May 2017 Assessing Financial Protection against Disasters: A Guidance Note on Conducting a Disaster Risk Finance Diagnostic Photo: Nugroho Nurdikiawan Sunjoyo / World Bank Assessing Financial Protection against Disasters: A Guidance Note on Conducting a Disaster Risk Finance Diagnostic Copyright © 2017 by International Bank for Reconstruction and Development / The World Bank and Asian Development Bank This work is a product of the staff of the World Bank and the Asian Development Bank (ADB) with external contributions. The opinions, findings, interpretations, and conclusions expressed in this work are those of the authors and do not necessarily reflect the views or the official policy or position of The World Bank, its Board of Executive Directors, or the governments they represent or of the Asian Development Bank (ADB), its Board of Governors, or the governments they represent. The World Bank and ADB do not guarantee the accuracy of the data included in this work and do not make any warranty, express or implied, nor assume any liability or responsibility for any consequence of their use. This publication follows the World Bank’s practice in references to member designations, borders, and maps. The boundaries, colors, denominations, and other information shown on any map in this work, or the use of the term “country” do not imply any judgment on the part of The World Bank or ADB, their respective Boards, or the governments they represent, concerning the legal status of any territory or geographic area or the endorsement or acceptance of such boundaries. The mention of any specific companies or products of manufacturers does not imply that they are endorsed or recommended by either The World Bank or ADB in preference to others of a similar nature that are not mentioned. Rights and Permissions The material in this work is subject to copyright. Because The World Bank and ADB encourage dissemination of their knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Assessing Financial Protection against Disasters: A Guidance Note on Conducting a Disaster Risk Finance Diagnostic Acknowledgements This note was authored by Martin Luis Alton and Olivier Mahul (Disaster Risk Financing and Insurance Program, World Bank) and Charlotte Benson (Asian Development Bank), with contributions from Benedikt Signer and Naomi Cooney (Disaster Risk Financing and Insurance Program, World Bank) and from Arup Chatterjee (Asian Development Bank).  The final note was greatly enhanced by reviews from Hiroyuki Aoki, Emre Balibek, Ugo Gentilini, Eugene Gurenko, Md. Golam Mortaza, Mayumi Ozaki, Rommel Rabanal, Joaquin Toro, Hanna Uusimaa, and Maria Vicedo Ferrer, and by inputs from Bianca Adam, Sevara Atamuratova, Antoine Bavandi, Ana Campos, Daniel Clarke, Julie Dana, Hideaki Hamada, and Jose Angel Villalobos. Anne Himmelfarb’s thorough editing ensured the final note remained readable. Design and layout were by Studiografik. The team gratefully acknowledges funding from the Global Facility for Disaster Reduction and Recovery (GFDRR) and the Swiss State Secretariat for Economic Affairs (SECO). ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 3 Table of Contents 2 Acknowledgments 6 Abbreviations 7 Executive summary 8 Introduction 11 A. Disaster risk and the impact of past disasters 11 1. Basic facts of past disasters 11 1.1 Hazard profile and history of disasters 11 1.2 Human toll of past disasters 11 1.3 Geographical distribution of impacts 12 2. Existing disaster risk assessments 13 3. Economic impact 13 3.1 Impact of past disasters by sector 14 4. Fiscal Impact 14 4.1 Contingent liabilities 15 4.2 Foregone revenue 16 5. Impact of past disasters on the poor 16 6. Key outputs 16 7. Resources 19 B. Existing legal and institutional framework and financial instruments for disaster risk finance 19 1. Objective 19 2. The legal and institutional framework for disaster risk finance 19 2.1 Laws governing the budget process and current practice 19 2.2 Post-disaster budget execution 20 2.3 Law(s) on disaster risk management and disaster risk finance 21 2.4 Institutional setup for disaster risk management and finance 21 2.5 Local government 4 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 21 3. Existing disaster risk finance mechanisms and instruments 21 3.1 Risk layering 22 3.2 Ex ante vs. ex post disaster risk finance instruments 23 3.3 Ex ante disaster risk finance instruments 23 3.4 Ex post disaster risk finance instruments 25 4. Key outputs 25 5. Resources 26 6. Indicative list of key counterparts to meet 27 C. Domestic insurance market review 27 1. Objective 27 2. Insurance penetration and insurance market 27 3. Regulatory environment 27 4. Property catastrophe insurance products 28 5. Agricultural insurance 28 6. Microinsurance 28 7. Resources 28 8. Indicative list of key counterparts to meet 29 D. Funding gap analysis 29 1. Objective 29 2. Estimating funding gaps 30 3. Performing a funding gap analysis in data-constrained environments 30 3.1 Short-term funding gap analysis 30 3.2 Long-term funding gap analysis 30 3.3. Funding gap estimates based on individual past disasters 31 4. Key output 31 5. Resources 33 E. Summary and options for consideration 33 1. Objective 33 2. Summary and gaps 33 3. Options for consideration 35 Glossary ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 5 37 References 39 Annex I: Disaster risk finance diagnostic—Questionnaire 39 A. Questions for national disaster management agency on DRM policy and strategy and disaster data 40 B. Questions for ministry of finance on fiscal management of disasters 40 Assessment of fiscal shocks associated with disasters 41 Ex ante disaster risk finance 42 Ex post disaster risk financing 43 Post-disaster expenditure 43 Budget execution 43 Sharing of responsibilities between federal/national and local governments 43 C. Questions for insurance regulator on the domestic catastrophe insurance market 45 Annex II: Template outline for a disaster risk finance country diagnostic 47 Annex III: Risk data and models in developing countries 48 Annex IV: Undertaking a basic financial risk assessment using historical data 48 1. Data collection 49 2. Data scaling 49 3. Basic statistical analysis 50 4. Basic risk assessment presentation 50 5. Further statistical analysis 51 Annex V: Contingent liabilities—Key concepts and good practice 51 1. Objective 51 2. Forms of liability 52 3. Measuring liabilities 52 4. Enhancing measurement of liabilities 52 5. Disclosing liabilities 54 Annex VI: Example of information on contingent liabilities due to disaster found in public documents 6 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC ABBREVIATIONS ADB Asian Development Bank AEL annual expected loss CIMA Conférence Interafricaine des Marchés d’Assurances DRM disaster risk management CPEIR Climate Public Expenditure and Institutional Review ECLAC Economic Commission for Latin America and the Caribbean FTS Financial Tracking Service GAR Global Assessment Report on Disaster Risk Reduction GDP gross domestic product GFDRR Global Facility for Disaster Reduction and Recovery GIS geographic information system IMF International Monetary Fund MoF ministry of finance PDNA Post-Disaster Needs Assessment PML probable maximum loss PEFA Public Expenditure and Financial Accountability PER Public Expenditure Review UNISDR United Nations Office for Disaster Risk Reduction ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 7 Executive summary The World Bank and Asian Development Bank have worked in ƒƒ Framework governing resource mobilization and more than 50 countries to (i) quantify the economic and fiscal execution. Reviews processes for the mobilization and impact of disasters; (ii) take stock of existing mechanisms to execution of disaster risk finance, including for both finance these costs and analyze their legal and institutional government and donor- funded mechanisms. underpinnings; (iii) review aspects of the insurance and ƒƒ Cost-sharing arrangements. Analyzes explicit capital markets that are relevant for disaster risk finance; and and implicit cost-sharing arrangements for (iv) estimate potential funding gaps following disasters. This disaster response between the national and note provides guidance on how to conduct such a diagnostic subnational governments. exercise in a systematic and comprehensive manner, covering the following areas: 3. Review of domestic insurance and capital markets. Reviews the legal and regulatory framework for 1. Assessments of the impact of past disasters: catastrophe risk insurance; reviews market-based ƒƒ The economic, fiscal, and social impact of disasters. financial instruments available to homeowners, famers, Collates information on the impacts of past disasters and businesses to protect their assets and livelihoods based on historical data and/or simulated impact data against disasters. from catastrophe risk models and relevant reports. 4. Funding gap analysis. Analyzes historical data in order to 2. Assessment of the current approach to disaster risk estimate the fiscal funding gap, defined as the difference finance, including between estimated fiscal costs and funds available to meet those costs; discusses timing of expenditure requirements ƒƒ Legal and institutional foundations for disaster for relief, recovery, and reconstruction (short-term and risk finance policies and instruments. Analyzes the long-term funding gaps), and timing of disasters relative existing legal and institutional framework that defines to the fiscal calendar. and governs mandates, responsibilities, budget execution, and accountability within a government for 5. Options for improved financial protection. Uses the financial protection against disasters. findings of the above sections to assess gaps and/or areas that need strengthening, and lays out potential next steps ƒƒ Portfolio of disaster risk finance mechanisms. Uses for the government’s consideration. the risk layering framework to document existing disaster risk finance instruments (including those used by development partners) and related funding amounts. 8 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC Introduction Disaster risk finance aims to improve the ability of governments The proposed diagnostics aim to serve as a basis for evidence- to clarify and meet obligations arising from shocks caused by based dialogue with governments. A diagnostic can be thought disasters while minimizing threats to development progress of as a living document, to be updated as new information and fiscal stability. Increasingly, governments seek to meet emerges and the dialogue advances. Rather than capturing this objective and manage the financial impact of disasters definite findings, it can serve as an evolving repository for triggered by natural hazards in a comprehensive and strategic key disaster risk finance information for a country. way. Development partners have stepped up their support for The main audience of a diagnostic is government officials, such efforts over the last 15 years, and have helped countries in particular from ministries of finance. While development estimate their financial exposure, take stock of existing partners will typically lead the diagnostic work, it is mechanisms to finance disaster response and reconstruction, important to identify a technical focal point in the ministry and develop policy frameworks, implementation plans, and of finance early in the process. Such a focal point will be financial instruments to manage the financial risk from crucial to validate preliminary findings and help coordinate disasters efficiently. the exercise. The purpose of this note is to help development practitioners In addition to the guidance note itself, this document gather relevant information, conduct analysis, and present comprises a questionnaire to help practitioners gather key both in a standardized diagnostic framework. Such a information for a diagnostic (annex I)1 and a suggested diagnostic serves three purposes: (i) it allows governments outline for a country diagnostic (annex II). to assess their level of financial protection against disasters and gives them an overview of current policies and This guidance note treats five main areas: impact of past mechanisms for financial protection; (ii) it serves as the disasters (section A); current approach to disaster risk foundation for identifying specific gaps and setting policy finance (section B); domestic insurance and capital markets priorities for implementing reforms and introducing new review (section C); funding gap analysis (section D); and financial instruments to strengthen financial resilience; and summary and options for consideration (section E). The (iii) it provides the basis for new or deepened engagements sections mirror the structure of the diagnostics to which this on disaster risk finance by international partners, as part document should give rise. Each briefly sets out the objective of the broader disaster risk management (DRM) and/or of the section and explains relevant concepts, discusses key public financial management dialogue. The findings of the information that should be captured, and lists expected diagnostic can feed into the development of disaster risk outputs (e.g., graphs or tables). Where relevant, sources for finance strategies, which set out policy priorities aimed at important data and information will be provided. meeting post-disaster financing needs in a strategic way. 1. Much information will be obtainable through desk research or expert interviews and can subsequently be validated by government officials. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 9 The proposed diagnostic approach is applicable in any Finally, in many countries data constraints may be severe country, regardless of income and capacity level. However, and hamper the analysis of numerous factors that determine depending on country context, different aspects of the a country’s level of financial protection against disaster outlined issues may need to be emphasized. In low-income risk. However, diagnostics based on the best available countries, for instance, foreign aid and humanitarian data are generally still useful for initial engagement with organizations in particular play a large part in the response governments on disaster risk finance, even if they will not be to disasters; thus their role in these countries will probably as comprehensive as suggested in the guidance note. receive greater emphasis than it will in middle-income countries, where a relatively greater focus on government is likely warranted. Nonetheless, the overall framework that underlies the proposed diagnostics is widely applicable, as are the questions practitioners should seek to answer at the beginning of an engagement on disaster risk finance. 10 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 11 A. Disaster risk and the impact of past disasters Governments and other stakeholders are often not aware the Philippines). The history of disasters should be told in a few of the past or potential future scale, exact nature, and paragraphs. This overview of prevalent risks and past disasters distributional implications of disaster impacts. To the can serve as a basis for initial conversations on disaster risk in extent that available data permit, a new engagement with a country. a government on the financial risk from disasters should be based on information about the hazards a country 1.2 Human toll of past disasters faces, the impact of past disasters and associated losses, Next, the human toll of past disasters should be described. and assessments of disaster risk. The first section of the This includes the number of lives lost, the number of people diagnostic note should provide this information. injured, and the number of displaced persons. An overview should be provided in a summary table. See for example table 1 , taken from a disaster risk finance note for Serbia. 1. Basic facts of past disasters This information, together with information on damage 1.1 Hazard profile and history of disasters and losses, is most likely held by the national statistics office or national disaster management agency. However, The first step in diagnosing a country’s financial vulnerability to the quality and depth of information will vary significantly disasters is to clarify what types of natural hazard the country across countries. faces and the frequency with which they occur at varying levels of intensity. A brief description of the country’s risk profile, 1.3 Geographical distribution of impacts including its geographical location, hydrometeorological and geophysical profile, topography, and surface characteristics Next, a description of the geographical distribution of (such as forestation), provides the necessary context for disasters should be provided. This step helps clarify where understanding what types of hazard can occur. The risk profile the highest impacts of disasters can be expected in the future, can be illustrated with historical examples, which may loom and may allow for more focused disaster risk finance efforts large in the national consciousness (e.g., Typhoon Haiyan in in particularly disaster-prone areas. 12 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC Table 1: Illustrative example - number of people affected by major disasters, by type of hazard: Serbia, 2000–2013 Type of hazard No. of events No. of deaths No. of people affected Contamination 4 0 2,650 Drought 45 0 9,100 Earthquake 1 3,106 9,164 Epidemic 12 0 2,230 Explosion 21 4 15,353 Fire 261 228 1,536 Flash flood 6 188 6,986 Flood 234 2 122,151 Forest fire 490 0 1,947 Frost 13 0 0 Hailstorm 134 0 46,652 Landslide 42 50 1,502 Leak 12 0 100 Snowstorm 106 12 140,275 Storm 24 0 101,953 Other 16 0 5,950 Total 1,421 3,590 467,549 Source: UNISDR, Desinventar database, http://www.desinventar.net/index_www.html. Adapted from World Bank and GFDRR (2016) Note: The Desinventar database contains information up to 2013. These figures do not include the catastrophic 2014 floods. Information is most likely available from a country’s national 2. Existing disaster risk assessments disaster management agency, from relevant technical/ This subsection of the diagnostic should describe what type of research agencies and institutes (including universities), risk assessments have been undertaken in the country, if any, and from international partners such as the United Nations and how up to date they are (see annex III for a discussion of Office for Disaster Risk Reduction (UNISDR). Key questions risk data and models in developing countries). It should focus that should if possible be answered in this subsection include on the quantification of disaster risk in monetary terms. the following: Does the assessment cover all types of natural hazard? Does it cover technological hazards as well? Or does This subsection should discuss geographical area of coverage, it cover only a subset of natural hazards? What exposure types of hazards assessed, quality of data, and the granularity data are captured in the model? How has vulnerability of existing risk assessments. It should also indicate if open been determined? access catastrophe models exist. Finally, it should provide information on how, and how extensively, the results of When risk assessments or risk models are not available, a any risk assessments have been communicated to the wider basic preliminary risk assessment can be completed using public, especially in the context of financial preparedness for limited historical data. This can be used to inform initial disasters. This information is needed to ensure that going discussions and analysis. The steps entailed are described in forward gaps can be adequately addressed. annex IV. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 13 Table 2: Template for table on economic impact of past disasters Economic impact (damages + losses), by sector US$ Local currency Social Housing Education Health Culture Nutrition Infrastructure Water & sanitation Community infrastructure Energy & electricity Transport & telecommunications Productive Agriculture, livestock, and fisheries Commerce & industry Tourism Total economic impact 3. Economic impact 3.1. Impact of past disasters by sector This subsection of the diagnostic should describe the If relevant data exist, this subsection should discuss the economic costs of recent disasters. This information may be disaggregated impact on different sectors such as housing, available from Post-Disaster Needs Assessments (PDNAs), transport, and agriculture, among others. Disaggregated data from regular economic monitoring reports such as the are often available only for disasters for which a PDNA has International Monetary Fund (IMF) Article IV reports, been conducted. Data on impacts of past disasters by sector or from annual budget speeches. In some cases, however, can be presented in the format of table 2. practitioners may need to estimate economic impacts themselves using available information on physical damage and economic disruption together with available economic data. Even crude estimates can be useful for illustrating the potential economic impact of disasters in the policy dialogue with government counterparts. Estimates of total damages and losses (see box 1) should be presented in monetary terms as well as relative to gross domestic product (GDP) and government expenditure. This approach puts damage and loss estimates in context and helps determine their wider macroeconomic significance. 14 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC Box 1: Damage vs. loss International organizations and governments increasingly follow a standard methodology when estimating the economic impact of major disasters. This methodology was originally developed by the United Nations Economic Commission for Latin America and the Caribbean in the 1970s and was subsequently refined by the European Union, World Bank, and the United Nations Development Programme. The methodology underpins Post-Disaster Needs Assessments and involves detailed examinations of the impacts and consequences of major disasters. The definitions used in PDNAs are as follows: • Damage is the replacement value of physical assets wholly or partly destroyed, built to the same standards that prevailed prior to the disaster. • Losses are the foregone economic flows resulting from the temporary absence of the damaged assets and/or due to any other disruption of economic activity caused by the disaster. Sources: ECLAC 2003; UNDP 2013. Information on past realizations of explicit contingent 4. Fiscal Impact liabilities after disasters4 can often be obtained from PDNAs, This subsection should describe the fiscal impact of major budget outturn documents, IMF Article IV reports, or other past disasters. Disasters impact government finances through official reports. PDNAs include assessments of the damage additional, unplanned spending for relief and reconstruction inflicted on public infrastructure and other public assets and through declines in expected revenues. A disaster’s and corresponding needs for repairs and reconstruction. local impact can also spread to the national economy, as The latter, and any costs incurred due to explicit contractual insolvencies and loan defaults create a domino effect. obligations that a government had to meet following the event, constitute the fiscal cost faced by the government 4.1. Contingent liabilities2 due to explicit contingent liabilities. Contractual obligations that are triggered by disasters can include social protection This subsection should describe government expenses payments linked to disasters, or payments linked to the incurred due to explicit and implicit contingent liabilities public backing of catastrophe insurance schemes and after past disasters.3 If sufficient information is available to institutions. They can also include the falling due of indirect distinguish between the two, they should be described in obligations related to government guarantees in support of separate subsections (see annex V for further background on other entities—for instance, liabilities relating to a publicly contingent liabilities). guaranteed bank that finds itself under pressure as a 4.1.1. Explicit contingent liabilities consequence of large-scale loan defaults in the aftermath of an earthquake. “Explicit contingent liabilities are obligations based on contracts, laws, or clear policy commitments” (Cebotari 4.1.2. Implicit contingent liabilities 2008, 6). Hence, they include the costs of repairing and Implicit contingent liabilities relate to moral or commonly reconstructing damaged public assets and infrastructure, recognized but nonlegal public obligations. In the context such as roads, bridges, hospitals, schools, and power and of disasters, they can include (i) search and rescue services water infrastructure. 4. That is, information on the explicit contingent liabilities that have actually fallen due after past disasters. The sum of these liabilities is not necessarily 2. See Cebotari (2008) for a thorough discussion of contingent liabilities. the same as the amount spent by a government on explicit contingent 3. When analyzing relevant data, care needs to be taken to ensure that liabilities after a disaster, since a government can renege on (some of) its actual expenditures are analyzed rather than allocations. obligations. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 15 and humanitarian relief; (ii) the provision of support, in 4.2. Foregone revenue particular to poorer households, to rebuild homes, replace productive assets, and provide compensation for injury and Disasters also impact government finances on the revenue loss of life; and (iii) measures to support the recovery of the side. Decreased economic activity due to disasters can translate private sector, such as tax breaks and the bailout of private into a reduction in tax receipts, at least initially. To get a sense institutions.5 of foregone revenue after past disasters, several metrics can be calculated. First, actual revenue outcomes can be compared Governments sometimes choose to absorb significant with projected outcomes. However, in countries where revenue implicit contingent liabilities in the aftermath of a disaster.6 projections tend to be poor, and where other factors may have Implicit contingent liabilities are not defined in advance plausibly caused a revenue shortfall, this metric may not be and are therefore hard to measure. An examination of past useful. In such instances, the deviation of actual outcomes expenditure may provide some indication of the cost incurred from the projected outcome could be compared to deviations by a government due to implicit contingent liabilities. Such in other years instead. If revenue shortfalls from pre-disaster information is most likely contained in budget (outcome) projections in the year after the disaster are significantly larger documents and reports by development partners, or it may be in percentage terms than revenue deviations in the past, this directly obtainable from ministry of finance (MoF) officials. result may give an indication of the revenue shortfall caused by the disaster. But again, such an interpretation would only 4.1.3. Measuring explicit and implicit be sensible in the absence of other factors that could have contingent liabilities plausibly caused or contributed to a poor revenue outcome. In practice, data on expected explicit and implicit contingent In practice, other factors influencing revenue outcomes are liabilities are often unavailable. In such cases, this subsection often prevalent, and it may not be possible to make accurate of the diagnostic should present whatever information is estimates of foregone revenue without expert knowledge. available on the actual total fiscal outlays due to disaster- Where such expert estimates exist, they should be presented.8 related contingent liabilities following recent events; if Table 3 provides a template for presenting the fiscal impact of possible, this information should be presented both in a past disaster. absolute terms and relative to reported damage and loss, to total annual expenditure, and to GDP. Recent IMF work estimates that, on average, these fiscal costs are equivalent Table 3: Template for table on fiscal impact of past to 1.6 percent of GDP.7 Information on the fiscal costs of disasters past disasters is again most likely found in PDNAs, budget Local outcome reports, IMF Article IV reports (see annex VI for an US$ currency example), or other relevant reports such as regular economic Expenditures and fiscal updates produced by the World Bank and the Asian Expenditures due to explicit Development Bank. contingent liabilities Expenditures due to implicit 5. In some countries, specific government agencies are mandated and contingent liabilities resourced to be first responders in the event of disasters, and statutory levels Foregone revenue and types of compensation (e.g., for loss of life, destruction of homes) are predetermined, implying that such responsibilities are explicit contingent Total liabilities. In consequence, the split between explicit and implicit contingent liabilities varies between countries. 6. For instance, following the 2011 Tōhoku earthquake and tsunami, which caused $360 billion in economic losses, the government of Japan spent more on economic and social support to affected communities and businesses than it did on the repair and reconstruction of public infrastructure. 8. In addition to the issues covered in this section, a more complete 7. See IMF (2016a). It is important to bear in mind that this average masks discussion of disasters’ impact on public finances would also include the wide variation in different countries’ fiscal costs due to the realization of issues such as (i) donor inflows following a disaster and how they change disaster-related contingent liabilities. Furthermore, it is based only on events the overall revenue impact of disasters; (ii) the impact of disasters on the leading to fiscal costs in excess of 1 percent of GDP. These are relatively government’s debt position and debt sustainability; and (iii) the impact of rare events, indicated by the fact that only 65 events, going back to the disasters on local government finances. These topics go beyond the scope late 1980s, were included in the analysis. This figure should therefore not be of the proposed diagnostics for which this note provides the template and taken as an indicator of likely disaster-related fiscal costs for any particular guidance. If task teams want to include these impacts and quantify them country. nonetheless, a public finance expert should be consulted. 16 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 5. Impact of past disasters 7. Resources on the poor Resources include the following: This subsection should present information on the impact ƒƒ EM-DAT Database of past disasters on the poor, including the number of people who fell into poverty as a result of them. The definition of http://www.emdat.be/ poverty applied should be explicitly stated in the diagnostic to This database, maintained by the Belgian Centre for help inform potential intercountry comparisons. The gathered Research on the Epidemiology of Disasters (CRED), information can be presented in the format of table 4. contains data on over 18,000 disasters globally since 1900. Data sources include governments, United Nations Table 4: Impact of past disasters on the poor agencies, nongovernmental organizations, insurance companies, research institutes, and press agencies. Impact on the poor ƒƒ DesInventar Number of poor affected Percentage of poor affected http://www.desinventar.net/index_www.html Number of non-poor affected Percentage of non-poor affected Hosted by the UNISDR, DesInventar is a disaster information management system that can be used to analyze disaster trends and their impacts in a systematic manner. By visualizing disaster information, it also 6. Key outputs facilitates dialogue on risk management between relevant actors, institutions, and sectors as well as provincial and Key outputs include the following: national governments. 1. Table on number of people affected by and number of ƒƒ Munich RE annual statistics on natural catastrophes deaths caused by major disasters, by date and type of hazard https://www.munichre.com/touch/naturalhazards/en/ natcatservice/natcatservice/index.html 2. Table on economic impact of past disasters; if data permit, this information can be broken down by damage and loss, Munich Re’s NatCatSERVICE database, which contains and by sector more than 30,000 entries, provides a platform with a wide range of information and services on various aspects of 3. Table on fiscal impact of past disasters; if data permit, natural catastrophes. The database analyzes and records this information can be broken down by expenditures up to 1,000 new natural hazard events each year. due to implicit and explicit contingent liabilities and foregone revenues ƒƒ Swiss RE Sigma Explorer 4. Table on impact of past disasters on the poor http://www.swissre.com/reinsurance/insurers/sigma_ explorer_the_data_you_need_at_your_fingertips.html Other potential outputs include a map illustrating the spatial distribution of past disaster impacts. This is likely The sigma explorer is an interactive, web-based app difficult to generate without good data and expertise that provides access to information about catastrophes in GIS (geographic information system) or a similar and world insurance premiums from 1970 onward. program, but in some countries such maps could be obtained from PDNAs or potentially from universities or technical institutes. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 17 ƒƒ Post-Disaster Needs Assessments ƒƒ Government disaster damage and loss databases http://www.recoveryplatform.org/pdna/key_documents ƒƒ Government economic reviews http://www.gfdrr.org/sites/gfdrr/files/urban-floods/PDNA. ƒƒ Fiscal risk statements html ƒƒ Government sector performance reviews http://www.cepal.org/cgi-bin/getprod.asp?xml=/noticias/ ƒƒ Annual budget statements paginas/7/37037/P37037.xml&xsl=/tpl/p18f.xsl&base=/tpl/ top-bottom.xsl ƒƒ IMF Article IV reports http://www.undp.org/content/undp/en/home/librarypage/ http://www.imf.org/external/np/sec/aiv/index.aspx crisis-prevention-and-recovery/pdna.html ƒƒ Sector-specific information from line ministries on the ƒƒ The Global Assessment Report on Disaster Risk Reduction impact of past disasters (GAR) ƒƒ Poverty data https://www.unisdr.org/we/inform/gar http://data.worldbank.org/topic/poverty The GAR is a biennial global assessment of disaster risk reduction and comprehensive review and analysis of the natural hazards that are affecting humanity. The linked website provides data and maps. 18 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 19 B. Existing legal and institutional framework and financial instruments for disaster risk finance 2.1. Laws governing the budget process and 1. Objective current practice This section should present an overview of the legal and This subsection should list and briefly review the laws institutional framework for disaster risk finance and of governing relevant budget processes. It should describe existing disaster risk finance instruments in the country. the responsibility of different actors with regard to post- This overview should provide a basis for discussing potential disaster relief, early recovery, and reconstruction (including legal, institutional, and financial reforms to strengthen any contingent budget line requirements); it should indicate the framework. the timeline for key steps in the budget process; and it should indicate whether the actors and their responsibilities and the timeline are clearly defined. It should also review 2. The legal and institutional procedures governing the reallocation of budgets in the framework for disaster aftermath of a disaster, the preparation and approval of supplementary budgets, and rules governing the release of risk finance public resources for disaster response purposes (typically linked to the declaration of a state of disaster at different To start, relevant laws and institutions that govern resource levels of government). If budget flexibility is constrained by a mobilization and execution and the country’s broader DRM limited proportion of discretionary expenditure, this should framework should be reviewed and described. Broadly, be stated. Finally, any important differences between actual three sets of laws provide the foundation for the legal and budgetary practice and formal processes and procedures regulatory framework for disaster risk finance: budgetary, should be noted. DRM, and insurance laws. Since relevant insurance laws will be discussed elsewhere in the diagnostic (see section C), this 2.2. Post-disaster budget execution subsection should focus on the two former sets of laws. After describing the relevant legal and institutional framework, To maximize the benefits of financial resources immediately this subsection should present the referenced laws and their available after a disaster, governments must be able to spend key provisions in a table at the end. them efficiently. Executing funds in a timely and efficient manner is a great challenge for many governments, even 20 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC under normal circumstances.9 After a disaster, the surge in ƒƒ Whether legal provisions exist for mechanisms to funds and the disruption wrought by the event combine to coordinate action. compound the challenges of execution. ƒƒ Whether there are stipulations for emergency Understanding budget execution processes and potential procurement procedures in the event that a national associated issues is a first step in improving post-disaster calamity is declared. spending. This subsection should review reports that ƒƒ Whether the circumstances under which a state of analyze the country’s expenditure efficiency, including national disaster/calamity can be declared, and the Public Expenditure and Financial Accountability (PEFA) process for declaring it, are well defined. assessments and Public Expenditure Reviews (PERs). Key issues to highlight are (i) the efficiency of public investment This point is particularly important in countries where management in a country, as this relates to government the declaration of a state of national disaster or calamity capacity to execute reconstruction projects; (ii) the efficiency is the trigger that activates catastrophe contingent credit of social welfare expenditure and delivery mechanisms, lines or disaster reserve fund withdrawals. as this relates to government capacity to channel relief payments through existing systems; and (iii) the expenditure ƒƒ What the process and methodology are for assessing efficiency of the various sectors through which emergency damages, losses, and needs after a disaster. funds might be channeled. Such assessments are a key part of disaster response since This subsection should also describe any special mechanisms they form the basis for post-disaster resource mobilization to expedite the execution of funds in emergency situations, and allocation. If government resources are insufficient, including emergency procurement processes; procurement they inform donor decisions on aid provision. The quality processes for disaster reserve funds; pre-agreed contracting and speed of past assessments should also be described, for the provision of emergency relief and early recovery; goods together with the prescribed methodology and (where it and services; and special (streamlined) budget execution differs) the actual methodology employed.10 processes. The described processes should be compared ƒƒ What the legal requirement is with regard to auditing to actual experience from recent disasters to identify any of post-disaster expenditure; the extent to which post- challenges to efficient post-disaster execution. disaster expenditure has been subject to audits in the past. 2.3. Law(s) on disaster risk management and ƒƒ What the requirements are for and practice of monitoring disaster risk finance and evaluation of post-disaster interventions. This subsection of the diagnostic should review laws and If a DRM act establishes a dedicated disaster reserve fund, regulations that govern post-disaster response more broadly, this subsection should say so explicitly and review any such as civil protection laws. It should also analyze/describe decrees that stipulate how the fund is financed, what it can the following: be used for, and how expenditures are accounted for. ƒƒ Who the key actors are in disaster relief, recovery, and In the unlikely case that a specific law on disaster risk finance reconstruction, along with their responsibilities. exists, this subsection should document its implications for ƒƒ Whether institutional mandates complement each other financial preparedness by government, households, and and avoid overlap. businesses. It should also discuss any other laws that are relevant for the mobilization and execution of resources in response to a disaster. 9. This is particularly true for public investment projects, under which post-disaster infrastructure reconstruction falls. According to the IMF, “the average country loses about 30 percent of the value of its investment to 10. See Box 1 for more information on the methodology for Post-Disaster inefficiencies in their public investment processes.” See https://www.imf. Needs Assessments promulgated by the European Union, World Bank, and org/external/np/fad/publicinvestment/pdf/PIMA.pdf. United Nations Development Programme. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 21 2.4. Institutional setup for disaster risk 2.5. Local government management and finance Local governments often have very limited revenue- This subsection should describe the institutional setup for raising capabilities of their own and therefore generate disaster response and related financing, and should analyze few resources that they can retain for post-disaster the capacity of relevant institutions and the level and quality response or use to procure risk transfer instruments. This of coordination among them. subsection of the diagnostic should discuss the role of local governments in disaster response, any rules that define It should first review reports that examine the functioning of cost-sharing arrangements between local and regional or the MoF and the state of public financial management in the national governments, financial mechanisms that local country, such as PEFA assessments or functional reviews of governments have at their disposal to respond to disasters, the MoF. This step should be complemented by interviews and recent disaster experience in practice from a budgetary with the heads of key units in the MoF, such as treasury, perspective, including with regard to the speed and adequacy budget, and macroeconomic management. of transfers from the national government. Gathering this Next, the diagnostic should describe the nature and quality information will require discussions with a few selected local of the relationship between the MoF and other ministries governments as well as with relevant national government involved in disaster response and related financing. The agencies, including the MoF. extent to which the MoF is willing to delegate oversight of funds to other ministries depends on the legal framework, but also on levels of institutional trust. Effective coordination 3. Existing disaster risk finance among different ministries and agencies is crucial for disaster mechanisms and instruments risk reduction, preparedness, and response, and its absence can have significant financial implications. This subsection should describe how the country currently finances disaster-related expenditures. This assessment If a formal coordination mechanism exists that brings should generate a list of any financing mechanisms explicitly together key DRM actors—e.g., a national disaster risk designed to mitigate disaster-related financial risk, as well management platform—the diagnostic should describe as other mechanisms that have helped to meet the financial whether such a platform is functional; whether it exists at burden from disasters in the past, or that might do so in the the political level, the technical level, or both; how often it future. holds meetings; what powers it has; and what its role is in determining the allocation of finance for DRM, including 3.1. Risk layering post-disaster response, and in making decisions around disaster risk finance policy. International experience has shown that governments should ideally combine different instruments to protect Furthermore, this subsection should review the presence and against events of different frequency and severity. Financing role of development partners in driving the country’s disaster mechanisms can be grouped into two main categories: risk finance agenda. It should describe the humanitarian response to disaster, including how this response is 1. Retention. The government assumes and manages disaster coordinated. In countries at risk of disasters, a wide array losses through its budgetary resources—e.g., budgetary of organizations, including nongovernmental organizations, reserves, post-disaster budget reallocations, or borrowing. are often involved in response efforts. An overview of active 2. Transfer. The government transfers potential future players should be provided. disaster losses to financial or insurance markets by paying a premium—e.g., through traditional insurance/ reinsurance. \ 22 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC Figure 1: Three-tiered risk layering strategy for governments International assistance Low frequency/high severity Risk transfer Sovereign risk transfer (E.G. Cat bond/cat swap, (Re)insurance) Insurance of public assets Post-disaster borrowing Contingent credit lines Increased taxation Risk retention High frequency/low severity Government reserves, Contingency budget/funds Budget reallocations Emergency funding Reconstruction Source: Adapted from World Bank and GFDRR 2014. Note: Cat = Catastrophe Combining different instruments to protect against events instruments are arranged before an event, whereas ex post of different frequency and severity is known as risk layering instruments are arranged after an event. The two should be (figure 1). A bottom-up approach is recommended: the combined to provide an effective package of instruments. government first secures funds for recurring disaster events Generally, there will be some scope for budget reallocations and then increases its post-disaster financial capacity following a disaster, as planned investments in disaster- to finance less frequent but more severe events. Risk affected areas have to be postponed. There may also be scope layering ensures that cheaper sources of money are used for borrowing, depending on market access and existing levels first, with the most expensive instruments used only in of domestic and external debt. If financing needs are large, exceptional circumstances. however, obtaining sufficient resources after an event can be difficult and may entail significant delays and economic and 3.2. Ex ante vs. ex post disaster risk finance developmental costs. Putting ex ante instruments in place instruments smooths disaster-related costs over time and can lead to better planning and processes for post-disaster expenditures. Disaster risk finance instruments can also be categorized as either ex ante or ex post financing instruments. Ex ante ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 23 3.3. Ex ante disaster risk finance instruments should describe any existing contingent credit lines for disaster response, and dates and purpose of any drawdowns. This subsection of the diagnostic should describe the MoF officials should be able to provide information on financing mechanisms governments have put in place to contingent credit lines in place, including their size and respond to disaster-related financial contingencies. under what circumstances they can be activated.11 3.3.1. Contingency budgets 3.3.4. Sovereign risk transfer solutions MoF officials and/or budget documents should be consulted This subsection should describe existing sovereign and to determine whether the budget includes any provisions for subsovereign risk transfer solutions. These can include disaster-related contingencies. If it does, further investigation products that protect the government’s budget against fiscal is required to determine (i) the amount of annual budget shocks from disasters by transferring the risk to international allocations in recent years and how these amounts were insurance, reinsurance, and capital markets, for instance agreed on; (ii) whether allocation amounts are discretionary via sovereign risk pools, parametric or index insurance or set in law; (iii) whether remaining balances can be rolled products, or catastrophe bonds. This description should over to the next year; (iv) procedures for allocating and include information on levels of cover and on any payouts. approving disbursements; (v) historical data on allocations This subsection should also discuss to what extent public to, and use of, the contingency budget for post-disaster assets are insured. Any policies that govern the insurance response; and (vi) broader rates of disbursement over the of public assets of different government agencies should be budget year if contingency budget lines are available for described elsewhere (see section C). Finally, this subsection non-disaster-related purposes as well. This last question is should specify whether insurance can be underwritten important because in practice, contingency budgets may only by an insurer licensed, registered, or authorized to do provide minimal support for disaster response if they are fully business in the country where the insured risk is located utilized for other purposes relatively early in the budget year. (admitted insurer), or whether coverage can be provided by an insurance company from another jurisdiction that is not 3.3.2. Reserve funds authorized by the host country’s regulation to cover risks in In some countries, general or disaster-specific reserve funds that country.12 are used to meet the costs of high-frequency, low-impact disasters. In some countries, they also pay for disaster 3.4. Ex post disaster risk finance instruments risk reduction and preparedness. If such a fund exists, this Governments typically meet part of their post-disaster subsection should describe (i) which ministry manages it financing needs through budget reallocations, tax increases, or and what role the MoF plays in its management; (ii) whether increased borrowing. This subsection should discuss ex post the fund is on- or off-budget; (iii) how the fund is financed; risk finance instruments used after past disasters and describe (iv) how funds are managed; (v) the fund’s current balance; any bottlenecks that were experienced in the mobilization (vi) its eligible uses; (vii) the procedures for accessing it; of funds. and (viii) the circumstances under which funds are released. Actual disbursement of the fund should also be reported, including dates and purpose. Should subsovereign disaster reserve funds exist, they should be analyzed and described along the same lines. 11. Contingent credit can typically be activated immediately following the occurrence of prespecified trigger events, such as the declaration of a 3.3.3. Contingent credit national disaster. 12. Many countries restrict transaction of insurance by entities without a Some development organizations and private creditors offer license. The same restrictions apply to local brokers who are intermediating on behalf of an affiliate without a license. Consequently, insurance contingent credit for disaster recovery and reconstruction, supervisors need to determine the conditions that must be met by the local subsidiaries or affiliates to insure local risks with a nonadmitted insurer, such facilitating rapid access to financing in times when liquidity as prior approval from regulators, a specific registration of the existence of tends to be constrained. This subsection of the diagnostic the insurance, use of a local placement broker, or the payment of insurance premium taxes locally by the local insured or local broker. 24 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 3.4.1. Budget reallocations of the diagnostic should indicate whether the government has increased taxes after past disasters, and provide details This subsection should describe the mechanics of budget on the specific taxes concerned, rate increase(s), amounts reallocation, including collected (and, for comparison, equivalent figures from past years), how long it took to implement the tax increase(s), ƒƒ Regulations regarding reallocations within existing and the time period for which the increased rate(s) remained budget lines at the discretion of the budget line holder in place. (e.g., a line agency), including any limits on amounts ƒƒ Regulations regarding reallocation between budget 3.4.3. Post-disaster borrowing lines and government agencies, including any limits on Governments may borrow to finance disaster costs, depending amounts on their access to capital markets and their creditworthiness. ƒƒ The approximate amount of time it takes to pass a This subsection should describe whether raising money supplementary budget from private lenders or official donors is an option for post-disaster financing, and to what extent this approach ƒƒ Historical data on the scale of post-disaster reallocations figures in official planning (e.g., in the government’s debt management strategy). This subsection should also describe ƒƒ Historical experience regarding any hurdles and delays any constitutional or other legal constraints on borrowing, faced by government in implementing post-disaster such as rules on the debt-to-GDP ratio or deficit spending. budget reallocations13 It should report as well on any past post-disaster borrowing, The process described should be compared to actual including any bond issues, and on the implied consequences experiences from recent disasters in order to identify existing for levels of debt and, if relevant, credit ratings. challenges and potential reforms to address them. Data on reallocations—and particularly on reallocations within budget 3.4.4. Donor presence and assistance14 lines (e.g., within operations and maintenance budgets of line Development partners and their activities affect how agencies)—may be difficult to obtain. Supplemental budget governments plan financially for disasters. This subsection statements and related speeches may contain information on should describe the financial resources and mechanisms any substantive reallocations between government agencies. donors can mobilize in post-disaster situations. In some low- Budget speeches for subsequent years may also provide an income countries, the number of international actors active in overview of post-disaster reallocations. post-disaster response might be large. For such environments, or environments where instruments target the population 3.4.2. Tax increases directly, this subsection should provide information on the Tax increases are another way to finance increased targeted groups and targeting mechanisms through which expenditure needs after disasters. Their effectiveness resources are channeled to recipients. It should also include depends on various factors, including a country’s tax base, details on international assistance provided in response tax compliance, and tax collection capacity. However, to recent disasters, including how long it took donors to they can be politically costly and potentially aggravate the mobilize and disburse pledged funds. economic stress caused by the disaster itself. This subsection 14. There is much innovation around donor-financed ex ante mechanisms for disaster response. For example, within the humanitarian community the provision of in-kind assistance sometimes occurs through forward purchase facilities, allowing purchases of commodities in advance of emergency needs. These have reduced delivery times and achieved greater cost- efficiencies, as well as supported the piloting of alternative procurement and financial management arrangements. Innovations such as storage pre- positioning, pre-contracts with financial institutions for cash delivery, and work around virtual food reserves might also be relevant in certain countries. 13 Minor reallocations are often within the discretion of individual line If so, such mechanisms could be described under a separate heading within agencies, particularly as regards recurrent spending. subsection 3.3. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 25 Table 5: Amount of funds available for disaster response – the example of Serbia in 2016 Disaster risk Financing source available Amount of funds available Unpredictable and unreliable (e.g., in 2014 the total Donor assistance commitment was €235 million, often in kind) High-risk layer (e.g., major Unpredictable (e.g., €227.5 million drawn from floods, major earthquakes) Emergency borrowing World Bank for 2014 floods emergency recovery) Insurance of public assets Unclear but very low Medium-risk layer (e.g., Not currently available (US$100 million CAT DDO is regional floods, minor Contingent financing in early preparation) earthquakes) Budget funds: permanent budgetary €17,000 (originally budgeted, increased one-off by reserve 2014 supplementary budget to almost €20 million) Budget funds: compensation for €700,000 (originally budgeted, increased one-off Low-risk layer (e.g., damage caused by the natural by 2014 supplementary budget to approximately localized floods, droughts, disasters (account 484) €1.5 million) landslides) Unclear (10 percent of each appropriation available Budget reallocation immediately; higher if supplementary budget is passed) Source: World Bank 2016. Note: CAT DDO = Development Policy Loan with a Catastrophe Deferred Drawdown Option 3.4.5. Tabulation of disaster risk ƒƒ Hyogo Framework for Action National Progress Reports finance instruments http://www.preventionweb.net/english/hyogo/progress/ A table with available financing sources and amounts for reports/ the different disaster risk layers should summarize the These reports assess levels of progress achieved in information of this subsection. An example from Serbia is implementing the five priorities of the Hyogo Framework given in table 5 . for Action 2005–2015. Going forward, they will likely be supplemented by progress reports on the implementation of the Sendai Framework for Disaster Risk Reduction 4. Key outputs 2015–2020. Key outputs include the following: ƒƒ Project documents of donors ƒƒ Table of relevant laws and their key provisions ƒƒ Government budget documents ƒƒ Table of existing disaster risk finance sources and ƒƒ BOOST databases amounts, organized by risk layer http://wbi.worldbank.org/boost/boost-initiative BOOST facilitates user-friendly access to budget data in 5. Resources about 40 countries as of mid-2016. Data sets typically Resources include the following: contain information on the approved budget, revised budget, and actual expenditure amounts broken down ƒƒ Legislative and regulatory databases by (i) government level; (ii) administrative unit; (iii) subnational spending unit; (iv) economic classification; ƒƒ Existing institutional reviews (v) functional classification; (vi) program classification 26 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC (if the country uses program-based budgeting); and (vii) 6. Indicative list of key financing source. counterparts to meet ƒƒ Public Expenditure Reviews 1. Ministry of Finance http://wbi.worldbank.org/boost/tools-resources/public- expenditure-review ƒƒ Budget Department ƒƒ Climate Public Expenditure and Institutional Reviews ƒƒ Macroeconomics/Fiscal (CPEIRs) ƒƒ Treasury/Debt https://www.climatefinance-developmenteffectiveness.org/ ƒƒ International cooperation CPEIR-Database 2. Ministry of Planning Using the PER methodology, CPEIRs carry out systematic qualitative and quantitative analyses of a country’s public 3. Ministry of Public Works expenditures as they relate to climate change. CPEIRs present evidence on public expenditures across all ministries, 4. Ministry of Transport and review a government’s climate change plans and policies, 5. Ministry of Energy institutional framework, and public finance architecture. Since 2016 the methodology has also been used in DRM 6. Ministry of Rural Development public expenditure and institutional reviews. 7. Ministry of Environment ƒƒ PEFA assessments 8. Ministry of Agriculture https://pefa.org/ 9. National Disaster Management Agency PEFA is a methodology for assessing public financial 10. Ministry or department that oversees management performance with quantitative indicators. subnational governance PEFA is designed to provide a snapshot of this performance at specific points in time using a methodology that can be 11. Department or agency that oversees state- replicated in successive assessments, hence providing a owned enterprises summary of changes over time. 12. Ministry or agency responsible for food security 13. Mangers of provident/pension funds15 14. Development partners 15. Local government (small sample) 15. In some countries, special arrangements may exist that allow for emergency withdrawal of retirement funds to expand social safety nets after a disaster. In such cases, the fund managers should be consulted. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 27 C. Domestic insurance market review indicate whether there are specific laws or regulations that 1. Objective require the purchase of catastrophe insurance, describe local This section of the diagnostic should review the state of the reinsurance regulations, and say whether premiums are tax country’s domestic insurance and reinsurance industry, and deductible. The subsection should also indicate whether the relevant legal and regulatory environment governing the the country’s regulations permit the sale of index-based sector. It should provide information on key aspects of the insurance products, and whether such products are based on regulatory framework, the non-life insurance penetration parametric triggers or aggregate output indicators. rate, property catastrophe insurance, agricultural insurance, and disaster microinsurance. It should assess a country’s level of uptake of disaster-related risk transfer through insurance 4. Property catastrophe and reinsurance. The availability of disaster risk assessments, which provide the basis for the development of risk transfer insurance products products, is discussed elsewhere (see section A). Property catastrophe insurance is a key instrument of financial protection against disaster risk. It protects households and businesses, thereby also reducing a government’s implicit 2. Insurance penetration and contingent liability.16 Furthermore, it offers direct cover for public properties. insurance market This subsection of the diagnostic should provide information This subsection should report on the country’s non-life on the perils and market segments for which property insurance penetration rate and trends over time, with a catastrophe insurance products are available in the country, specific focus on catastrophe insurance. Comparisons with along with premium rates for different perils (e.g., residential, countries that are in the same region or that have similar industrial and commercial fire, earthquake, wind, etc.). characteristics, together with views of credit rating agencies, For each product line, information should be provided on can provide useful context. This subsection should also whether catastrophe coverage is embedded, an endorsement, provide information on the number of private insurers, or stand-alone. Policy exclusions, deductibles, and limits of captives, and, if applicable, government-owned insurance coverage should also be documented. companies. If parametric products are available, this subsection should describe the parametric triggers. It should also note any 3. Regulatory environment mandatory insurance products and, in the case of residential insurance, indicate whether mandatory purchase is tied to This subsection should provide information on whether mortgage eligibility. It should also provide information on insurance regulations are set domestically or by a regional regulatory body, such as the West African Conférence Interafricaine des Marchés d’Assurances (CIMA). It should 16. This is true insofar as government would otherwise support the restoration of private homes. 28 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC the percentage of commercial and residential properties that 7. Resources are insured against catastrophe risk. Resources include the following: Finally, this subsection should discuss whether public assets (e.g., buildings, infrastructure) are insured against disasters ƒƒ Axco reports and under which conditions. http://www.axcoinfo.com/countries.aspx ƒƒ Reports from the Access to Insurance Initiative 5. Agricultural insurance https://a2ii.org/en/knowledge-centre/reports This subsection should discuss the availability and uptake of agricultural insurance, including crop, livestock, forestry, ƒƒ Reports from the Micro Insurance Network and aquaculture insurance. It should examine government http://www.microinsurancenetwork.org/resources involvement in agricultural insurance, including any public or public-private schemes, premium subsidization, or loss co-sharing arrangements. It should also provide information 8. Indicative list of key on the cost of any public support (subsidy), if available. counterparts to meet 1. Insurance regulator/supervisor 6. Microinsurance 2. Insurance industry association Microinsurance can be an effective risk transfer mechanism and an integral part of an overall disaster risk finance 3. Banking regulator framework. It can provide access to post-disaster liquidity, thus protecting assets and livelihoods as well as providing funds for early recovery. This subsection should describe the kind of disaster microinsurance that exists in the country, the penetration rates, and the segments of the population and business community that are served. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 29 D. Funding gap analysis bear in mind that even where all identified needs for disaster 1. Objective response, early recovery, and reconstruction have been met, This section should present estimates of the potential they may not have been met in the most cost-effective or shortfall of funds a government might face after a disaster. timely manner. The objective is to provide the key information required for However, even simplified analysis, focusing solely on ex entering into dialogue with government on potential funding ante tools, can be useful in providing a first approximation gaps and options for strengthening financial preparedness of funding gaps as a basis for reviewing and strengthening for disasters. While estimating funding gaps precisely disaster risk finance arrangements and for exploring the requires extensive data that are not readily available in wider fiscal consequences of these gaps. If ex ante tools are many countries, calculations based on historical data can too limited to serve as the sole basis for an analysis, it will provide a reasonable approximation of the potential funding be necessary to make assumptions about ex post options for gaps, particularly if combined with available information on financing post-disaster needs and to incorporate these into probable maximum losses. The following paragraphs provide the analysis. some guidance on estimating disaster-related funding gaps and explain how to conduct a funding gap analysis with It is important to develop a clear picture of the likely relatively little data. relative spread of funding needs and resources over time, distinguishing between the relief, early recovery, and reconstruction phases of the response effort for each type of 2. Estimating funding gaps natural hazard (see figure 2). The consequences of the timing Ideally, a funding gap analysis should be based on estimated Figure 2: Timing of needs and execution of contingent liability (see section A) combined with full financial instruments information on available disaster risk finance (see table 5) and Resource requirements (US$) should derive estimated funding gaps for different disaster scenarios. The funding gap can be derived by subtracting available disaster risk funds from the costs associated with each scenario. Estimating funding gaps precisely requires comprehensive data on public contingent liabilities for all types of hazards experienced by a country, the hazards’ return periods at varying intensities, and all existing ex ante disaster risk financing mechanisms (both formal and informal), together with assumptions on the scale of fund flow from ex post disaster risk financing instruments. In Relief Recovery Reconstruction Time the case of ex post analysis, it requires data on actual flows of funding.17 Anyone undertaking this latter analysis should Source: World Bank and GFDRR 2014. 17. It is important to remember that contingency budgets may be intended to cover all events over a particular period, and so it cannot be assumed that the full amount of funds is available for each individual event, particularly in countries that experience multiple disaster events each year. 30 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC of a disaster relative to the fiscal calendar should also be built 3.2. Long-term funding gap analysis into the analysis. Should a disaster occur at the beginning of a fiscal year, some governments may have to wait a year To estimate longer-term funding gaps, post-disaster or more before significant funding can be disbursed for reconstruction needs have to be considered. If detailed data reconstruction.18 on contingent liabilities are not available, data on historical damages and on related government reconstruction spending can serve as a basis to estimate longer-term reconstruction needs. Costs can be adjusted to reflect likely increases 3. Performing a funding in exposure in the intervening period (due to economic gap analysis in data- expansion) by multiplying the losses incurred in year X by a factor equivalent to GDPcurrent year/GDPyear X, which constrained environments would also control for inflation if nominal GDP figures are Various funding gap metrics can be calculated from relatively used. Future levels of exposure should be based on GDP and limited data to provide a useful first approximation of the capital asset forecasts. funding gap for initial discussions with governments. Two metrics are discussed below: (i) a short-term funding gap 3.3. Funding gap estimates based on individual analysis to assess the government’s readiness to finance past disasters immediate and near-term emergency relief and early Funding gap analyses based on an individual past disaster, or recovery costs; and (ii) a longer-term funding gap analysis to on several individual past events, and associated information assess a country’s ability to finance potential reconstruction on impact can serve as an alternative starting point for needs. The note also discusses short- and long-term funding conversations with officials on the current state of financial gap estimates based on (i) ex post analysis of individual past protection against disasters. Depending on the available data, events, and (ii) analysis of hypothetical events with varying both the short- and long-term funding gaps can potentially return periods derived from historical data. be estimated. As discussed above, inflation and changes of exposure due to population growth and economic growth 3.1. Short-term funding gap analysis should be accounted for. Past data on per capita relief and early recovery spending can serve as the basis for estimating the potential future Figure 3: Post-disaster emergency funding gap short-term cost of responding to various types of hazard. US$ million Multiplying past emergency relief per person by the number 0 5 10 15 20 25 30 35 40 45 50 of people a hypothetical event affects (after adjusting for growth and inflation over time) will yield an estimate of 2013 Flood emergency response cost emergency relief costs. 1-In-10 year event emergency A funding gap analysis based only on emergency relief and response cost early recovery costs may be useful for an initial conversation Annual average emergency with government counterparts. However, it is important response cost to emphasize that a comprehensive funding gap analysis is US$ 24 million Available funds required to determine the full extent of the funding gap and Potential funding gap to provide sufficient basis for an in-depth examination of disaster risk finance options. 18. It is also important to keep in mind that funding gaps at the local level may be felt particularly acutely, even if funding gap analyses can be difficult to conduct at that level due to data constraints. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 31 3.4. Funding gap estimates based on hypothetical events with varying return periods If statistical estimates of the impact of past disasters with varying return periods are available, an analysis of funding gaps associated with events of different return periods should also be presented. A simple graphical presentation of the post-disaster emergency funding gap for events with varying return periods (i.e., with different probability of occurrence) is illustrated in figure 3. Both short-term and long-term funding gaps associated with events of varying return periods should be presented if possible. 4. Key output There is one key output, a funding gap analysis graph. 5. Resources Resources include the following: ƒƒ United Nations Office for the Coordination of Humanitarian Affairs (UN OCHA) http://www.unocha.org/ ƒƒ Financial Tracking Service (FTS) https://fts.unocha.org/ FTS records all reported humanitarian aid contributions, with a special focus on humanitarian response plans and appeals. ƒƒ Global Facility for Disaster Reduction and Recovery (GFDRR) Disaster Aid Tracking http://gfdrr.aiddata.org/dashboard ƒƒ See sections A, B, and C for other potential data sources. 32 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 33 E. Summary and options for consideration risk finance instruments. The diagnostic should also help 1. Objective practitioners engage with governments on policies and This section of the diagnostic should summarize the specific instruments to improve financial protection. current status of the disaster risk finance framework in The diagnostic note should therefore conclude with the country, identify gaps, and briefly outline possible next recommendations and potential next steps for improving the steps the government could take to advance the disaster risk country’s disaster risk finance framework. The diagnostics finance agenda. will often recommend further analysis to close identified knowledge gaps and the development of a disaster risk finance strategy that sets out key reforms in a prioritized, sequenced, 2. Summary and gaps and comprehensive way. In general, but especially in low- The following building blocks characterize a well-functioning capacity environments, it is important to clearly prioritize disaster risk finance framework: potential reforms, and to develop a realistic sequence and time frame for implementing them. Recommendations 1. Availability of relevant data and risk assessments should account for contextual factors such as government capacity, relevant political economy factors, and overall 2. A sound legal and institutional framework for disaster government priorities. risk finance Finally, it is worth recalling that the diagnostic is only a small 3. Financing solutions to manage disaster risk step in the overall dialogue with the government on disaster 4. Mechanisms to deliver funds to targeted beneficiaries and risk finance. The diagnostic should help structure the the ability to execute relief, recovery, and reconstruction dialogue and anchor it in the best initially available data and activities in a timely and efficient manner analysis. Eventually, deeper engagement can lead to a disaster risk finance strategy, which in turn will need to be followed This subsection should briefly summarize key information by an implementation plan. Throughout, technical assistance gathered and presented in the previous sections, and discuss may be necessary depending on the country’s capacity and gaps in the existing disaster risk finance framework. Based needs. As mentioned at the outset, the diagnostic may be on the identified gaps, possible next steps and options for most useful if treated as a living document that is updated consideration can be presented to the government. as new information emerges, rather than a document that is published and then shelved. 3. Options for consideration The purpose of a disaster risk finance diagnostic is to help governments understand their current level of financial protection against disasters, relevant legal and institutional gaps, and the limitations of their current basket of disaster 34 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 35 Glossary 19 19. annual expected loss (AEL). Expected loss per year when insurance endorsement. An amendment to an insurance averaged over a very long period (for example, 1,000 years). contract which modifies the terms or scope of the Computationally, AEL is the summation of products of event original policy. losses and event occurrence probabilities for all stochastic parametric insurance. A form of insurance that makes events in a loss model. indemnity payments based not on an assessment of the captive insurance. The arrangement whereby a subsidiary policyholder’s individual loss, but rather on measures of a company provides insurance or reinsurance for its parent. parametric index that is assumed to proxy actual losses. catastrophe bond (CAT bond). A high-yielding, insurance- probable maximum loss (PML). The largest loss believed linked security providing for payment of interest and/ to be possible for a certain type of event in a defined return or principal to be suspended or cancelled in the event of period, such as 1 in 100 years or 1 in 250 years. a specified catastrophe, such as an earthquake of a certain return period. A statistical estimate of the average magnitude or above within a predefined geographical area. recurrence interval of a particular event—that is, the average catastrophe model. A computerized model generating length of time in which a particular event can be expected to a set of simulated events to calculate losses arising from occur once. a catastrophe. risk layering. The process of separating risk into tiers to exposure. Measure of vulnerability to loss, usually expressed allow for more efficient financing and management of risks. in terms of sum insured, dollars, or units. stand-alone policy. An insurance policy that a business or indemnity insurance. An insurance policy that pays individual purchases to cover a specific risk or cost. It is claims based on the actual economic losses incurred by the opposite of an insurance policy with broad coverage the policyholder. that applies to a number of risks in different scenarios. For example, a standard homeowner's insurance policy index insurance. An insurance policy that pays claims based covers most common sources of damage, including fire, on an index. Indexes are typically chosen to be a good proxy wind, and hail. But homeowners who want protection of the economic losses incurred by the policyholder. against earthquake (for example) can purchase stand-alone earthquake insurance to cover their homes in such an event. 19. Several of the definitions in this glossary are from Clarke and Dercon (2016) and from World Bank (2012). 36 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 37 References Benson, Charlotte. 2009. “Developing and Targeting IMF (International Monetary Fund). 2016a. “Analyzing and Economic and Financial Information in Support of Managing Fiscal Risks: Best Practices.” IMF, Washington, DC. Effective Disaster Risk Management.” Draft. United ———. 2016b. “Country Report No. 16/243: St. Vincent Nations Development Programme, Geneva. and the Grenadines.” International Monetary Fund, Bova, Elva, Marta Ruiz-Arranz, Frederik Toscani, and H. Elif Washington, DC. Ture. 2016. “The Fiscal Costs of Contingent Liabilities: A UNDP (United Nations Development Programme). 2013. New Dataset.” IMF Working Paper WP/16/14, International “Post-disaster Needs Assessments, Volume A, Guidelines.” Monetary Fund, Washington, DC. United Nations Development Program, New York. Cebotari, Aliona. 2008. “Contingent Liabilities: Issues and World Bank. 2012. "Advancing Disaster Risk Financing and Practice.” IMF Working Paper WP/08/245, International Insurance in ASEAN Member States: Framework and Monetary Fund, Washington, DC. Options for Implementation. Main report." Washington, Clarke, Daniel J. and Dercon, Stafan. 2016. “Dull Disasters? DC: World Bank. http://documents.worldbank.org/ How planning ahead will make a difference.” Oxford curated/en/265831468205180872/Main-report. University Press, New York. World Bank and GFDRR (Global Facility for Disaster ECLAC (Economic Commission for Latin America and the Reduction and Recovery). 2014. "Financial Protection Caribbean). 2003. "Handbook for Estimating the Socio- against Natural Disasters: An Operational Framework for economic and Environmental Effects of Disasters." Disaster Risk Financing and Insurance." Washington, DC: Economic Commission for Latin America and the World Bank. Caribbean and World Bank. http://www.preventionweb. ———. 2016. “Disaster Risk Finance Country Note: Serbia.” net/files/1099_eclachandbook.pdf. World Bank, Washington, DC. 38 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 39 Annex I: Disaster risk finance diagnostic— Questionnaire This questionnaire is designed to (i) assess a country’s preparedness to manage disaster-related fiscal risks; and A. Questions for national (ii) identify potential challenges to developing disaster risk disaster management agency finance strategies (e.g., legal and regulatory barriers). on DRM policy and strategy and The three sets of questions are intended to be directed toward three key government agencies:20 disaster data 1. Questions on disaster risk management policy and 1. Concerning the national DRM strategy, strategy and disaster data should be directed to the a. Does the country have such a strategy? national disaster management agency. b. If so, what is the legal status of this strategy (i.e., is it a 2. Questions on fiscal management of disasters should be law or advisory)? directed to the ministry of finance. c. To what extent are the actions/policies outlined in this 3. Questions on domestic catastrophe insurance markets strategy being implemented by the government? should be directed to the insurance regulator. d. Does the strategy mention financial arrangements The questions listed below can be addressed (separately) for post-disaster response, including contingency both to central and local governments. The link between budgets, reserves, and/or insurance at any scale?21 central and local governments in the financing of disaster relief, early recovery, and reconstruction should be carefully e. If yes to question d, how well developed are discussed in meetings with government counterparts. these arrangements (e.g., country is investigating options, country is developing reserve funds, country is using financial/insurance instruments)? 20. This does not mean, however, that the questions should be directed exclusively at those three agencies. Depending on the context, some of the questions could also be directed at entities such as the central bank or securities regulator. 21. See section B3 for information on disaster risk finance instruments. 40 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 2. Concerning the legal framework for emergency declaration, B. Questions for ministry of a. Which government entities are vested with the finance on fiscal management of authority to declare states of emergency and disaster, including regional and national declarations? disasters22 b. What criteria must be satisfied before declarations can Assessment of fiscal shocks associated be made? with disasters 3. Concerning disaster response, 1. Concerning contingent liability of the government, a. What government entities execute the government’s a. What are the government’s legal, stated contingent efforts in (i) disaster relief; (ii) early recovery; and (iii) liabilities associated with disasters (public assets, low- reconstruction? income housing, guarantees, etc.)? b. In cases where the government uses ex ante risk b. Historically, what implicit (i.e., socially/economically financing instruments, is there a defined process enforced) contingent liabilities has the government by which the receiver of funds transfers them to the assumed (i.e., approved expenditure for) in the event entity executing the government’s disaster response? of a disaster? 4. Concerning damage and loss data, c. How much has the government spent annually on post-disaster response over the past 10 years a. Is there a central database on historical damage and or other relevant period (ideally broken down loss? If so, how complete are the records and how can according to humanitarian relief, early recovery, the database be accessed? and reconstruction)? b. What methodologies and systems are in place to assess 2. Concerning fiscal risk assessment of disaster shocks, and record damage and losses? Are these applied? a. Does the government have data on historical fiscal 5. Concerning disaster response funding requirements, are revenue loss as a consequence of disasters? records on historical disaster relief, early recovery, and response funding requirements available? If so, who holds b. How are losses estimated by the government and the data? communicated to relevant authorities? 6. Concerning disaster risk assessment and modeling, c. What data are included in these records? a. Has any risk assessment and modeling been done in d. For how many years (and/or for how many disaster the country? events) are records available? b. If so, (i) by whom; (ii) when; (iii) focusing on what 3. Concerning public disclosure of disaster-related types of natural hazard; (iv) focusing on what unit of fiscal exposure, analysis; and (v) at what level of resolution? a. Does the government assess and disclose its fiscal c. Which national and/or regional bodies are involved in exposure to disasters in its fiscal risk assessment, the monitoring and collation of data on natural hazards? either voluntarily or because it is required to? If so, d. Are networks in place to monitor activity (e.g., ground motion recorders, anemometers, river gauges)? 22. In countries with significant devolution of post-disaster financing responsibilities, many of the questions within this section are also relevant for the highest tier of local government. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 41 ƒƒ Does the government conduct this analysis by 3. Concerning a dedicated disaster reserve fund, sector (e.g., agriculture, transport infrastructure, a. Does the government maintain such a fund? hydraulic infrastructure, education, health, etc.)? b. If so, what is the current balance? ƒƒ Does the government consider short-term and long-term fiscal risk from disasters? c. What amounts have been allocated to this fund over the last five years, and from what source(s) ƒƒ Does the government account for potential fiscal (e.g., government budget, private donations, shocks related to disasters? development partners)? ƒƒ Has the government identified any funding gaps d. For what purposes can the fund be used (e.g., disaster in its post-disaster response (i.e., funding shortfall risk reduction, preparedness, relief, early recovery, during relief, recovery, or reconstruction phases)? reconstruction)? Ex ante disaster risk finance e. Historically, how often has this fund been exhausted at 1. Concerning the annual contingency budget, or before the end of the year? a. What portion of the government’s annual national/ f. Please provide data on actual use of the fund over the federal budget is allocated to a contingency line for past five years (or other relevant period). unforeseen events? 4. Concerning line agency funding, b. Do local governments maintain annual a. Do line ministries have a dedicated budget line contingency budgets? for disasters? c. Are allocations set in statute? b. Are related budget allocations set in statute? d. For what purposes can the resources be used (including c. For what purposes can the funds be used (e.g., risk any non-disaster-related purposes)? reduction, preparedness, relief, early recovery, e. Who manages/controls access to these budget lines? reconstruction)? f. Who can access them? 5. Concerning contingent credit, g. How long does it take for allocations from the a. Does the government use any contingent credit contingency budget to be approved and disbursed? instruments for disaster response purposes? h. Can any remaining funding be rolled over across b. If so, what are the terms, conditions, and loan periods budget years? (including trigger type/level) of these instruments? i. Please provide data on actual use of the national/ c. What are the return periods of the events that these federal contingency budget over the past five years (or funds are designed to cover? other relevant period). d. Have the contingent credit instruments ever been 2. Is there a dedicated budget line for disaster risk reduction triggered? If so, when and releasing what amount (as opposed to a contingency line for disasters, which was of funding? covered under question 1)? 42 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 6. Concerning insurance for public assets, does the government c. How long does it take to pass a supplementary budget, purchase any disaster insurance for public assets? If so, and if such budgets can be passed only according to fixed schedules, how often and when can they a. Are line ministries and local governments legally be passed? required to purchase insurance for their public assets? If so, are they required to purchase cover based on d. Are historical data on the scale of post-disaster replacement value? reallocations readily available and, if so, from where? b. Who is responsible for purchasing insurance (i.e., are e. Has the government faced any major hurdles or risks pooled and insured aggregately or insured by delays in determining and approving post-disaster individual managers)? budget reallocations? c. Is there any central entity that coordinates purchase 2. Concerning external assistance, of cover and provides line ministries and local a. How much external assistance has been provided in governments with technical assistance in this regard? response to recent disaster events (over the past 10 d. What are the current amounts of insurance cover, years or other relevant period)? Please provide data premium rates, and associated premium payments? for each relevant disaster event and donor, if available. Are data available on specific assets insured? b. What proportion of this assistance has been in the form 7. Concerning other insurance, does the government of (i) grants; (ii) loans; and (iii) technical assistance? purchase any other forms of insurance (e.g., sovereign c. Approximately what portion of the disaster response parametric cover)? If so, financing has international assistance provided? a. Who is responsible for purchasing cover? d. What portion of external assistance is targeted directly b. What are the terms (including triggers) and amount of at beneficiaries in the form of cash transfers or in- the cover? kind assistance? 8. Concerning risk transfer through capital markets, does e. What are the targeting mechanisms through which the government utilize any capital market instruments cash or in-kind assistance is delivered to beneficiaries? to transfer risk directly to the capital markets (e.g., f. What are the delivery mechanisms for cash transfers catastrophe bonds)? If so, (e.g., the payment mechanisms of existing social a. Who is responsible for purchasing cover? protection programs) or in-kind assistance? b. What are the terms (including triggers) and amount of 3. Concerning other ex post mechanisms, the cover? a. What other ex post financing sources has the government Ex post disaster risk financing used to finance disaster response (e.g., domestic and/or external borrowing, tax increases, etc.)? 1. Concerning post-disaster budget reallocation, b. How much financing has each relevant mechanism a. Are there any regulations regarding the reallocations generated each time it has been used? within existing budget lines at the discretion of the budget line holder (e.g., a line agency), including any c. Approximately what portion of the disaster response limits on amounts? financing has each relevant source provided? b. Are there any regulations (including approval processes) regarding reallocation between budget lines and government agencies, including any limits on amounts? ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 43 Post-disaster expenditure C. Questions for insurance 1. Are historical data available on government expenditure regulator on the domestic on post-disaster relief, early recovery, and reconstruction? catastrophe insurance market 2. If so, how complete is this information and how can it be accessed? 1. What is the non-life insurance penetration rate in the country in terms of premiums as a percentage of GDP? Budget execution 2. Do insurance companies report on their property 1. Is there a special (streamlined) budget execution system premium separately by personal and commercial lines? in place for use in the event of disasters (e.g., regarding Yes or no? abbreviated procurement procedures)? 3. If yes, what is the amount of property premiums 2. What are the key steps in the budget execution process? written for households? For businesses? For small and medium enterprises? 3. Have any challenges or impediments in post-disaster execution been experienced? If so, of what nature and 4. How many personal fire (FLEXA) policies exist today in with what consequences? the country? Sharing of responsibilities between federal/ 5. How many of those policies have catastrophe coverage national and local governments endorsement? Which perils are covered by endorsements? 1. What are the legal financial responsibilities of central 6. Concerning catastrophe insurance cover for property, and local governments with regard to disaster relief, early a. Can catastrophe insurance coverage be bought as a recovery, and reconstruction? stand-alone policy? 2. Does the national government have formal rules on b. Are there any requirements for catastrophe insurance sharing risk with local governments (e.g., on providing imposed by law and/or regulation (e.g., compulsory funding when disaster damage and losses exceed a catastrophe insurance for mortgages in disaster-prone predetermined percentage of the local budget)? areas, compulsory insurance requirements for all 3. If so, what is the formal process for requesting national property dwellings)? government support? 7. Concerning the regulatory and tax environment, a. Are insurance regulations set domestically, or are they set by a regional regulatory body (e.g., West African CIMA)? b. Does the country’s insurance regulation permit the sale of index-based insurance products (parametric and aggregate output)? c. Are premiums tax deductible? d. Are premiums subject to a sales tax? 44 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 8. Concerning agricultural insurance, a. Is agricultural (crop, livestock, forestry, aquaculture) insurance offered? If so, for what specific types of production, and applying what type of insurance instrument (indemnity/parametric)? b. What are the levels of uptake (number of policies written) and penetration? c. Does the government support this insurance in any way—for instance, via ƒƒ Premium subsidies? ƒƒ Co-sharing of losses? d. What is the annual budget appropriation, if any, in support of agricultural insurance? 9. Concerning disaster microinsurance, a. Is disaster microinsurance offered? b. What is the level of uptake and penetration? c. Does the government support this insurance in any way—for instance, via ƒƒ Premium subsidies? ƒƒ Co-sharing of losses? d. What is the annual budget appropriation, if any, in support of disaster microinsurance? ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 45 Annex II: Template outline for a disaster risk finance country diagnostic The outline below, which follows directly from the guidance Existing legal and institutional framework and B. note, is the recommended structure for the disaster risk financial instruments for disaster risk finance finance diagnostic. However, depending on the type and 1. Objective quantity of available information, not all subsections (e.g., 1.1–1.3) need to be reproduced separately, i.e., with their 2. The legal and institutional framework for disaster own subheadings. The important point is that the relevant risk finance information is presented and discussed. 2.1 Laws governing the budget process and current EXECUTIVE SUMMARY practice INTRODUCTION 2.2 Post-disaster budget execution A. Impacts of past disasters 2.3 Law(s) on disaster risk management and disaster risk finance 1. Basic facts of past disasters 2.4 Institutional setup for disaster risk management 1.1 Hazard profile and history of disasters and finance 1.2 Human toll of past disasters 2.5 Local government 1.3 Geographical distribution of impacts 3. Existing disaster risk finance mechanisms 2. Existing disaster risk assessments and instruments 3. Economic impact 3.1 Risk layering 3.1 Impact of past disasters by sector 3.2 Ex ante vs. ex post disaster risk finance instruments 4. Fiscal impact 3.3 Ex ante disaster risk finance instruments 4.1 Contingent liabilities 3.4 Ex post disaster risk finance instruments 4.2 Foregone revenue 5. Impact of past disasters on the poor 46 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC C. Domestic insurance market review 1. Objective 3. Insurance penetration and insurance market 2. Regulatory environment 4. Property catastrophe insurance 5. Agricultural insurance 6. Microinsurance D. Funding gap analysis 1. Objective 2. Estimating funding gaps 3. Data 4. Short-term funding gap analysis 5. Long-term funding gap analysis E. Summary and options for consideration 1. Objective 2. Summary and gaps 2. Options for consideration GLOSSARY REFERENCES ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 47 Annex III: Risk data and models in developing countries Disaster risk finance solutions are only as reliable as the risk Where financial risk models exist, they are usually not models that support them, and the latter are only as good as tailored to answer governments’ specific disaster risk finance the data used to develop them.23 Unfortunately, developing questions and information needs, such as those regarding the countries often lack adequate data to build and validate risk scale of collapsed buildings or fatalities, impact on crops and assessment tools, not least because gathering the necessary food security, or consequences for the homeless population. data sets requires large investments. Exposure data—such Almost always developed for the insurance industry, these as information on public and private assets—are the hardest models often assess only the impact on “insurable” assets, and most expensive to gather and organize. Use of satellite excluding, for example, low-income housing. Exposure data imagery is often the only way to gather up-to-date exposure may also rely heavily on official census data that often exclude data, but the cost of acquiring such images can be prohibitive infrastructure and public buildings, and that disregard for developing countries, unless companies such as Google unofficial settlements (such as shantytowns or squatter provide information already in their possession free of towns), which regularly suffer the most damage in a disaster. charge for disaster risk management and other development Even where countries can access risk modeling tools, these purposes. Insofar as data on exposure exist, they may be tools go out of date quickly; some are even born obsolete or scattered among different government ministries and other inaccurate. For example, many models rely on census data organizations that do not readily share data with each from 10 or more years ago. other, and they may not be backed up in a way that ensures their safety. 23. This annex is adapted from World Bank and GFDRR (2014, box 11, p. 53). 48 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC Annex IV: Undertaking a basic financial risk assessment using historical data A basic risk assessment can be undertaken with limited ƒƒ Declaration of a state of emergency historical information. The steps involved in undertaking a ƒƒ Call for international assistance basic risk assessment are as follows: Data on recent major disaster events can also be drawn 1. Data collection from Post-Disaster Needs Assessments; these data can 2. Data scaling complement or substitute for the data available from EM- DAT (or other data sources). The EM-DAT database is the 3. Basic statistical analysis best available database on historical losses, but its records 4. Basic risk assessment presentation are not complete (there are particular gaps in data on total damage in monetary terms), so they should be supplemented 5. Advanced statistical analysis by information in national databases and DesInventar where available. 1. Data collection The data can be collected and presented as an initial data series in several ways, depending on the user requirements Information from historical disaster events that can be used and on the amount and quality of available data. For example, in a basic risk assessment include the initial data series could be presented as follows: ƒƒ Estimates of the number of people affected ƒƒ On a per event basis or on an annual basis. For an annual basis, the historical data estimates for all individual events ƒƒ Estimates of the total damages (in monetary terms) in the same calendar year are summed. Note that when Databases such as EM-DAT contain information on historical annual data are presented, years with zero events should disaster events. EM-DAT records events across the world also be included in the initial data series. that satisfy at least one of the following criteria: ƒƒ On an individual peril basis or on an all-perils basis, e.g., flood risk only or flood, tropical cyclone, and earthquake ƒƒ Ten or more people reported killed risk. Often, where limited records are available (e.g., for only 20–50 events), more useful statistical insights can be ƒƒ One hundred or more people reported affected gained by initially looking at the data on an all-perils basis. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 49 2. Data scaling 3. Basic statistical analysis The initial data series should be scaled to allow for an increase A number of basic statistical measurements can be made using in population, inflation, or exposure over time, as follows: the scaled data series. These can be calculated in Microsoft Excel using simple inbuilt formulas as set out in table 6. When the initial data series is based on the number of people affected by a disaster, the series should be recalibrated Table 6. Basic statistical measurements relative to national population in the year of each event: Measure Description Excel function ƒƒ For each data point, a corresponding total population figure Mean The central tendency of AVERAGE() should be assigned. For example, if the data point reflects (average) the data points in question. events from the year 2000, the 2000 total population Determined by adding all the number will apply to this data point. data points in a series and then dividing the total by the ƒƒ Each data point in the data series should be adjusted by number of data points. Total Population Current year a factor of . This adjusts for the Median A simple measure of central MEDIAN() Total Population Event year tendency that is determined increase in population between the date of the event and by arranging the observations the current date. in order from smallest to largest value. If there is an odd When the initial data series is based on the total damages number of observations, the caused by a disaster, the series should be recalibrated relative median is the middle value. If there is an even number of to GDP in the year of each event: observations, the median is the average of the two middle ƒƒ For each data point, a corresponding GDP figure should values. be assigned. For example, if the data point reflects events Standard A statistic used as a measure STDEV() from the year 2000, the 2000 GDP will apply to this deviation of the dispersion or variation data point. in a data series, equal to the square root of the arithmetic ƒƒ Each data point in the data series should be adjusted by mean of the squares of the GDP deviations from the arithmetic a factor of GDPCurrent year . This adjusts for inflation and the Event year mean. increase in exposure between the date of the event and the Minimum The smallest value in a data MIN() current date. series (this could be zero). Historical total population and GDP statistics can be found at Maximum The largest value in a data MAX() series. http://databank.worldbank.org/data/home.aspx. It may be the case that total population and GDP statistics are available for a shorter time period than the available data points. For example, disaster events could be recorded from 1963 but total population or GDP statistics available only from 1990. In such a case, the user may choose to ignore the earlier data points for which no population or GDP statistics are available (the data points between 1963 and 1989 in the example above) and remove them from the data series. The resulting series of new data points is called the scaled data series. 50 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 4. Basic risk 5. Further statistical analysis assessment presentation When sufficient data are available, a more technical statistical analysis can be completed to create a simulated risk profile. The results of the basic risk assessment can be presented in This could include the fitting of a distribution to the scaled data the form of various economic measures depending on the series using a suitable software program. Once a distribution requirements of the user. Some suggestions for presentation is fitted to the scaled data series, 10,000 years of events (or of results are as follows: more) can be simulated. Advanced statistical measurements ƒƒ Costs in monetary terms. Results based on the number can then be determined based on this simulated risk profile. of people affected by a disaster can be converted into a Probable maximum loss measurements for events of different monetary amount by applying a cost per person assumption severity and frequency can be determined—e.g., losses from to the number of people affected. For example: Average events with a 1-in-10-year return period (that is, events that number of people affected annually by flood * Cost of emergency occur once every 10 years on average and so have a 10 percent response per person following a flood event = Average annual probability of occurrence in any one year). emergency response cost for flood risk. Note that an advanced statistical analysis should be ƒƒ Damage costs as a proportion of total GDP. For example: undertaken by someone with a background in statistics and Annual average damage cost / Current year GDP. experience in fitting distributions to data series. ƒƒ Population affected as a proportion of total population. For example: Annual average number of people affected / Current year population. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 51 Annex V: Contingent liabilities—Key concepts and good practice as a consequence of large-scale loan defaults in the aftermath 1. Objective of an earthquake. Governments do not bear full responsibility for disaster risk Implicit contingent liabilities relate to moral or commonly or for the implied post-disaster relief, early recovery, and recognized but nonlegal public obligations. In the context reconstruction costs should a disaster occur. Instead, both of disasters, governments are widely regarded as having a disaster risk and related responsibilities and costs are shared moral and ethical imperative to provide search and rescue across government, the private sector, and civil society. A services and humanitarian relief to affected populations. key step in assessing disaster-related fiscal risk is therefore Governments are also perceived as insurers of last resort, to determine national and subnational government roles, often leading them to provide further support, in particular responsibilities, and implied public contingent liabilities in to poorer households, to rebuild homes, replace productive the event of disasters of varying magnitude. assets, and provide compensation for injury and loss of life. Such support can be crucial in minimizing welfare losses and keeping government poverty reduction targets on track. 2. Forms of liability Other government actions are driven by economic growth Governments face both explicit and implicit disaster-related concerns, leading to measures to support the recovery of the liabilities. They can also experience a decline in revenue as private sector, such as tax holidays and breaks and the bailout a consequence of a disaster-related slowdown in economic of private institutions as a consequence of disaster-triggered activity,24 which they must manage simultaneously with the near failure. Such actions can help speed up the pace of disaster. economic recovery, in turn leading to a faster recovery of fiscal revenue. Explicit contingent liabilities include the costs of repairing and reconstructing damaged public assets and infrastructure, Governments sometimes choose to absorb significant such as roads, bridges, hospitals, schools, and power and implicit contingent liabilities in the aftermath of a disaster. water infrastructure. They also include sovereign liabilities For instance, following the 2011 Tōhoku earthquake and linked to the public backing of catastrophe insurance tsunami, which caused $360 billion in economic losses, the schemes and institutions, and, indirectly, to government government of Japan spent more on economic and social guarantees in support of other entities that could be called as support to affected communities and businesses than it did a consequence of a disaster—for instance, liabilities relating on the repair and reconstruction of public infrastructure. to a publicly guaranteed bank that finds itself under pressure 24. It is also conceivable that a disaster can lead to a short-term decline in tax administration capacity, leading to lower revenues after a disaster. 52 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC compensation. Past levels of compensation payments, 3. Measuring liabilities including payments not required by law, indicate likely levels To develop appropriate country-specific disaster risk of total compensation. For other purposes—e.g., support for financing strategies, it is important to determine the the recovery of livelihoods and businesses—examination nature and scale of disaster-related explicit and implicit of past behavior provides an indication of likely levels liabilities and of the fiscal consequences of disaster- related of compensation. macroeconomic instability. Further implicit contingent liabilities can be monitored via The measurement of explicit contingent liabilities stemming regular assessments of potentially vulnerable institutions. from a disaster is straightforward in theory. National and These liabilities should be addressed first and foremost local governments maintain updated lists of liabilities. through efforts to reduce the fiscal risk posed to government These lists, including information on guarantee ceilings, can (for instance, via industry regulations requiring proper be reviewed to determine liabilities that could be called in risk pricing of insurance premiums, adequate reserves, and either directly or indirectly as a consequence of a disaster. insurance of mortgaged properties), rather than simply by Contingent liabilities relating to the loss of public assets making appropriate provision for potential bailouts. and infrastructure can also be calculated relatively easily in countries where a detailed inventory and comprehensive disaster risk assessment of public property and infrastructure 4. Enhancing measurement are available, although the calculation is slightly more complicated where the private sector is heavily involved in of liabilities basic services provision. Disaster risk assessments should Governments can enhance the measurement of contingent take into account both the location of public assets relative liabilities by passing legislation and regulations regarding to potential natural hazard events and the vulnerability of post-disaster compensation to both households and public assets. Implied potential levels of damage should be businesses (and related conditionalities). This approach has valued in terms of repair and replacement costs. These cost the additional indirect benefit of clearly signaling the extent estimates should also reflect build-back-better principles and and limitations of public support in the event of a disaster, the precise nature of specific actions required to strengthen thereby encouraging households and businesses to recognize, resilience to future natural hazards (e.g., relocation of public and take action to manage, their share of responsibilities. assets), since these will also affect the cost of reconstruction. Critical infrastructure that might undergo both temporary repairs to restore service and longer-term reconstruction 5. Disclosing liabilities should also be identified; this two-step process will have additional cost implications. Governments should be encouraged to disclose their explicit disaster-related contingent liabilities. Disasters will inevitably A significant share of contingent liabilities can be directly occur; disaster risk can be estimated with increasing determined in countries where governments have defined accuracy; and governments in more hazard-prone countries their roles and responsibilities in relation to disaster-affected will periodically incur substantial related expenditure. The communities and businesses. Many governments have disclosure of explicit contingent liabilities thus constitutes established statutory levels of compensation under different a key step in strengthening fiscal resilience. It highlights disaster scenarios, in particular relating to death or injury the extent of disaster risk and the associated potential fiscal and loss of homes. However, political pressure often pushes burden for which the government is explicitly liable, in turn governments to go beyond the statutory requirements in strengthening accountability, encouraging governments to providing compensation. Expectations of compensation establish disaster risk financing strategies, and improving the payments can therefore constitute an important implicit quality and robustness of those strategies. Disclosure also liability. An analysis of statutory requirements provides forces an articulation of the forms and levels of contingent a basis to estimate the explicit contingent liability from liabilities accepted by government, encouraging other ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 53 stakeholders to acknowledge and take responsibility for their own liabilities. A government may be reluctant to disclose its explicit contingent liabilities because of concerns regarding long- term confidence in the country’s economy. However, if the government releases information on its disaster risk financing strategy along with information on explicit contingent liabilities, it can assuage concerns and limit adverse market reactions. Governments may be more circumspect about disclosing implicit disaster-related contingent liabilities, as such disclosure may create or exacerbate moral hazard. While it is important for fiscal planners to take the likely absorption of implicit contingent liabilities into account when assessing disaster-associated fiscal risk, they may prefer not to disclose this information. 54 ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC Annex VI: Example of information on contingent liabilities due to disaster found in public documents The text below (box 2) is taken from the IMF’s July 2016 Article IV consultation report on St. Vincent and the Grenadines. The report provides no disaggregated information on past fiscal costs due to explicit and implicit disaster-related contingent liabilities. However, it mentions that about three-quarters of annual damages from disasters have been the responsibility of the government in the past. It also mentions that average annual damages from disasters triggered by natural hazards are equivalent to 1.2 percent of GDP. From these two facts, we can infer that, on average, the government has faced annual fiscal costs associated with disaster reconstruction of around 0.9 percent of GDP, with additional humanitarian response costs roughly equivalent to an additional 0.2 to 0.3 percent of GDP. ASSESSING FINANCIAL PROTECTION AGAINST DISASTERS: A GUIDANCE NOTE ON CONDUCTING A DISASTER RISK FINANCE DIAGNOSTIC 55 Box 2: Example of publicly available information on disaster-related contingent liabilities: St. Vincent and the ST. Grenadines VINCENT AND THE GRENADINES Medium-Term Fiscal Framework 14. More ambitious fiscal consolidation is needed over the medium-term to meet the authorities’ public debt target of 60 percent of GDP by 2030. Public Debt: Active and Baseline Policy Scenarios, with and  The baseline scenario is projected to without Natural Disasters1 (percent of GDP) result in the public debt overshooting the 100 84.2 target. The medium-term primary 90 85.9 79.2 surplus is projected to reach 1.6 percent 80 78.8 71.3 70 of GDP by 2019 and result in a debt to 60 73.7 59.7 GDP ratio of 71.3 percent of GDP in 50 45.6 40 2030. This reflects the projected full year 30 Debt (Baseline, no natural disasters) Debt (Baseline, with natural disasters) impact of current revenue measures, and 20 Debt (Active, no natural disasters) assumes continued expenditure restraint 10 Debt (Active, with natural distasters) 0 and no more natural disasters. 2014 2016 2018 2020 2022 2024 2026 2028 2030 Source: Fund staff estimates. 1 Assumes that natural disasters occur at 5-year intervals, lowering GDP  Furthermore, if natural disasters growth in the first year and causing a 3-year primary balance deterioration associated to fiscal costs. materialize at their historical magnitude and frequency, staff projects that public debt would be even more elevated, at 85.9 percent of GDP in 2030. This is based on historical data indicating that average annual damages from natural disasters in St. Vincent and the Grenadines are 1.2 percent of GDP, of which about ¾ have typically been the responsibility of the government. Frequency of Natural Disasters since 1990 Average Annual Losses from Disasters, 1990-2013 (pct. deviation from median number of disasters/person; top 25 independent countries) (pct of GDP; top 25 fund members) Tuvalu Samoa Dominica St. Kitts and Nevis Vanuatu Grenada American Samoa Haiti Antigua and Barbuda Antigua and Barbuda Tonga Guyana Micronesia Fed States Maldives St Vincent and the Grenadines Tonga Marshall Is Belize St Lucia Tajikistan St Kitts and Nevis St. Lucia Guam St. Vincent and the Grenadines Seychelles Bahamas, The Grenada Lao PDR Georgia Bahamas Cambodia Others Samoa Chile Belize Fiji Fiji Dominica Solomon Is Liberia Other Caribbean Barbados Others Oman Iceland Pakistan Kiribati Jamaica Comoros Other Caribbean Dominican Republic St. Vincent and the Djibouti Mauritius Grenadines Cape Verde Is Madagascar 0 1,000 2,000 3,000 4,000 5,000 6,000 0 5 10 15 20 Sources: EMDAT; World Development Indicators; and Fund staff estimates. Sources: EMDAT; USAID; WDI; and Fund Staff Estimates. 15. Source: this backdrop, the government would need to target a higher primary surplus Against IMF 2016b, 12. over the medium-term to meet its public debt target, especially given recurrent natural disasters. A primary surplus reaching 3.3 percent of GDP by 2019 and sustained over the long-term is deemed adequate to help the authorities achieve their public debt target in 2030, even if natural disasters materialize. This would require well-specified tax and expenditure policy measures with annual yield of about 1.7 percent of GDP beyond the projected yield from the government’s current About the World Bank The World Bank comprises the International Bank for Reconstruction and Development (IBRD) and the International Development Association (IDA). It provides financing, knowledge, and convening services that help its clients address many of their most important development challenges. With 189 member countries, staff from more than 170 countries, and offices in over 130 locations, the World Bank is a unique global partnership with the mission of ending extreme poverty and promoting shared prosperity. About the Asian Development Bank ADB’s vision is an Asia and Pacific region free of poverty. Its mission is to help its developing member countries reduce poverty and improve the quality of life of their people. Despite the region’s many successes, it remains home to a large share of the world’s poor. ADB is committed to reducing poverty through inclusive economic growth, environmentally sustainable growth, and regional integration. Based in Manila, ADB is owned by 67 members, including 48 from the region. Its main instruments for helping its developing member countries are policy dialogue, loans, equity investments, guarantees, grants, and technical assistance. World Bank contact details: The World Bank 1818 H Street NW Washington, DC 20433 www.worldbank.org ADB contact details: ASIAN DEVELOPMENT BANK 6 ADB Avenue, Mandaluyong City 1550 Mtero Manila, Philippines www.adb.org