53252 IDA16 Enhancing IDA's Capacity to Respond to Severe Crises International Development Association IDA Resource Mobilization Department (CFPIR) February 2010 Selected Abbreviations and Acronyms CRED Center for Research on the Epidemiology of Disasters CRW Crisis Response Window ERLs Emergency Response Loans FDI Foreign Direct Investment GDP Gross Domestic Product GFDRR Global Facility for Disaster Reduction and Recovery GFRP Global Food Crisis Response Program IBRD International Bank for Reconstruction and Development IDA International Development Association IEG Independent Evaluation Group IFC International Finance Corporation IMF International Monetary Fund LICs Low-Income Countries MICs Middle-Income Countries MTR Mid-term Review NGO Non-governmental Organization ODA Official Development Assistance OECD Organization for Economic Co-operation and Development PBA Performance-Based Allocation UN United Nations WBG W orId Bank Group Table of Contents Executive Summary ....................................................................................................................................... i I. Introduction ......................................................................................................................................... I II. Review of Selected Issues for the Establishment and Design of a Dedicated Crisis Response Window ............................................................................................................................................... 3 A. What is the Impact of Crises on IDA Countries? .................................................................. 3 B. What is IDA's Current Role in Crises and Emergencies? .................................................... 6 C. What Would Be the Value Added of an Enhanced Capacity within IDA for Responding to Severe Crises? ...................................................................................................................... 7 D. How Would a CRW Complement Global Capacity to Respond to Crises? .......................... 9 E. How Would the Establishment of a CRW Interact with IDA's Performance Based Allocation (PBA) System? ................................................................................................. II F. What Would Be the Scope of a CRW and Criteria for Accessing Resources? ................... 11 G. Which Countries Would Be Eligible to Access a CRW?.................................................... 14 H. How Would CRW Resources Be Allocated? ..................................................................... 14 I. What Kinds of Activities Would a CRW Finance? ............................................................ 15 J. How Would a CRW Be Financed? ..................................................................................... 15 III. Issues for Discussion ......................................................................................................................... 17 Boxes Box 1: IEG evaluations highlight the effectiveness of IDA's crisis support and provide valuable lessons of experience ...................................................................................... 8 Box 2: IBRD Catastrophic Risk Deferred Drawdown Option (CAT DDO) ............................................... 12 Box 3: Possible Option: IBRD financing combined with an IDA interest subsidy credit scheme ............. 17 Figure 1: From Hazard to Reduced Welfare ............................................................................................... 13 Executive Summary i. Successive global crises relating to food, fuel and finance have severely impacted IDA countries' progress in achieving sustained high growth rates and reducing poverty. Major natural disasters such as the recent devastating earthquake in Haiti and recurrent threats of global health pandemics have further underlined the vulnerability of many IDA countries to exceptional shocks. In light of this, the international community called on IDA to explore a new crisis response mechanism that would provide assistance to low-income countries affected by severe crises. ii. With increasing globalization, countries and economies have become more interconnected. This brings many benefits. but at the same time, problems that develop in one part of the world can spread rapidly to others. Examples include the transmission of a financial crisis from the industrialized countries to the rest of the world or the rapid global spread of communicable diseases. More often than not, it is developing countries that are most vulnerable to the globalization of financial. environmental, epidemiological and other threats. Ill. The on-going global financial crisis provides an example of how quickly problems can be transmitted from one country to the next. The IBRD, based on its ability to borrow from the markets, quickly scaled up its support to middle-income countries. IDA also responded rapidly to the crisis, but unlike IBRD, IDA resources are pre-set by replenishment levels. As a result, IDA concentrated on front loading of existing resources, which has recently been complemented with the establishment of the pilot Crisis Response Window (CRW), approved by IDA's Board of Executive Directors on December 10,2009. iv. IDA plays an important role in a broad range of crises responses, including for example the 2004 Asian Tsunami, Hurricane Mitch, and the more recent establishment of the Global Food Crisis Response Program, Rapid Social Response Program, and Pilot IDA Crisis Response Window. lEG reports indicate that the World Bank is an important lifeline for countries struck by crises and that it has responded effectively to countries in distress. Addressing future, often unpredictable crises, the Bank, and IDA, must continue to improve agility and strengthen the instruments available to it. v. In most cases, IDA acts as part of a broader collaborative response from the international community. This includes bilateral donors, the private sector and NGOs, the IMF, and other multilateral development organizations, acting under the leadership of recipient governments. In past replenishment periods, some 7 to 8 percent of IDA resources have been used for crises response. However, in the absence of a dedicated funding mechanism, this assistance has been provided through an ad hoc approach, generally by restructuring existing projects or reallocating resources within country portfolios, diverting resources from long-term development and poverty reduction activities. Other response mechanisms have included exceptional allocations in the case of natural disasters, which effectively "tax" other IDA countries' allocations. IDA countries also benefit from the creation of multi-donor trust funds which, however, can be slow to set up and become operational and whose funding is contingent on donors' budgetary situation at the time of crisis. - ii vi. A small but dedicated crisis response mechanism would introduce a more systematic approach to IDA funding in response to major~ unanticipated, temporary crises, replacing current ad hoc funding arrangements. The main benefits would include: (a) availability of a rapid financing mechanism that would further enhance the effectiveness of IDA's role in crisis response by reducing the need to engage in ad hoc resource mobilization every time a major crisis occurs; (b) provision of needs-based IDA resources in case of a crisis, that would be additional to IDA~s PBA based country allocations, avoiding the crowding out of IDA support for long-term development; (c) reducing the need for ad hoc, parallel funding during times of crisis; and (d) greater transparency and predictability of IDA funding for crisis response. vii. The design of a Crisis Response Window (CRW) would involve a range of issues that will need to be addressed. These include defining the scope of a new crisis response window, identifying a set of criteria that would guide the types of crisis events for which the new window could be accessed as well as principles that would guide the allocation of resources from the window, specifying the terms on which assistance would be provided and what could be financed using CRW resources. Vlll. The level and sources of funding are key questions in considering the establishment of a new crisis response window. The two basic options that are reviewed are upfront provision of resources that would be used in case of an eligible crisis event and contingent funding, where IDA would seek bridge financing that would be settled in the replenishment round following the crisis event. Past levels of IDA funding for crisis response suggest a potential funding level of the CRW of up to 5 percent of the overall replenishment. ENHANCING IDA'S CAPACITY TO RESPOND TO SEVERE CRISES I. INTRODUCTION 1. The recent global financial crisis, on the heels of the earlier food and fuel crises, has led to an increasing recognition that there is a gap in the international aid architecture to systematically assist poor countries in dealing with the impact of large, external shocks on government core spending and on the poor and vulnerable. The World Bank Group rapidly responded to the onset of these crises. In the case of the financial crisis, ffiRD scaled up lending to middle income countries; IFC introduced a number of new facilities to support the private sector; and IDA front-loaded resources, using fast track procedures. In this context, concerns have been raised about the need to improve crisis response, especially in low-income countries (LICs) where counter-cyclical instruments are scarce. At the October 2009 World Bank-IMF Annual Meetings in Istanbul, the Development Committee "committed to explore the benefits of a new crisis response mechanism in IDA to protect LIes from crises, to be considered as part of the IDA 15 Mid-term Review."l Similarly, at the Pittsburgh Summit on September 24-25, 2009, G-20 leaders asked their ministers to "explore the benefits of a new crisis support facility in IDA to protect LICs from future crises".2 2. IDA has responded to these requests, introducing a two-stage process. In the first stage, IDA developed a pilot Crisis Response Window (CRW) that would assist eligible IDA countries in their efforts to mitigate the immediate impact of the on-going global financial crisis. 3 The pilot CRW was discussed at the IDAI5 Mid-Term Review (MTR) in November 2009, and subsequently approved by IDA's Executive Directors in December 2009. The pilot CRW is now under implementation and funds are expected to be largely committed by the end of the current fiscal year in June 2010. 3. The second stage of the process is to consider how a more systematic approach to severe crisis response might he introduced within the IDA framework so that future crisis response is strengthened. As agreed during the IDAI5 MTR, this discussion would take place in the context of the IDA 16 replenishment negotiations. The present paper provides a starting point for this discussion, and is focused on responding to questions raised during the IDA15 MTR, Board discussions, and subsequent discussions with stakeholders. Subject to the feedback received at the first IDA16 meeting, Management would prepare a subsequent paper for the second IDA meeting, containing a proposal for the establishment of a dedicated crisis response window. 4. IDA has long played an important role in post-disaster recovery and reconstruction, often including taking the lead in damage and needs assessments. IDA is increasingly involved in risk reduction as welL IDA provides critical assistance in response to natural disasters, such as the 2004 tsunami and the recent earthquake in Haiti. IDA also supports responses to economic crises, such as through the 2008 Global Food Crisis Response Program World Bank (2009). Development Committee Communique, Annual Meetings, October 5, 2009. 2 0-20 Leaders' Statement, The Pittsburgh Summit, September 24-25,2009. World Bank (2009). "Proposal for an IDA Crisis Response Window," November 2009. -2 and the ongoing pilot CRW. In recent years IDA played an active role in the control of the spread of avian flu, including through the Global Program for Avian Influenza. Given the fixed IDA envelope, crisis response is typically funded by restructuring of projects or reallocation of existing country portfolios, including front-loading. IDA14 introduced additional flexibility, allowing reprogramming of under-utilized resources within the broader IDA envelope to help with responses to natural disasters. In addition to these mechanisms, trust funds may be mobilized to help IDA countries recover from crises. 5. lEG evaluations indicate that the Bank's crisis related interventions have had higher success ratings than the broader Bank portfolio, but there are downsides to the instruments currently available. Restructuring projects and/or reallocating resources within a country portfolio diverts resources from long-term development and poverty reduction activities, the primary objective of PBA based IDA allocations. A 2006 lEG evaluation of natural disaster responses reinforces this, stating that "reallocations fill an important niche in the Bank's ability to respond to natural disasters ... however, reallocation can be impractical (for countries with few Bank loans) and can be overused (in countries hit by frequent disasters).,,4 Reallocations within the broader IDA portfolio also effectively "tax" other IDA countries and resources can be unpredictable as the amount available at any time depends on overall demand for resources. Finally, trust funds can leverage significant resources for crises relief but may require significant time to set up and receive funds from donors. 6. As a result, the financing of IDA's crisis response activities has tended to be ad hoc, rather than systematic. To be effective, crisis response needs to provide rapid access to countercyclical resources. Consequently, an important consideration for the discussion of how IDA crisis response can be made more effective is how it can be made more systematic. The creation of a crisis response window within IDA is a promising solution. The objectives behind establishing such a mechanism include: (i) timely support in the case of exceptional (i.e., high risk, low probability) crises; (ii) additionality - minimizing the diversion of resources from long term development objectives; (iii) mainstreaming support through IDA, reducing the need for ad hoc, parallel funding during times of crises; and (iv) greater transparency and predictability. 7. In considering how IDA crisis response can be strengthened further, this paper is organized around 10 key questions that would need to be fully fleshed out in a subsequent paper based on guidance from Deputies. They are: · What is the impact of crises on IDA countries? · What is IDA's current role in crises and emergencies? · What would be the value added of an enhanced capacity within IDA for responding to severe crises? · How would a CRW complement global capacity to respond to crises? · How would a CRW interact with the PBA system? World Bank (2006). "Hazards of Nature, Risks to Development: An lEG Evaluation of World Bank Assistance for Natural Disasters." The Independent Evaluation Group. -3 · What would be the scope of a CRW and criteria for accessing resources? · Which countries would be eligible to access a CRW? · How would CRW resources be allocated? · What kinds of activities would a CRW finance? · How would a CRW be financed? 8. The answers to these questions are not meant to be exhaustive, but rather to provide an initial sense of how Management could approach the task of designing a Crisis Response Window within IDA. The paper concludes by setting out key issues for discussion. II. REVIEW OF SELECTED ISSUES FOR THE ESTABLISHMENT AND DESIGN OF A DEDICATED CRISIS RESPONSE WINDOW A. What is the Impact of Crises on IDA Countries? 9. IDA countries are particularly vulnerable to crises and have limited capacity to address them given inadequate access to financial resources, environmental vulnerabilities, infrastructure limitations, limited economic diversification, and often poorly developed formal safety nets. Severe crises can destroy decades of development gains, often contributing to lower growth, destruction of public and private assets, and declining government revenues, which in turn can lead to lower spending on poverty reducing expenditures and social safety nets. This compounds the effects on the poor, who are typically the most vulnerable to crises and need more time to recover because the poor are more likely to live on the most fragile lands, in unsafe structures, and have fewer resources with which to protect themselves and smooth consumption. 10. IDA countries are subject to a variety of crises and emergencies that undermine their social and economic development efforts. These include: (i) natural disasters such as earthquakes, floods, droughts, tsunamis and storms; (ii) health pandemics such as the avian flu, SARS, and the HINl virus; (iii) spill-overs from conflict and political crisis in neighboring countries; and (iv) economic shocks such as the recent global food, fuel, and financial crises. With increasing globalization, countries and economies have become interconnected. This brings many benefits, but at the same time, problems that develop in one part of the world can spread more rapidly to others. This is evident in the transmission of a financial crisis from industrialized countries to the rest of the world and in the spread of communicable diseases. More often than not, it is developing countries that are most vulnerable to the globalization of financial, environmental, epidemiological and other threats. s 11. Natural disasters: The economic and social impact of natural disasters is on the rise, and is more severe in developing countries than in developed ones. According to Haitian authorities, an estimated 220,000 or more inhabitants have died as a result of the January 2010 earthquake, which would rank it among the top natural disasters ever. By comparison, in 2008 the global death toll from natural disasters tripled to 235,000 from an annual average of more World Bank (2009). "Review of IBRD and IFC Financial Capacities - Working with Partners to Support Global Developmcnt through the Crisis and Bcyond." Background paper prepared for the October 5, 2009 meeting of the Development Committee. -4 than 66,000 over the period 2000-2007. Economic losses totaled US$190 billion, more than double the annual average of US$82 billion over the 2000-2007 period. Six of the 10 countries with the highest number killed by natural disasters (per 100,000 inhabitants) in 2008 were IDA countries Afghanistan, Angola, Haiti, Kyrgyzstan, Mongolia and Myanmar. Similarly, five of the ten countries with the highest number affected (per 100,000 inhabitants) were IDA countries - Djibouti, Eritrea, Guyana, Somalia, and Tajikistan. The most frequent natural disasters are floods and storms, while droughts, floods, and insect infestations tend to affect IDA countries with disproportional frequency. 6 . 12. As a share of GDP, natural disasters tend to be more costly in low-income countries and small states. Hurricane Mitch, for example, produced damages of more than 40 percent of GDP in Honduras, and contributed to a rise in extreme poverty from 46 to 49 percent. 7 Nonetheless despite the large number of natural disasters, those where damages exceed 1 percent of GDP are relatively rare. Of an estimated 2,088 disasters that occurred during the period 2006-2008, only 19 resulted in damages of more than 1 percent of GDP; and of that number, 12 took place in IDA countries. 8 13. Health pandemics: Pandemics are particularly frequent in IDA countries. Increased global mobility has also raised the threat of the global spread of pandemics, the Avian Flu, SARS, and the HINl virus being recent examples. Besides their impact on health, pandemics can create immediate economic shocks by their impact on the workforce. However, the main economic effects arise from the uncoordinated efforts by large numbers of individuals to avoid infection. These efforts, such as fleeing from the area of an outbreak or reducing contact with other people, lead to sharpl~ reduced demand for many service sector activities, as occurred during the SARS outbreak. When large, pandemics' health and economic impacts may justify significant globally coordinated preventive measures supported by the World Bank. For example, at the onset of the Avian Flu pandemic, its potential cost was estimated at up to 2 percent of world GDP. IO 14. Economic crises: The financial impact of economic shocks tends on average to be larger than that of natural disasters and generally affects a larger number of countries. The recent financial crisis has also demonstrated that events in high and middle-income countries can have important spillover effects on IDA countries. While advanced and middle-income countries are now showing tentative signs of recovery, low-income countries continue to suffer the consequences of the global economic crisis. Net private capital flows to low-income countries 6 Center for Research on the Epidemiology of Disasters (CRED), Various. World Bank (2006). "IDA Countries and Exogenous Shocks," November 7,2006. Center for Research on the Epidemiology of Disasters (CRED). Various. 9 Brahmbhatt, Milan and Arindam Dutta (2008). "On SARS-Type Economic Effects during Infectious Disease Outbreaks," Policy Research Working Paper 4466, World Bank, Washington, DC. 10 Brambhatt, Milan (2005). "Avian Influenza: Social and Economic Impacts" Mimeo, World Bank, September 23,2005. -5 declined from US$30 billion in 2007 to US$21 billion in 2008 and are projected to have declined further to US$13 billion in 2009. These conditions have significantly reduced growth in low income countries. The global financial crisis has led to a median growth slowdown of 1.8 percent in IDA countries over the past year,l1 while the growth rate in Sub-Saharan Africa is expected to decline from 5.2 percent in 2008 to 1.7 percent in 2009. 15. Economic crises have considerable human consequences, especially for the poor. World Bank estimates l2 suggest that lower economic growth rates in 2009 will trap 46 million more people on less than US$1.25 a day in developing countries than was expected prior to the crisis. Slower economic growth also results in severe fiscal pressure and financing shortfalls, placing core spending requirements for education, health, infrastructure and social protection at risk. Core spending at risk for IDA-only countries was estimated at US$I1.6 billion in 2009 alone. 16. The 2008 global food and fuel crises also highlighted the vulnerability of IDA countries to commodity price shocks. Higher prices pushed many more people into poverty. Without an adequate response, the rise in global food prices threatened to push an additional 100 million people in low-income countries below the poverty line. However, the increase in the aggregate number of poor is only part of the costs of the crises. The more profound consequence is the impact of rising prices on households who were already poor. For those already struggling to meet their daily food and nutrient needs, the double shock of food and fuel price rises threatened basic survival. 17. Economic crises have persistent effects on human development and long-term development outcomes. Such crises force the poorest households to reduce consumption of food, schooling, and basic services, leading to irreparable damage to the health and education of millions of children. The effects of the food and fuel crisis on malnutrition and schooling can undermine years of progress on the MDGs. For example, estimates suggest that the total number of malnourished worldwide might have increased up to 44 million in 2008 alone. A recent World Bank working paper found that economic shocks contribute to pro-cyclical outcomes in child schooling and health, especially in Africa and low-income Asia. 13 In other words, aggregate economic shocks translate into higher infant mortality, malnutrition, and declining school enrollment. Similarly, the 2000/2001 World Development Report noted that "most social indicators either deteriorate or improve at a slower pace during a macroeconomic crisis." Such findings underscore the importance of protecting social expenditures targeted at the poor during times of crises. II World Bank (2009). "Proposal for a Pilot IDA Crisis Response Window" November 23,2009. 12 Ravallion, Martin (2009). "The Expected Impact of the Global Financial Crisis on the World's Poorest," Mimeo, The World Bank, March 2009. 13 Ferreira, Francisco and Norbert Schady (2008). "Aggregate Economic Shocks, Child Schooling and Child Health," Policy Research Working Paper No. 4701, The World Bank, August 2008. -6 B. What is IDA's Current Role in Crises and Emergencies? 18. Reducing vulnerability to disasters and other crises is an integral part of the fight against poverty. Given the vulnerability of IDA countries to natural disasters, a focus on prevention is critical to long-term poverty reduction. In recent years, the World Bank has shifted its focus on disaster management from support after an event to a combination of mitigation support and ex-post disaster assistance to accelerate the transition from relief back to development. Working in close coordination with the UN, the creation of the Global Facility for Disaster Risk Reduction helps to support the development of comprehensive disaster risk reduction programs in client countries. As preventive measures need to be incorporated into longer-term development planning, a particular focus has been on supporting the inclusion of appropriate measures in poverty reduction strategies designed by governments and IDA country assistance strategies (such as Bangladesh. Cape Verde, Guinea-Bissau, Guyana, India, Mozambique, Sri Lanka and Vietnam). IDA also supports countries in strengthening resilience to economic shocks through country assistance programs aimed at reducing structural vulnerabilities, supporting export diversification. strengthening public expenditure management, and developing social safety nets, to cite just a few examples. IDA has also helped introduce innovative approaches to catastrophic risk management products in IDA countries, such as the first-ever weather risk management contract to help Malawi protect itself against the risk of severe drought. 14 19. IDA is one of the largest donors for disaster reconstruction and recovery. Between 1984 and 2009, 389 disaster-related projects (reconstruction and prevention/mitigation) were approved for IDA countries, for a total commitment amount of US$23.6 billion. While the World Bank has stepped up prevention efforts, IDA countries remain vulnerable to crises and will continue to rely on IDA to provide short-term financing for post-crisis recovery and reconstruction and protecting core expenditures. According to information from the OECD's Development Assistance Committee, humanitarian aid offsets less than 10 percent of a country's disaster losses on average. 20. IDA often leads the international response to global crises. This has included the Avian Flu program, the Global Food Crisis Response Program (GFRP), the Rapid Social Response Program, the IDA Financial Crisis Fast Track Facility, and the IDA Pilot Crisis Response Window. The GFRP has delivered to date about US$l billion supporting food crisis response programs in 34 IDA countries. Under the IDA Financial Crisis Fast Track Facility, 11 operations for a total amount ofUS$1.5 billion have been supported. Under the Pilot IDA Crisis Response Window, about US$I.6 billion!5 are expected to be committed by the end of FYII for operations helping to mitigate the impact of the global financial crisis in 56 eligible IDA-only countries. 14 Syroka, Joanna and Antonio Nucifora. 2010 National Drought Insurance for Malawi. World Bank Policy Research Paper No. 5169. January 2010. 15 The US$1.6 billion consists of the US$l.3 billion identified in the pilot Crisis Window Paper (World Bank (2009). "Proposal for an IDA Crisis Response Window," November 2009) and additional donor contributions. -7 21. IDA is uniquely positioned to support effective recovery. In supporting recovery and reconstruction, IDA has demonstrated flexibility, leadership, and technical expertise to build on policy reform and draw in local communities for successful project results. IDA has close working relationships with the governments of member countries, and because of its unique convening power, IDA is well positioned to bring together relevant stakeholders, and coordinate and assemble leading global specialists in various fields of expertise that are needed for crisis recovery. C. What Would Be the Value Added of an Enhanced Capacity within IDA for Responding to Severe Crises? 22. IDA plays an important role in assisting IDA countries affected by natural disasters, health pandemics, or economic crises. However, due to the absence of a dedicated funding mechanism, ad hoc funding arrangements have to be mobilized whenever a country requests IDA's crisis support. When crises occur, resources are often front-loaded from the IDA country allocation and/or reallocated through restructuring the country portfolio and/or ongoing projects. If a natural disaster is particularly severe, IDA can also, under the terms of the IDA replenishment agreements, provide an exceptional allocation. In addition to these mechanisms, trust funds may be mobilized to help IDA countries recover from crises. 23. A systematic approach to crisis response through the introduction of a CRW would ensure greater transparency and predictability in the allocation of resources than the current ad hoc redeployments. The primary objective of IDA resources is to support the efforts of IDA countries to invest in their long-term development. As IDA is limited to the resources available in any given replenishment period, restructuring projects and/or reallocating resources within a country portfolio diverts resources from long-term development and poverty reduction activities for that country. Exceptional allocations for natural disasters can also be impractical as the practice effectively "taxes" other IDA countries. In addition, the lack of a systematic approach to IDA financing of crises response implies limited predictability with regard to IDA's ability to provide crises support in individual cases for countries affected by crises and other development partners. 24. Trust funds can leverage significant resources for crisis relief, but may require significant time to set up and receive funds from donors. While crisis related trust funds will continue to provide an important vehicle for international cooperation for crisis response, the establishment of a CRW would enable IDA to launch an immediate response which could form the basis for any subsequent additional support through trust funds. Furthermore, a CRW would reduce the need for donors to provide funding immediately following a crisis event, independent of their fiscal situation and budgetary timetable or at times when they may be under budgetary pressure themselves and find it difficult to respond. -8 Box 1: lEG evaluations highlight the effectiveness of IDA's crisis support and provide valuable lessons of experience A 2006 lEG evaluation of the Bank's support to countries affected by natural disasters concluded that: · The World Bank is an important lifeline for countries struck by natural disasters and it has responded effectively to countries in distress, but the challenge is to be more proactive. · Natural disasters are risk factors in development rather than interruptions to it and need to be factored into the design of projects and country programs. · The Bank, along with other members of the international community, needs to develop more effective financing mechanisms for emergency assistance and vulnerability reduction. When disaster strikes, immediate funds are needed and often get diverted from development because no contingency funding is available. Loan reallocations are used much more frequently than other types of Bank disaster response in highly vulnerable countries, putting development goals at risk. A 2008 lEG review of lessons from World Bank Group responses to past financial crises notes that programs that focused on areas of World Bank strength - whether public finance, social safety nets, or infrastructure were much more successful than those that tried to cover a broad range of topics. Collaboration across sectors and among the WBG and partners, preparedness, timeliness, appropriateness of instruments, and continuity of follow-on operations have been key determinants of effectiveness. An ongoing lEG evaluation of the WBG's response to the global crisis notes that WBG is in a unique position to lead a crisis response based on its global reach, and as it spans all sectors and covers public and private spaces. It further states that WBG's response to the global crisis has been quick and sizable with a stronger focus on poverty. Of particular relevance for the establishment of a CRW is the lEG finding that IDA countries are heavily affected by the crisis, but that "IDA funding arrangements constrain flexibility to increase disbursements - especially in the outer years of the crisis". lEG thus recommends the establishment of contingent capital arrangements for greater flexibility of response (especially by IDA and IFC). Sources: lEG, 2006, Hazards of Nature, Risks to Development: An lEG Evaluation of World Bank Assistance for Natural Disasters; lEG, 2008, Lessons from World Bank Group Responses to Past Financial Crises; Evaluation Brief 6; and lEG, 2010, The WBG's Response to the Global Crisis: An Ongoing lEG Evaluation. CODE Briefing. January 20, 2010. 25. With the establishment of a dedicated IDA crisis finance mechanism, IDA's response to large-scale crises could be substantially simplified, further improving response time. The 2006 IEG evaluation highlighted in Box 1 above shows that the fastest response to crises has been achieved when: existing operations are used to help with the crisis response (and thus no new operation needs to be prepared); a relationship with the agencies contributing to the disaster response already exists; and no new funding needs to be identified. Other response mechanisms, such as mobilization of trust funds, can contribute to significantly longer response times. In the case of IDA's response to large-scale natural disasters, the lack of additional resources can mean that crucial time is lost in the process of restructuring projects and country portfolios, as well as in mobilizing trust fund resources from donors. The net effect on IDA countries is a delay and/or disruption of ongoing country programs. 26. Providing IDA with the capacity to provide truly additional and contingent resources would both safeguard development expenditures, and ensure a measure of predictability to IDA countries regarding the response they can expect from IDA. To summarize, the main benefits of establishing an enhanced capacity within IDA to respond to crises would include: -9 · Introduce a systematic approach to the financing of IDA support in response to major crises which would replace current ad hoc arrangements, promoting greater transparency and predictability to IDA's crisis response efforts; · Provide additional financing for IDA support in the case of exceptional crises and emergencies, thus ensuring that such support does not come at the cost of diverting IDA resources from longer-term development objectives; · Ensure that a rapid financing mechanism is available that would further enhance the effectiveness of IDA's role in responding in a timely manner to crises and emergencies; · Complement IDA's PBA system that seeks to allocate resources based on expected long-term development impact with a transparent and predictable contingent financing mechanism that would be responsive to extraordinary needs in case of a crisis. A CRW would not address preventive measures. As discussed in the previous section, such measures should be linked to longer-term development programs and, as such, should continue to be addressed through core IDA financing. D. How Would a CRW Complement Global Capacity to Respond to Crises? 27. An effective crisis response requires concerted efforts and collaboration by the international community. Financing needs following severe crises are typically very large, so that no one institution can address the entire response. The World Bank, as the leading global development institution, has been called upon to provide support in many crises and emergencies at the country, regional, and global level. IDA works closely with other agencies engaged in crisis response, including the United Nations, bilateral agencies and NGOs, who frequently focus on the immediate humanitarian relief needed in the wake of a crisis, and other multilateral development banks. In addition, IDA collaborates closely with the IMF in the case of economic crises. Under a CRW, IDA would continue to act in concert with other members of the international community involved in crisis response. 28. Each institution involved in crisis response brings its own expertise to the table. In the case of natural disasters, IDA typically finances reconstruction and recovery. In the case of economic crisis, IDA focuses on protecting core spending on health, education, social safety nets, infrastructure, and agriculture. Examples of the role of other institutions in crisis response include: · UN agencies, including UNDP, UNHCR, UNICEF, and WFP typically focus on immediate emergency and humanitarian assistance. The UN Office for the Coordination of Humanitarian Assistance COCHA) established the Central Emergency Response Fund in 2006, with the objective of providing more timely assistance in response to natural disasters and armed conflicts, with a special emphasis on under-funded crises. . · European UnionfEuropean Commission - the EU established the Fluctuations of Exports scheme (EU FLEX) in 2000 to provide fast disbursing assistance to African, Caribbean, and Pacific countries suffering from shocks leading to loss of export - 10 earnings (10 percent drop for developing countries; 2 percent for the least developed) and deterioration of the public deficit (10 percent). The European Commission Humanitarian Office (ECHO) was created to provide emergency humanitarian assistance, such as medical supplies, foodstuffs, and fuel, to victims of natural disasters and armed conflict. · IMF provides balance of payments support to countries suffering from shocks or natural disasters. In 2005, the Fund established the Exogenous Shocks Facility (ESF) for low-income countries suffering from terms-of-trade shocks, natural disasters, or conflicts in neighboring countries but which do not have a current IMF program in place. Given its relatively less concessional resources, most IMF crisis response has been designed with the expectation that other donors would provide additional concessional resources. As oflate 2009, the ESF has been replaced by two new, relatively more concessional instruments - the Rapid Credit Facility, replacing the rapid access component of the ESF, and the Standby Credit Facility, replacing the high access component of the ESP. · African Development Bank uses its Special Relief Fund, financed through net income transfers, to provide emergency grants of up to US$5oo,000 to countries emerging from conflict or natural disasters. 29. IDA's comparative advantage in crisis response lies in its capacity to offer countries multi-sectoral support focused on the poor, extending from the immediate crisis response to helping countries regain their pre-crisis development path. IDA's experience in areas such as infrastructure rehabilitation, social protection, agricultural production and public expenditure management as well as its platform role are central elements of its comparative advantage for crisis response. This experience is also leveraged in IDA's well established partnership with the IMF in providing crisis support, where the Fund's focus on macroeconomic stabilization and IDA's sectoral and public expenditure focus have important synergies in crisis response. 30. Furthermore, IDA's support to governments in carrying out analytical work during a crisis situation often plays a critical role in assuring an effective crisis response which focuses on the poor and which is consistent with long term development. For example, in the East Asia and Pacific Region, many countries turned to IDA for financing in the aftermath of the financial crisis following IDA's support in designing stimulus packages (Vietnam), or analytical work on issues like labor market development, poverty targeting of social expenditures, and income support programs. More flexible mechanisms of financing from IDA would help further complement IDA's analytical and knowledge work just as in the case of IBRD. 31. An important additional comparative advantage of IDA with respect to crisis response is its global reach. While some of the above institutions have limited country or regional coverage, IDA is in the position to respond to crisis in virtually all of the poorest countries, ensuring that every poor country has quick access to support when a crisis hits. This global coverage would also be important in responding to global crises and pandemics. Recent IDA programs such as the Global Food Crisis Response Program, the Avian Flu Program, or the Global Financial Crisis Response have highlighted the importance of IDA's global reach for an effective crisis response. - 11 E. How Would the Establishment ofa CRW Interact with IDA's Performance Based Allocation (PBA) System? 32. The CRW would complement IDA's PBA system and would represent a deliberately small share of the overall IDA envelope, so that resources continue to be linked largely to likely long-term development outcomes. 16 The PBA system is designed to maximize the impact of IDA resources by distributing resources based on current policy and institutional performance which are good predictors of long-term development outcomes. This is a robust approach to ration limited resources in order to maximize outcomes, but it was neither designed nor intended to support client countries in responding to short-term divergences from long-term development trajectories. There are two major issues which arise in a crisis context: · Within a fixed envelope, countries have to redeploy resources to protect development outcomes already achieved and defer efforts to achieve additional development outcomes. Such redeployment of PBA based resources undermines the effectiveness of IDA's resource allocation mechanism for long-term development. · The allocation of crisis response resources needs to be based on the immediate crisis impact and resulting needs and not on current development performance in order to avoid a situation of "double jeopardy". For example, in the case of fragile states, which have the least flexibility to respond to a crisis from a policy and institutional perspective, small PBA allocations would be quickly exhausted in crisis response. The situation is similar for small states, which have higher fixed costs of economic production and public administration and increased economic and natural vulnerabilities 17 and receive modest volumes of PBA allocations. F. What Would Be the Scope of a CRW and Criteria for Accessing Resources? 33. While the pilot CRW was narrowly focused on assisting countries in mitigating the impacts of the global financial crisis, a dedicated CRW could meaningfully address a wider range of crises and emergencies. Historically, IDA has provided support in a wide variety of crises, including natural disasters, pandemics, and economic crises. Considering IDA's comparative advantages such as its multi-sectoral approach, deep sectoral and country knowledge, and convening power, a wider scope of eligible crises may be appropriate. 34. Regardless of the scope, a CRW would be small relative to the overall IDA envelope. As such, it would focus only on severe crises, where support cannot be provided using regular IDA resources or where the use of regular IDA resources would have a significant impact on the. achievement of CAS objectives (e.g., support for an event like the current global economic crisis v. an average recession; a global food crisis v. seasonal drought; or the effects of a Hurricane Mitch or tsunami v. seasonal floods). The following spells out some of the key considerations 16 If a CRW were truly needs based, it would need to be relatively large in size and scope. The intent behind the creation of a CRW within IDA is to provide some degree of flexibility to improve crisis response and effectiveness. It is purposely envisioned to be small so that the core focus of IDA remains long-term development. 17 For further details on the challenges of allocating resources to small states, see Section IV of World Bank (2007). "IDA's Performance-Based Allocation System: Simplification of the Formula and Other Outstanding Issues," September 2007. - 12 that could guide eligibility for accessing CRW resources. They include the temporary and unanticipated need for exceptional IDA assistance for crisis mitigation, the degree of exogeneity of a crisis, and the one-off nature of a crisis event. 35. ffiRD's Catastrophic Risk Deferred Drawdown Option (CAT DDO) (Box 2) provides some relevant experience regarding the identification of qualifying events, the magnitude of support, and the treatment of moral hazard issues, which could inform the design of the CRW. However, CAT DDOs themselves would not be appropriate in the IDA context, as with performance based country allocations, contingent country specific funding such as a CAT DDO would mean setting aside resources that are better used for long-term development needs. A CR W would allow the pooling of risks, where individual countries do not have to put aside IDA resources for an uncertain crisis event in the future (such as would be the case with a CAT DDO), but where the international community provides some assurance of support in the event of a sever crisis through an IDA CRW. Box 2. IBRD Catastrophic Risk Deferred Drawdown Option (CAT 000) IBRD offers to middle-income country governments the CAT DDO. Its purpose is to make financing immediately available after a natural disaster, like an earthquake or hurricane. It is intended to fill the gap while other sources of funding, such as emergency relief aid, are being mobilized. Countries can access funds from the facility if they declare a state of emergency as a result of a natural disaster. Countries that sign up for the CAT DDO must have an adequate hazard risk management program in place that is monitored by the World Bank. The maximum amount available will be limited to US$500 million or 0.25 percent of a country's gross domestic product, whichever is smaller. 36. Beyond meeting the test of being a "qualifying event," it will also be necessary to identify when thresholds have been met to trigger exceptional support. However, triggers would need to be considered carefully, with an emphasis on overall impact rather than specific parametric triggers like those used for insurance mechanisms (e.g., exact wind speed exceeding specified kilometers per hour or rainfall in excess of specified centimeters) to avoid situations where support is withheld despite extensive damage. Questions to consider include: · How deep is the crisis in terms of fall in GDP, employment, and poverty impact? Research suggests that the effects of disasters on income tend to be non-linear, with severe disasters having disproportionally large effects. Based on a review of the size of disasters and crisis events, a cut-off point of damages and losses exceeding one percent of GDP or resulting in a decline of more than one percent of GDP growth appears reasonable. Most IDA-only countries have allocations that exceed one percent of GDP, which would provide in most cases some headroom to mobilize resources for crisis response from the regular IDA allocation, without entirely derailing progress towards the CAS objectives. While IDA assistance is targeted at mitigating the impact of the crisis on the poor, a precise measurement of the poverty incidence of a crisis event is rarely possible. However, whenever the impact of a crisis on poverty can be assessed, this would also influence the decision as to whether a crisis event warrants access to CRW resources. - 13 · Have all existing support mechanisms been sufficiently employed and exhausted? · What is the Bank's role in the country and what are the other multilateral and bilateral organizations doing to respond to the crisis? 37. Access to a CRW would also depend on the extent to which a crisis is home-grown or externally caused. Recent work has highlighted the difficulties in distinguishing between home-grown and externally caused crises. In many cases, the extent to which an external shock translates into a disaster or a crisis depends on countries' measures for disaster prevention (Figure 1).18 The determination of whether a crisis event warrants access to CRW resources would include some judgment of the role of domestic policies in causing the deterioration of a country's economic situation at the time of a crisis. For example, an economic crisis caused by poor economic management andlor domestic political economy constraints would not be considered as a crisis that merits CRW support. On the other hand, a crisis caused by factors beyond the control of a government (e.g., commodity price drop, major recession by leading trading partner) would support a case for considering the use of CRW resources. Figure 1. From Hazard to Reduced Welfare Hazard (Storms. EarthqU.1kf'S. Flood'>, Droughts) Disaster PREVENT REACT ormatioo Insurance IudiYidUfl15< Remittances lncmtive5 Relief and Aid .l' Institutions '" Governments < L1'N5. Regulation:;, l\lblic Policy l1l.fuIstructuref'Public Goods Source: UN and World Bank, 2009. 38. Attention to Moral Hazard Concerns. A CRW would focus on unanticipated, one-off shocks only since providing support for recurring or ongoing crises would create incentive issues. A systematic capacity within IDA to respond to crises needs to take account of moral hazard issues so that it does not lead countries to take unnecessary risks by reducing spending on prevention and preparedness efforts. Indeed it may be desirable to build into the design of a CRW, incentives for investing in crisis prevention and preparedness. 18 See for example, Keefer, Philip, Eric Neumeyer, Thomas Plumper (2010). "Earthquake Propensity and the Politics of Mortality Prevention," Policy Research Working Paper. World Bank, January 2010. Keefer et al. examine factors that influence governments' decisions to invest in ex-ante preventative measures and mortality in earthquake zones. - 14 G. Which Countries Would Be Eligible to Access a CRW? 39. Eligibility for access to a CRW needs to be considered at two levels. The first level concerns eligibility in principle. In view of the vulnerability of all IDA countries to crises, it seems reasonable that all IDA recipients would have access to a CRW (instead of limiting support to only a subset such as IDA-only countries). General eligibility would be consistent with current access to exceptional allocations in the aftermath of natural disasters, which are available to all IDA countries. For example, in the wake of the 2004 Tsunami, India (a blend country) was granted an exceptional (though modest) IDA allocation. On the other hand, circumscribing eligibility more narrowly (e.g., IDA-only countries) would ensure that limited CRW resources would be focused on countries that have no access to alternative sources of World Bank support such as IBRD assistance. However, experience with the Pilot CRW showed that even blend countries may not be able to access IBRD funds in time of crisis for a variety of reasons. J9 40. The second level relates to the determination of which countries should benefit from CRW resources, once a crisis of sufficient magnitude has been determined. This is of particular relevance for crises and emergencies that affect several countries, such as the global financial crisis. In determining country eligibility for access to a CRW in a specific crisis event, Management would consider the impact of the crisis on individual countries as well as countries' ability to mobilize alternative domestic and external resources for crisis mitigation. H. How Would CRW Resources Be Allocated? 41. Once it has been decided that a particular event justifies access to a CRW and countries which would be eligible in the specific crisis case have been identified, decisions have to be made on the appropriate amount of CRW support, both the overall level of CRW allocation for a particular crisis event and, in the case in which the crisis event affects more than one country, how resources should be distributed among eligible countries. 42. Determination of allocations for individual crises events. In the case of natural disasters, the appropriate level of support from the international community to an affected country is determined based on the basis of damage and loss assessments. For other crisis events such as health epidemics or shocks to current and capital account flows, the World Bank, IMF, or the UN typically prepare assessments of the overall impact of the shock. Based on such assessments, the determination of the appropriate contribution by IDA to crisis recovery efforts would also consider alternative domestic or external resources available for crisis mitigation. 43. Allocations to multiple countries for a specific crisis event. In cases where a crisis affects multiple countries, Management would prepare a framework document that spells out the allocation mechanism to individual countries for approval by the Board. Criteria that would be considered in the allocation of CRW resources would include the severity of the effects of the crisis, a country's access to other domestic and external resources for crisis mitigation, pre existing needs such as the level of income or extent of poverty, and absorptive capacity for the 19 This could include countries that have reached their IBRD exposure limits or that temporarily do not meet IBRD's creditworthiness requirements. - 15 use of CRW resources. Allocations to individual countries will have to be made in consideration of the overall CRW resource envelope. A proposal for a CRW would also need to define upper limits on what share of the CRW an individual country can receive. 44. Experience with the food and financial crises demonstrated that the impact of crises can be very heter<~geneous across countries, which makes it difficult to allocate resources based on a few pre-determined indicators. Given that most IDA countries suffer from weak statistical and monitoring and evaluation capacity, it is difficult to obtain comparable data on the crisis impact and national response capacity in a relatively short time requiring the use of alternative proxy triggers. In many cases, allocation mechanisms will thus need to employ quantitative indicators that measure the impact of the crisis consistently and reliably across all affected countries as well as additional country specific information. The exact nature of an allocation mechanism can only at best be determined once a crisis has commenced. Management would propose to seek case-by-case approval of specific resource allocation criteria in the case of a multi-country crisis response. I. What Kinds of Activities Would a CRW Finance? 45. IDA's country-based development model will be central to a CRW, in order to ensure that IDA's contribution to crisis response is well aligned with a country's specific crisis-related needs. As IDA generally provides crisis response as part of a concerted international effort, the activities supported may depend on the division of labor among the actors involved. Experience has shown that the type of assistance needed depends on the nature of the crisis and specific country circumstances. For example, country demand for support under the Global Food Crisis Response Program has covered a wide range of sectors and instruments. In some countries, crisis support financed seed and fertilizer programs while in others the focus was on social safety nets such as school feeding programs. Since many IDA countries have only very weakly developed formal social protection systems, a poverty focused crisis response often requires IDA to take an opportunistic approach that uses existing structures to protect the poor from the impact of the crisis. This could include the scaling up of community driven development programs, the expansion of labor-intensive work programs, or the expansion of school feeding programs. Consistent with the use of the country-based model, there will thus be no sectoral ear-marking of CRW resources beyond the general requirement that CRW resources be used to help mitigate the negative effects of a crisis. J. How Would a CRW Be Financed? 46. Basic principles for financing the CRW include the need to ensure that the resources countries receive are additional to their long-term development assistance, and that the creation of large, idle pools of "contingent" financing is avoided. Furthermore, resources for crises response should be additional to IDA's overall finances, so that other IDA countries would not implicitly finance another country's crisis recovery through a tax on their IDA allocations. 47. Overall size of the facility. Given the unpredictability of the occurrence and magnitude of crisis events, there is no established methodology for estimating resource needs in advance. · 16 Data on historical crises as well as IDA crisis support shows large year-to-year fluctuations. 2o At the same time resources for crises response should be kept at a level which ensures that IDA resources continue to be allocated primarily based on country performance. A benchmark could be the level of resources devoted to crisis response in the past, which is roughly 7.5 percent of IDA commitments. 21 As support for crises with a limited impact would continue to come from regular IDA allocations, an initial funding level of no more than 5 percent of the overall IDA replenishment might be appropriate. 48. During the pilot CRW phase, financing was provided by redeployment from existing IDA funds and voluntary donor contributions. For a more permanent CRW, financing would need to come through the normal burden sharing arrangements. The basic options include: · Ex-ante funding of the window to have resources available in the case of an eligible crisis event. Such funding would include donor contributions through the replenishment and supplemental resources, but could also include private contributions, contributions from within the WBG, etc. Unused CRW resources could be rolled over into the next replenishment period. The main drawback of this approach is that substantial resources may be sitting idle for a considerable time, rather than being deployed for much needed development purposes. · Ex-post funding, where in the case of an eligible crisis event IDA would mobilize resources by drawing on internal resources or bridge financing (an example of how this could be done from IBRD is discussed in Box 3, further options would be explored in preparation of a CRW). The expectation would be that donors would provide compensation for such resources during the replenishment following the crisis event. Donors could authorize a ceiling for the CRW in advance. In the event in which needs exceed this ceiling, IDA would consult IDA Deputies and the Board as to whether an increase in the ceiling would be appropriate. · A combination of these financing sources could also be envisaged. For example, some ex-ante funding could support crisis events that are more limited in scale, while ex-post funding could be more appropriate for systemic and large crisis events, such as the ongoing global financial crisis. 20 The EM-Dat database on disasters records 7,788 events over the period 1970-2008 with large annual fluctuations in the number of people affected, deaths, and economic damage. Over the period 1997-2008, average annual commitments for emergency recovery loans were US$904 million with a standard deviation of US$320 million. 21 In recent years, the share of IDA commitments for emergency recovery loans was between 7 and 8 percent of total IDA commitments. The 2006 IEG evaluation of the Bank's response to natural disasters estimates that less than 50 percent of the Bank's crisis response are delivered through Emergency Response Loans (ERLs), which would imply that the share of crisis response is significantly higher than 8 percent. On the other hand, in recent years about 45 percent of ERLs provided assistance to post-conflict countries, which is funded from a dedicated IDA window for post-conflict countries. However, the underestimate due to the unavailability of comprehensive data of crisis support delivered through other instruments and the inclusion of post-crisis funding seem to broadly offset each other. - 17 Box 3. Possible Option: IBRD financing combined with an IDA interest subsidy credit scheme The CRW financing would be undertaken through: · An IBRD loan to provide full or partial bridge financing until the beginning of the next replenishment cycle. The expectation would be that donors would provide additional resources to allow recipients to fully refinance the IBRD loans for IDA-only countries and partially refinance (to be defined) for blend countries. The IBRD grace period on principal exceeds the IDA replenishment horizon of 3 years, and therefore no principal payment should normally become due during the time that the crisis loan is on IBRD books. · Interest charges on the IBRD loan could be covered by an IDA creditlgrant. These supplemental IDA credits would be additional to the PBA allocation for the affected country during the replenishment period under implementation. Supplemental IDA credits could cover a high percentage of the IBRD interest charges. The level of subsidy could depend on (i) the poverty status of country, and (ii) debt sustainability, and could go up to 100 percent. · The combination of an IBRD bridge loan and an interest subsidy would minimize the risk to the IBRD portfolio. For this to work, donors would need to agree on (i) the maximum amount IBRD could provide in bridge-financing (e.g., up to 5 percent of the total three-year IDA envelope); under IDA15, this would have amounted to US$2.1 billion; (ii) that IDA would provide an interest subsidy for the IBRD loans; and (iii) the compensation modalities for IDA during the next replenishment. IBRD could consider making such loans available if: · For IDA-only countries: Full donor commitment exists to replace the IBRD loans at the beginning of the next replenishment period. · For IDA Blend countries: Full or partial replacement would be decided as a function of (0 the country's debt sustainability, and (ii) how close the country is from graduation from IDA. 49. Funding for a CRW would be part ofthe overall IDA replenishment (similar to contributions for regional projects) and imply the same rights and responsibilities for donors. Due to the fact that potentially part of a CRW funding could be contingent, the overall replenishment size would consist of donor pledges for unconditional contributions plus a ceiling for contingent contributions that would be paid at the time of the subsequent replenishment. With respect to contingent CRW funding, IDA donors would only benefit from the rights and obligations associated with IDA contributions when actual contributions are made to compensate for any bridge financing. III. ISSUES FOR DISCUSSION 50. Guidance from Deputies is sought on the following key elements of a CRW: · Do Deputies agree that IDA should further strengthen its capacity for crisis response? In particular, do Deputies agree that the introduction of a systematic approach would be beneficial and that a CRW could make IDA's crisis response more effective, transparent, and predictable? - 18 · Do Deputies agree that a CRW should support only severe crises where the scale prevents using regular IDA resources without significant impacts on CAS objectives? In this context, should a CRW cover a wide variety of crises, such as economic, natural disasters and health pandemics? · Do Deputies agree that triggers should focus on impact (e.g., declines in GDP) rather than parametric indicators such as those used in insurance schemes? What should be the balance between quantitative triggers and a principles-based judgment to determine access to a CRW? · Do Deputies agree that 5 percent of the overall IDA replenishment would be an appropriate ceiling for a CRW? What would be an appropriate balance between ex ante and contingent funding of a CRW? Should IDA explore bridge financing mechanisms with IBRD? 51. Based on feedback received during the first consultation with IDA Deputies, IDA management will present the features for a CRW during the second IDA16 replenishment meeting.