56773 IDA16 Technical Note on the Establishment of a Crisis Response Window in IDA16 International Development Association IDA Resource Mobilization Department (CFPIR) September 2010 A B B R EV IA T IO N S AND ACRONYMS CRW Crisis Response Window DaLA Damage and Loss Assessment DSA Debt Sustainability Analysis FY Fiscal Year GDP Gross Domestic Product GFDRR Global Facility for Disaster Reduction and Recovery GFRP Global Food Crisis Response Program GNI Gross National Income HIPC Heavily Indebted Poor Country IBRD International Bank for Reconstruction and Development IDA International Development Association IMF International Monetary Fund JMAP Joint Management Action Plan LIC Low Income Country MDRI Multilateral Debt Relief Initiative PDNA Post-Disaster Needs Assessment OECD Organization for Economic Cooperation in Development RSR Rapid Social Response SDR Special Drawing Right UN United NationsWEO IMF World Economic Outlook TABLE OF CONTENTS I. Introduction .........................................................................................................................1 II. Summary of the Objectives and Basic Design Parameters of the Proposed CRW..............1 III. Review of Past Economic Crises .........................................................................................3 IV. Approach to Triggering and Allocating CRW Resources ...................................................6 V. Coordination with IMF ......................................................................................................12 VI. Financing the CRW............................................................................................................13 VII. Consolidation of Trust Funds for Crisis Response ............................................................14 VIII. Evaluation of the Pilot-CRW Implementation ...................................................................15 Charts: Chart 1: Share of OECD Countries Experiencing a Recession ..................................................4 Chart 2: Share of IDA Countries Experiencing a Given GDP Growth Decline ........................5 Chart 3: Average GDP Growth by Country Category ...............................................................6 Annexes Annex 1: Impact of the Global Economic Crises on LICs.........................................................16 Annex 2: Fund Balances for Programs and Trust Funds related to Crises (US$ million) .........17 Technical Note on the Establishment of a Crisis Response Window in IDA16 I. Introduction 1. At the first IDA16 meeting held in Paris in March 2010, IDA Deputies began to consider the establishment of a more systematic approach to respond to severe crises within the IDA framework. At that meeting, IDA Deputies discussed the broad principles that should guide the design of a Crisis Response Window (CRW). Based on these broad principles, Participants requested a specific proposal for the CRW within IDA16 for consideration at the second IDA16 meeting in Bamako. Participants at the Bamako meeting provided comments and suggestions regarding the proposal to establish the CRW. The purpose of this technical note is to follow-up on this guidance by addressing the key questions that were raised and summarize the proposed CRW operational framework revised in light of that guidance. 2. The note is organized as follows. Section II provides a summary description of the objectives and basic design parameters of the proposed CRW. Sections III through VIII respond to the guidance and questions from Participants. The issues that are addressed include: a review of past economic crises (Section III); the approach to triggering and allocating CRW resources in response to economic crises and natural disasters (Section IV); coordination with the IMF (Section V); the financing of the CRW (Section VI); the consolidation of trust funds for crises response (Section VII); and evaluation of the pilot-CRW implementation (Section VIII). 3. In addition to this paper, the CRW is described in Section III of the main text of the draft IDA16 Deputies' Report. The draft IDA16 Deputies' Report, which will be reviewed at the third IDA16 meeting on October 11-12, 2010, also contains an annex which describes the specific step-by-step procedure for accessing CRW resources in case a natural disaster or an economic crisis of exceptional severity occurs during the IDA16 period. The text of the report includes the clarifications set out in this technical note and can be revised further after the third meeting should that prove necessary. II. Summary of the Objectives and Basic Design Parameters of the Proposed CRW 4. Objectives. Over the years, IDA has been involved in various aspects of crises response, including the provision of exceptional financial support. Such support has been provided in the aftermath of both major natural disasters and when countries are hit by severe exogenous economic shocks such as in the recent global economic crisis. IDA's comparative advantages in crisis response have included its ability to work closely and collaboratively with other organizations (including the UN and IMF), to build on previous analytical work and operational portfolio, its flexibility and global reach, and its ability to link short-term crisis mitigation and long-term development objectives. 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, thus diverting resources from long- term development activities, or through trust funds which may be slow to set up or, in the case of natural disasters, exceptional allocations which effectively decrease the allocations of other IDA countries. -2- 5. The design of the CRW builds on and is consistent with IDA's role in past crises, including most recently the response to the global economic crisis through the pilot-CRW. The main objectives of the CRW are to establish a more systematic approach for IDA to respond to severe crises caused by exogenous shocks; enhance IDA's capacity to provide rapid response and effectively participate in global disaster response efforts; and provide additional and predictable financing to countries hit by crises. · In the case of natural disasters, the CRW would target events that are exceptionally severe. The proposed procedures for responding to such events build on the authority already provided under IDA replenishments to allocate resources on an exceptional basis. · In the case of economic crises, the CRW would target severe economic crises caused by exogenous shocks that affect a number of countries. The objective of CRW support would be to protect core development spending affected by the crisis. The design of the CRW in response to economic crises draws on the process and methodology used for the pilot-CRW, which were endorsed by Participants at the IDA15 Mid-Term Review and subsequently approved by the IDA Board of Executive Directors, but have been updated to reflect further guidance received from Deputies as part of the IDA16 discussions. 6. Principles. To meet these objectives, the proposed design of the CRW is for a transparent and robust process with strong Board involvement and oversight, in which resources would be accessed only as a last resort to assist countries in coping with clearly severe crises and emergencies. Access to CRW resources would be granted only if alternative sources of funding are insufficient and where IDA can participate as part of a concerted international response to a broadly recognized crisis. CRW interventions would be designed to limit moral hazard with resources provided only in the context of crises that are clearly caused by exogenous factors. Furthermore, unless covered by existing operations, those financed by the CRW would need to include components or features designed to help prevent future crises or that would mitigate their impact. 7. Country eligibility. All IDA countries are in principle eligible for CRW support. When dealing with a specific crisis, access to CRW resources would be linked to country-specific circumstances (e.g., the magnitude of the impact of the crisis, access to alternative sources of financing, including IBRD, and ability to finance recovery from natural disasters using the country's own resources). As a result, some countries ­ or categories of countries ­ could be deemed not eligible for CRW support for specific crises events. 8. CRW and role of IDA. The primary purpose of IDA resources is to support long term development in low income countries, which includes building resilience so that they are less prone to being affected by crises and better able to handle the impacts of those that do occur. The CRW, and its resources deployed for emergencies and crises, are a necessary complement to this primary mission and will help countries return to their long-term development paths. Increasing globalization, the threats arising from climate change and environmental degradation, and population pressures that force people to live in areas exposed to natural disasters are all contributing to the need for IDA to be able to support and reinforce a strong international response when emergencies do occur. Creation of the CRW will provide an important additional -3- tool for the international community to respond to severe natural disasters or severe economic crises. 9. Coordination with other donors. As in all of IDA's work in responding to crises, coordination both with the affected countries' governments and with other agencies is critical. IDA will continue to work closely with all its partners and focus its interventions on areas where it has an existing role and key comparative advantages. This means that in the case of severe natural disasters, IDA would complement the primary mandate of the UN for providing emergency relief, with IDA's efforts aimed at supporting safety nets for affected groups and restoring basic physical assets destroyed by the crisis. In the case of severe economic crises, IDA's role would be to help mitigate the impact on vulnerable groups and protect core development spending at risk in health, education and infrastructure. The CRW would thus complement IMF's role as the first responder to the macro-economic effects of severe economic shocks with balance of payments implications. III. Review of Past Economic Crises1 10. At the second IDA16 meeting, Management proposed that access to CRW resources would be triggered in cases when an exogenous shock would lead to at least a 3 percent decline in projected GDP growth in a group of countries, based on the IMF World Economic Outlook (WEO) data. Participants asked for a review of how this benchmark would have fared if it were applied during past economic crises. This section presents the results of a review of past economic crises and provides evidence that the trigger proposed by Management would correspond to a selective use of CRW funds, i.e., that these funds would be deployed only for severe crises caused by exogenous shocks and affecting a significant number of countries. 11. Since 1980, the world economy has experienced three global recessions: 1982, 1991, and 2 2009 (see Chart 1 for the share of OECD countries experiencing recessions). Based on WEO data, these three episodes also correspond to episodes when 30 percent or more of IDA countries suffered a growth decline in line with the proposed trigger of 3 percent (see Chart 2), suggesting the trigger coincides broadly with crises episodes driven by factors exogenous to IDA countries.3,4 More specifically, these three recessionary episodes corresponded to: 1 Unless otherwise indicated, the charts in this section do not include IDA oil exporter countries (Angola, Chad, Republic of Congo and Nigeria), inactive countries (Myanmar, Somalia, Sudan, and Zimbabwe), India and Pakistan. 2 Kose, Loungani and Terrones identify four global recessions during the post World War II period (1975, 1982, 1991 and 2009; see "Out of the Ballpark," Finance and Development, June 2009). In this paper, a global recession is defined as a contraction in world real per capita gross domestic product (GDP) accompanied by a broad decline in various other measures of global economic activity. The current recession is easily the most severe of the four with no previous global recession having so many countries in a state of recession simultaneously. 3 See Claessens, Kose and Terrones "What Happens during Recessions, Crunches and Busts?" November 2008. 4 Note: in more recent years, roughly 10 percent of IDA countries have experienced growth declines of 3 percent or more in any given year; however, Chart 2 demonstrates that a spike in the number of IDA countries experiencing a growth decline of this level, such as the illustrative 30 percent level used in Chart 2, corresponds broadly to exogenous shocks. -4- · The 1982 recession, associated with a variety of factors, including tight monetary policies in several advanced economies, the rapid increase in oil prices, and the debt crisis experienced by a number of Latin American countries. · The 1991 recession, which reflected problems in various regions of the world: difficulties in the U.S. saving and loan industry, banking crises in several Scandinavian economies, adverse effects of an exchange rate crisis on a large number of European countries, challenges faced by the East European transition economies, and the uncertainty stemming from the Gulf War and the subsequent increase in the price of oil. · The 2009 recession linked to the global financial crisis. Chart 1: Share of OECD Countries Experiencing a Recession (percent of OECD countries) Source: Claessens, Kose and Terrones "What Happens during Recessions, Crunches and Busts?" November 2008. 12. Providing CRW support during these three episodes would have satisfied the principles informing the design of the CRW, notably responding to major, unanticipated, and temporary crises.5 These episodes, as shown in Chart 3, correspond to instances where IDA countries have experienced a significant GDP growth decline. During these episodes, the aggregated GDP growth of IDA countries declined on average by 2.0 percentage points compared to the preceding peak. The aggregated GDP growth decline during the most recent episode (coinciding with the fuel, food and financial crisis) is estimated at 3.4 percentage points. 5 The proposed trigger refers to WEO forecasts of GDP growth declines, not actual ones. Evidence from applying the proposed trigger to WEO forecasts over the period 2002 to 2009, however, suggests that it would adequately filter events deserving CRW support. -5- Chart 2. Share of IDA Countries Experiencing a Given GDP Growth Decline 50 (percent of IDA countries meeting the benchmark) 45 40 35 30 25 20 15 10 5 0 Alternative benchmark: at least 2.0 percentage points Proposed benchmark: at least 3.0 percentage points Sources: April 2010 GDF, April 2010 WEO, Staff estimates 13. Using a lower benchmark for the GDP growth decline (i.e., less than 3 percent) may result in opening up the CRW to a less selective filtering of events. The dotted line in Chart 2 illustrates the outcome of using a 2 percentage points GDP growth decline as a first-tier benchmark for granting access to CRW resources. Using this alternative benchmark would have resulted in seven episodes where at least 30 percent of IDA countries experienced a GDP growth decline, significantly more than could be explained by transmission effects driven by recessions in OECD countries. The more frequent number of incidents in which a significant number of IDA countries would meet the threshold would raise the issue of whether triggering the use of the CRW in the case of a projected 2 percent decline in growth would be responding to shocks induced by exogenous factors, as intended, or to endogenous factors or the policy conditions in these countries. -6- Chart 3. Average GDP growth by Country Category 8 6 (percent) 4 2 0 IDA non-oil exporters OECD countries IDA oil exporters Sources: April 2010 GDF, April 2010 WEO, Staff estimates IV. Approach to Triggering and Allocating CRW Resources Process for Responding to Economic Crises 14. Triggering and Allocating CRW Resources. A proposal to IDA's Executive Directors for accessing CRW resources would be triggered by evidence of an economic crisis that is caused by an exogenous shock and affects multiple IDA countries. Specifically, the crisis should be expected to result in a widespread or regional year-on-year GDP growth decline of at least 3 percentage points in a significant number of IDA countries. 6, 7 15. Country eligibility for CRW support will be determined first by the expected impact of the crisis on GDP. A projected year-on year decline of GDP growth of 3 percentage points or more would be the threshold to identify countries that could be eligible for CRW support. This preliminary ring-fencing would be further vetted by an analysis of available fiscal data and other relevant data in line with the CRW objective to protect core spending in the short-term and avoid derailing long-term development objectives. As a result, countries where the crisis did not have a significant fiscal impact could be excluded from CRW support eligibility, even if they did experience the 3 percentage points decline in GDP growth. 16. Management's analysis and recommendations for triggering access to the CRW and on the selection of countries eligible for support would be provided in a note submitted for 6 Limiting CRW support to events affecting a significant number of countries helps address moral hazard concerns. 7 The projected year-on-year GDP growth decline will be assessed using data from the IMF's World Economic Outlook (WEO). -7- discussion by IDA's Executive Directors. Individual projects would follow for formal Board approval. Specifically, the initial note would: · Demonstrate that responding to the crisis is in line with the CRW objectives and its guiding principles.8 In particular, the note would demonstrate that the crisis has a severe impact on a group of IDA countries and that it has been caused by exogenous factors. · Propose the overall volume of CRW resources to be allocated in response to the event and present its rationale. The proposal will need to factor in the nature and scope of the crisis as well as the resources available in the CRW. · Propose the framework for allocating the approved resources across countries and present its rationale. In designing such framework, Management will follow a two-stage approach to allocations akin to the one used under the pilot-CRW, which would provide flexibility to adjust the response in light of additional country-specific information on the crisis impact.9 The framework will be based primarily on the fiscal analysis described in the paragraph below. When designing the crisis-specific allocation framework, Management will also consider including: (i) a base allocation to ensure a meaningful response, particularly for small states; and (ii) a cap to the resources allocated to any one country or group of countries (in the case of the pilot-CRW the cap was set at 5 percent of total resources); such a cap could be particularly relevant in cases where the same event affects countries or groups of countries with different lags to avoid the risk of a first-come first-served approach that leads to depletion of finite resources. 17. The fiscal analysis required at different stages of the CRW process (eligibility, overall envelope and country allocation framework) would cover government revenues, spending and financing plans to estimate the "core development spending at risk". Core development spending at risk is defined as the amount needed to maintain the pre-existing path of spending on education, health, and operations and maintenance of existing infrastructure, and to maintain, or potentially increase depending on the nature of the crisis, spending on safety nets. Countries with the greatest impact as measured by core development spending at risk would receive proportionately more resources than those with a lower impact. 18. The proposed approach is in line with the guidance provided by a number of Participants at the Bamako meeting. The impact of a crisis on core development spending is better captured 8 As outlined in "Enhancing IDA's Capacity to Respond to Severe Crises" presented to IDA Deputies in Paris in March 2010. 9 The pilot-CRW was designed in the midst of the global economic crisis, while the impact was projected to be extremely widespread. At the time, countries with larger pre-existing needs were anticipated to require more support in protecting core development spending. As more information becomes available, one of the lessons learned from the pilot-CRW is that resources were provided to some countries that that were not severely impacted by the global economic crisis in the end. For this reason, future allocation models would not include indicators of pre-existing needs as the pilot-CRW allocation model did. However, pre-existing need could be considered as a factor in the second stage allocation. As noted in the "Proposal for a Permanent Crisis Response Window in IDA" presented in Bamako, the second stage allocation (15 percent of resources in the case of the pilot-CRW) has been particularly useful in providing resources to the most-affected countries where the magnitude of the crisis was not easily detectable during the first stage allocation. -8- by changes in fiscal variables than by changes in GDP growth.10, 11 However, changes in fiscal variables could also reflect factors that are not exogenous (e.g. the country's policy environment). In view of this, Management proposed the above approach, which would ensure that the impact of an economic crisis on both GDP growth and fiscal variables is used. 19. The process for allocating CRW resources would build on the experience of the pilot. In the first stage of this process the bulk (at least 75 percent) of the resources would be allocated, with those countries that have been most severely impacted receiving the highest levels of support. In the second stage, allocations would be adjusted (using the share of resources not allocated in the first round) in light of additional country specific information related to crisis impact, resource requirements and capacity to mobilize an effective response through the use of additional resources. The allocation framework would calculate allocations on a per capita basis (to take account of country-size). 20. A question was raised in the Bamako meeting regarding how the second stage allocations were determined for the pilot-CRW and on the need to further extend this flexibility. While Management has not yet conducted a full review of the implementation of the pilot-CRW (see below), the following two examples give a sense of the kinds of factors that influenced the second stage allocations. · Mongolia. The decision on the Stage 2 allocation (SDR19.7 million) was based on one of the highest crisis impact (GDP growth declined from 8.9 percent in 2008 to 1.6 percent in 2009). · Lesotho. Lesotho was granted an additional Stage 2 allocation of SDR3.3 million over and above the Stage 1 allocation of SDR2.9 million already approved. This decision was based on the very high growth decline (-4.5 percent), the second highest in African CRW-eligible countries. 21. As under the pilot-CRW, the present proposal envisages that such flexibility be allowed during the second allocation round, in order to be able to direct CRW resources to the countries that have been the hardest impacted by the crisis. Management has proposed that the share of CRW resources to be allocated in the second round be increased to 25 percent, an increase of 10 percentage points from the 15 percent level that was used in the pilot-CRW. Management believes this is warranted in order to take account of both country circumstances not captured in the data used to allocate resources during Stage 1, and also to respond to specific high pay-off opportunities. 10 The relation between changes in the GDP and amount needed by a country to maintain its core development spending depends on the component of the GDP affected by the crisis as well as on how that component contributes to government revenues. The first element can vary from crisis to crisis, depending on the main channel of transmission of the crisis (e.g. imports or remittances). The second element varies from country to country. 11 During crises, countries experience significant fiscal challenges: as growth weakens, revenues suffer while at the same time governments face pressure to maintain or increase spending, including on core development spending. Because of these pressures, the countries' fiscal positions tend to deteriorate (in the case of the global financial crisis, macroeconomic projections underpinning the LICs debt sustainability analysis (DSAs) provide evidence of the fiscal impact of the crisis ­ see Annex 1). -9- 22. Exogenous Price Shocks. In addition to global recessions, IDA countries can be affected by exogenous price shocks. Price shocks can have an adverse impact on the poor and serious macroeconomic effects, including on GDP growth, inflation and the balance of payments.12, 13 Measures to mitigate the impact of exogenous price shocks on the most vulnerable segments of the population typically result in increased fiscal outlays.14 At the country level, the poverty and macroeconomic impact of price shocks will ultimately depend on the country's specific circumstances and the policy mix used to address the shocks. 23. The occurrence of exogenous price shocks may not necessarily trigger access to the CRW. For instance, in the event of sharp fluctuations in commodity prices, which primarily affect balance of payments, affected countries can draw on existing IMF facilities. However, access to the CRW may be warranted when price shocks have widespread impact across IDA countries, have a direct impact on poverty, and call for targeted interventions. For instance, food price increases of the magnitude and scope experienced in the first half of 2008, when prices almost doubled, significantly impacted poverty in several IDA countries.15 Higher food prices impact poverty through consumption channels and particularly affect the poor who spend around half of their income in food. The overall impact in any given country would vary depending on whether the county is a net producer or a net consumer of food (which depends on the relative size of the urban population) and whether households are able to substitute between food items. 24. The CRW approach for dealing with price shocks will reflect the heterogeneous impact that these shocks can have across countries. Introducing a CRW trigger to deal specifically with price shocks would raise a number of complicated design issues, including the definition of what would constitute a qualifying shock, equality of treatment across countries and moral hazard. In addition, adding such trigger would add further complexity to the design of the CRW. Considering this, the proposed approach for responding to price shocks through the CRW would be as follows: · The CRW would provide support if a price shock results in a GDP growth decline in line with the proposed trigger (i.e., a 3 percentage points year-on-year GDP growth decline or 12 The 2009 Global Economic Prospects estimates that the 2008 episode of food prices increase has increased global poverty by some 130­155 million people (mostly in urban areas) as well as the poverty deficit (the annual cost of lifting the incomes of all of the poor to the poverty line) by US$38 billion, or 0.5 percent of developing-country GDP. 13 For a detailed discussion of the impact of the recent increase in food and fuel prices see: 2009 Global Economic Prospects, "The Implications of the Global Crisis for Low-Income Countries," IMF, March 2009 and "The Implications of the Global Financial Crisis for Low-Income Countries--An Update," IMF, September 2009. 14 During the 2008 food and fuel price increase (which has been the most marked in the past century in terms of magnitude, duration, and the number of commodity groups whose prices have increased), most countries have reacted by some combination of increased government spending on existing social safety net programs (e.g., subsidies, conditional transfer systems, or food distribution schemes). Others have responded by seeking to hold domestic prices down by reducing taxes or instituting temporary restrictions on exports. 15 Impact of the 2008 food/fuel crises (from GEP 2009): (i) the food bill increased by an average of 2.4 percent of GNI and 8 percent of government expenditures, and in some countries as much as 20 percent of GNI (equal to the total of government expenditures); (ii) overall additional fiscal costs of measures aimed at offsetting higher fuel and food prices were estimated at zero-4.8 percent of GDP, mainly through price subsidies; and (iii) the total cost of reversing the poverty impact of higher food prices in IDA countries was estimated at US$2.4 billion. - 10 - higher for a significant number of IDA countries) as was the case for the oil shock in the early 1980s; · If a price shock does not result in a GDP growth decline in line with this trigger, CRW support could be considered on an exceptional basis if: (i) the shock is broad based and deemed severe in terms of fiscal impact (i.e., additional spending for targeted interventions to protect vulnerable groups); (ii) there is consensus that a concerted international response is needed; and (iii) the existing IDA country allocations are deemed insufficient to provide an adequate response. · As in the case of global and regional economic crises, Management would present to the Board for discussion by IDA Executive Directors its analysis and recommendations for triggering access to the CRW. Individual projects would then follow for formal Board approval. Natural Disasters 25. Triggering and Allocating CRW Resources. In keeping with previous practice IDA would only provide CRW resources in the aftermath of a major natural disaster where existing resources and allocations are insufficient to mount a credible response.16, 17 The existing exception to IDA's allocation rules that allows IDA to provide additional resources in response to major natural disasters has been used sparingly, and this practice will be maintained under the CRW.18 Furthermore, unless addressed by pre-existing projects, CRW resources would be provided under the presumption that a portion of these resources would be used to strengthen disaster resilience. This will offset the incentive for governments to neglect ex ante disaster risk reduction investments. 26. Only disasters that are exceptionally severe and intense would be considered for CRW eligibility. Parametric data on disaster frequency and impact would be used to corroborate the extent to which an event would qualify for CRW resources, but would not be the only basis of eligibility.19 For minor and/or recurring events IDA countries would rely on existing flexibility 16 See "Additions to IDA Resources: Fifteenth Replenishment" Annex, paragraph 7, bullet 4. This exception to IDA's allocation rules has been included in the Replenishment Agreements since IDA14. 17 For example, small states typically have small IDA allocations and the few ongoing credits and grants are often small compared to the magnitude of the disaster. Furthermore, while reallocations can make sense, especially where the disaster has reduced the relevance of the original financing plans, there may be cases where reallocations can undermine the original purpose of a project. 18 Since FY05, 10 countries have received exceptional IDA allocations in the aftermath of major natural disasters. They are India, Indonesia, Maldives, Nepal, Sri Lanka (all in FY05 in the aftermath of the Indian Ocean Tsunami); Haiti and Yemen (in FY09); and Central African Republic, Samoa, Haiti and Bhutan (in FY10). The aggregate amount provided is approximately US$380 million. 19 Parametric data ­ e.g. the magnitude of an earthquake on the Richter's Scale ­ may not always reflect adequately the impact of a disaster. In the case of an earthquake, for example, disaster preparedness and proximity to human settlements are just as important factors that determine the severity of a disaster. For example, while the 1965 Rat Islands earthquake with magnitude of 8.7 was the 10th largest earthquake in human history, it had little impact (no deaths, damage of less than US$10,000). The 2010 earthquake in Haiti, in contrast, had a magnitude of 7.0 but its low depth and proximity to the capital city of Port-au-Prince - 11 - to reallocate resources within the existing project portfolio, and/or reprogram resources in the project pipeline. 27. The process would build on current practice but would also add a number of elements intended to ensure a more systematic determination of whether the impact of a specific natural disaster is sufficiently severe so as to warrant accessing CRW resources. Furthermore, it would also add guidance regarding the amount of resources that would be made available. It is not, however, practical to limit this guidance to a simple formula. This is because IDA countries have very different characteristics, including with respect to their size and absorptive capacity, and may also face different constraints in terms of mobilizing resources for reconstruction, including from their existing portfolio of IDA projects. CRW financing would need to take account of these factors which may vary considerably. 28. IDA management would follow a two stage process that takes account of the need to provide an early signal20 regarding the potential availability and quantum of resources, while also reserving the flexibility to adjust decisions as more information becomes available. In the immediate aftermath of catastrophic natural disasters Management would review available impact data to form an early assessment regarding the need to access CRW resources. As immediate post-disaster impact data will tend to be limited and evolving, this assessment may also take account of whether the affected country has (a) issued a declaration of emergency; (b) requested CRW resources; and (c) requested a Post-Disaster Needs Assessment (PDNA) or a Damage and Loss Assessment (DaLA).21 Lastly, it would take account of the capacity of the World Bank Group to respond without accessing the CRW. If initial information indicates a sufficiently severe impact, Management would include as part of its briefing to the Board on the situation and intended plan of action its view regarding whether the crisis should be eligible for CRW support, and, if so, provide an initial estimate of the size of the allocation from the CRW. This estimate will be subject to adjustment in Stage 2 as better information becomes available. In the second stage, the initial impact data would be validated with the outcomes of PDNA/DaLA and other information, in order to calculate a final allocation. Project documents submitted to the Board for approval would clearly spell out the rationale for accessing CRW resources and for the size of the CRW allocation. 29. The final decision on the size of the CRW allocation will be informed by IDA's past practice and would take account of the following factors: (i) information on the severity of the crisis and cost of recovery from PDNA/DaLAs; (ii) number of affected persons (defined as persons rendered homeless and/or incurred loss of income or livelihood); (iii) estimates of impact on GDP; (iv) availability of resources to respond to the crisis from: (a) the IDA portfolio; compounded by limited disaster preparedness made it one of the deadliest earthquakes on record (220,000 killed and 1,500,000 injured) with damages estimated at US$7.9 billion or 120 percent of Haiti's 2009 GDP. 20 In the case of the recent earthquake in Haiti, where the evidence of massive destruction was available almost immediately, the Bank publically indicated that it would provide exceptional support from IDA within 24 hours of the disaster. 21 PDNAs/DaLAs provide a reliable, internationally recognized and government-owned mechanism to verify the impacts (damage and losses) of a disaster. They would also (a) provide a comprehensive estimate of overall and multi-sectoral disaster recovery needs; (b) incorporate disaster risk reduction as an agreed element of the disaster recovery framework; and (c) reflect multi-stakeholder consensus over sectoral recovery strategies. - 12 - (b) domestic sources; (c) other external financing (including IBRD); and (d) the amount of resources left in the CRW; (v) absorptive capacity; (vi) issuance of UN Flash Appeal; and (vii) country size (e.g., small states status). 30. Use of Flash Appeal to Trigger CRW Resources for Natural Disasters. Management believes that the issuance of a UN Flash Appeal can be a useful input into the determination of the size of a CRW allocation following a natural disaster. There are however several reasons why Flash Appeals are not suitable as an initial (and sole) trigger for access to CRW resources. First, Flash Appeals can be issued for disasters that would not necessarily qualify for CRW funding. This is the case for instance with respect to humanitarian crises where the need for support for reconstruction is limited. Second, Flash Appeals are often issued with a substantial delay. In the case of the Haiti earthquake for instance, the Flash Appeal was issued almost a month later. In contrast, the World Bank announced its intention to provide exceptional IDA resources the day after the earthquake (after the Board had been briefed). Third, explicitly linking CRW resources to Flash Appeals may give rise to an expectation that IDA resources would be used every time one is issued. This could lead to misallocation of CRW resources as well as institutional duplication, as high humanitarian needs may not always correspond to high needs for reconstruction and recovery assistance that CRW resources are targeted at. V. Coordination with IMF 31. Some Participants asked for close collaboration with the IMF in the implementation of the CRW in the case of economic crises. The previous papers on the CRW already describe the distinct and complementary roles of IMF and IDA in the context of economic crisis.22 This note examines more closely the current processes for collaboration and coordination between the two institutions. 32. Current Bank/Fund efforts to strengthen their collaboration arise from the recommendations contained in the Report of the External Review Committee on Bank-Fund Collaboration ("Malan Report"), which was presented to the Boards of the Bank and the Fund in 2007. This report identified three broad priority areas for improvement in Bank-Fund collaboration: · Improving collaboration on country work, including through new procedures for country team coordination; · Enhancing communication between staff of the two institutions working on common thematic issues; · Improving incentives and support for collaboration on policies, review, and other issues. 33. The managements of the Bank and the Fund responded to the recommendations of this Malan report with a Joint Management Action Plan (JMAP), the implementation of which was 22 See "Proposal for a Permanent Crisis Response Window in IDA". IDA policy paper, May 2010. Para 21; and "Enhancing IDA's Capacity to Respond to Severe Crises". IDA policy paper, February 2010. Para 28. - 13 - assessed in a recent paper that was discussed by the Boards of both institutions in March 2010.23 Overall, the report finds that: "JMAP implementation has varied, but has nevertheless had a positive impact on Bank- Fund collaboration. In particular, consultations between country teams, though still far from a universal practice, have improved information flow and helped coordinate work programs. Other factors have also contributed to enhanced collaboration. In particular, the rapid succession of global crises beginning in 2008 has forced the staffs to work together more closely than previously."24 34. This report describes a number of actions to be taken by both institutions to move the JMAP and collaboration forward. Consistent with the recommendations of the Malan report, and in view of the progress made to date these actions are focused inter alia on giving joint country- team consultations a new push and on enhancing clarity on information sharing.25 35. The proposed CRW would follow established procedures for Bank/Fund collaboration, coordination, information-sharing and Board consultation, in particular building on the experience with collaboration since the rapid onset of subsequent crises in 2008. Management would ask IMF staff to provide their views as an input to the paper in which Executive Directors would be asked to trigger access to CRW resources in the aftermath of a severe economic crisis. Furthermore, IMF management would be invited to the Board discussion of this paper so that Executive Directors could seek such clarifications from IMF as they may need. VI. Financing the CRW 36. Choice of financing option. As outlined in the two previous papers on the proposed CRW, there are essentially two ways in which it can be financed. The first is to set aside resources ­ so-called ex ante financing. The second option is ex post financing, where IDA would rely on bridge-financing to provide the resources to countries affected by crisis, which would then be reimbursed by IDA in the subsequent replenishment period. 37. The ex ante option is attractive because of its simplicity, for the fact that the resources will be readily available and that donors will not need to reimburse IDA in the subsequent replenishment. The downside is that IDA resources will be set aside as a contingency, rather than being deployed immediately. The ex post option has the opposite pros and cons: it is a more complex option and donors would be called on to reimburse outlays in a subsequent replenishment; but this option would avoid idle resources and also ensure that crisis-related resources would be additional within the context of a single replenishment period. 38. At the Bamako meeting, views were mixed on the choice of the financing option for the Crisis Response Window (CRW), with some IDA Deputies indicating that they would not be able to support the ex post funding options given their internal budgetary provisions. After careful consideration, and taking account the expressed difficulty some Participants had with 23 See "Implementation of the Joint Management Action Plan on Bank-Fund Collaboration", March 3 2010. SecM2010-0120. 24 Ibid., page 2. 25 Ibid., para 55. - 14 - committing to reimburse IDA under the ex post option, Management proposes that the CRW be financed ex ante for an amount of SDR1,335 million (4.2 percent of the preferred scenario). This amount would include SDR329 million for Haiti. 39. Should additional resources prove necessary, up to the 5 percent maximum cap, Management proposes to solicit additional donor contributions and possibly also to advance reflows on an exceptional basis from the subsequent replenishment period. Should future developments point to a need to raise additional resources within a replenishment period that may not be available through additional donor financing or advance reflows, Management may bring the option of a bridge financing from IBRD back for further consideration through the Bank's formal governance structure (including through prior consideration by the Audit Committee). 40. Allocation for Haiti: On the issue of exceptional financing for Haiti, Management proposes to include the exceptional allocation of SDR329 million within the CRW. This allocation would be financed from the SDR1,335 million financed ex ante. For details on this amount, please refer to the paper "Updated IDA16 Financing Framework and Key Financial Variables" distributed for the third IDA16 meeting on October 11-12, 2010. VII. Consolidation of Trust Funds for Crisis Response 41. A number of Participants at the Bamako meeting recommended that the CRW be accompanied by efforts both to consolidate existing trust funds and reduce the future creation of new trust funds aimed at responding to crises and disasters. 42. Since FY05, donors have provided approximately US$1.57 billion to 16 trust fund programs aimed at different aspects of work related to disasters and emergencies in IDA countries.26 Of this amount, US$1.3 billion (82 percent) has been contributed to five programs: the Multi Donor Trust Fund Program for Aceh Indonesia (US$595 million); Global Food Crisis Response Program (US$472 million); Java Reconstruction Program (US$93 million); the Global Facility for Disaster Reduction and Recovery (GFDRR)27 (US$72 million); and the Haiti Reconstruction Fund (US$64 million). Remaining trust fund programs are smaller and mostly targeted at specific crises (e.g., post-tsunami recovery in Sri Lanka and the Maldives, the earthquake in Pakistan in 2006, and drought management in Africa). 43. Of the resources that have been contributed to these trust fund programs, approximately US$394 million of the funds yet to be committed to specific projects could be considered for reallocation to IDA. About 91 percent of these resources are in three trust funds programs, those 26 Note that this analysis is focused on trust fund programs that have received new resources in the last five years and that are targeted at crises and emergencies similar to those that may be supported by CRW resources. It therefore excludes trust funds aimed at reconstruction after conflict, such as the Afghanistan Reconstruction Trust Fund, and also the Avian Flu and Influenza program. Furthermore, note that the largest single recipient is Indonesia, which graduated from IDA in FY08. 27 Note that the GFDRR is primarily aimed at disaster preparedness, whereas the CRW would be aimed at supporting recovery after a natural disaster has occurred. - 15 - for reconstruction in Aceh and Haiti, and for the food crisis.28 Annex 2 provides detailed information about unused funds. Management plans to engage donors to these trust funds to encourage them to consider reallocating the resources to IDA in the event the funds are no longer needed in the various trust fund programs. However, the extent to which trust funds can be reallocated to IDA will depend in large part on individual donor appropriations processes and/or other legal or regulatory requirements. For some donors, funds earmarked for a specific country may not be available for use by different countries. 44. With respect to the future creation of trust funds for crisis and emergencies that are eligible for CRW financing, Management will encourage donors that would like to provide additional resources for such crises and emergencies do so through the IDA CRW. Such contributions would take the form of "additional contributions" which can be made at any time during the replenishment period (i.e., independent of the replenishment negotiations). Such additional contributions could have a "soft" earmark for a specific crisis or emergency (i.e., the resources would be included in the pool of regular CRW resources, but Management could report back to specific donors on the utilization of these resources so that information on specific crises responses or countries is available). A presumption would be that the decision on whether a specific crisis or emergency meets the CRW eligibility criteria would have already been taken in advance of the consultation on additional contributions from donors. Furthermore, these contributions would be additional to the allocation determined by IDA (i.e., they would not substitute for the allocation determined through the CRW allocation methodology). This would recognize that in many cases, the CRW can only meet a small share of the resources needed to recover after exceptionally severe emergencies and disasters. VIII. Evaluation of the Pilot-CRW and Review of the Proposed CRW 45. Management agrees with the request by several Deputies for a full review of the implementation experience with the IDA15 pilot-CRW and proposes to present the findings for discussion at the time of the IDA16 Mid-Term Review (MTR). The review of the pilot CRW will include an assessment of its timeliness and effectiveness, as well as of its allocation methodology (including stage 2 allocations). Management also agrees with the suggestion to review the proposed CRW at the time of the MTR. This would consist of an update on the implementation so far in the IDA16 period, and also indicate plans for the reallocation of any unused ex ante resources during the last year of the IDA16 period. As the financing for the CRW would cover only the IDA16 period, continued funding would be reviewed within the context of subsequent replenishment negotiations. 28 Note that no new IDA-funded operations can be funded under the Global Food Crisis Response Program (GFRP) after the end of FY10, while operations financed from the GFRP trust fund can be approved after FY10 (i.e., does not have a sunset clause). The sunset clause for the Rapid Social Response (RSR) program, which was also included in the Vulnerability Financing Framework, is end-FY11. - 16 - Annex 1: Impact of the Global Economic Crisis on LICs Given the LIC's low degree of financial integration, the direct financial transmission of the global financial crisis has only played a limited role for most of them. However, reflecting their increased integration in the world economy, the global recession had a major impact on LICs through a sharp contraction in export growth, FDI inflows and remittances. This is reflected in the macroeconomic projections underpinning the Debt Sustainability Analysis (DSAs) of 36 LICs for which a post-crisis DSA was available by March 2010. Specifically: · On average, real GDP growth was revised downwards by 2.5 percentage points in 2009 and 1 percentage point in 2010, but is expected to return to pre-crisis levels in 2011-12. · The new growth and price outlook implies a permanent reduction in the level of nominal GDP, by an average of 10.5 percent in the long-term (relative to pre-crisis DSAs); exports and government revenues follow a similar pattern. · Primary fiscal deficits were revised upwards by an average of about 2 percentage points of GDP in 2009. This reflects both the operation of automatic stabilizers (mostly lower revenues) and, in a number of LICs, discretionary fiscal stimulus (mostly higher non- interest expenditures), both contributing to mitigating the impact of the crisis.1, 2 · Current account deficits, excluding interest payments, are larger over a prolonged period, reflecting mainly lower exports and remittances. The initial deterioration averages about ¾ percentage point of GDP in 2009 and peaks at about 2 percentage points of GDP in 2012. Current account deficits only return to within one percentage point of GDP of their pre-crisis levels around 2021. · FDI, a non-debt creating source of external financing, is substantially reduced in 2009 and 2010 (as a percentage of GDP) but is projected to recover to pre-crisis levels by 2011. Despite this, the DSAs suggest that debt vulnerabilities would remain manageable in LICs. In particular, while the share of LICs facing higher debt vulnerabilities is significant, it has not increased significantly with the crisis. Moreover, post-crisis DSAs project that, on average, the deterioration in debt ratios resulting from the crisis tapers off over the long term with the ratios returning to a downward trend. This conclusion, however, needs to be treated cautiously as it hinges on a few key assumptions: (i) the crisis will not adversely affect medium- and long-term growth; (ii) LICs, after having appropriately adopted accommodative fiscal policies in response to the crisis, will reduce their fiscal and current account deficits to more sustainable levels in the medium and long-term; and (iii) LICs will have access to adequate financing. ________________________ 1 Revenues are projected to decline with GDP, but, in addition, most LICs are projected to see revenues fall relative to GDP, due to the disproportionate impact of the crisis on trade and commodity tax revenues as well as lower compliance. Expenditures are also expected to increase as planned spending increases are maintained in the face of the crisis and about one-third of countries implement discretionary fiscal stimulus. 2 Primary fiscal deficits gradually return to their pre-crisis levels by 2015. Thus, DSAs assume that the policy space used during the crisis is progressively restored as revenues recover and non-interest expenditures (as a share of GDP) are reduced to levels slightly below those projected in pre-crisis DSAs. Sources: The Implications of the Global Financial Crisis for Low-Income Countries--An Update (September 28, 2009) and Preserving Debt Sustainability in Low-Income Countries in the Wake of the Global Crisis (March 29, 2010). - 17 - Annex 2: Fund Balances for Programs and Trust Funds related to Crises (US$ million) Fund Balance as of FY10-Q4 Program Name Trustee Name Donor Recipient Country Trustee level Sub-fund level Grand Total AVIAN AND HUMAN INFLUENZA TRUST FUNDS Mul ti Donor Trus t Fund for the Avi a n a nd Huma n Infl ue nza Fa ci li ty (AHI Fa ci li ty) Mul ti Donor Worl d 9 9 FREE-STANDING TF FOR ECA SD SECTOR UNIT Southea s t a nd Centra l Europe SECO TA-TF for Ca ta s trophe Ri s k Ins ura nce Fa ci l ity (SECE CRIF) Swi tze rl a nd ECA 0 0 FREE STANDING - COFINANCING Ma l di ves Pos t Ts una mi Recovery a nd Recons truction Project EC/EU Ma l di ves - - Trus t Fund for the Ts una mi Emergency Recovery Project for Democra ti c Soci a l i s t Re publ i c of Sri La nka Swi tze rl a nd Sri La nka 0 0 FREE STANDING - SINGLE PURPOSE TF Ca ri bbe a n Ca ta s trophe Ri s k Ins ura nce Fa ci l ity Trus t Fund Mul ti Donor Ca ri bbea n 21 21 GLOBAL FOOD CRISIS RESPONSE PROGRAM Food Pri ce Cris i s Res pons e Trus t Fund IBRD Worl d 3 3 The Mul ti Donor Trus t Fund for Food Price Cris is Res pons e Core - FPCR Core Mul ti Donor Worl d 149 6 155 Rus s i a Food Price Cris i s Ra pid Re s pons e Trus t Fund for Ta ji ki s ta n a nd Kyrgyz Republ ic Rus s i a n Fede ra ti on Worl d 8 - 8 GLOBALFACILITYFORDISASTERREDUCTION&RECOV Ma i ns trea mi ng Di s a s ter Reducti on of the Gl oba l Fa ci l ity for Di s a s ter Reducti on a nd Recovery Mul ti Donor Worl d 17 0 17 Spa ni s h Trus t Fund for Ma ins trea mi ng Di s a s ter Reducti on Ini ti a ti ve of the Gl oba l Fa ci l ity for Di s a s ter Re ducti on a nd Recove ry (GFDRR) - Tra ck 2 Spa i n Worl d 0 - 0 Aus tra l i a n Trus t Fund for Ma i ns tre a ming Dis a s ter Re duction of the Gl oba l Fa ci li ty for Di s a s ter Reducti on a nd Recovery Aus tra li a Worl d 0 0 0 Ca ll a bl e Funds for the Sta ndby Recovery Fina nci ng Fa cil i ty of the Globa l Fa cil i ty for Dis a s te r Reduction a nd Re covery (GFDRR) Mul ti Donor Worl d 1 - 1 Sta ndby Recovery Fi na nci ng Fa cil i ty of the Globa l Fa cil i ty for Di s a s te r Reducti on a nd Recovery Mul ti Donor Worl d 1 0 1 South-South Coope ra ti on for Ma i ns trea mi ng Di s a s ter Reducti on Mul ti Donor Trus t Fund unde r the Gl oba l Fa ci l i ty for Di s a s ter Reducti on a nd Recovery Mul ti Donor Worl d 0 - 0 MULTIDONOR TF PROGRAM FOR ACEH,INDONESIA Mul ti Donor Trus t Fund for Ace h a nd North Suma tra ( MDTFANS) Mul ti Donor Indones i a 129 2 131 Indones i a : Ma i ns trea mi ng Envi ronment i n Aceh a nd Nia s Project Denma rk Indones i a - - JAVA RECONSTRUCTION PROGRAM Ja va Recons tructi on Fund (JRF) Mul ti Donor Indones i a 19 2 21 IFC TA TRUST FUNDS The Fi na ncia l Support of the Globa l Fina nci a l Cri s i s Progra m Aus tri a Worl d 0 0 PAKISTAN EARTHQUAKE DISASTER RECOVERY Trus t Fund for Ea rthqua ke Di s a s ter Recovery i n Pa ki s ta n IBRD Pa ki s ta n - - RAPID SOCIAL RESPONSE PROGRAM Ra pi d Socia l Res pons e Mul ti-Donor Trus t Fund Mul ti Donor Worl d 25 25 RAPID SOCIAL RESPONSE CATALYST Ra pi d Socia l Res pons e Ca ta l ys t Trus t Fund Uni te d Kingdom Worl d 0 0 TSUNAMI RELIEF FUND Trus t Fund for Ts una mi Dis a s te r Recove ry i n India IBRD Indi a 1 1 VIETNAM FREE STANDING TRUST FUND PROGRAM Vi etna m Na tura l Di s a s ter Ri s k Ma na gement Project Netherl a nds Vi etna m - - HAITI RECONSTRUCTION FUND Ha i ti Recons tructi on Fund Mul ti Donor Ha iti 64 64 384 10 394