordination. In addition, there is experience in managing World Bank financed projects within MOFED & MOARD. MOFED's new Project Coordinating Unit will facilitate and monitor FM aspects of the project, among their other duties. Project Level H As above. S N The project is complex Capacity building trainings/workshops to project with the involvement of a accountants and relevant internal auditors will be large number of dispersed planned and conducted. entities with a mix of large and small amounts of disbursement. Inherent risk H S Control Risk Budgeting S Variance analysis training should be offered by M N Wide number of MOFED and development partners will follow- implementing entities may up. A review of the draft budget will identify this lead to delays in problem before it is too late. The review will be appropriate budgeting done every quarter. MOFED will issue an annual process. Lack of budget calendar circular and ensure that all satisfactory variance project participants have enough time to produce analysis to monitor budget budgets. The budget codes will be updated and implementation may also the use of subsidiary ledgers, where necessary, be a risk. has been agreed. Accounting M The current FM manual will be reviewed and L N Cutoff and some updated to outline clearly the cut-off procedures accounting problems were and areas where accounting processes are not noted in the past. There is explicit, as well as to address FM issues also the risk of delays in pertaining to additional components like the recording of transactions. support to HABP. Regular training to regional and woreda staff and close follow-up of woreda accounts will help with timely recording of transactions. Regular supervision and close 99 follow-up will be made by MOFED and BOFEDs. Internal Control S In addition to the risk mitigation measures M N Internal audit function is mention in "Accounting" above, internal auditors weak. Further, it was noted of Regions/woredas will review the project at that satisfactory action has least once per year. It is expected that the capacity not being taken on issued building component will enhance the management letter points. effectiveness of the current regional/BOFED Cash and in-kind support internal audit departments. The interim audit may not reach the target focusing on internal control and transfer of funds beneficiaries and resources (including food and HABP capacity building items) to beneficiaries will be conducted thereby helping to ensure that funds are used for the purposes intended. The need to act on management letter weaknesses was agreed during Negotiation and the new Channel 1 Donor Coordination Unit at MOFED will be involved in facilitating actions and follow-up at regional level. The strengthened use of PASS (payroll system) in tracking beneficiaries would be given attention to ensure that the target beneficiaries are being addressed. Funds flow S Receiving entities will be asked to report to M N There may be delays in MOFED when they receive funds, which will be flow of funds to the lowest summarized every six months to determine implementation levels. delays. Based on the outcomes, a discussion will be held with MOFED to review and update the service standards for funds flow for the project. Close supervision by development partners will continue. The interim audit is to continue focusing on internal control and transfer of funds. Financial Reporting S All woredas will be asked to submit monthly M N Delays in reporting, which reports to BOFEDs, which will consolidate was noted in the past, has reports every quarter. Regular training on the improved recently but preparation on IFRs will be conducted by could worsen as a result of MOFED. Close follow-up by MOFED and the the introduction of HABP development partners will be made. and pastoral areas, which have weak capacity. Auditing S Follow-up on timely recruitment of auditors and M N Because of the involvement timely closure of accounts will be made. There of a number of will be a direct link between the interim audit and implementing entities, the financial audit whereby the interim audit will delays in the preparation of supplement and facilitate the process of the final financial accounts for the financial statement audit. consolidation necessary for submission of external audit reports could be encountered. Control Risk S M Total Project FM Risk H S 100 351. In the view of the above table, the inherent risk of the project is Substantial while the control risk is Moderate. The overall financial management risk rating of the project (APL III) continues to remain Substantial as in APL II. Strengths and Weaknesses of Proposed Financial Management System 352. As noted in various reports, the country's discipline in executing budget and compliance with existing government regulations are major strengths in implementing the project. The Government's existing arrangements are already being used in a number of projects, including PSNP APL II and PBS. 353. The main weaknesses for the FM arrangements continues to be turnover and shortage of qualified accountants and auditors mainly at woreda-level, delays in reporting, limited attention of the internal audit, for various reasons combined with the system's limited capacity to effectively function. The long process involved in producing reports from woredas to regions, and from regions to MOFED may delay the timely submission of financial reports to the development partners. Delayed submission of reports (both audited and unaudited) is, however, significantly improving, although the need for constant follow up remains. 354. Lessons from APL II: Audit reports noted weaknesses in the accounting and internal control areas. These include: (i) repeated problems in relation to supporting payments with adequate documents; (ii) cutoff problems/errors; (iii) posting/coding errors; (iv) issues with cash certificates and bank reconciliation; (v) some control weaknesses in connection with payroll payments to beneficiaries; (vi) inconsistencies between woreda and regional reports; (vii) idle resources, (viii) lack of ledgers and transaction registers including not being up-to-date; and (ix) lack of analysis and follow up of long outstanding debtors and creditors. MOFED is taking steps to address these weaknesses but much remains to be resolved. These efforts have been detailed in action plans and the World Bank and development partners will monitor the implementation of the agreed actions. In addition, MOFED and BOFEDs should intensify efforts to support the WOFEDs. This will include regular field visits to WOFEDs and robust reviews and checks on the reports. Ongoing trainings would also address accounting problems noted. Financial Management Implementing Entities 355. At the federal level, MOFED will continue to be responsible for the overall financial management of the project. This includes, but is not limited to, the management of the designated and the pooled Birr account, the transferring funds to BOFEDs and MOARD (based on the direction of MOARD), the responsibility for producing regular financial reports and facilitating the annual audit of the project account. MOFED will ensure that acceptable financial management systems are in place and are well documented in FM manuals. The MOARD is responsible for oversight and coordination of the project. It will also be responsible for the funds transferred to it from MOFED. 356. At the regional level, BOFEDs continue to be responsible for ensuring that a suitable accounting system covering both regional and woreda levels is established. BOFEDs will continue to collect and aggregate all financial data and information from BOARD and woredas 101 on the Project, review the effective use of accounting procedures by woredas, and provide technical support and assistance to them. Each region will prepare quarterly and annual reports, which will be sent to the federal level. BOARDs are responsible for the management of the funds transferred to them for implementation. 357. At the woreda level, a suitable accounting system is established for the disbursement of funds for activities financed under the PSNP including the HABP component. The records of funds utilized will be maintained in accordance with sound accounting practices that are capable of generating accurate and timely information for verification. Woreda accounting personnel will receive training on how to maintain accurate accounts for the funds. In case woredas face difficulties in accounting or handling financial records, the region will provide timely assistance and training to resolve such difficulties. WOFEDs (i) ensures that the budgets for the PSNP are received in a timely manner at the woreda level to guarantee smooth implementation of approved plans and activities; (ii) undertakes timely PSNP (and RF) payments for beneficiary households, supervising personnel, and the purchase of relevant equipment and materials; and (iii) exercises necessary fiduciary controls and reports on fund utilization to Regional BOFEDs. 358. MFIs at regional level or below and RuSACCOs at woreda or kebele levels will benefit from the capacity building activities financed under the HABP. They will not directly receive development partner financing from BOFEDs or WOFEDs but they will benefit from the capacity building activities of the program. 359. There should be strong coordination and communication between MOARD and MOFED in implementing the project. A number of committees like the Joint Strategic Oversight Committee, Regional Steering Committees, and Technical Committees at federal and regional levels will help foster linkages between Government implementing agencies and development partners. The details of overall implementing arrangements are stated in the Annex 6. Budgeting 360. The overall budgetary arrangements of APL II will continue to prevail for APL III. The Ethiopian budget system is complex, reflecting the fiscal decentralization structure. The budget is processed at federal, regional, zonal (in some regions), and woreda and municipality levels. The federal budgeting process usually starts with the three-year Macroeconomic Fiscal Framework, which is the basis for issuing the budget preparation note to the Budgetary Institutions. Based on the budget manual, the Budgetary Institutions prepare their budgets in line with the budget ceilings and submit these to MOFED within six weeks following the budget call. The budgets are reviewed at first by MOFED and then by the Council of Ministers. The final recommended draft federal budget is sent to Parliament in early June and is expected to be cleared at the latest by the end of the EFY. 361. The PSNP is on budget and the project budget will be proclaimed in the budgets of the Federal Government in the name of MOARD. 362. Each region should prepare a consolidated PSNP work plan and budget for all components for each budget year based on inputs from lower levels and submits the same to MOARD, which 102 shall be reviewed/updated/consolidated. Finally, a consolidated budget will be submitted to MOFED that should be broken down by appropriate project category or components along with quarterly classifications. Based on the Program budget, a detailed and comprehensive fund transfer schedule by Region and woreda will be prepared and disseminated by MOARD to all relevant stakeholders in July of each year. 363. Although the overall budgeting is satisfactory, it was observed that budgets are prepared and reported on a lump sum basis with no breakdown, without proper identification of the category such as Public Works, Direct Support, Capital and Administrative. Furthermore, there was limited use of variance analysis and budget as a management control tool. The Government will work on these to improve the situation as agreed in the plan of action stated below. 364. The FM Manual of PSNP will be reviewed and revised, as appropriate, to reflect the current situations in terms of budgeting, accounting, fund flows, internal controls, financial reporting, and auditing issues and to include new developments like the procedures pertaining to the HABP component and other relevant aspects. 365. The Government's functional classification system follows a tiered structure (e.g., programs, sub-programs, projects). The classification system allows for budgeting and accounting of funds by region and by source of financing. The economic classification coding structure allows for a breakdown of expenditures by types of expenditure (e.g., salaries, fuel, educational supplies, etc.). The classification system is contained in the Government's Chart of Accounts. The classification structure thus allows for the project components to be specifically identified within the functional classification system. Activities would be uniquely identified within the Government's Chart of Accounts, thereby facilitating budgeting, accounting and financial reporting for the project funds, as is the current case for APL II. However, agreement was reached to update the current budget codes (account codes) as a result of the new features of APL III and this would be reflected in the revised FM Manual. Accounting 366. As in APL II, the Government's accounting policies and procedures are expected to be largely used for the accounting of the project. The Ethiopian Government follows a double entry bookkeeping system and modified cash basis of accounting. This is documented in the Government's Accounting Manual. This has been implemented at the federal-level and in many regions. The main elements of the accounting reform are the adoption of: (i) comprehensive Chart of Accounts consistent with the budget classification; (ii) a system of ledgers accommodating all types of accounts (including transfers, assets, liabilities and fund balance in addition to revenues and expenditures); (iii) double entry book-keeping (thus, self balancing set of accounts); (iv) a system of control of budgetary commitments (recording commitments as well as actual payments); (v) modified cash basis transaction accounting; and (vi) revised monthly report formats to accommodate double entry bookkeeping and commitment control and permit better cash control. The Government's Accounting Manual provides detailed information on the major accounting procedures. On the basis of the Government Accounting Manual, the APL II FM Manual was developed. This FM manual is expected to be used for APL III. However, as 103 noted above, the FM manual will be reviewed and revised, as appropriate, to include the new HABP component, food resources, aid in-kind accounting, and strengthen other relevant aspects. 367. The existing accounting software (IBEX) and the support program (MS EXCEL) at all levels will be maintained in order to allow production of all accounting and financial data required. The Chart of Accounts (budget codes) described above will facilitate the preparation of relevant quarterly and financial statement including information on the total project expenditures. The linkages of the accounting software with the beneficiary payroll software (PASS) will be assessed and reviewed through the interim audit. 368. There are weaknesses and problems observed in the accounting and internal control areas, which need attention, as discussed in the "Strengths and Weaknesses" section above. MOFED is taking steps to address these weaknesses but still much remains to be done. MOFED and BOFEDs should intensify efforts to support the WOFEDs. Regular field visits to WOFEDs and robust reviews and checks on the reports should be continued. Ongoing capacity building trainings would also address the accounting problems noted. 369. The current project has recruited and maintained a significant number of accountants and cashiers at federal, regional, and woreda levels. It is expected that the current staffing levels will continue to operate for APL III. However, reviews are being undertaken with MOFED and MOARD to assess the FM staffing arrangements both in terms of quantity and quality so that optimal staffing arrangement can be determined. MOFED will also review and, as appropriate, update the current job descriptions of FM staff. Internal Control and Internal Auditing 370. Internal control comprises the whole system of control, financial or otherwise, established by management in order to: (i) carry out the project activities in an orderly and efficient manner; (ii) ensure adherence to policies and procedures; (iii) ensure maintenance of complete and accurate accounting records; and (iv) safeguard the assets of the project. Regular government systems and procedures would be followed, including those relating to authorization, recording and custody controls. 371. The project's internal controls, including processes for recording and safeguarding of assets, are also documented in the FM Manual. These procedures will continue to be applicable. However, as discussed in above, the FM Manual will be reviewed and revised. 372. Internal audit (post audit reviews) will be carried out by the Internal Audit Departments of the respective entities. MOFED, BOFEDs and WOFEDs have internal audit departments that perform this function, including an assessment of whether the budget utilization is in line with the intended purposes. Furthermore, there are Inspection Departments in MOFED and BOFEDs that (i) ensure the quality of internal audits in the Ministries at federal level and Sector Bureaus at region level; (ii) follow-up on the audit recommendations noted by audit reports at different Ministries and Sector Bureaus; and (iii) provide training and improving manuals, among other responsibilities. The staffing of Departments varies from region to region. However, the lack of effective and value adding internal audit function at all levels was noted in APL II. In general, 104 PFM analytic work notes that internal audit is weak, and there is a need for significant improvements. This will need to be improved for APL III as noted in the action plans. Funds Flow and Disbursement Arrangements 373. The flow of funds and disbursement arrangements for APL II will continue to be applicable for APL III, subject to minor modifications, as shown below. Segregated US Dollar Designated Accounts will be opened at the National Bank of Ethiopia for each of the development partners. Then, the existing pooled local currency account at federal level and separate bank accounts at regional and woreda level will be maintained for disbursement purposes in accordance with the project objective. 374. The project will use report based disbursement arrangements. After the initial advance, which will be the equivalent of six months budget/forecast, to be deposited into the project's US Dollar Designated Account, MOFED will have the funds of the Designated Account replenished on the basis of Withdrawal Applications. These are submitted along with the quarterly IFR, which is prepared for actual expenditures and includes the cash forecast required for the next six months. The World Bank in its own capacity and on behalf of other development partners for Bank-administered Multi-Donor Trust Funds (MDTFs) will then deposit their share of financing to a Designated Account that Government has designated for that purpose in the National Bank of Ethiopia. MOFED will then draw resources from this account to its ETB account opened at MOFED, which will use for activities under its control. MOFED will also disburse the remaining funds as per instructions of MOARD and disbursement plans to BOFED and MOARD. BOFED will disburse as requested to WOFED and BOARD. WOFEDs will in turn pay beneficiaries in collaboration with WOARD. WOFEDs will notify the BOFEDs when they receive funds. These reports will be summarized every six months by BOFEDs and reported to MOFED to determine if there are delays in fund transfer. Based on the outcomes, necessary action would be taken to facilitate timely fund transfers to woredas. 375. The Government and development partners, including the World Bank, will agree on the annual budget and work plans. Definitive proportions of financing between the Government and the development partners will be established each quarter. The World Bank Task Team Leader (TTL) will advise the World Bank's Loan Department of the share of financing to be disbursed by the World Bank for the project by linking it to the projected cash flow. 376. The project will follow advances method using Designated Accounts as outlined above as well as the Reimbursement, Direct Payment, and Special Commitment methods. 377. The following chart illustrates the funds flow system: 105 DPs Other DPs MDTF for IDA PSNP Separate US$ segregated Designated Accounts at NBE MOARD Pooled Birr account at MOFED BOARD BOFEDs WOARD WOFED beneficiaries/ local suppliers Fund flow Report flow Financial Reporting 378. The project will continue quarterly preparation of the Consolidated Interim unaudited Financial Report (IFR). This will be submitted to the World Bank within 60 days of the end of the quarter. The format and the content, which are consistent with the World Bank's standards, have been agreed with MOFED and MOARD as well as development partners and are documented in the minutes of Negotiation. 379. The format of IFRs will be produced from the existing government accounting system. The IFR will include: (a) a statement of sources and uses of funds, opening and closing balances for the quarter and cumulative; (b) statement of uses of fund that shows actual expenditures, which are appropriately classified by main project activities (categories, components and sub- 106 components) and include an actual versus budget comparisons for the quarter and cumulative; (c) a statement of cash forecast/ requirement- for six months (d) notes and explanations; (e) a statement on the movement of project's Designated Account, including opening and closing balances and the movements (inflows and outflows); and (f) other supporting schedules and documents. The annual financial statements will adopt the same format as the quarterly reports and may also include other issues. However, the annual financial statements do not need to include a statement of cash forecast/ requirement. 380. Although there have been significant improvements as a result of the steps taken by MOFED and MOARD to enhance the timeliness of reporting and auditing, the program continues to experience some delays in the completion and submission of reports which are of the desired quality. The reports coming from woredas and regions to MOFED are usually delayed and are sometimes incomplete with inadequate documentation. Training, adequate support and follow up of woreda staff on the preparation of reports and stringent review of woreda reports by MOFED and BOFED could help alleviate this problem. 381. Financial reports are sent from WOFEDs on a monthly basis to the BOFEDs. The BOFEDs will ensure the reports received from the lower level are complete and meet all expected standards. After performing this quality control, BOFED will consolidate and submit quarterly financial reports to MOFED. MOFED will in turn check, consolidate, and submit quarterly IFRs to development partners within 60 days of the end of the quarter. Auditing 382. MOFED will have the project financial statements audited by the Office of the Federal Auditor General (OFAG) or an auditor assigned by OFAG. Should OFAG decide to assign an independent audit firm to handle the financial audit, efforts will be made to ensure that the auditors are recruited or appointed within two months of effectiveness. 383. The auditor will submit the audit report in a form and content satisfactory to the World Bank within six months of the end of the Ethiopian Fiscal Year. As part of this annual audit, the same audit firm will conduct a review of the financial transactions of the program at woreda level. This is what is referred to as an interim audit a process similar to the previous "roving audit". This interim audit will not be an end on its own but will supplement the finalization of the final financial audit and is an integral part of it. 384. The audit Terms of Reference has been revised to include primarily the link between the interim and financial final audits and other relevant issues. Some of the issues incorporated include: (i) the need for robust sampling techniques; (ii) examination of systems and controls; (iii) testing of a sample financial transactions selected on the basis of the auditor's evaluation of internal controls; (iv) verifications of services financed by the program including payments to beneficiaries as documented in payrolls; (v) reconciliation between the beneficiaries data as per the payroll system and the accounting system; and (vi) provide an opinion as to whether funds are being used for the intended purposes. 107 385. Reports summarizing the findings of the interim audit will be submitted on a quarterly basis to the World Bank and development partners within 60 days of the end of the quarter. The auditor will use, among other tools, the results of the interim audit in forming an opinion on the Project Financial Statements. The auditor will plan and perform the interim audit in such a manner that it will add value and reduce the time it takes to produce the final financial audit report. The findings of the interim audit reports will be followed up by the Government, World Bank and other development partners. The Audit Terms of Reference has been agreed to at Negotiations. 386. Lessons from APL II indicate that there is significant improvement in the timely delivery of financial audit reports. However, there are still delays in conducting the roving audits as agreed. The latest audit report submitted is qualified. The qualification points are: (i) lack of supporting documents and inadequate explanation of a prior year adjustment; and (ii) lack of cash count certificates, significant retention of cash with woreda cashiers, and cut-off errors. In addition, there are a number of internal control weaknesses noted in the management letter. These include issues such as: (i) inadequate supporting documents for payments; (ii) cut-off problems/errors; (iii) posting/coding errors; (iv) problems concerning cash certificates and bank reconciliation; (v) control weaknesses with payroll payments to beneficiaries; (vi) inconsistencies between woreda and regional reports; (vii) the repeated existence of idle resources; (viii) occasional lack of ledgers and transaction registers including not being up to date; and (ix) lack of analysis and follow up of long outstanding debtors and creditors. MOFED in collaboration with MOARD has issued a plan of action to address these weaknesses. Development partners including the World Bank will monitor implementation of the agreed actions. 387. The movement of commodities such as food and other resources that are either purchased or donated in-kind from various development partners for the Program will be audited. Relevant reports include the quarterly Commodity Flow and Status Report. MOARD will be responsible for conducting this audit and following-up on findings thereof. MOARD will submit this audit report within six months of the end of the fiscal year ending on July 7. The Commodity Audit Terms of Reference have been agreed at Negotiations. Financial Covenants 388. MOFED will submit the audited Program accounts to the World Bank 6 months after the end of each fiscal year, which ends on July 7. The audited financial statement will include all sources of funds for the Program, including other development partners and the Government. In addition, reports on the findings noted during the interim audit will be submitted quarterly to the World Bank and development partners within 60 days of the end of the quarter. 389. MOFED will submit quarterly IFRs to the World Bank 60 days after the end of each quarter period. 390. MOARD will submit annual commodity audit report within 6 months of the end of the fiscal year ending of July 7 of each year. 108 Financial Management Action Plan 391. The action plan below indicates the actions to be taken for the project to strengthen its financial management system and the dates that they are due to be completed by. Action Date due by Responsible 1 Review and revise the FM Manual to reflect the current situation in 2 months after MOFED/MOARD terms of budgeting, accounting, fund flows, internal controls, Effectiveness financial reporting, and auditing issues and to include new developments such as the procedures pertaining to the HABP component, Chart of Accounts (budget codes), and other relevant aspects. 2 Trainings on the FM manual with particular emphasis on budget Ongoing MOFED/MOARD preparation and variance analysis; all relevant issues on accounting, reporting and fund flow arrangements will be provided. 3 If OFAG decides to assign an independent audit firm to handle the Within two months OFAG/MOFED financial and interim audit then the recruitment/appointment of of effectiveness. auditors at early stages of the project. 4 Ongoing trainings will be conducted, including budget analysis Ongoing MOFED training, IFR preparation training, etc. 5 MOFED and BOFED should conduct regular field visit to support Ongoing MOFED as well as monitor the performances of WOFEDs. 6 Increased engagements of Internal Audits at all levels to identify Ongoing MOARD/MOFED control weaknesses early. In this respect, workshops or capacity building activities/training will be conducted for auditors at regional and woreda level. 7 MOFED and BOFEDs should undertake adequate robust reviews Ongoing MOFED and checks on the reports submitted to them from Regions and woredas, respectively. MOFED/MOARD should take action on woredas that delay reports. 8 Implement action plans to address weaknesses noted in the audit As per the MOFED/MOARD reports of APL II. deadlines set in the Action plans 9 Close supervision by the World Bank and development partners. Ongoing World Bank and development partners Supervision Plan 392. The project will be subject to a minimum of two annual supervision missions to be conducted jointly with development partners. Supervision activities will include: review of quarterly financial management reports; review of annual audited financial statements, and timely follow up of any resulting issues; transaction review; participation in project supervision missions as appropriate; and updating the FM rating in the Implementation Status Report (ISR). 109 Annex 8: Procurement Arrangements ETHIOPIA: Productive Safety Net APL III 393. Procurement Environment: A new public procurement law: The Federal Government of Ethiopia Procurement and Property Administration Proclamation No. 649/2009 was enacted by parliament on July 7, 2009 repealing the existing public procurement Proclamation No. 430/2005, "Determining Procedures of Public Procurement and Establishing its Supervisory Agency Proclamation of the Federal Democratic Republic Government of Ethiopia", dated January 12, 2005. The World Bank will assess the new procurement regime and based on the new procurement proclamation make any necessary modifications to the Recipient's competitive bidding procedures to make the procedures acceptable for use in IDA-financed procurement. The Public Procurement Agency (PPA) which was established under the old law had achieved the following: (a) preparation and distribution of National Standard Bidding Documents (SBDs) for the procurement of goods, works and non-consultant services, as well as for the selection of consultants, including development of guidelines for their use; (b) organization of regular public procurement workshops including training of trainers; (c) development of a capacity building strategy (currently in progress); and (d) preparation of national public procurement manual and training modules to facilitate further training. The World Bank has reviewed the bidding documents for procurement of goods and works and found them to be generally acceptable. The World Bank is now supporting, under the PBS II project, the revision of these documents in view of the new law. 394. At the Regional and woreda levels, public procurement is governed by proclamations enacted by the respective regional governments derived from the federal model law. Currently, all nine regions and the city administration of Addis Ababa have public procurement proclamations adopted from the old procurement law while Dire Dawa city administration applies the federal proclamation and directives. However, in general the proclamations ratified by regions do not establish independent procurement oversight agencies (regional PPAs). At operational level, Tender Committees are established in most procuring entities but the majority of entities have yet to create functional procurement management units. 395. The "Ethiopia 2002 Country Procurement Assessment Report" (CPAR) identified weaknesses in the country procurement system and recommended actions to address these areas. The Government has implemented many of the CPAR recommendations but challenges remain in the areas of: coordination of procurement reforms, shortage of qualified procurement staff, institutional structures for procurement management, weak institutional capacity, appeals mechanisms for addressing stakeholder complaints, systematic procurement monitoring and evaluation of performance, and capacity building. The new federal procurement law addresses some of these shortcomings but the Bank will need to assess the new procurement regime when the new directives and revised SBDs are available and the relevant institutions are in place. General 396. The Government has committed to ensure that procurement under the Productive Safety Net Program to be financed through funding contributed by IDA as well as by other development partners, whether through a MDTF administered by the Bank or directly into the Government's 110 pooled birr account at the federal level, will be carried out in accordance with the World Bank's "Guidelines: Procurement under IBRD Loans and IDA Credits" dated May 2004; revised October, 2006; "Guidelines: Selection and Employment of Consultants by World Bank Borrowers" dated May 2004; revised October, 2006; and the provisions stipulated in the IDA Financing Agreement, MDTF Grant Agreements, and bilateral grant agreements. 397. The relevant directives and national SBDs have not yet been amended based on the new procurement proclamation. The competitive bidding procedures of the Federal Government have been reviewed by the Bank. Based on this review, contracts that will not be procured under International Competitive Bidding (ICB) and consulting assignments that will not involve international competition may follow the Recipient's procurement procedures, subject to the following additional procedures: (i) the Recipient's standard bid documents for procurement of goods and works shall be used; (ii) if pre-qualification is used, the World Bank's standard prequalification document shall be used; (iii) margin of preference shall not be applicable; (iv) bidders shall be given a minimum of 30 days to submit bids from the date of availability of the bidding documents; (v) use of merit points for evaluation of bids shall not be allowed; (vi) foreign bidders shall not be excluded from participation; and (vii) the results of evaluation and award of contract shall be made public. Procurements at the regional level would generally involve contracts procured through National Competitive Bidding (NCB) and Shopping procedures whilst the small purchases at the woreda level would invariably be through the shopping procedures. The NCB and Shopping procedures at the regional and woreda levels would be carried out in accordance with regional procurement proclamations and using national SBDs acceptable to the World Bank. 398. The above mentioned modifications are based on the existing procurement procedures and do not take account of the new procurement law that was approved by Parliament on July 07, 2009. The World Bank will review the procurement law when the directives and the SBDs are made available to it. For procurement under APL III, the Recipient would continue to follow the procedures under the repealed Proclamation 430/2005 with the above modifications. When the World Bank has issued new modifications, if any, to apply to the new procurement regime under Proclamation 649/2009, the new modifications would be incorporated to into a revised PIM for PSNP and would become applicable to procurement under PSNP and HABP after the Bank's approval of the revised PSNP PIM and HABP PIM. 399. In accordance with para.1.14(e) of the Procurement Guidelines, each bidding document and contract financed out of the proceeds of the Financing shall provide that: (i) the bidders, suppliers, contractors and subcontractors shall permit the Association, at its request, to inspect their accounts and records relating to the bid submission and performance of the contract, and to have said accounts and records audited by auditors appointed by the association; and (ii) the deliberate and material violation by the bidder, supplier, contractor or subcontractor of such provision may amount to an obstructive practice as defined in paragraph 1.14(a)(v) of the Procurement Guidelines. 400. Under APL III, the Recipient is obliged to continue to follow the NCB modifications to its procurement procedures as stated above notwithstanding the adoption of the new procurement proclamation. The World Bank will review the new procurement proclamation together with the 111 new directives and revised SBDs in order to determine the modifications needed for Ethiopia's revised NCB procedures to be acceptable for use in World Bank-financed contracts. The PIM for the Project will be revised as necessary to capture the World Bank's comments. 401. The various items under different expenditure categories are described in general below. For each contract to be financed by the Grant/Credit: (i) the different procurement methods or consultant selection methods; (ii) the need for pre-qualification; (iii) estimated costs; (iv) prior review requirements; and (v) timeframe are agreed between the Borrower and the World Bank in the Procurement Plan. The Procurement Plan will be updated at least annually, or as required, to reflect the actual project implementation needs and improvements in institutional capacity. 402. Procurement under Component 1 Safety Net Grants: This component makes cash or in-kind transfers to chronically food insecure families in either of two forms: (i) through payment for unskilled labor provided by adults for public works; or (ii) through direct support to labor poor households that cannot undertake public works due to age, sickness, pregnancy, etc. The public works focus is on soil and water conservation works, with significant investments also made in rural roads, irrigation, and social infrastructure. Regions and woredas will receive money for payment for labor, administration, as well as for capital goods. Procurement will be conducted to provide public works inputs including, tools and materials like cement, sand, aggregates, steel, etc. The regions will as far as practicable consolidate requirements into packages valued at more than US$50,000 equivalent to be procured through NCB procedures either for stocking and subsequent delivery or for direct delivery to the beneficiaries. Purchases for materials valued at below US$50,000 equivalent would be conducted either at the regional or woreda level using Shopping procedures in accordance with the provisions of paragraph 3.5 of the Procurement Guidelines. This will involve comparing price quotations obtained from several suppliers with a minimum of three, to assure competitive prices. The request for quotations shall indicate the description and quantity of the goods as well as the desired delivery time and place. The PIM shall include a sample Request for Quotations (RFQ) for use by the regions and woredas. 403. Procurement of Goods: Other than capital goods for the public works program, goods and equipment to be procured under the Project include: vehicles, generators, computers, motorcycles, Geographic Information System equipment, small furniture and office equipment, etc needed for institutional support to PSNP and to build capacity for the Household Asset Building Program (HABP). A list of the items together with a cost estimate and a preliminary procurement plan were completed at Appraisal. Goods will be packaged whenever possible in packages higher than US$500,000 and tendered under ICB following the procedures described under Section 2 of the Procurement Guidelines. In addition, the Bank's standard bidding documents would be used for all ICB contracts. Contracts for goods procurement under the monetary value of US$500,000 equivalent will be tendered through NCB. Goods procurement contracts below the monetary threshold of US$50,000 equivalent will be awarded under shopping. Procurement of goods other than through ICB would use the national procedures and SBDs as agreed with and deemed satisfactory to the World Bank. Direct contracting will be used where it is to the benefit of the project and in accordance with the provisions of paragraph 3.6 of the Bank's Procurement Guidelines. 112 404. Procurement of Food: DRMFSS will undertake the procurement of bulk food items. The current practice is that the bulk food grain procurement under APL II is outsourced to WFP. The procurement capacity of DRMFSS will be strengthened to enable it to undertake procurement of food grain by itself starting in 2010. The PSNP will then distribute both food and grants to targeted beneficiaries. Food procurement will be undertaken only when there is a defined need. The procurement of food would follow the Modified ICB method for the Procurement of Commodities in accordance with provisions of para. 2.68 of the Bank's Procurement Guidelines. The DRMFSS will undertake annual prequalification of food Suppliers and a list of prequalified bidders will be maintained to whom periodic invitations are issued whenever food procurement becomes necessary during the year. 405. Procurement of Works: Under the Project there will be small value works contracts (such as construction of small food stores, culverts, small bridges, etc) that shall be planned and implemented at the woreda level as part of the public works program. The procurement of such small works will be implemented using the Regional procurement systems. For simple civil works of small value procured through shopping procedures, the request for quotation will indicate the specifications of works as well as the completion time and the contract award will be based on comparing price quotations from several qualified contractors, with a minimum of three, to ensure competition. When the value of the contract of such civil works exceeds the Shopping threshold and when procured through NCB procedures, the national SBDs issued by the Federal PPA and acceptable to the World Bank will be used. Direct contracting shall be used where it is to the benefit of the project and in accordance with paragraph 3.6 of the Bank's Procurement Guidelines. 406. Procurement of non-consulting Services: Depending on the nature of the services, procurement of non-consulting services, such as transport, will follow procurement procedures similar to those stipulated for the procurement of goods. NCB procedures acceptable to the Bank would be used for contracts above an estimated monetary amount of US$50,000. Contracts valued at less than US$50,000 equivalent would use shopping procedures in accordance with the provisions of paragraph 3.5 of the Bank's procurement Guidelines or in accordance with established Commercial practices of common use in Ethiopia acceptable to the Bank. 407. Consulting Services: The project will make use of consultant services for studies, baseline surveys, impact assessment, automation of MIS and other services. Contracts above US$100,000 will be awarded through the use of the Quality and Cost-Based Selection (QCBS) described under Sections 2 of the Consultant Guidelines. Consulting Services for audit and other contracts of a standard or routine nature may be procured under the Least Cost Selection method (LCS) described under Section 3.6 of World Bank Consultants Guidelines. Consulting assignments costing less than US$100,000 may be procured through the Selection Based on Qualifications method: shortlists of consultants for services estimated to cost less than US$200,000 equivalent per contract may be composed entirely of national consultants in accordance with the provisions of paragraph 2.7 of the Consultant Guidelines. Contracts for individual consultants will be advertised on national papers of wide circulation to allow for the drafting of shortlists. Single- source selection may be used where it is to the benefit of the project in accordance with the provisions of paragraphs 3.9 or 5.4 of the Bank's Consultant Guidelines. 113 408. Operating Costs: Expenditures made for operational costs such as fuel and stationery, cost of operation and maintenance of equipment, communication charges, transportation costs, and travel allowances to carry out field supervision will follow Ethiopian Government practices that have been found acceptable to the Bank and included in the PIM. 409. Training and workshops will be based on capacity building needs. Venues for workshops and training and purchase of materials will be done on the basis of at least three quotations. The selection of institutions for specialized training will be done on the basis of quality and therefore use the Based on Qualifications method. Annual training plans and budget shall be prepared and approved by the World Bank in advance of the training. 410. Margin of Preference for Domestic Goods: In accordance with paragraph 2.55 and 2.56 of the Procurement Guidelines the borrower may grant a margin of preference of 15% in the evaluation of bids under ICB procedures to bids offering certain goods produced in the Country of the Borrower, when compared to bids offering such goods produced elsewhere. Assessment of the Agency's Procurement Capacity 411. Procurement Capacity Assessment for PSNP APL III has mainly relied on the report the "Assessment of Procurements Conducted by Regions and Woredas for PSNP, January 2009". The World Bank had also employed a consultant to conduct an independent assessment of regional and city administrations procurement management capacity, which was submitted in March 2009. In addition to these recent documents, the World Bank Country Office's Procurement Specialist conducted a procurement capacity assessment on two representative PSNP implementing agencies: FSCD and the Oromiya Food Security, Disaster Prevention & Preparedness Commission. Their procurement capacity was assessed from 27-30 April 2009. In general, these reports found that although there are encouraging efforts at all levels to recruit procurement staff and introduce public procurement procedures, there are major procurement performance limitations in all Ethiopian public procurement implementing entities. These reports and other concurrent procurement capacity assessments indicate that there are significant capacity limitations and systemic constraints in the Ethiopian public procurement environment. In particular, due to its highly decentralized implementation arrangements and high turnover of procurement staff common in public organizations, the procurement risk of the proposed project is rated high risk. 412. The procurement assessment of the two implementing agencies included a review of the following aspects: (a) Legal Aspects and Procurement Practices; (b) Procurement Cycle Management; (c) Organization and Functions; (d) Support and Control Systems; (e) Records Keeping; (f) Staffing; and (g) General Procurement Environment. 114 413. The general procurement capacity situation of the two assessed agencies, as revealed through the above parameters, is weak and procurement tasks need streamlining. The agencies' procurement processes, organizations and staffing, quality of internal administrative practices, staff capacity, and completeness of procurement records have been found weak. The major findings of the procurement assessment of procurements conducted by the Regions and woredas for PSNP confirm this conclusion. 414. Institutional Setup for Procurement: Ethiopia has had several years of experience with multi-development partner collaboration in several sectors related to the present project. In addition, the PSNP has been under implementation for five years (APL I and II). The new project therefore will build on previous institutional experience. The Ministry of Agriculture and Rural Development (MOARD) is responsible for oversight and coordination of the PSNP through the Food Security Coordination Directorate (FSCD) (see below). The Ministry provides necessary technical support for PSNP planning and implementation. The Program will be largely implemented through decentralized arrangements. The more decentralized implementation arrangements coupled with the general public procurement performance gaps in the country and high procurement staff turnover in all public organizations may undermine project procurement implementation unless dedicated capacity building actions are designed and implemented at early stages of the Program. 415. As the result of the recent Ethiopian Civil Services Reform and Business Process Reengineering (BPR) study recommendations, the three working units of MOARD (Disaster Prevention & Preparedness Agency, Food Security & Coordination Bureau and Early Warning part of MOARD) are now merged under the State Minister for Disaster Risk Management & Food Security Sector (DRMFSS) of MOARD. At the federal level, the FSCD and the Early Warning and Response Directorate (EWRD) are responsible for the majority of procurement of APL III. As appropriate, NRMD and AED will undertake procurement. 416. The procurement capacity of these implementing agencies has been assessed. The MOARD Directorates have some experience working on World Bank financed projects and have procurement units in their institutional setups. There is a Finance and Procurement Case Team serving the FSCD and EWRD. The Finance and Procurement Team Coordinator will have primary responsibility for the procurement of ICB contracts for goods and for the selection of international consulting services of the project to be implemented by DRMFSS. The FSCD has currently one external TA (Procurement Specialist) to support procurement activities, such as procurement planning, management, processing and supervision. The overall procurement capacity assessment indicates that the procurement arrangements at the Federal, Regional and Woreda levels need further strengthening. There is a lack of functional procurement management units under each Directorate. However, according to the recent reform in the Ministry, there is a procurement unit and technical assistance to support the two Directorates in procurement management. There is also a Tender Committee for bid evaluation, with an office and fulltime chairperson and secretary. As the chairperson and the secretary are assigned full-time, they also handle some of the procurement process activities of the two Directorates. 115 417. At Regional level, Regional Food Security Coordination Offices of the BOARDs of the participating Regions will be responsible for overall implementations, including developing and consolidating annual implementation plans, procurement plans, mobilizing technical assistances from other line agencies. The procurement section of BOARD (and other line ministries) will have the primary responsibility for NCB procurement for goods and procurement of consulting services. Offices have also recruited Regional Procurement Coordinators under the previous phases of the PSNP and these will continue to provide technical assistants for the planning, processing and supervision of procurement activities. At woreda levels, those woredas with adequate procurement capacity may also undertake procurement for goods and larger civil works following established NCB procedures. Woredas will have primary responsibility for shopping and small scale procurement of works. Contract awards will follow established government procedures regarding composition and mandate of the tender committees. 418. Procurement Process Flow Arrangement: The procurement process arrangement at Federal and Regional level is currently as follows: (a) Procurement Steps followed at Federal Level Step Activity Responsibility 1 Preparation of Procurement Plan Procurement Specialist 2 Approval of Procurement Plan Director, FSCD 3 Preparation of bidding documents/RFPs Procurement Specialist 4 Approval of bidding documents/RFPs Tender Committee 5 Bid/proposal opening Tender Committee 6 Bid/proposal evaluation Technical/Tender Committee 7 Approval of evaluation reports FSCD Director/DRMFSS State Minister 8 Preparation of contract document Procurement Specialist 9 Contract award/signature FSCD Director/DRMFSS State Minister (b) Procurement Steps followed at Regional Level Step Activity Responsibility 1 Preparation of Procurement Plan Procurement Officers 2 Approval of Procurement Plan Head/Process Owner/Commissioner 3 Preparation of bidding documents/RFPs Procurement Officers 4 Approval of bidding documents/RFPs Tender Committee 5 Bid/proposal opening & Evaluation Tender Committee 6 Approval of evaluation reports Head/Process Owner/Commissioner 7 Preparation of contract document Procurement Officers 8 Contract award/signature Head/Process Owner/Commissioner 419. Procurement risk rating and mitigation plans: The key systemic issues and risks concerning procurement for implementation of the project have been identified. The overall project risk for procurement is rated high and the thresholds for prior review, for ICB, including the maximum contract value for which the short-list may comprise exclusively national firms in the selection of consultants are agreed (Table 11). 116 Table 11: Table of Thresholds Category Prior Review ICB National Short-List Threshold (US$) Threshold Max Value (US$) (US$) Works 5,000,000 5,000,000 NA Goods 500,000 500,000 NA Consultants (Firms) 200,000 NA <200,000 Consultants (Individuals) 100,000 NA NA 420. The first two (2) contracts of each procurement method, irrespective of their amount, will be subject to IDA prior review in accordance with paragraphs 2 and 3 of Annex 1 of the World Bank's Procurement Guidelines. All ICB contracts shall be subject to IDA prior review. All single-source selection and all direct contracts, irrespective of the amount, will be subject to IDA prior review. 421. Table 12 summarizes the procurement issues identified and the proposed action plan to enhance the capacity of the executing agencies to implement project procurement. The matrix covers findings and actions to be taken at both the federal and regional/woreda levels. Table 12: Summary of Findings and Actions (Risk Mitigation Matrix) Major Actions proposed Responsibility Targeted date findings/issues 1 Inadequate The FSCD will recruit an additional procurement FSCD October 2009 planning and specialist at the federal level and provide more scheduling technical support to the sub-national implementing capacity agencies. 2 Lack of Annual independent procurement reviews. MOARD Annually two procurement (DRMFSS) months after supporting and the end of the control systems. FY 3 Lack of staff 1. In addition to the recruitment of a second MOARD/WB At an early skilled in procurement staff as indicated in proposed action no. stage of project procurement 1 above, each region will have a Regional implementation management Procurement Coordinator and continued support will be provided from the Federal Government. 2. The procurement staff and the tender committee members should undertake basic procurement training. The course shall be equivalent to the courses provided in regional procurement training centers and shall be provided by qualified staff. 4 Lack of written The existing PIM of APL II will be reviewed and FSCD Before procedural updated, as necessary, to review the procedures and Effectiveness manuals/systems lay out the steps for conducting efficient procurement in place to assist and guide the practitioners including code of 117 ethics 5 Inadequate 1. Training on procurement records keeping will be FSCD/RFSCO During project Procurement provided to all Regions and woredas. launch records keeping 2. The regional offices are to be supported with necessary office equipment and supplies (scanners, computers and printers, box files, file folders, etc) from the project. 6 At the woredas, The pool in each woredas will assign one existing Each region Before project the procurement staff from the pool to be responsible for PSNP launch arrangement is procurement. "pool" system and this seems to create delays in the project procurement 7 The Regional Each region to provide field transportation vehicles to FSCD/RFSCO At an early Food Security support procurement monitoring under the project. stage of the Coordination The Regional Food Security Coordination Office will project Offices have be capacitated with manpower and training implementation limited capacity and facilities Procurement Plan 422. The Borrower has drafted a procurement plan for the Project that will provide the basis for the procurement methods and implementation schedule. The final plan will be included in the project database and made available for inspection at each Regional bureau and at the office of the FSCD. The Plan will be updated in agreement with the Project Team annually or as required to reflect the actual project implementation needs and improvements in institutional capacity. 423. An independent procurement review of the project procurements will be conducted annually covering 40 of the woredas and at least 30 percent of the contracts subject to post review. The woredas will be selected to ensure that each woreda will have undergone a procurement audit at least once over the life of the PSNP. Frequency of Procurement Supervision 424. In addition to the prior review of procurement actions under ICB and QCBS to be carried out from the World Bank Country Office, at least two supervision missions per year will carry out post reviews of procurement on a sample basis. 118 Details of the Procurement Arrangements - For Works, Goods, and Non-consulting services. (a) List of contract packages for goods to be procured Estimated Prior/ Bid Date Domestic Proc. Description Amount Post Closing/ Contract Preference Comments Method (US$) Review Opening Signature Requirements for Procurement of Food EFY 2002 Modified Pre- 1 1st Round (30,000 MT) 13,000,000 ICB Prior Feb. 2010 May 2010 Yes qualification Modified Pre- 2 2nd Round (30,000 MT) 13,000,000 ICB Prior Mar. 2010 June 2010 Yes qualification Modified Pre- 3 3rd Round (30,000 MT) 13,000,000 ICB Prior Apr. 2010 July 2010 Yes qualification Modified Pre- 4 4th Round (30,000 MT) 13,000,000 ICB Prior May 2010 Aug. 2010 Yes qualification Modified Pre- 5 5th Round (30,000 MT) 13,000,000 ICB Prior June 2010 Sept. 2010 Yes qualification Sub-total: 65,000,000 Procurement of Other Goods 6 Vehicles (74 Units) 3,047,000 ICB Prior Feb. 2009 Aug. 2010 Yes 7 Motor Bikes (1830 Units) 7,185,000 ICB Prior Mar. 2010 Sept. 2010 Yes 8 Bi- Cycles (1500 Units) 118,000 NCB Post Jan. 2009 Mar. 2010 No Rub Hall- 10mX32 9 (35 Units) 700,000 ICB Prior Mar. 2010 Sept. 2010 Yes Computers and acc. 10 (1377 sets) 2,163,000 ICB Prior Feb. 2010 Aug. 2010 Yes 11 Photo copiers (220 Units) 432,000 NCB Post Mar. 2010 Sept. 2010 No Field equipment (small 12 items in 5 tenders btw $50,000-$60,000) 280,000 NCB Post Dec. 2009 Mar. 2010 No 13 70 Generators 165,000 NCB Post Jan 2010 June 2010 No Small office equipment 14 (7 sets) 24,000 Shopping Post Jan 2010 Mar. 2010 No 15 Safe boxes (4800 units) 2,639,000 ICB Prior Feb. 2010 Aug. 2010 Yes Sub-total: 16,753,000 Total: 81,753,000 119 (b) List of consulting assignments 1 2 3 4 5 6 7 Expected Estimated Date Selection Prior or Post Proposals Description Amount Contract Method Review Submission (in US$) Signature Date Disaster Risk Management Food Security Sector 1 Appeals Review EFY 2002 100,000 QCBS Post May 2010 Sept. 2010 Independent Procurement Review 2 EFY 2002 100,000 QCBS Post Sept 2010 Dec 2010 Woreda Procurement Capacity 3 Training 50,000 IC Post Jan 2010 Dec. 2010 4 Impact Evaluation 150,000 QCBS Post March 2010 Dec 2010 5 Financial Audit- EFY 2002 56,000 QCBS Post Oct. 2009 Dec 2010 6 Financial Audit- EFY 2003 56,000 QCBS Post Sept 2010 Dec 2011 7 Commodity Audit EFY 2002 100,000 QCBS Post Aug 2010 Dec. 2010 Preparation training program pastoral 8. areas 100,000 QCBS Post Nov 2009 Jan 2010 Sub-total: 712,500 NRMD (Public Works related services) 9 Public Work Review (1st 2010) 80,000 QCBS Post Jan 2010 April 2010 10 Public Work Review (2nd 2010) 80,000 QCBS Post July 2010 Oct.2010 11 Preparation Training Materials 210,000 QCBS Prior Nov. 2009 June 2010 12 PW guidelines for pastoral areas 150,000 QCBS Post Dec 2009 April 2010 13 PW Impact Evaluation 240,000 QCBS Prior June 2010 Oct. 2010 Sub-total: 760,000 Extension (HABP related services) 14 PIM preparation 110,000 QCBS Post Oct 2009 Feb.2010 Training on market and technical 15 analysis 175,000 QCBS Prior Sept. 2009 Feb. 2010 16 HAB woreda training support 35,000 IC Prior Oct. 2009 Feb. 2010 17 Financial sector capacity assessment 55,000 QCBS Post Oct. 2009 Dec. 2009 18 HAB MFI linkages study 130,000 QCBS Post Nov. 2009 May 2010 19 Financial product development 50,000 QCBS Post Dec. 2009 Feb. 2010 20 Training input providers 85,000 QCBS Post May 2010 Aug. 2010 21 Market linkages/value chain 55,000 QCBS Post May 2010 July 2010 22 HAB strategy for pastoral areas 100,000 QCBS Post March 2010 July 2010 Sub-total: 795,000 Total: 2,267,000 120 Annex 9: Governance Dimensions of the PSNP ETHIOPIA: Productive Safety Net APL III 425. Any public program implemented through different tiers of administration with specific targeting criteria, involving millions of beneficiaries and a large volume of resources requires strong checks and balances to protect against manipulation for personal ends or special interests. The same is true of the PSNP. Establishing and upholding the reputation of the PSNP is particularly important in the contested political environment leading up to the general elections in 2010. The Program's ability to demonstrate that it is fair and delivering the expected results without systematic leakage is critical to sustaining broad-based support in Ethiopia, internationally and among development partners. In Ethiopia, as elsewhere, this requires a commitment to transparency, participation and accountability. 426. Promoting citizen participation. The PSNP is designed to encourage strong citizen participation, particularly in targeting beneficiaries and planning public works, in order to ensure transparency and accountability in program delivery at local levels. These participatory processes determine the allocation of resources at local level and are the foundation of the Program. 427. The targeting of the PSNP is carried out through participatory processes that build on the history of community targeting in rural areas of the country. Taskforces comprised of community members and local officials use the PSNP targeting criteria to choose which households will take part in the Program. This is then verified through broader community meetings, where both beneficiaries and non-beneficiaries debate and agree on which households to include in the Program under public works or direct support for the next year. An independent study noted that the inclusion and exclusion of individuals have been raised through these public meetings, leading to adjustments in the list of beneficiaries. There is some indication, however, that women can experience difficulties expressing their opinions in such public forums. 428. There is strong evidence that the community-based targeting system is fair and transparent86. A recent study found that poverty was understood to be the reason for household participation in the PSNP among implementers, non-beneficiaries and beneficiaries alike87. Citizen perceptions of households being targeted for the PSNP on the basis of religious or ethnic affiliation or patronage are negligible88. Overall, a number of studies have concluded that the PSNP is well targeted to the poorest households, which have significantly lower incomes, fewer assets, and farm less land than non-beneficiaries. 429. There are, however, reports of exclusion in program woredas, given the extent of rural poverty and Program budget constraints. The difference between households in the Program and those that are excluded is often slight. Currently, a pilot is underway to determine the possibility of providing variable levels of transfers to households (i.e. transfers for three, six or nine 86 Over 85% of respondent considered the PSNP selection process to be fair. Urban Institute and Birhan Research and Development Consultancy. The Ethiopia Productive Safety Net Program, In the Financial Transparency and Accountability Perception Survey. Washington and Addis Ababa, 2008. 87 Devereux S et al. Ethiopia's Productive Safety Net Program: 2008 Assessment Report. Brighton and London: Institute of Development Studies and Overseas Development Institute, 2008. 88 Gilligan D. et al. Ethiopia Food Security Program: Report on 2008 Survey. Washington D.C.: IFPRI, 2009. 121 months). If administratively feasible, rolling out this approach is anticipated to improve the ability of the Program to respond to differing households' needs. 430. Each year, the PSNP creates thousands of public works projects in targeted communities. These are identified through a community-based participatory public works planning process each September/October to ensure that works are valuable to the community in general and to balance the competing interests of various interest groups, including the different interests of men and women. 431. Studies confirm high levels of community involvement in the selection of public works. A high and increasing number of households report participation in the project identification and planning exercises89. Within the community, public works are widely perceived to be beneficial. In 2008, 92 percent of households indicated that their community benefited from the construction of roads, while 88 percent reported the same for soil and water conservation on communal lands. Public works are increasingly perceived to benefit individual households as well90. 432. The HABP is designed to promote community participation in the identification of investment and income-generating opportunities. This will complement and feed into market and value-chain analysis carried out by woreda officials. This recognizes that there are many stakeholders within communities who have valuable knowledge of potential investment opportunities and experience with production and marketing. 433. Enhancing transparency. The Government is committed to strengthening overall transparency of the Program. This includes the public disclosure of key program information. The annual woreda-by-woreda resource allocation plan for the PSNP is posted on the MOFED website. While there is broad agreement that key program information will also be disclosed at local levels, implementation has been variable. The safety net budget and public works plan are posted for public review in some woredas or are posted for parts of the year. Similar trends are seen with the posting of beneficiary lists and list of appeals and appeal resolutions in woredas and kebeles. This process of disclosing key program information at local levels will be systematized across woredas in the next phase of the Program. 434. Steps have also been taken to ensure widespread understanding of program objectives, processes and procedures, including the targeting and appeals systems. Posters are being rolled out in woredas and kebeles to build further awareness of the PSNP among target communities while a newsletter designed to share experience among implementers was launched in 2008. 435. The importance of these initiatives is demonstrated by the fact that while the level of satisfaction with the Program is high (75 percent), households that perceive they have enough information to understand how the program works report even higher rates of satisfaction (90 89 IFPRI/CSA. Ethiopia Food Security Program: Report on the 2008 Survey. Washington D.C. and Addis Ababa: IFPRI and CSA, 2009. 90 For example, the percentage of households indicating that their household benefited from the construction of roads increased from 64% in 2006 to 82% in 2008. IFPRI/CSA 2009. 122 percent)91. APL III will therefore continue to promote widespread understanding of program objectives and procedures, particularly at local levels, as a core element of the governance agenda. 436. Complementing the use of posters to promote awareness of the Program among communities, the Government will explore the use of community radio to reach a high proportion of illiterate people and remote communities. The possibility of linking up with the PBS 2 initiative of supporting the purchase and distribution of wind-up radios to a significant number of kebeles and using local community radio to disseminate information on the PSNP will be explored. In addition, the Program will produce a beneficiary charter of rights and responsibilities. This will be disseminated widely in communities, including as part of the new system of client cards. These activities are all built into the PSNP Communication strategy, which will be broadened to include the objectives, procedures, and processes related to the HABP. This will include complementary initiatives that are designed to ensure that citizens have the information required to hold local decision-makers to account. 437. Strengthening Accountability. The above measures are designed to improve accountability by ensuring that citizens are informed and have the opportunity to engage in key decision-making processes. Available evidence suggests that high rates of community involvement can effectively hold local decision-makers to account, with few reports of resources being manipulated for personal ends or special interests92. 438. In addition to these, a number of specific structures and processes have been established with the aim of deepening local accountability. To guarantee timely and objective treatment for those who might have a grievance, an appeals system was introduced in 2007. By design, this system is separate from that for targeting and reports directly to locally elected councils. Appeals committees were established in 95 percent of the woredas surveyed in 2008, although weaknesses in record keeping have been observed93. For APL III, the Appeals Review (see below) will continue to provide advice and guidance to Kebele Appeals Committees (KAC) and relevant woreda and regional decision-makers in order to strengthen the overall appeals system. The mandate of the KACs will also be widened to address grievances related to the HABP. 439. While the design of accountability mechanisms for the targeting and appeals of the Program appear to be relatively sound, some concerns remain over the payment system. To date, the Program has relied on payroll and attendance sheets prepared by DAs and certified by woreda officials as the mechanisms to ensure that beneficiaries receive their payments in full. There are some indications, however, that this is not always the case, and survey worked raised the issue of accuracy of payments in some woredas. As a result, close follow-up was initiated by Government and development partners, firstly through an internal audit by MOFED and then through a joint field review. Both of these exercises suggested that the problem was not 91 Urban Institute and Birhan Research and Development Consultancy. The Ethiopia Productive Safety Net Program, In the Financial Transparency and Accountability Perception Survey. Washington and Addis Ababa, 2008. 92 Various Rapid Response Team reports 93 WABEKBON Development Consultants PLC. Roving Appeals Audit of PSNP in Ethiopia. Addis Ababa: Government of Ethiopia, 2008. 123 significant and that beneficiaries understanding of their entitlements under the Program may have been a significant factor in the original survey findings. Nevertheless, a number of additional steps will be taken in APL III to tighten the control of payments. 440. Under APL II, the payroll and attendance sheets - a critical internal control mechanism at woreda level ­ were computerized with the launch of the Payroll and Attendance Sheet System (PASS), supported through IT Helpdesks at Regional level. While the use of the PASS has progressed, further work is needed to ensure that it is used systematically across the Program. To this end, use of PASS will be made mandatory in APL III. Additionally, "client cards" will be introduced in all Regions to provide evidence of entitlements under the Program and proof of payment. This will enable beneficiaries, local decision-makers and Federal level officials to better track receipt of payments over time. 441. The financial audit system verifies that resources are used for the purpose intended. Under APL III, the financial audit will be strengthened in the area of reconciliation of the payroll with household level data, including a review of the PASS to provide a further check on the accuracy of payments to households. The audit function will also be extended to a review of food management systems and practices through a Commodity Audit, which will significantly improve the accountability of in-kind resources. At the same time, the roving audit function implemented under APL II will continue to provide an ongoing review of the effectiveness of the appeals and procurement processes and disseminate best practices across kebeles and woredas. 442. The program monitoring system, including initiatives to provide real-time data, provide a continuous overview of progress in implementation. This is one mechanism to promote the accountability of decision-makers in the civil service and among partner agencies. Evidence to date suggests, however, that this information is not being used to inform management decisions to ensure consistence performance across regions. APL III will aim to strengthen both the monitoring system and, perhaps more importantly, how this information is used by all levels of management. 443. The initiatives to strengthen accountability at local levels and within the civil services are designed to augment formal systems of political accountability in Ethiopia. MOARD already reports on the PSNP to the Rural Development Standing Committee and the Pastoral Areas Standing Committee of the Federal Parliament. The PSNP is also subject to post-audit review by the Standing Committee on Public Accounts. 124 Annex 10: Economic and Financial Analysis ETHIOPIA: Productive Safety Net APL III 1. Scope of Economic and Financial Analysis 444. The economic benefits of the Productive Safety Net Program (PSNP) are represented by (i) improvements in household wellbeing as a result of consumption smoothing, asset protection and the avoidance of negative coping behaviors; (ii) enhanced livelihoods through asset accumulation and increased productivity; and (iii) increased use of social services, market access and agricultural productivity as a result of the infrastructure created through the community public works. Therefore, PSNP provides both protective and productive benefits at the household and the community levels. Additional economic impacts are also expected from the Household Asset Building Program (HABP). 445. Since this is the third phase of the PSNP APL series, this economic analysis draws on experience from the first two phases in addition to other relevant evidence. This includes the recently completed impact evaluation and other empirical evidence94. There are often methodological challenges associated with valuing the costs and benefits attributable to interventions for projects of this nature. Therefore, a full-fledged cost-benefit analysis with an estimated rate of return for the PSNP as a whole is not applicable. However, where possible, benefits and costs are quantified to assess the economic and financial feasibility of the Program. 446. To analyze the economic and financial effects of this program, this Annex is divided into the following sections: (a) Beneficiary population and transfer level (b) Household level benefits (c) Community level benefits (d) Overall program efficiency (e) Economic analysis of the Household Asset Building component (f) Fiscal and macroeconomic implications 2. Analysis of Beneficiary Population and Transfer Level 447. The PSNP currently reaches 7.57 million beneficiaries95. This level of coverage is comparable to the proportion of the population reported to be food insecure in the 2004/05 94 The main sources of data for this Annex include: Gillian D et al. An Impact Evaluation of Ethiopia's Productive Safety Nets Program. Washington D.C: IFPRI, 2009. This impact evaluation draws on longitudinal household and community data on a sample of 3,467 households in 68 woredas. Both PSNP beneficiaries and non-beneficiaries were included in two rounds of surveys (2006 and 2008). Matching methods are used to estimate net impact. Devereux S et al. PSNP 2008 Assessment Report. Brighton and London: Institute of Development Studies and Overseas Development Institute, 2008. This study covered 960 PSNP beneficiary and non-beneficiary households in 16 kebeles, complemented by community questionnaires, focus group discussions and market monitoring. Government of Ethiopia. Detailed Report for the Productive Safety Net Programme for the Food Security Programme Review, Addis Ababa: Government of Ethiopia, 2008. Government of Ethiopia, PSNP Medium-Term Expenditure Framework and cost tables. 95 This figure does not include Somali Region, which is projected to be included in the Program in 2010 and onwards. 125 Welfare Monitoring Survey. At its current scale the PSNP is one of the largest safety net programs in a low-income country. 448. The estimated annual transfers for both direct support and public works beneficiaries per household are US$137 in 2009 (Table 1396). Analysis indicates that PSNP transfers represented 40 percent of the annual food needs of households participating in the PSNP. Table 13: Estimated Annual Direct Benefits Generated by PSNP (Based on 2009 program parameters) Total PSNP Total number of program beneficiaries97 7,574,480 Total number of households98 1,514,896 Total value of transfer (US$)99 206,783,304 PSNP Public Works Beneficiaries Number of households participating in public works 100 1,272,513 Estimated number of participation days per average household per year 150 Total number of days of public works generated per year 190,876,896 Estimated value of wages earned per average household (US$) 137 Value of wage transfer through public works (US$) 173,697,975 PSNP Direct Support Beneficiaries Number of households benefiting from direct support101 242,383 Estimated value of annual transfer per average household (US$) 115 Total value of transfer through direct support (US$)102 27,791,676 Source: World Bank estimates based on 2009 PSNP program data. 3. Analysis of Household Level Benefits (Protective and Productive Benefits) 449. The PSNP provides protective and productive benefits to food insecure households. As a safety net, PSNP transfers are designed to reinforce household income during the agricultural slack season to help households smooth their consumption and avoid asset depletion. Any 96 This table presents the estimated annual direct benefit generated by the PSNP in terms of cash transfers. The Program offers transfers in cash and in-kind, which are not shown here. The actual value of the cash benefit level was affected by the rise in food prices between 2006 and 2008, which rose much more rapidly than the wages for agricultural labor. PSNP transfers were 6 Birr/day from 2005 and 2007, and then increased to 8 Birr/day for 2008 and 10 Birr/day in 2009. Conversely there was a proportional increase in the real value of the in-kind transfer during the same period. 97 The number of program beneficiaries is based on an extensive community targeting exercise carried out in preparation for the 2006 program. This figure has since been revised to include additional chronically food insecure woredas in Oromiya and Somali for the pastoral pilot, woredas reallocated from Somali to Oromiya Regions and households severely affected by drought in SNNPR The actual number of beneficiaries will be confirmed each year through an update to this community targeting exercise and an assessment of household graduation. The actual beneficiary number is expected to decrease during the 5-year APL III. 98 Based on an average household size of 5 persons (source: CSA), which is consistent with the findings of the IFPRI/CSA 2008 household survey. 99 The cash wage rate is set at 10 birr per person per day. This calculation is based on all households receiving cash transfers. 100 It is estimated that 84 percent of households participate in public works and 16 percent in direct support (PSNP Review 2008). 101 Based on the IFPRI/CSA 2008 household survey, the average size of households receiving direct support payments is 4.2 persons. 102 Transfers under direct support are set at 50 birr per month per person. 126 current assessment of the protective benefits of the Program must be put in context: Ethiopia experienced rising food prices, falling (in real terms) asset prices, widespread drought shocks, and difficulties in accessing inputs in 2007/08. Therefore, an analysis of this period assesses how this safety net functions during times of extreme multiple shocks. 450. A recent impact evaluation concluded that the Program is smoothing household consumption and protecting assets, even during times of crisis. Evidence from 2008 shows that households tend to spend 75 percent of cash transfers on consumption and 25 percent on investments. Investments included debt alleviation, accumulation of livestock, agriculture investments, and utilization on health and education services. This suggests that PSNP cash transfers are enabling households to make investments in their livelihoods that were previously impossible. This use of cash transfers for consumption and investments purposes is supported by evidence from other countries. In rural Mexico, households participating in Oportunidades were found to use 75 cents of every peso transferred to purchase consumption goods and services. The remaining funds were invested in microenterprises and agricultural production, which led to improvements in their living standards103. Table 14: Utilization of Cash Transfers by PSNP Households (2008) Number of Average annual birr Consumption Uses Total Birr Ranking households per household Bought staple food (e.g. grain) 194 248.1 48,131 1 Bought clothes or cloth 98 123.7 12,123 2 Bought groceries (e.g. salt) 141 49.2 6,937 3 Social obligations 41 33.8 1,386 10 Bought other food (e.g. meat) 25 44.4 1,110 11 Paid taxes 41 17.1 701 12 Gave some cash to help others 3 42.4 127 15 Investment Uses Debt repayment 36 161.3 5,807 4 Bought livestock 23 226.5 5,210 5 Paid for education costs 65 63.1 4,102 6 Paid for health costs 28 101 2,828 7 Bought seeds for farming 31 68.1 2,111 8 Other 17 99.4 1,690 9 Used for business (e.g. trading) 5 86.4 432 13 Bought fertilizer for farming 16 10.1 162 14 Note: The number of households (n=230) refers to the number of households surveyed that reported consumption or investment uses of cash transfers Source: Devereux S. et al. 2008. 103 Gertler P et al. Investing Cash Transfers to Raise Long Term Living Standards. World Bank Policy Research Working Paper 3994, August 2006. 127 451. The 2008 Impact Evaluation quantified the protective benefits of the Program, namely: (i) reduced household food gap, (ii) increased caloric consumption, (iii) protection from distress sales of assets, and (iv) other negative coping behaviors: (a) Household food gap104: The impact evaluation found that PSNP public works participants105 had improved food security by 0.40 month versus comparable households. (b) Caloric availability: Growth in caloric acquisition was 17 percent higher for those PSNP households that received recent and regular transfers. When transfers were irregular, no beneficial effect was found. When exposed to shocks, PSNP households that received regular transfers of a high value had 30 percent higher caloric acquisition growth as compared to non-beneficiaries. (c) Distress sale of assets: Overall, PSNP public works recipients reported a level of distress sale of assets 4.4 percentage points higher than the control group. However, in looking at the absolute figures, distressed sale of assets decreased for both PSNP and non-PSNP households, though the decrease was slightly larger for non-PSNP households. Interpretation of this finding is difficult given the overall net increase in total livestock holdings among PSNP beneficiaries. (d) Other negative coping strategies: A second study found differences in coping strategies between PSNP beneficiaries and other households (Table 15). Non-PSNP beneficiaries saw a significantly larger increase in negative coping strategies as compared to trends among PSNP beneficiaries from 2005/06 to 2007/08. These negative coping strategies may have longer- term impacts on household economic status. Table 15: Coping Strategies by PSNP Status 2005/06 and 2007/08 PSNP respondents Non-PSNP respondents Coping strategy 2005/06 2007/08 2005/06 2007/08 Ate less food (smaller portions) 76% 78.1% 59% 72.3% Reduced number of meals per day 70% 72.3% 56% 65.2% Reduced spending on non-food items 28% 31.8% 18% 31.2% Sold livestock to buy food 22% 29.3% 27% 33.3% Borrowed food or cash to buy food 14% 21.8% 9% 23.4% Household members migrated for work 13% 12.8% 13% 10.6% Relied on help from relatives/ friends 10% 8.1% 11% 5.6% Rented out land to buy food 10% 6.9% 9% 0.0% Sent children to work 5% 2.7% 9% 6.3% Sold other assets to buy food 4% 2.1% 3% 4.2% Sent children to stay with relatives 4% 1.7% 2% 4.2% Withdrew children from school 2% 0.9% 1% 2.1% Source: Devereux S. et al. 2008. 104 The Impact Assessment uses the following definition of food gap: number of months, out of the preceding 12 months, that household report that they had "no problems satisfying the food needs of the household." The difference between this number and 12 is what is called the food gap. 105 Defined as receipt of at least 100 Birr in payments over the first five months of 2006, 2007 and 2008. 128 452. PSNP transfers have also provided productive benefits to households, including increased asset holdings, productivity and accumulation of human capital: (a) Availability of household assets: Participation in PSNP public works increased growth in livestock holdings by 0.28 Tropical Livestock Units (TLU) over comparable households. According to studies elsewhere in Ethiopia, such an increase in livestock holdings could decrease food insecurity by about 7 percent106. Again, results were sensitive to program performance. Households that received low levels of transfers experienced no net increase in holdings. In contrast, households receiving recent high value transfers had an increase of 2.6 TLUs. (b) Farm productivity: Access solely to public works transfers had no effect on output, acreage, productivity or fertilizer use among beneficiary households, which is consistent with the evidence on utilization of cash transfers by PSNP households. This suggests that the PSNP is having no disincentive effects on household investments in own production. Interestingly, access to the household investments provided by the Government for inputs and technical assistance also did not impact output, acreage, productivity or fertilizer use among beneficiary households. There were, however, large impacts on productivity ­ increases of more than 200 kg/ha ­ when households had access to both public works transfers and household investments. (c) Increased human capital: There is evidence that PSNP beneficiaries have increased their use of social services. In 2006, 35 percent of PSNP beneficiary households reported that they used health facilities more that year than the year before, with a further increase of 13 percent in 2008. This information, together with reports that PSNP beneficiaries use some of their cash transfers to invest in health and education, suggests that the PSNP has a positive impact on human capital. 453. In addition, there is emerging evidence that participation in the PSNP supports households in adopting high risk/high return strategies, such a taking credit leading to higher rates of agricultural productivity. For example, from 2006 to 2008, the proportion of households in food insecure areas that did not take a loan because they were afraid that they could not pay it back declined from 37 to 25 percent. This is in addition to the finding that households use cash transfers to invest in their livelihoods, which suggests that the PSNP provides households with cash when they require liquidity to make productive investments. A study of the PROCAMPO program in Mexico concluded that the program's cash transfers to farmers in the ejido sector relaxed liquidity constraints for households, thereby allowing these households to take advantage of existing assets, particularly irrigated land. This resulted in a multiplier effect ranging from 1.5 to 2.6107. These findings may explain, in part, the increase in agricultural productive among households receiving both PSNP and household investment packages described above. 454. Calculating a single economic rate of return for the Program is not appropriate for this type of safety net operation. This is because there are methodological limitations in imputing economic value for all of the PSNP transfer benefits. For instance, the short timeframe under 106 When TLU holdings increase by one unit, the probability of food insecurity declines by 24 percent. Source: An Empirical Analysis of Food Insecurity in Ethiopia: the Case Of North Wello. Africa Development, Volume XXVII, No. 1 & 2, 2002. 107 Sadoulet E, De Janvry A, Davis B. "Cash Transfer Programs with Income Multipliers: PROCAMPO in Mexico." World Development, 2001, 29(6):1043­56. 129 consideration cannot capture one of the most important aspects of the Program's impact: a reduction in the long-term transmission of poverty and destitution. In addition, increases in human capital are difficult to quantify in monetary terms without data on educational attainment, labor market outcomes, and the value of improved health108. 455. It is important to note that the economic direct benefits from the transfers would need to be added to those from the infrastructure created through public works to arrive at economic benefits of the total program. 4. Analysis of Community Level Benefits (Productive Benefits) 456. The PSNP initiates about 34,000 public works projects each year. The individual public works subprojects are not pre-selected, but are rather planned on an annual basis within each woreda and kebele. Table 16 gives some examples of the types of community infrastructure completed in 2007. This data suggests that the bulk of investments are concentrated in soil and water conservation (SWC) and rural feeder roads, with selected investments in natural resource management and social services. Unfortunately, the PSNP information system currently tracks only the outputs of public works by sector. This will be addressed during APL III (see Annex 8). Table 16: Sample of PSNP public works completed in 2007 Activity Result Activity Result Soil Embankment Construction (km) 482, 542 Seedlings Produced (no.) 301,778,607 Stone Embankment Construction (km) 443,148 Seedlings Planted (no.) 12,883,657 Pond Construction & maintenance (no.) 88,936 School Classroom Construction (no.) 340 Spring Development (no.) 598 Rural Road Maintenance (km) 20,458 Hand-dug well construction (no.) 491 Rural road construction (km) 8,323 Land rehabilitated Area Closure (Ha) 530 Farmer Training Centre Const. (no.) 119 Small-scale irrigation canals (km) 2,679 Animal Health Post Construction (no.) 71 Tree nursery site establishment (no.) 285 Source: FSCD 2007 Annual Report, January 2008. 457. These projects have a potential economic effect on local communities. A composite economic rate of return is not possible to calculate for PSNP APL III since the specific public works projects are not known in advance and, for most of the soil and water conservation and rural road projects, expected rates of return vary widely depending on location109. Moreover, analysis cannot adequately capture the full economic benefits for several types of investments. 458. Economic analysis has been carried out for a sample of completed public works projects despite these limitations. In 2008, an impact assessment was conducted in ten different watersheds to determine the actual and potential impacts of the public works. Economic analysis was carried out for each type of project and, wherever possible, a benefit-to-cost ratio 108 For example, malnutrition's economic costs are substantial: productivity losses to individuals are estimated at more than 10 percent of lifetime earnings, and gross domestic product (GDP) lost to malnutrition runs as high as 2 to 3 percent. World Bank. Repositioning Nutrition as Central to Development. Washington D.C: World Bank, 2006. 109 For example, plots with stone bunds are more productive than those without such technologies in semi-arid areas but not in higher rainfall areas because the moisture conserving benefits of this technology are more beneficial in drier areas. 130 calculated110. The overall conclusion of the impact assessment is that the public works created through the PSNP are delivering significant economic benefits to communities. 459. The detailed findings of the impact assessment are summarized below and in Table 17: (a) Soil and Water Conservation: Ex-post visits to these projects found significant and visible increases in wood and herbaceous vegetation cover and a broader diversity of plant species. The increase in herbaceous vegetation has already had a positive and visible impact on the increased supply of livestock feed, bee forage, and medicinal plants. Of the households interviewed, 34 percent reported significant benefits from the areas enclosures in terms of forage for livestock. Cost-to-benefit ratios for the SWC measures were calculated that covered the benefits of soil loss reduction, woody biomass and forage production and carbon sequestration. The results range from 1.5 to 2.6, with an average of 1.8. (b) Water Supply Projects: In addition to health gains from greater access to clean water and resulting increased labor productivity, water projects were found to reduce the distance women and children travel to fetch water. This time savings allows household members to engage in other value-added household activities like crop production. Using the willingness- to-pay method, the discounted benefit-to-cost ratio of a typical single developed spring was calculated to be 3.7. (c) Small-scale irrigation: Small-scale irrigation from water sources developed by PSNP has helped to expand livestock for 4-12 percent of households and increase incomes by 4-25 percent. It is estimated that even very small irrigated plots (190 m2) are capable of generating gross margins of between ETB 4,200 to 6,000/yr (double cropped). Thus, even micro-scale irrigation can have a significant impact on household livelihoods and food security. (d) Health Projects: Potential impacts of investments in health facilities will be through improved household labor productivity and a reduction in health care costs from not being ill. Two approaches to benefit-to-cost analysis for this type of project were used: (i) willingness to pay for medical services and (ii) the opportunity costs of labor lost due to illness. The former method provided an estimated benefit-to-cost ratio of 1.8 and the latter one of 2.2. (e) Rural roads: PSNP rehabilitated roads are providing better access, particularly for vehicles, carts and mules. As the roads constructed by the PSNP generally consist of small additional sections to existing roads, are not commercial or not always passable in the rainy season, it did not prove possible to determine benefit-to-cost ratios. 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. 110 Results vary considerably between watersheds and between agro-climatic zones. Thus in order to be able to arrive at aggregate results, it is necessary to be able to weight the findings according to the details of public works created to date across the entire program. In the absence of reliable data of this type, it has not been possible to calculate definitive figures for the entire program. Nonetheless, wherever possible, broad conclusions for the program as a whole have been arrived at based on average results weighted according to the estimated number of watersheds of various types in the program. 131 Table 17: Sample of Public Works Outputs Completed in 2007 and Benefit/Cost Ratio Benefit/Cost ratio Activity 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. 460. These findings are consistent with evidence on the economic returns of other investments in Ethiopia. A recent World Bank project found general SWC economic rates of return (ERR) of 10-17 percent (World Bank/GEF)111. Other research in Ethiopia on stone bunds found yield increases of 20-50 percent, with an ERR of 46 percent112. Soil bund returns were more variable by location with marginal or negative returns in high rainfall areas, for instance113. 5. Analysis of Overall Program Efficiency 461. This section assesses the efficiency of the PSNP in delivering the transfers. Table 18 estimates the potential savings from the introduction of cash transfers. It assumes that the value of the food and cash transfers is equivalent, meaning the cash transfer would purchase the equivalent amount of food at the retail parity price in Addis Ababa markets. In this way the estimated savings are not due to any difference in value between the monthly cash or food transfers but rather reflect the efficiency gains from not having to transport food. As shown in Table 18, given the 2008 mix of cash and food beneficiaries, the estimated annual savings to the Program is almost US$11.0 million. An additional US$11.0 million in efficiency gains would be possible if the Program were to provide all transfers in cash and result in a cumulative savings of US$22.0 million. 462. Recent evidence also suggests that the shift to cash transfers in the PSNP can have significant efficiency gains for beneficiaries in terms of the time and costs associated with collecting the transfers. In Regions that provided transfers in cash, 80 percent of beneficiaries reported that they received their transfer in the same place as they live. As a result, the distance 111 Economic analysis of the Ethiopia Sustainable Land Management Project (World Bank Project Appraisal Document Report No. 42927-ET) found an ERR of 10 percent for physical investments in SWC (stone bund, soil bund, fanyajuu, and grass strips), and 17 percent for SWC physical structures combined with fodder on bunds and intercropping. 112 Pender J., Gebremedhin B. 2006. "Land Management, Crop Production, and Household Income in the highlands of Tigray, Northern Ethiopia: An Econometric Analysis." In Pender, J. et al. (eds). Strategies for Sustainable Land Management in the East African Highlands. Washington D.C.: IFPRI., 2006. Kassiea et. al. "Estimating Returns to Soil Conservation Adoption in the Northern Ethiopian Highlands" Environmental Economics Policy Forum for Ethiopia, Addis Ababa, Ethiopia. 2007. Nyssen, et al., "Interdisciplinary on-site evaluation of stone bunds to control soil erosion on cropland in Northern Ethiopia" Soil and Tillage Research, Volume 94, Issue 1, May 2007, Pages 151-163. Stochastic Dominance Analysis of Soil and Water Conservation in Subsistence Crop Production in the Eastern Ethiopian Highlands: The Case of the Hunde-Lafto Area" Journal of Environmental and Resource Economics Volume 32, Number 4 / December, 2005. 113 Kassie M et al. "Economics of Soil Conservation Adoption in High-Rainfall Areas of the Ethiopian Highlands" Environment for Development Discussion Paper Series March 2008. Herweg K., Ludi E. "The performance of selected soil and water conservation measures--case studies from Ethiopia and Eritrea" CATENA Volume 36, Issues 1-2, June 1999. 132 to the payment site was 5.5 kilometers and the average cost incurred by the household in collecting their transfer was 0.5 birr. In comparison, among Regions providing a significant proportion of transfers in food, 47.6 percent of beneficiaries reported that payments are made in the place that they live. The distance to the payment site was 11.4 kilometers on average and cost 2.8 birr114. Table 18: Cost Savings of the Shift from Food to Cash transfers (ETB) Scenario 1: Scenario 2: Scenario 3: 2008 Figures All Food Current Mix All Cash Beneficiaries receiving cash transfers 3,556,499 7,355,049 Beneficiaries receiving food transfers 7,355,049 3,798,550 Total 7,355,049 7,355,049 7,355,049 Monthly cash transfer per beneficiary a 50 Equalized to retail value of food ration 67 Value of monthly food ration (at Addis retail parity price) in ETBb 67 Transport costs to regions per beneficiary per month in ETBc 5.55 ---d Projected monthly transfer cost, in birr 535,079,815 515,341,245 494,259,293 Total annual cost (6 months of benefits), in birr 3,210,478,889 3,092,047,472 2,965,555,75 Annual Savings, in ETB (as compared with Scenario 1) 118,431,417 244923132 Annual Savings, in US$ (as compared with Scenario 1) $10,574,233 $21,868,136 a Based on daily wage rate of 10 birr for 5 days per month = 50 birr/month. b Based on Addis Ababa retail parity price of 4.48/kg of wheat at 15 kgs/month as reported in: Ethiopian Development Research Institute. The Implications of Inflation on the Purchasing Power of the Productive Safety Net Cash Transfer, September 2008. c Trade and transport costs estimated as the difference between the Addis Ababa retail parity price and the average regional import parity price of wheat assuming average margin rates (4.85/kg) as reported in "The Implications of Inflation on the Purchasing Power of the Productive Safety Net Cash Transfer" Ethiopian Development Research Institute (EDRI), September 2008. Trade and transport cost = 0.37 birr/kg x 15 kg/month. d Program management and administrative costs are assumed to be equal between the cash and food options. As these are the only costs associated with transferring cash, no incremental cost of transferring cash is included. Source: World Bank estimates; ETB 11.2=US$1. 463. Comparisons can also be made with other international experience and benchmarks for public works programs. The efficiency of public works programs is largely a function of (i) effectiveness of targeting the poor (i.e. greater share of transferred resources reaching target population); (ii) labor intensity of the public works (resulting in a higher share of resource transferred through as direct economic benefit), (iii) the economic effects of the public works themselves, and (iv) efficient program administration costs. PSNP compares favorably with international experience on public works programs, both for its targeting well as the high wage intensity coupled with low administrative cost, which is achieved by using existing government systems. These elements in turn are described below: (a) Wage targeting performance: This measures the proportion of the wages that are paid to poor workers. A study carried out in 2006 found that 87 percent of PW beneficiaries had a 114 Gilligan D et al. 2008. It is noted that there are additional costs for beneficiaries who receive cash specifically with regards to the time and effort required to purchase food. Conversely, evidence shows that beneficiaries receiving food tend to sell at least part of their transfer to meet their non-food needs, which would impose similar costs. 133 food gap of three months or more. International experience suggests that the best performing programs targeting 80 percent of benefits to the poorest households115. (b) Labor intensity: This measures the share of total program costs allocated to wages. Program rules dictate that at least 80 percent of program costs at woreda level go to wages, with 15 percent for capital and 5 percent for administration. This is quite high by international standards. The labor intensity of large-scale public works program averaged 60-70 percent in India (Maharashtra Employment Guarantee Scheme), 70 percent in Korea's public works program, 40-50 percent in Argentina's Trabajar program, and 60-70 percent in Bangladesh's Food for Work Program. (c) Benefit to cost ratio for the PW projects: Where ex-post data is available, for example for soil and water conservation projects, the rates of return for PSNP public works are acceptable, as detailed in paragraph 461 above. (d) Program administrative costs: The overall breakdown of program costs is provided in Table 19. An estimated 92 percent of program costs are direct benefit to households and communities through transfers and public works created, while 7.8 percent is absorbed by administrative costs. This is very efficient by international standards. For example, experience in developing countries finds that safety net programs can be run well for modest administrative costs, with a rule of thumb of roughly 10 percent of overall program costs116. Table 19: Cost Breakdown for PSNP Food or cash transfers 77.7% PW capital expenditures 14.6% Woreda level administration 4.9% Regional Administration 1.9% Federal Administration 1.0% Total 100.0% Note: This includes contract staff but not civil servants. Source: Authors calculations 6. Economic Analysis of Household Asset Building Program (HABP) 464. There is some evidence to suggest that the combination of the PSNP with HABP activities can provide a pathway to food security for some households (Annex 1). In developing an ex-ante economic appraisal of this component, this Annex draws on the experience and evaluations of the World Bank-financed Ethiopia Food Security Project and a Review of the Government's Household Asset Building Program117. 115 Coady D et al. Targeting of Transfers in Developing Countries: A Review of Lessons and Experience. Washington D.C.: World Bank and the International Food Policy Research Institute, 2004. 116 Grosh M et al. For Protection and Promotion: The Design and Implementation of Effective Safety Nets. Washington D.C.: World Bank, 2008 117 Coulter L, Sutcliffe JP, Household Extension Packages: Modeling Impacts and Comparing Approach. Prepared for the Review of the Government of Ethiopia's Other Food Security Programme, 2008. The main source of data is the Baseline Survey of the USAID-funded Livelihoods Integration Unit (LIU) of the Ministry of Agricultural and Rural Development (MoARD). This data has been collected using the "Household Economy Assessment" framework, including data on sources of food, income, and expenditures. In addition, household case studies of household adoption household investment packages were undertaken in 10 woredas in the four Regions. 134 465. The Household Asset Building Program (HABP) main objectives are to: - Increase assets and income; - Increase agricultural productivity; and - Diversify income. 466. Financial analysis on HABP investments is based on the incremental production and incremental cash flow calculated for five years taking into account whether produce was consumed by the household or sold, and adding credit payments or subtracting loans where necessary. Table 20 below presents cost components of the packages assessed ­ including credit taken for purchase of assets, additional capital and annual maintenance costs ­ and calculates the ratio of total income generated over a five-year time period to amount spent on the package (in ETB). This ratio can be considered a rough indicator of cost effectiveness. The income to cost ratio for the various types of packages was found to range from 1.97 to 5.63. Table 20: Cost Components and Cost Effectiveness of Packages ­ 5-Year Totals (ETB) Additional Annual Income Credit Total Change in Scenario Capital maintenance to cost taken cost income costs costs ratio Ox fattening (Oromiya) 3,200 0 2,350 5,550 10,933 1.97 Ox-Shoat (SNNP) 2,440 163 1,519 4,122 8,791 2.13 Bees-Irrigation/vegetables (Tigray) 1,400 0 7,410 8,810 23,444 2.66 Ox/cow-Crop-Irrigation/vegetables (Oromiya) 2,700 2,915 2,660 8,275 22,709 2.74 Crop-Irrigation/fruit-Camel-Ox-Shoat- Dairy (Tigray) 4,600 0 4,119 8,719 28,961 3.32 Poultry-Shoat-Dairy (Amhara) 2,760 200 175 3,135 16,656 5.31 Irrigation/fruit-Shoat (Amhara) 2,100 500 762 3,362 18,924 5.63 Source: Coulter L, Sutcliff JP 2008. 467. The FSP Review and other assessments118 of the household investments revealed that success rates were high with 85 to 90 percent of households reporting positive rates of financial return. In a few cases, where households had adopted more than one component, failure of one component (e.g. death of a fattening animal) was offset by revenues from the other components. Total failures were found to be among those households that had adopted only one component, generally with large animals (oxen, camels). The other package that experienced total failure was bee keeping, where the entire bee colony had absconded. 7. Analysis of Fiscal and Macroeconomic Implications 468. There are several potential macroeconomic and fiscal issues related to the PSNP, including: (a) The potential inflationary effects on food prices (b) The Program's relative weight in overall public spending (c) The Program's fiscal effects at the woreda level 118 Legese M et al. 2007; REST 2008; Tefera M et al. 2007; Tafesse B. 2008. 135 (d) Perspectives on the financial sustainability of the Program 469. Inflationary Effects. The PSNP has two hypothetical inflationary effects: (i) the demand side price effects, where providing cash payments to participants increase household incomes and push up demand for food, or (ii) the supply side, where short-term reductions in the amount of food aid translate into reduced grain supplies in the country, thereby eliminating the price- dampening effects of food aid119. Economic estimates on the potential inflationary effects of PSNP found that from an economy-wide perspective PSNP transfers are very small120. The total safety net cash transfer program in 2005-2006 was equivalent to only 1.1 percent of GDP. In simulations, this additional income shifted demand for wheat by only 0.8 percent and raised wheat prices by about 1.8 percent. 470. However, the net effect of PSNP on wheat prices through a reduction in the supply of food aid was estimated to be a 12 percent increase in price. This higher price would create an incentive for increased local grain production. In this case, the medium-term effect of the PSNP would be to raise wheat production slightly by 2-3 percent and to raise domestic wheat prices by only about 8 percent. Recent analysis that assumed about two-thirds of in-kind transfers were in the form of imported food found that the net effect of increased supply of cereals plus increased demand due to higher household incomes resulted in insignificant increase in food prices of less than 0.5 percent. This analysis found that the recent food price inflation in Ethiopia can be largely explained by overall nominal increases in prices, which have, in turn, been very closely associated with increases in the money supply121. A 2008 woreda-level analysis similarly concluded that there is no evidence to suggest that increasing the size of PSNP cash transfers can fuel inflation122. 471. Relative Weight in Public Spending. PSNP, the main public safety net in the country, currently represents 1.2 percent of GDP. The relative weight of PSNP compared with other social sector spending as a share of GDP is presented in Table 21. By means of comparison, most developing countries spend in the range of 1-2 percent of GDP on safety nets123. 119 Recent analysis of agricultural distortions in Ethiopia concluded that the high influx of food aid resulted in market disincentives that negatively impact domestic prices by 2-26 percent for wheat, 3-13 percent for maize and 2-11 percent for teff (Rashid, Assefa and Ayele, 2006). 120 Dorosh P, Subran L. "Food Markets and Food Price Inflation in Ethiopia", November 2007. 121 Explaining Sources of Food Price Inflation in Ethiopia, World Bank, June 30, 2007. 122 Rashid S. et al. Grain Markets and Large Social Transfers: An analysis of the Productive Safety Net Program (PSNP) in Ethiopia. Washington D.C.: IFPRI, 2008. 123 Grosh M. et al. 2008. 136 Table 21: Budgetary Allocations of Social Sectors as percent of GDP Sector 2008/09 Public Education 3.7 Public Health 1.4 124 PSNP 1.2 Source: World Bank estimates. 472. Relative Weight in Woreda Spending. Given the relatively slow pace of fiscal decentralization in Ethiopia, PSNP resources continue to represent a significant share of local resources among target woredas. As shown in Figure 5, the mean value of PSNP resources as compared with annual total expenditure in woredas that are receiving PSNP transfers is 62 percent with a range of 14 to 175 percent. In terms of generating additional recurrent costs from the new infrastructure created there is little available analysis. Many of the soil and water conservation and roads projects consist of rehabilitations at existing sites, which would generate few new recurrent cost obligations. Figure 5: Distribution of ratio of woreda PNSP resources to total woreda expenditure, 2007 16 14 12 Number of woredas 10 8 6 4 2 0 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 110% 120% 130% 140% 150% 160% 170% 180% Source: World Bank estimates. This analysis is based on a sample of 87 woredas that receive PSNP resources solely in cash. PSNP resources refers to: cash transfers, capital and administrative budgets and woreda contingency budget. 473. Fiscal Sustainability. A fundamental aspect of the fiscal sustainability of the PSNP is how the beneficiary numbers evolve over time. This is a function of current weather and economic conditions as well as the ability of PSNP beneficiaries to graduate out of food insecurity. The Government has proposed a scenario for graduation based on highly ambitious targets with up to 80 percent of beneficiaries graduating by 2014. A modeling analysis using data from the Livelihood Integration Unit of the DRMFSS suggests that 50-55 percent of the current PSNP 124 Includes cash and food transfers 137 caseload could graduate from food insecurity by 2014. This analysis assumes that each household receives timely and predictable PSNP transfers and accesses successive loans from the HABP. In contrast, if the more structural challenges of reducing food security are not met, core program beneficiaries will continue to rise, reaching over 9 million, without any assumptions about contingencies needed for weather or other shocks. However, it is important to also consider experience that shows that the PSNP remains better value for money than the emergency system. Thus, any short-term gains achieved by graduating households from the PSNP will be lost if these same households fall quickly into the emergency system. 474. A further issue related to fiscal sustainability is the trends in food prices, which will help determine the ultimate cost of the Program. Until 2008, raising food price in Ethiopia had been relatively consistent with overall inflation. Since early 2008, however, the price of food has accelerated to an unprecedented extent. This is important for the PSNP because rising food prices erode the purchasing power of cash transfers in terms of food (Figure 6) and the primary intention of PSNP cash transfers is to provide market access to food. These recent food price trends indicate that the PSNP is highly sensitive to food price inflation. Figure 6: Rural Real Wage (Wheat Equivalent) and Cash Transfers 4.5 Wheat equivalent of daily wage (kgs) 16 Nominal daily wage rate (Birr) 14 4.0 12 3.5 10 3.0 8 6 2.5 4 2.0 2 1.5 0 Food Equivalent of cash Transfer Wheat equivalent of daily wage (Cash Woreda) Food transfer (3 kgs wheat) Nominal daily wage rate (Cash Woreda) Source: Rashid S. et al. 2008. 138 Annex 11: Safeguard Policy Issues ETHIOPIA: Productive Safety Net APL III Environmental Assessment Procedures 475. This Annex provides more specific details on Environmental Assessment procedures in use under the Program. Additionally, it provides an overview of safeguards issues arising from implementation of PSNP Phase II as well as issues arising from the adoption of the new support to HABP. The principal instrument for ensuring that projects are designed to avoid or minimize negative environmental impacts under both the World Bank's safeguard policies and those of the Government is Environmental Impact Assessment (EIA)125. The World Bank's Environmental Assessment policy (OP 4.01) is applicable. However, being community-based, on a small scale, and not identified in advance, public works in the PSNP will not normally require a separate EIA. Instead, the EIA requirements will be addressed through an Environmental and Social Management Framework (ESMF), which has been developed and publicly disclosed by the Government. 476. Under the PSNP, a number of potential public works are identified at kebele-level through a community-based watershed management planning process that prioritizes and selects activities contributing to improved watershed management and infrastructure. The ESMF specifies criteria for all types of public works eligible for PSNP support, avoiding locations or project designs which might give rise to unmanageable impacts, and recommending design modifications where appropriate. 477. The community having thus developed a shortlist of desired public works, design work is carried out by the DA with assistance as required from woreda technical staff. The resultant kebele PSNP public works plan is then submitted to the woreda offices where a simple ESMF screening procedure is adopted in order to ensure that negative impacts are avoided or minimized and that any public works that may require an EIA are identified. Given the nature of the public works such cases are infrequent. 478. DAs and technical staff at woreda level are guided by the MOARD Community Participatory Watershed Management Guidelines. Thus the design of public works such as soil and water conservation and road works will have had standard environmental mitigating measures already incorporated into the design of the public works by the time they are screened by the ESMF procedures. Thus the screening procedure will generally be quite rapid. 479. Only in exceptional cases should it be necessary for a public works project to be reviewed at a higher level in which case the Regional Environmental Protection Authority (REPA) will be the responsible authority for deciding whether an EIA is required. 480. All environmental safeguards procedures are documented in full in the ESMF. Responsibility for their implementation and monitoring are designated at federal, regional, woreda, and kebele levels. In addition, joint Government-development partner monitoring of 125 In keeping with Ethiopia's Environmental Impact Assessment proclamation, the term "environment" in this context covers biophysical, social and cultural heritage impacts. 139 ESMF implementation will be conducted and followed by any necessary corrective measures. This monitoring is undertaken through: (i) the public works component of the Program M&E system, which tracks the nature and extent of implementation of the ESMF; and (ii) the twice- annual joint Government-development partner Public Works Reviews, in which samples of public works from across the Program are examined for quality, sustainability, impact, and ESMF implementation. Any required rectification works (both labor and non-labor) are conducted using Project resources in the form of repair and rehabilitation works under the following year's public works. Ensuring that this happens is the responsibility of the DA involved in the community PSNP public works planning process and the Natural Resource Expert in the Natural Resource Woreda Desk. Safeguards Issues Arising in PSNP Phase II 481. The principal safeguards issue arising from public works reviews conducted during Phase I was a reportedly low implementation rate of the ESMF procedures in the kebeles and woredas examined. This led to steps being taken by the federal Government to ensure compliance at the woreda level and to a stronger focus being put on support to the Regional Public Works Focal Units for monitoring compliance. The outcome was that in Phase II the reported rate of ESMF implementation increased considerably and reached acceptable levels. However, it was noted that the quality of ESMF implementation was frequently below the standard required. 482. Evaluations conducted during Phase II identified that quality improvements would require enhancement of the capacity of the PWCU to review and streamline the design of the ESMF with a view to making it easier to implement at both DA and woreda level, as well as monitor at the regional level the 34,000 separate ESMF implementations conducted and documented every year. 483. Thus the major upgrading of the PWFU that took place in April-May 2009 incorporated significant new capacity to review and strengthen ESMF design, implementation, and monitoring of outcomes (ref. Section (iv), e and f). As a result, the ESMF has been reviewed and a number of new sections added. In addition, the PWCU will monitor the performance of the ESMF on a regular basis throughout the course of Phase III so that the ESMF implementation procedures can be appropriately streamlined. 484. In addition, Government has made a renewed commitment to develop ESMF and monitoring & evaluation capacity in the regional Public Works Focal Units, which following organizational changes during the course of 2008 had declined. 485. The ESMF as disclosed for Phase III has been updated and expanded to incorporate the following changes: (a) Improvements have been made in the guidance provided on Screening SWC sub-projects; further work on making this guidance region-specific will be carried out during the first year of implementation by the PWCU. (b) Clearer guidance is given to ensure that sub-projects involving large dams, large irrigation schemes, land use changes potentially affecting natural habitats or forests, or having the 140 potential for involuntary loss of assets or access to assets, and other projects ineligible under the PSNP, are eliminated by the screening procedure. (c) Explicit guidance is provided on how the kebele and woreda authorities should address the issue of voluntary land loss, in the event that it may occur, in the form of a Voluntary Land Loss Procedure (see below). (d) More detailed guidance is provided on the roles and responsibilities of each concerned party at each stage of implementation of the ESMF procedures. (e) The Government's Guide for Integrated Pest Management in Small-Scale Irrigation Projects and the Government's Medical Waste Management Guide for Rural Health Clinics, which were previously separate documents, have now been incorporated into the ESMF. 486. Staff of the Regional EPAs and the woreda Environmental (Natural Resources) focal persons will participate in training courses for the PSNP public works, which includes ESMF training. These training courses, which are subject to an ongoing program of upgrading by the expanded PWCU, will be provided by teams drawn from MOARD at Federal and Regional level, with technical assistance from the Natural Resources Management personnel of MOARD, the regional EPAs and agencies such as WFP. 487. The cost of implementing the environmental and social safeguard measures are covered partly by the program management budget at federal, regional, and woreda levels, and by the regular government staffing and overhead costs at all levels. For this reason a precise costing of this specific aspect of the coordination, management, and implementation of the public works is not possible. Nonetheless an approximate estimate of the costs can be made as follows in Table 22. Table 22: Cost Estimate/Annum of Implementing Safeguard Measures Level Project Government Approx. Total No. Approx. Est. Staff per Staff per No. of of Staff- Staff Cost Total Staff Team Team Teams Equiv. (Birr/staff) Cost (incl. o'head) (Birr) Federal 2 1 2 109,000 218,000 Region 1 8 8 88,000 704,000 Zone 1 20 20 88,000 1,760,000 Woreda 0.5 320 160 29,000 4,640,000 Kebele 0.25 4,800 1,200 16,000 19,200,000 Source: World Bank calculations 488. Based on the above Table, the following costs were calculated, reflecting in-kind contribution of Government staff time: (a) Estimated Staff Cost (above) = ETB26,522,000/annum = US$ 2,122,000 (b) Estimated Non-Pastoral Annual Training Cost (PSNP) = ETB746,400/annum = US$60,000126 (c) Estimated Pastoral Annual Training Cost127 = US$21,000/annum 126 Based on 20 percent of the cost of WFP proposals to train 560 woreda staff per year in 178 non-pastoral woredas. 141 (d) Estimated Incremental Training Cost (HABP) = US$19,000/annum. (e) Estimated Total Implementation Cost = US$2,222,000/annum, of which $1,907,000/annum will be covered by Government, and US$315,000/annum will be covered by the project, under the Institutional Support Capacity component. 127 The cost of training staff from 62 woredas is calculated pro-rata based on the WFP cost of training staff from 178 woredas. 142 Annex 12: Project Preparation and Supervision ETHIOPIA: Productive Safety Net APL III Preparation Schedule: Planned Actual Project Concept Note Review March 16, 2009 March 16, 2009 Initial PID to PIC March 24, 2009 March 24, 2009 Initial ISDS to PIC March 31, 2009 March 31, 2009 Appraisal September 3, 2009 September 3, 2009 Negotiations September 11, 2009 September 11, 2009 Board Approval October 22, 2009 Planned Date of Effectiveness November 30, 2009 Planned Date of Mid-Term Review June 15, 2012 Planned Closing Date June 30, 2015 Key institutions responsible for preparation of the project: Government of the Federal Democratic Republic of Ethiopia: - Federal Ministry of Finance and Economic Development - Regional Bureaus of Finance and Economic Development - Ministry of Agriculture and Rural Development (Disaster Risk Management and Food Security Sector, Food Security Coordination Directorate, Early Warning and Response Directorate, Natural Resources Management Directorate, and Agricultural Extension Directorate) - Regional Bureaus of Agriculture and Rural Development (Disaster Prevention and Preparedness and Food Security Bureaus) - Federal Cooperative Agency Development Partner Agencies: - Canadian International Development Agency (CIDA) - Department for International Development, U.K. (DFID) - European Commission (EC) - Irish Aid - Royal Netherlands Embassy (RNE) - Swedish International Development Agency (SIDA) - United States Agency for International Development (USAID) - World Bank - World Food Program - The PSNP Donor Coordination Team World Bank staff and consultants who worked on the project included: Name Title Unit Harold Alderman Advisor AFTSP Tesfaye Ayele Procurement Specialist AFTPC Muderis Abdulahi Social Protection Specialist AFTSP Abiy Demissie Belay Financial Management Specialist AFTFM Marylou Bradley Sr. Operations Officer AFTSP Ian Leslie Campbell Consultant, Sr. Environment & Safeguards Specialist AFTSP 143 Sarah Coll-Black Consultant, Social Development Specialist AFTSP Endeshaw Tadesse Gossa Sr. Operations Officer AFTSP Laketch Mikael Imru Sr. Rural Development Specialist AFTAR Renate Kloeppinger-Todd Rural Finance Adviser ARD Josiane Luchmun Program Assistant AFTSP Richard Olowo Sr. Procurement Specialist AFTPC Jonathan Pavluk Sr. Counsel LEGAF Luis Schwarz Sr. Finance Officer LOAFC Carolyn Winter Sr. Social Development Specialist AFTCS William Wiseman Sr. Economist and Task Team Leader AFTSP Shimelis Woldehawariat Procurement Specialist AFTPC 489. World Bank funds expended to date on project preparation: (a) Bank resources: US$120,000.00 (b) Trust funds: US$225,000.00 (c) Total: US$345,000.00 490. Estimated Approval and Supervision costs: (a) Remaining costs to approval: US$70,000.00 (b) Estimated annual supervision cost: Total US$500,000 of which US$200,000 from Bank Budget and US$300,000 from Partnership MDTF. 491. PSNP co-financing development partners will continue to provide resources via the current Bank-administered Partnership MDTF to support the enhanced supervision, coordination, and monitoring of the third phase of Program. Program Supervision Strategy 492. The third phase of the PSNP provides an important opportunity to build on strong partnerships with joint accountability for program results between Government and development partners. While Government and its implementing agencies are responsible for program implementation, responsibility for resolving implementation issues is continuously shared by all stakeholders including the Government and development partners. The third phase of the PSNP will continue to use common planning and budgeting mechanisms as well as joint reviews based on a single set of program indicators developed during the preparation phase. 493. Guiding Principles for Program Supervision. Program supervision will be guided by the following principles: (a) Use of existing Government systems and a harmonized and collaborative approach by development partners. The supervision strategy for the third phase of the PSNP is aligned with existing government planning and budgeting systems, timelines and procedures. PSNP development partners will continue to use joint mechanisms for organizing missions, reviews and assessments. (b) Supervision will be flexible and responsive. Joint supervision will be structured as a flexible process that is able to identify and respond quickly to implementation challenges, while also 144 providing a mechanism for close collaboration and dialogue between the Government, development partners, and other stakeholders. (c) Ensure technical rigor. To respond effectively to the demands of program implementation, joint supervision missions will comprise team members with appropriate technical skills and experience commensurate with Program requirements. (d) Genuine partnership. Much of the success of the Program will depend on the ability of stakeholders to collaborate effectively, and requires Government and development partners to share information openly and regularly. (e) Joint accountability. Government, with the support of the Donor Working Group, is responsible for the successful implementation of the Program. Both parties are jointly accountable to ensure regular dialogue on how to further strengthen program performance. 494. World Bank Program Coordination. Day-to-day handling of Program matters, including all issues related to IDA support, World Bank policies or Bank-administered MDTFs will be the responsibility of the World Bank Task Team Leader who will be supported by a multi-disciplinary core team of both internationally and locally recruited staff. 495. Development partners, including the World Bank, will: (i) collaborate closely with the PSNP Donor Coordination Team; (ii) participate in various PSNP thematic Technical Committees and taskforces to ensure that appropriate targets, technical norms/standards and policies are supported as part of the PSNP dialogue; and (iii) liaise directly as necessary with relevant programs, as well as Government departments and units including sector agencies on program planning and implementation issues. 496. PSNP Donor Coordination Team (DCT). The PSNP DCT supports development partners in their information and analytic needs, while monitoring and coordinating activities of the program, facilitating communications between Government and PSNP partners, and managing the large volume of studies and technical assistance mobilized in support of the Program. The functions, roles, and responsibilities of the PSNP DCT will evolve commensurate with the design and requirements of the third phase of the program. 497. Program Supervision. The supervision strategy will continue to use a number of instruments to review progress and respond to implementation issues. These are a mix of joint Government-development partner initiatives and independent reviews and assessments. These are described below. 498. Joint Review and Implementation Support (JRIS) Missions. Semi-annual JRIS reviews will be conducted around April/May and September/October of each year to review overall program implementation performance and progress towards the achievement of program development objectives. Technical reviews, including audits will be regularly carried out as described in Annex 3 on Results Framework and Monitoring, Annex 7 on Financial Management and Disbursements Arrangements, and Annex 8 on Procurement Arrangements. 499. Mid-Term Review (MTR). A Mid-Term Review will be carried out mid-way in the implementation phase. The MTR will include a comprehensive review of the overall progress with implementation and achievement of program objectives. The MTR will also serve as the 145 forum for reviewing any design issues that may require adjustments to ensure the satisfactory achievement of program objectives. 500. Other External Reviews. The World Bank's Governance and Anti-Corruption Strategy calls for selected external reviews. These reviews are currently being undertaken for the PSNP and will continue through the third phase of the program (these include various social assessments and reviews of program targeting). 501. Complementary Analytic and Advisory Services. During supervision, PSNP development partners will complement their financial support with analytical and advisory services as well as technical support in selected areas. These services may include policy analysis and evaluation, knowledge sharing and dissemination, etc. 146 Annex 13: Documents in the Project File ETHIOPIA: Productive Safety Net APL III Government Documents 1. Ministry of Agriculture and Rural Development. Review of the PSNP: Addis Ababa, Government of Ethiopia, 2009. 2. Ministry of Agriculture and Rural Development. Review of the Other Food Security Program. Addis Ababa: Government of Ethiopia, 2009. 3. Ministry of Agriculture and Rural Development. Food Security Program 2010-2014. Addis Ababa: Government of Ethiopia, 2009. 4. Ministry of Agriculture and Rural Development. Productive Safety Net Program 2010- 2014. Addis Ababa: Government of Ethiopia, 2009. 5. Ministry of Agriculture and Rural Development. Household Asset Building Program 2010- 2014. Addis Ababa: Government of Ethiopia, 2009. 6. Food Security Coordination Directorate. Mid-Term Review Report for the Productive Safety Net Program. Addis Ababa: Ministry of Agriculture and Rural Development, October 2008. 7. Productive Safety Net Program, Program Implementation Manual (PIM), July 2006. 8. Environmental and Social Management Framework (ESMF); Federal Democratic Republic of Ethiopia Productive Safety Net Programme Phase II, October 2006. World Bank Documents 1. Minutes of PSNP Informal Quality Enhancement Review, February 17, 2009. 2. World Bank. Country Assistance Strategy for the Federal Democratic Republic of Ethiopia. Washington D.C.: World Bank, 2008. 3. Mid-Term Review Aide-Memoire of PSNP APL2, October 2008. 4. Project Concept Note, March 16, 2009. 5. Minutes of Project Concept Review, Mach 16, 2009. 6. Formulation Support Mission, PSNP APL III, Aide-Memoire, May 10-21, 2009. Studies and Assessments 1. IFPRI/CSA. Ethiopia Food Security Program: Report on the 2008 Survey. Washington D.C. and Addis Ababa, IFPRI and Government of Ethiopia, 2009. 2. IFPRI/CSA: Ethiopia Productive Safety Net Program Impact Assessment. Washington D.C. and Addis Ababa, IFPRI and Government of Ethiopia, 2009. 3. M.A Consulting Group. Impact Assessment Study Report for PW Component of the PSNP in Ethiopia. Addis Ababa: Government of Ethiopia, 2009. 4. Urban Institute and Birhan Research and Development Consultancy. The Ethiopia Productive Safety Net Program, In the Financial Transparency and Accountability Perception Survey. Washington and Addis Ababa, 2008. 5. Ministry of Agriculture and Rural Development. 2008 1st PSNP Public Works Review. Addis Ababa: Government of Ethiopia, 2008. 147 6. Ministry of Agriculture and Rural Development. 2008 2nd PSNP Public Works Review. Addis Ababa: Government of Ethiopia, 2008. 7. WABEKBON Development Consultants PLC. Roving Appeals Audit of PSNP in Ethiopia. Addis Ababa: Government of Ethiopia, 2008. 8. Devereux S et al. Ethiopia's Productive Safety Net Program: 2008 Assessment Report. Brighton and London: Institute of Development Studies and Overseas Development Institute, 2008. 148 Annex 14: Statement of Loans and Credits ETHIOPIA: Productive Safety Net APL III 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 P113156 2009 ETHIOPIA GLOBAL FOOD CRISIS 0.00 250.00 0.00 0.00 0.00 9.41 -89.33 0.00 RESPONSE PRO P106855 2009 ET-General Educ Quality Improv. (FY09) 0.00 50.00 0.00 0.00 0.00 51.60 0.00 0.00 P096323 2008 ET-Tana &Beles Int. Wat Res Dev Project 0.00 45.00 0.00 0.00 0.00 37.91 -3.00 0.00 P101474 2008 ET-Urban Local Govt Development 0.00 150.00 0.00 0.00 0.00 122.67 -6.56 0.00 (FY08) P101556 2008 ET-Elect. Access Rural II SIL (FY07) 0.00 130.00 0.00 0.00 0.00 132.72 39.39 0.00 P074011 2008 ET/Nile Basin Initiative:ET-SU Interconn 0.00 41.05 0.00 0.00 0.00 36.49 4.92 0.00 P106228 2008 ET- Ethiopia Nutrition SIL (FY08) 0.00 30.00 0.00 0.00 0.00 26.69 -3.00 0.00 P107139 2008 ET-Sustainable Land Mngt SIL (FY08) 0.00 20.00 0.00 0.00 0.00 18.10 -0.50 0.00 P108932 2008 ET-Pastoral Community Develpt II 0.00 80.00 0.00 0.00 0.00 73.44 9.33 0.00 (FY08) P101473 2007 ET-Urban WSS SIL FY07) 0.00 100.00 0.00 0.00 0.00 89.61 6.99 0.00 P098093 2007 ET-Productive Safety Nets II (FY07) 0.00 200.00 0.00 0.00 0.00 24.30 -31.66 0.00 P098031 2007 ET-Multi-Sectoral HIV/AIDS II (FY07) 0.00 30.00 0.00 0.00 0.00 17.60 17.54 0.00 P092353 2007 ET-Irrigation & Drainage SIL (FY07) 0.00 100.00 0.00 0.00 10.00 77.03 30.17 0.00 P091077 2007 ET-APL3-RSDP Stage III Proj (FY07) 0.00 225.00 0.00 0.00 0.00 205.88 30.96 0.00 P079275 2006 ET- Cap. Building for Agric. Serv (FY06) 0.00 54.00 0.00 0.00 0.00 40.64 5.41 -1.64 P094704 2006 ET-Financial Sector Cap Bldg. Project 0.00 15.00 0.00 0.00 0.00 12.51 6.54 0.00 P097271 2006 ET-Electricity Access (Rural) Expansion 0.00 133.40 0.00 0.00 0.00 112.35 94.50 0.00 P074015 2006 ET-Protection of Basic Services (FY06) 0.00 430.00 0.00 0.00 0.00 20.48 -209.56 0.00 P082998 2005 ET-Road Sec Dev Prgm Ph 2 Supl 2 0.00 248.20 0.00 0.00 0.00 142.51 25.98 -5.12 (FY05) P078692 2005 ET-Post Secondary Education SIL (FY05) 0.00 40.00 0.00 0.00 15.00 8.60 21.52 0.00 P078458 2005 ET-ICT Assisted Dev SIM (FY05) 0.00 25.00 0.00 0.00 0.00 16.12 14.39 6.47 P050272 2005 ET-Priv Sec Dev CB (FY05) 0.00 24.00 0.00 0.00 0.00 17.00 8.32 2.06 P074020 2004 ET-Pub Sec Cap Bldg Prj (FY04) 0.00 100.00 0.00 0.00 20.00 15.26 28.38 0.00 P076735 2004 ET-Water Sply & Sanitation SIL (FY04) 0.00 100.00 0.00 0.00 13.00 43.73 37.98 0.00 P049395 2003 ET-Energy Access SIL (FY03) 0.00 132.70 0.00 0.00 0.00 96.19 72.10 67.04 P044613 2003 ET-RSDP APL1 (FY03) 0.00 126.80 0.00 0.00 0.00 21.55 7.01 0.00 P050383 2002 ET-Food Security SIL (FY02) 0.00 85.00 0.00 0.00 0.00 35.06 16.57 -14.13 Total: 0.00 2,965.15 0.00 0.00 58.00 1,505.45 134.39 54.68 149 ETHIOPIA 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. Total portfolio: 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Approvals Pending Commitment FY Approval Company Loan Equity Quasi Partic. Total pending commitment: 0.00 0.00 0.00 0.00 150 Annex 15: Country at a Glance ETHIOPIA: Productive Safety Net APL III S ub- P O V E R T Y a nd S O C IA L S a ha ra n Lo w- D e v e lo pm e nt dia m o nd* E t hio pia A f ric a inc o m e 2007 P o pulatio n, mid-year (millio ns) 79.1 800 1,296 Life expectancy GNI per capita (A tlas metho d, US$ ) 220 952 578 GNI (A tlas metho d, US$ billio ns) 17.6 762 749 A v e ra ge a nnua l gro wt h, 2 0 0 1- 0 7 P o pulatio n (%) 2.6 2.5 2.2 GNI Gro ss Labo r fo rce (%) 2.9 2.6 2.7 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 1- 0 7 ) capita enro llment P o verty (% o f po pulatio n belo w natio nal po verty line) .. .. .. Urban po pulatio n (% o f to tal po pulatio n) 17 36 32 Life expectancy at birth (years) 52 51 57 Infant mo rtality (per 1,000 live births) 77 94 85 Child malnutritio n (% o f children under 5) 35 27 29 A ccess to impro ved water so urce A ccess to an impro ved water so urce (% o f po pulatio n) 42 58 68 Literacy (% o f po pulatio n age 1 5+) 36 59 61 Gro ss primary enro llment (% o f scho o l-age po pulatio n) 91 94 94 Ethio pia M ale 97 99 100 Lo w-inco me gro up Female 85 88 89 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 7 19 9 7 2006 2007 E c o no m ic ra t io s * GDP (US$ billio ns) 10.4 8.9 15.2 19.4 Gro ss capital fo rmatio n/GDP 16.1 19.8 24.2 25.0 Trade Expo rts o f go o ds and services/GDP 6.0 1 1 .4 13.8 12.8 Gro ss do mestic savings/GDP 10.5 13.2 1.5 5.5 Gro ss natio nal savings/GDP 1 1 .9 17.8 15.1 20.7 Current acco unt balance/GDP -4.2 -2.2 -9.1 -4.5 Do mestic Capital Interest payments/GDP 0.6 0.5 0.4 .. savings fo rmatio n To tal debt/GDP 70.5 1 1 3.3 15.3 .. To tal debt service/expo rts 38.3 9.5 7.1 .. P resent value o f debt/GDP .. .. 5.9 .. P resent value o f debt/expo rts .. .. 38.7 .. Indebtedness 19 8 7 - 9 7 19 9 7 - 0 7 2006 2007 2 0 0 7 - 11 (average annual gro wth) GDP 2.0 6.2 10.9 1 .1 1 .. Ethio pia GDP per capita -1.3 3.3 8.0 8.4 .. Lo w-inco me gro up Expo rts o f go o ds and services 1.2 12.8 -0.2 10.2 .. S T R UC T UR E o f t he E C O N O M Y 19 8 7 19 9 7 2006 2007 G ro wt h o f c a pit a l a nd G D P ( %) (% o f GDP ) 30 A griculture 54.3 57.6 47.9 46.3 Industry 13.3 10.7 12.7 13.4 20 M anufacturing 5.5 5.0 4.5 5.1 10 Services 32.5 31.7 39.4 40.3 0 Ho useho ld final co nsumptio n expenditure 79.0 78.8 86.4 83.9 - 10 02 03 04 05 06 07 General go v't final co nsumptio n expenditure 10.6 8.0 12.1 10.6 Impo rts o f go o ds and services 1 .7 1 17.9 36.5 32.2 GCF GDP 19 8 7 - 9 7 19 9 7 - 0 7 2006 2007 G ro wt h o f e xpo rt s a nd im po rt s ( %) (average annual gro wth) A griculture 3.1 4.3 10.9 9.4 40 Industry -1.5 7.8 10.2 1 1 .0 30 M anufacturing -2.8 5.7 10.6 10.5 20 Services 1.5 7.7 12.9 13.9 10 Ho useho ld final co nsumptio n expenditure 3.1 6.4 14.3 8.9 0 General go v't final co nsumptio n expenditure -2.9 6.4 8.1 -3.8 -10 02 03 04 05 06 07 Gro ss capital fo rmatio n -0.8 6.6 18.5 17.4 Exports Imports Impo rts o f go o ds and services 0.7 10.4 18.0 3.8 No te: 2007 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. 151 Ethiopia 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 7 19 9 7 2006 2007 Inf la t io n ( %) D o m e s t ic pric e s 20 (% change) Co nsumer prices -9.5 -6.4 12.3 17.8 10 Implicit GDP deflato r -6.1 4.5 1 1 .6 16.8 G o v e rnm e nt f ina nc e 0 (% o f GDP , includes current grants) 02 03 04 05 06 07 Current revenue 14.5 15.4 16.7 16.3 -10 Current budget balance 2.3 5.5 5.2 5.9 GDP deflat or CPI Overall surplus/deficit -4.1 -1.3 -5.5 -4.9 TRADE 19 8 7 19 9 7 2006 2007 E xpo rt a nd im po rt le v e ls ( US $ m ill.) (US$ millio ns) To tal expo rts (fo b) 391 599 1,000 ,1 1 85 6,000 Co ffee 253 355 354 424 P ulses and o il seeds 9 23 248 258 4,000 M anufactures 74 75 94 105 To tal impo rts (cif) 1,081 1,309 4,592 5,1 26 Fo o d 194 17 333 259 2,000 Fuel and energy 109 147 861 875 Capital go o ds 466 528 1,552 2,020 0 01 02 03 04 05 06 07 Expo rt price index (2000=100) 15 1 132 10 1 1 18 Impo rt price index (2000=100) 99 87 139 151 Exports Import s Terms o f trade (2000=1 00) 1 16 152 79 78 B A LA N C E o f P A Y M E N T S 19 8 7 19 9 7 2006 2007 C urre nt a c c o unt ba la nc e t o G D P ( %) (US$ millio ns) Expo rts o f go o ds and services 623 ,01 1 1 2,105 2,486 0 Impo rts o f go o ds and services ,21 1 7 1,589 5,548 6,266 01 02 03 04 05 06 07 -2 Reso urce balance -594 -578 -3,443 -3,780 Net inco me -48 -97 -38 13 -4 Net current transfers 204 484 2,095 2,885 -6 Current acco unt balance -437 -191 -1,386 -882 -8 Financing items (net) 416 -94 989 1,050 Changes in net reserves 21 285 397 -168 -10 M emo : Reserves including go ld (US$ millio ns) 224 558 ,1 1 58 1,326 Co nversio n rate (DEC, lo cal/US$ ) 2.1 6.5 8.7 8.8 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 7 19 9 7 2006 2007 C o m po s it io n o f 2 0 0 6 de bt ( US $ m ill.) (US$ millio ns) To tal debt o utstanding and disbursed 7,364 10,077 2,326 .. IB RD 57 0 0 0 G: 114 IDA 601 1,532 553 711 B: 553 F: 316 To tal debt service 249 99 164 .. IB RD 13 0 0 0 IDA 9 26 39 6 Co mpo sitio n o f net reso urce flo ws Official grants 346 357 5,015 .. Official credito rs 422 104 306 .. P rivate credito rs 72 23 -45 .. E: 783 Fo reign direct investment (net inflo ws) -3 288 364 .. D: 560 P o rtfo lio equity (net inflo ws) 0 0 0 .. Wo rld B ank pro gram Co mmitments 50 0 281 520 A - IBRD E - Bilateral Disbursements 86 65 179 132 B - IDA D - Other mult ilateral F - Privat e P rincipal repayments 13 15 23 0 C - IM F G - Short -t erm Net flo ws 73 50 157 132 Interest payments 10 11 16 6 Net transfers 64 38 140 126 No te: This table was pro duced fro m the Develo pment Eco no mics LDB database. 9/24/08 152 IBRD 37166 32°E 36°E 40°E 44°E ETHIOPIA ERITREA R PRODUCTIVE SAFETY NET REP. PROGRAM III e OF d ETHIOPIA Axum Adigrat YEMEN TOTAL PSNP BENEFICIARIES: Above 40,000 Humera S 14°N 14°N 30,001 ­ 40,000 T I G R AY e 20,001 ­ 30,000 a Mekele 10,000 ­ 20,000 Below 10,000 No beneficiaries Gonder SELECTED CITIES AND TOWNS AMHARA NATIONAL CAPITAL WEREDA BOUNDARIES Lake Tana A FA R REGION BOUNDARIES Debra Weldiya Bahir Dar Tabor DJIBOUTI INTERNATIONAL BOUNDARIES SOURCE: Government of Ethiopia data courtesy of the World Food Program. Dese SUDAN 48°E Debre BENSHANGUL Markos 10°N 10°N DIRE DAWA Dire Dawa This map was produced by the Map Design Unit of The World Bank. Jijiga The boundaries, colors, denominations and any other information ADDIS Harer shown on this map do not imply, on the part of The World Bank Gimbi ABABA Group, any judgment on the legal status of any territory, or any Nekemte Awash endorsement or acceptance of such boundaries. ADDIS ABABA HARARI Nazret SOMALIA Gambela Degeh Bur Aware Gore Welkite GAMBELA Asela Domo Jima Hosaina O R O M I YA Bonga Shashemene Goba SOMALI Awasa Dodola Warder Sodo Kebri Dehar Wendo Imi 6°N SNNPR Arba 6°N Minch Negele Ferfer Yavello Dolo Odo Lake Mega 0 50 100 150 200 Kilometers Turkana Moyale I ND I A N UGANDA KENYA 0 50 100 150 Miles OC E A N 32°E 36°E 40°E 44°E 48°E SEPTEMBER 2009